1246966

research-article2024
PSIXXX10.1177/15291006241246966Flusberg et al.Psychological Science in the Public Interest

ASSOCIATION FOR
PSYCHOLOGICAL SCIENCE
Psychological Science in the

The Psychology of Framing: How Everyday Public Interest

2024, Vol. 25(3) 105­–161
© The Author(s) 2024
Language Shapes the Way We Think, Article reuse guidelines:
sagepub.com/journals-permissions

Feel, and Act DOI: 10.1177/15291006241246966

https://doi.org/10.1177/15291006241246966
www.psychologicalscience.org/PSPI

Stephen J. Flusberg1, Kevin J. Holmes2, Paul H. Thibodeau3,

Robin L. Nabi4, and Teenie Matlock5
1
Department of Cognitive Science, Vassar College; 2Department of Psychology, Reed College;
3
Independent Scholar; 4Department of Communication, University of California, Santa Barbara;
and 5Department of Cognitive Science, University of California, Merced

Abstract
When we use language to communicate, we must choose what to say, what not to say, and how to say it. That is, we
must decide how to frame the message. These linguistic choices matter: Framing a discussion one way or another can
influence how people think, feel, and act in many important domains, including politics, health, business, journalism,
law, and even conversations with loved ones. The ubiquity of framing effects raises several important questions
relevant to the public interest: What makes certain messages so potent and others so ineffectual? Do framing effects
pose a threat to our autonomy, or are they a rational response to variation in linguistic content? Can we learn to use
language more effectively to promote policy reforms or other causes we believe in, or is this an overly idealistic goal?
In this article, we address these questions by providing an integrative review of the psychology of framing. We begin
with a brief history of the concept of framing and a survey of common framing effects. We then outline the cognitive,
social-pragmatic, and emotional mechanisms underlying such effects. This discussion centers on the view that framing
is a natural—and unavoidable—feature of human communication. From this perspective, framing effects reflect a
sensible response to messages that communicate different information. In the second half of the article, we provide a
taxonomy of linguistic framing techniques, describing various ways that the structure or content of a message can be
altered to shape people’s mental models of what is being described. Some framing manipulations are subtle, involving
a slight shift in grammar or wording. Others are more overt, involving wholesale changes to a message. Finally, we
consider factors that moderate the impact of framing, gaps in the current empirical literature, and opportunities for
future research. We conclude by offering general recommendations for effective framing and reflecting on the place
of framing in society. Linguistic framing is powerful, but its effects are not inevitable—we can always reframe an issue
to ourselves or other people.

Keywords
framing, framing effects, communication, language, reasoning, attitudes, affect, pragmatics, metaphor, decision-
making, risky choice

Language is used for doing things. (Clark, 1996, government of Oceania created a simplified form of
p. 3) English—“Newspeak”—to keep the population in
check. By whittling down the meaning of words such
We are captivated by stories about the arcane power as “freedom,” this language suppresses the inclination
of language. Saying just the right words at just the right to entertain subversive thoughts, reducing a person’s
moment can help you find hidden treasures, mesmerize will to resist. Other fictional worlds depict a vast array
your enemies, defeat true evil, and unlock the mysteries
of time. One popular trope involves using language to Corresponding Author:
manipulate and control people. Consider George Stephen J. Flusberg, Department of Cognitive Science, Vassar College
Orwell’s dystopian classic 1984. The totalitarian Email: [email protected]
106 Flusberg et al.

of coercive linguistic techniques, from magical spells researchers have sought answers to these questions,
(“Imperio! yelled Voldemort”) to hypnotic brainwashing which bear on a broad set of issues in the public inter-
(“You are getting very sleepy . . .”). A recent example est—the impact of partisan media and political rhetoric,
comes from the 2013 sci-fi novel Lexicon by Max Barry. the efficacy of environmental and health communica-
The story follows a shadowy organization whose mem- tion, and support for systemic policy reforms, among
bers, known as “Poets,” are masters of verbal persua- many others. Our goal in this article is to synthesize
sion. New recruits, screened for their intuitive charisma, this literature, much of which is rooted in the core
learn a secret, ancient vocabulary that lets them control construct of framing.
other people’s minds and manipulate their memories, When we use language to communicate with others,
thoughts, and behavior. we must choose what to say, what not to say, and how
It is tempting to dismiss these stories as mere fantasy, we want to say it. That is, we must decide how to frame
but speculative fiction often mirrors real cultural anxiet- the message. Framing is an intrinsic feature of linguistic
ies. The focus on linguistic control in popular media communication because every message must take one
may reflect a deeper concern—and fascination—with particular form and not another, and the same event or
the force of language in our daily lives. We routinely issue can always be described in many different ways.
use words to try to influence what other people are We can emphasize the positives or negatives (or both),
thinking, feeling, and doing. You may attempt to con- the risks or sure bets (or both); we can use loaded or
vince a hesitant friend to attend a party (“You only live neutral language, concrete metaphors or abstract
once!”) or persuade your spouse to forgive you for descriptions, active or passive voice; and we can com-
spending your savings on an unusable plot of land (“It’s municate our own perspective and preferences, obscure
an investment!”). Cable news pundits work to drum up our group identity, or tailor our message for each audi-
support for their preferred politicians (“She’s a true ence. These linguistic choices matter: Framing a discus-
patriot!”) and vilify their ideological opponents (“He’s sion one way or another can influence how people
a fascist!”). Think tanks, advertising companies, media think, feel, and act in many important domains, includ-
corporations, and political campaigns spend billions ing politics, health, business, journalism, law, and even
each year to generate the perfect slogan, catchphrase, everyday conversations with loved ones. As a result,
or headline to drive votes, clicks, and sales. And then research on framing spans a wide range of fields,
there are the influencers, self-help gurus, and grifters including sociology and anthropology (Bateson,
whose livelihoods depend on robust verbal influence. 1972/1987; Goffman, 1974), economics and political
Amidst this endless barrage of linguistic messaging, science (Chong & Druckman, 2007; de Bruijn, 2019;
it makes sense that we would be drawn to fantastical Farrow et al., 2018), linguistics (Lakoff, 2010; Matlock,
stories about the manipulative powers of language. The 2012), media studies and communication (Iyengar,
stakes are high, with financial and political fortunes on 1991; Lecheler & de Vreese 2019; Nabi, 2003), and psy-
the line. And there is something enticing about the chology and neuroscience (De Martino et al., 2006;
prospect of honing our own powers of verbal persua- Flusberg et al., 2022; Tversky & Kahneman, 1981). The
sion. Being more persuasive will surely solve our prob- multidisciplinary study of framing effects examines
lems, right? If we learn to say the right things at the when, how, and why different forms of language—
right times, won’t we achieve greater success in work including differences in word choice, grammar, con-
and love? There is clearly a market for this belief. creteness, and emphasis—affect how people respond
Bookshelves are lined with promises to unlock the to a message.
secrets of linguistic influence, including titles such as Sometimes framing involves glaring contrasts in mes-
Words That Work: It’s Not What You Say, It’s What People sage content. This is apparent to anyone attuned to the
Hear by pollster Frank Luntz (2007) and Magic Words: ongoing culture war in the United States. The escalating
What to Say to Get Your Way by marketing professor conflict between “blue” and “red” America is unfolding
Jonah Berger (2023). as a series of linguistic battles in all forms of public
But are we really so easy to manipulate, like the discourse. Were the events of January 6, 2021, a violent
denizens of Orwell’s Oceania, unconsciously pushed insurrection or legitimate political discourse? Is legal
this way and that by the words and phrases we encoun- access to abortion about the right to bodily autonomy
ter in everyday life? Or are we somehow immune to and reproductive health care or about the state-spon-
the effects of language? Why are some messages so sored murder of unborn children? Do companies have
captivating or insidious and others so mundane or an obligation to serve everyone equally, or do they main-
benign? Can we learn to use language more effectively tain the freedom to reject same-sex couples and other
to promote causes we believe in and achieve outcomes customers on the basis of religious liberty? The two
we seek, or is this overly idealistic? For decades, sides have framed each issue differently to promote a
Psychological Science in the Public Interest 25(3) 107

particular set of values and way of thinking and to consider what all of this means for navigating our lan-
shape public sentiment and voter turnout. Such heavy- guage-rich social environments. Throughout the article,
handed rhetoric may seem cynical and manipulative, we emphasize the relevance of framing to the public
but framing can also be useful and productive, helping interest, highlighting issues such as environmental and
us approach an issue from different perspectives and health communication, political rhetoric, and policy-
improve our reasoning (Bermúdez, 2020). making. If we have framed this article effectively, your
At other times, framing is more inconspicuous. For curiosity should be piqued and you’ll want to read on.
example, you may not have noticed that we used the Go right ahead!
language of war to describe both our modern predica-
ment (“endless barrage of linguistic messaging . . .”) and
What’s in a Frame? A Brief History
the divide between American liberals and conservatives
(“The escalating conflict . . . a series of linguistic bat- For millennia, humans have been using language to
tles”). Among other psychological effects, war meta- influence others’ attitudes, decisions, and actions.
phors have been shown to heighten the sense of Persuasive communication, or rhetoric, has been a topic
urgency surrounding an issue (Flusberg et al., 2017, of interest since classical antiquity. Aristotle famously
2018). Would it make any meaningful difference if we argued that effective rhetoric rests on three pillars:
used a different metaphor to describe political polariza- logos (appeals to logic and reason), pathos (appeals to
tion? For instance, we could say that Republicans and emotions), and ethos (appeals to the authority or status
Democrats are “playing political theater” and “perform- of the speaker; Rapp, 2022). Semantic drift has bleached
ing dramatic monologues” in the media, but would this the term “rhetoric” of some of its original cachet. Today,
matter? We return to this issue later in the “Figurative it often refers to divisive commentary that reflects
Framing” section, in which we take a closer look at the “party-line” thinking, at least in the context of political
nature, impact, and limits of metaphor framing. discourse. Even so, Aristotle’s insights resonate with
But first we need to lay some foundations. We begin contemporary theories of persuasion and framing, as
with a brief history of research on frames and framing. we discuss throughout this article.
Interest in these ideas developed independently in dif- The concept of “framing” developed over the course
ferent academic disciplines during the 20th century of the 20th century, spurred by interest across various
(K. Sullivan, 2023; Tannen, 1993). Consequently, terms disciplines, including psychology, computer science,
such as “frame” carry several different, albeit interre- anthropology, sociology, linguistics, communication,
lated, connotations. Unpacking this history is helpful media studies, artificial intelligence (AI), and political
for situating contemporary research on framing. We science (for historical reviews, see K. Sullivan, 2023;
then discuss research on specific framing effects, exam- Tannen, 1993). This literature is vast, messy, and com-
ining how language can influence attention and percep- plicated (for pointed critiques, see Cacciatore et al.,
tion, beliefs, attitudes, memory, judgment and 2016; Entman, 1993; van Dijk, 2023). One problem is
decision-making, and real-world behavior. From there, that there is no single, good definition of frames or
we explore where these effects come from, beginning framing. Sometimes “frame” has referred to a feature of
with a primer on the psychology of language compre- memory or cognition; other times it has referred to a
hension: How do we make sense of what we hear and feature of language or communication (Druckman,
read? This discussion illuminates the cognitive, social- 2001; K. Sullivan, 2023). Tracing the evolution of these
pragmatic, and emotional mechanisms underlying many ideas is illuminating. The linguist K. Sullivan (2023)
linguistic framing effects and shows that framing is a recommends differentiating among frames that operate
natural—and unavoidable—feature of human commu- at the level of thought (cognitive frames), linguistic
nication. From this perspective, framing effects reflect meaning (semantic frames), and social communication
a sensible response to messages that communicate dif- (communicative frames). Following her lead, in this
ferent information. section we discuss the nature and origins of each con-
With these pieces in place, we then provide a tax- ception of “frame” across different academic disciplines
onomy of linguistic framing techniques, which we orga- (see Table 1).
nize around the elements of language manipulated in
messages. These techniques include various labels,
Cognitive frames
metaphors, and grammatical forms. We go on to
describe some factors that moderate the impact of fram- Psychologists and early AI researchers originally used
ing, along with gaps in our current understanding that the word “frame” to refer to a type of memory or knowl-
would benefit from additional research. We conclude edge structure, closely related to concepts such as
with advice on how to frame messages effectively and “scripts,” “schemas,” “prototypes,” and “cognitive
108 Flusberg et al.

Table 1. Different Conceptions of “Frame” in the Academic Literature

Type of frame Definition Example

Cognitive Relatively abstract, structured knowledge about Your “university” frame includes elements such as a
some aspect of reality; includes information about campus, academic buildings, professors, students,
typical objects, participants with specific roles, degrees, majors, dorms, a library, and administrators.
causal relationships, and sequences of events; If you are in the United States, it may also include
used to interpret experience, guide behavior, and fraternities, sororities, and football. This frame is
anticipate how situations will unfold; related to complex, including other features such as when
other psychological constructs such as “schemas,” students generally apply, how to behave in the
“scripts,” “prototypes,” and “cognitive models”; also classroom, and much more.
called “frames in thought”
Semantic Basic cognitive frames that are evoked by language; The “competition” frame includes elements—and is
clusters of related words will evoke the same evoked by words—such as competitors, winners,
semantic frame, which facilitates comprehension; losers, venue, rank, place, and score. It can apply
according to “frame semantics,” we understand to descriptions of many events, such as basketball
the meaning of words by virtue of the conceptual games, presidential elections, and spelling bees.
knowledge—the semantic frames—they evoke; also
called “linguistic frames”
Communicative The structure or content of a linguistic message that The frames “limited English proficient,” “English
communicates a particular viewpoint or cognitive/ learner,” and “emergent bilingual” communicate
semantic frame to the receiver; also called “message different ideas about the language skills and potential
frames” or “frames in communication” of U.S. children whose first language is not English.

models” (Bartlett, 1932; Bower et al., 1979; Gentner, blowing out candles, eating cake, unwrapping presents,
1983; Kintsch & van Dijk, 1978; Lakoff, 1987; Mandler, and so on. It does not include details such as everyone
2014; Minsky, 1974; Rosch, 1983; Rumelhart, 1975; dresses in a fur catsuit, even if that’s how your sister’s
Schank & Abelson, 1977). Although these terms have kid celebrated last year (don’t ask).
subtly different connotations, they all refer to what the Of course, not all birthdays contain every element
linguist Deborah Tannen (1993) calls “structures of of your birthday frame. Maybe your sister is afraid of
expectation”: general knowledge about the world that balloons, so she does not purchase them for your
we use to interpret experiences, guide our behavior, niece’s party. Even so, your birthday frame will still be
and anticipate how events will unfold. A situation that invoked by other party features, which will allow you
is inscrutable at first—such as seeing a dozen children to easily identify the event and what people will do
waltz into your neighbor’s house holding strange pack- there. Things would be different if no one ever used
ages—becomes meaningful once you apply the proper balloons at kids’ birthday parties. The specific content
frame (“Oh, it’s little Chloe’s birthday party”). of your frames is based on general patterns your brain
A related conception of frames was developed inde- extracts from your everyday experiences and culture
pendently by the sociologist Erving Goffman. In his (DiMaggio, 1997; Rogers & McClelland, 2004; Strauss &
influential book Frame Analysis: An Essay on the Quinn, 1997). If you grew up in a community in which
Organization of Experience (Goffman, 1974), Goffman birthday parties featured furry outfits, spinach tarts, and
described frames as the organizational structures in the 90 min of silence, you would expect to see such fea-
mind that enable us to make sense of the natural and tures at any future party. And so would others from
social world. He explains: “When the individual in our your community. It is important that our expectations
Western society recognizes a particular event, he tends, for how the world works—our frames—align with those
whatever else he does, to imply in this response (and of other community members. This “common ground”
in effect employ) one or more frameworks or schemata forms the foundation for successful language compre-
of interpretation” (p. 21). hension and social coordination (Clark, 1996), leading
Importantly, the composition of a frame is relatively some scholars to view cognitive frames as key to under-
abstract. It includes information about typical objects, standing culture (DiMaggio, 1997; Swidler, 1986).
participants with specific roles, causal relationships Note, however, that this sense of the word “frame”
between them, and sequences of events—but not idio- concerns what we know and how we think, not neces-
syncratic details. For example, your “child’s birthday sarily what we say. We can tap into our knowledge of
party” frame includes prototypical elements such as a birthday parties without using or hearing words such as
group of kids gathering in a single place, brightly colored “presents” and “cake.” That is why this type of frame is
balloons, giving wrapped presents to the birthday kid, called a “frame in thought” (Druckman, 2001) or a
Psychological Science in the Public Interest 25(3) 109

“cognitive frame” (Fillmore, 2008; K. Sullivan, 2023). language. This should not be confused with “linguistic
Cognitive frames are not stored in isolated files in your framing,” which is our generic term for the use of a
brain and activated on command like an mp3 on your particular aspect of language to describe an issue or
laptop. Rather, our world knowledge is organized into situation (see the discussion of communicative frames
a vast web of interconnected conceptual networks at below). Semantic frames represent “story fragments” in
varying levels of abstraction that function together as the mind (Ruppenhofer et al., 2016)—not complex
we navigate our environments. Fully understanding a scripts, narratives, or schemas—because they often
birthday party also requires knowing how people keep comprise narrow slices of more elaborate cognitive
track of someone’s age, basic addition, parent–child frames (K. Sullivan, 2023). For example, you can under-
relationships, behavioral norms for social gatherings, stand the word “bought” in the sentence “José bought
and much more. As Goffman (1974) observed: “During apples at the store” because it evokes the frame of
any one moment of activity, an individual is likely to commercial transactions, with a buyer, seller, and
apply several frameworks” (p. 25). Thus, cognitive exchange of goods for money. Going to the grocery
frames form an integrated system of conceptual knowl- store, on the other hand, activates a much more elabo-
edge, derived from experience, that shape how we pro- rate set of cognitive frames that structure your expecta-
cess information, make sense of our surroundings, and tions and behavior for the entire sequence of events,
structure our behavior. This is why many policy inter- none of which necessarily involves language. All your
ventions aimed at improving educational and economic tacit knowledge of commercial transactions, nutrition,
outcomes seek to reshape people’s cognitive frames. credit cards, ripe fruit, budgeting, waiting in line, and
For example, much work has gone into helping students much more is needed to successfully navigate this
adopt a “growth mindset” in school—a cognitive frame excursion. By contrast, a semantic frame is more lim-
that stipulates intelligence and academic skills can ited, consisting only of the knowledge structures
improve through practice and hard work (e.g., Yeager evoked by hearing or reading a specific word in
et al., 2019; but see Macnamara & Burgoyne, 2023). context.
Fillmore and colleagues have cataloged hundreds of
semantic frames evoked by thousands of English words
Semantic frames and the relationships among them, revealing a rich
In linguistics, interest in frames arose from a desire to tapestry of interconnected meanings (see https://
explain how people understand language in context. framenet.icsi.berkeley.edu; see also Ruppenhofer et al.,
This is evident within the “frame semantics” approach 2016). Frame semantics is not the only theory in town
to word meaning pioneered by Charles Fillmore (e.g., when it comes to word meaning, but it is consistent
Fillmore, 1982, 2008; Fillmore & Baker, 2009). Frame with various psychological theories of language com-
semantics draws heavily on the concept of cognitive prehension, as discussed in the “Mechanics of
frames described in the previous section (Fillmore, Communication and Linguistic Framing” section.
1982; K. Sullivan, 2023). In so doing, it rejects the view Notably, this view of language seems to entail that all
that understanding a word involves retrieving a unique linguistic communication involves framing because
definition or mental image. Rather, according to frame every message will evoke certain semantic frames and
semantics, “people understand the meaning of words not others. One exception might be a conventionalized
largely by virtue of the frames which they evoke” greeting, such as saying “hello” to a coworker. Yet
(Ruppenhofer et al., 2016, p. 7; emphasis in original). “hello” would elicit a very different response than “salu-
That is, the “meaning” of any given word is grounded in tations, your highness.” The latter greeting evokes the
our world knowledge and coupled with clusters of other “monarchy” frame and may be interpreted as sarcasm
words that evoke the same frame. For example, words (unless you happen to work for the royal family). If
such as “bride,” “groom,” “wedding,” “divorce,” “wife,” you are interested in framing, it is important to recog-
“husband,” and “honeymoon” all evoke the “marriage” nize that different linguistic forms, whether wildly dis-
frame; it is difficult to understand any one of these words similar or almost interchangeable, may invite different
without also knowing the others because they all specify reactions to a message (for discussions of how frame
roles or events within a single knowledge structure. semantics informs theories of framing in contemporary
Reading that “Mary honeymooned in Paris after she wed political discourse, see Lakoff, 1996, 2008, 2014).
Bill” evokes the marriage frame, which helps you under-
stand the relationship between the two individuals and
Communicative frames
thus the meaning of the sentence.
Frames that are evoked by words in this way are We described earlier how liberal and conservative pun-
called “linguistic” or “semantic” frames because they dits talk about hot-button issues such as abortion in the
are directly tied to how we derive meaning from United States. We suggested this was no accident:
110 Flusberg et al.

Politicians, journalists, and activists all use language—as are critical for effective communication. According to
well as images, sounds, and presentation style—with Bateson, frames are “metacommunicative” because they
the specific intention of shaping social discourse and contain information about how the interlocutor should
influencing public opinion. This is what many people make use of the communicative signal: “Any message,
view as a paradigmatic instance of framing. As political which either explicitly or implicitly defines a frame,
communication scholar Robert Entman (1993) explained: ipso facto gives the receiver instructions or aids in his
“Framing essentially involves selection and salience. To attempt to understand the messages included within
frame is to select some aspects of a perceived reality and the frame” (p. 145). We offer this useful way of thinking
make them more salient in a communicating text, in about communicative frames throughout the article.
such a way as to promote a particular problem defini- Interestingly, Goffman (1974) credited Bateson for
tion, causal interpretation, moral evaluation, and/or inspiring his own thinking about this subject. Although
treatment recommendation” (p. 52). Note the emphasis Goffman focused on cognitive rather than communica-
here is on how (and perhaps why) a frame is commu- tive frames (Druckman, 2001), his frame analysis book
nicated in a message, not how it is represented in peo- highlighted different ways that people send signals to
ple’s minds. That is why this is referred to as a “message transform how others organize their understanding of
frame,” a “frame in communication” (Druckman, 2001), reality. For example, advertisers can promote the view
or a “communicative frame” (K. Sullivan, 2023). that cars represent a life of pleasurable leisure by pair-
To be sure, the ultimate goal for many communica- ing them with certain images or labels. This social,
tors is to instantiate a particular cognitive or semantic communicative conception of a frame closely aligns
frame in the target audience to influence attitudes (e.g., with contemporary work on framing in politics, media,
the “murder” frame for thinking about abortion). But law, and communication. The FrameWorks Institute, for
this goal may or may not be achieved. The key distinc- instance, is a nonprofit, progressive think tank that
tion, then, is that a communicative frame describes researches the efficacy of different communicative
message structure or content that cues “a particular frames for explaining complex sociopolitical issues and
problem definition” rather than the internal thought promoting systemic reforms.
processes of the person exposed to the message.
Sometimes speakers intentionally craft a message to
provide the audience with a particular “causal interpre-
Framing effects
tation” or model of a target issue, as is often the case For decades, researchers have focused on measuring
in political, business, and legal communication. This and explaining the impact of different (communicative)
deliberate use of language in service of persuasion is frames on cognition and behavior. This is the study of
the archetype of communicative framing and of particu- framing effects—a topic that traverses a range of theo-
lar interest to many social scientists (Druckman, 2001; retical and empirical methods across disciplines.
Entman, 1993; Schwartzstein & Sunderam, 2021). In Research on framing effects is primarily concerned with
other cases, however, a message may be generated linguistic features of a message that influence how
spontaneously without careful consideration of the people respond to that message (although framing can
information it communicates to the audience. This can extend to other modalities as well, such as imagery;
result in unanticipated framing effects, with unintended e.g., Powell et al., 2015). In this way, research on fram-
consequences (see, e.g., our discussion of subject-com- ing is both broader and narrower than research on
plement framing in the “Beliefs” and “Pragmatic rhetoric and persuasion. It is broader because it
Inference” sections). We consider both types of mes- addresses how any language encountered impacts atti-
sages to be examples of communicative frames. tudes and behavior, not just language used in the con-
Anthropologist and cyberneticist Gregory Bateson text of explicitly rhetorical communication. But it is
was among the first to consider the communicative narrower because the focus is on language specifically,
function of frames, which he discussed in an influential not other elements of persuasion such as a speaker’s
article on play and fantasy (Bateson, 1972/1987). style or charisma (although language is known to affect
Bateson was struck by the way young monkeys engage how speakers are perceived; see “Social-Pragmatic
in “play fighting” without devolving into real conflict. Mechanisms” section). In the next section, we survey
Somehow, he surmised, signals must be exchanged to various ways that language can influence an audience.
ensure the “play” frame governs their interaction—and For this article, we take a big-tent approach to linguistic
something similar must be going on with human play framing effects. We include any research in which a
and other complex social rituals. Communication is change in linguistic structure or content has a measur-
critical for social coordination, and, in his view, frames able impact on a receiver’s thoughts, feelings, or actions.
Psychological Science in the Public Interest 25(3) 111

What Does Language Do? Measuring in visual “noise” (Lupyan & Ward, 2013; see also Holmes
the Effects of Linguistic Framing & Wolff, 2013; Lupyan & Spivey, 2010a). In fact, linguis-
tic framing also figures into how we resolve sensory
One way to approach the study of language is to exam- ambiguity, which can shape our conscious experiences.
ine how it differs from other cognitive abilities. For A few years ago, for example, a viral video showed
example, researchers often aim to identify unique com- someone activating a toy that emitted a garbled English
putational processes and brain networks that support phrase. What you hear changes depending on whether
language comprehension and use (e.g., M. D. Hauser you have just read the word “brainstorm” or “green
et al., 2002; Malik-Moraleda et al., 2022). Another needle” (to experience this delightful auditory illusion
approach is to examine how deeply connected language for yourself, see https://tinyurl.com/2dvp8rn8).
is to other elements of our cognitive ecology. This is Most effects of language on perception are not
where research on linguistic framing lives. Below, we described as examples of framing because their locus
discuss different ways researchers have measured the is in immediate sensory processing rather than “higher
impacts of language, revealing the wide range of influ- level” conceptualization or reasoning. In line with
ences language can have on cognition and behavior. Lupyan et al. (2020) and other cognitive scientists, how-
ever, we see these processes as tightly connected. If
language drives our attention and guides what we per-
Attention and perception ceive, this can be expected to have downstream con-
Have you ever been engrossed in a conversation at a sequences for other aspects of cognition. And, as we
crowded event, discussing the ins and outs of Colombian discuss in the following subsections, it does.
architecture or Japanese whiskey (or whatever floats
your boat)? With your full attention on your interlocu-
Beliefs
tor, other sounds in the space fade into a background
murmur. But then, to your surprise, you suddenly hear Language is often used to influence beliefs about the
someone across the room say your name, even though state of the world: When I tell you it’s raining outside,
you hadn’t even realized they were there. This “cocktail you’ll know to bring an umbrella. Language can also
party effect,” a classic phenomenon in the attention shape beliefs about how the world works, and often in
literature, shows how certain linguistic signals—those subtle ways. For example, we indicated earlier that war
especially pertinent to us—automatically capture our metaphors—which are prevalent in everyday discourse
attention even when our focus is elsewhere (Moray, (Flusberg et al., 2018)—can increase our sense of
1959; Shapiro et al., 1997; Wood & Cowan, 1995). This urgency about social and political issues (Flusberg
is one reason why people are more likely to click on et al., 2017). War metaphors can also shape our beliefs.
marketing emails with their name in the subject line Several studies have contrasted the use of “battle” and
(Sahni et al., 2018) and to give careful thought to prod- “journey” metaphors in discourse about cancer, both of
ucts or resumes that share a name similar to their own which are common in descriptions of the disease
(Howard & Kerin, 2011). (D. J. Hauser & Schwarz, 2020; Hendricks et al., 2018;
Language can also help you find things in your envi- Magaña & Matlock, 2018; Semino et al., 2017, 2018). In
ronment. Hearing a noun such as “chair” or a preposi- one study, participants read a story about a man recently
tion such as “above” will automatically direct your diagnosed with cancer, framed using either battle or
attention toward the corresponding object or region of journey metaphors (D. J. Hauser & Schwarz, 2020;
space (Logan, 1995). If your task is to search for a spe- vignette adapted from Hendricks et al., 2018):
cific item—such as a vertical red line among lines of
other colors and orientations, or a specific probe Joe was just diagnosed with cancer. He knows
appearing next to a set of 5s and 2s—then hearing a that for the foreseeable future, every day will be
word that cues the target—such as “red” or “five”—will a battle against [journey with] the disease. The
make your search more efficient (Hommel et al., 2001; battle [road] he has to fight [take] will not always
Lupyan & Spivey, 2010b; Reali et al., 2006). Even quanti- be an easy one. Many people have written about
fier words such as “each” and “every” guide attention, their experiences on the battlefield [path], and he
and in slightly different ways: “Each” leads to a focus can turn to those for consolation. His friends and
on individual objects, whereas “every” facilitates group- family want him to know that he will not be alone
ing (Knowlton et al., 2022). in his battle [journey]. Even though sometimes he
The tight connection of language to attention and might not feel like talking, other times he may
perception means that simply hearing a label can help want to share stories of his battle [journey] with
you detect otherwise imperceptible objects shrouded others, and they will be there for those moments.
112 Flusberg et al.

Participants then rated their agreement with a series research has focused on the process by which this
of statements expressing fatalistic beliefs about cancer occurs (Albarracín et al., 2018; Perloff, 2017; Petty &
(e.g., “If someone is meant to get cancer, they will get Cacioppo, 1986). The structure and content of the mes-
it no matter what they do”). Those who had read the sage matter: Linguistic framing has been shown to
battle-framed story endorsed fatalism more strongly shape attitudes toward other people, objects, events,
than those who had read the journey-framed story. As and social problems (e.g., Chatruc et al., 2021; Flusberg,
we discuss further in the “Figurative Framing” section, van der Vord, et al., 2022; Landau et al., 2009; Landau
metaphors are especially effective for shaping beliefs & Keefer, 2014; Levin & Gaeth, 1988; Nelson & Kinder,
about abstract and complex issues such as cancer 1996; Rook & Holmes, 2023; Thibodeau, Crow, &
because they leverage what you know—your cognitive Flusberg, 2017).
frames—about more concrete, familiar domains such For example, people express more favorable atti-
as battles and journeys. tudes toward ground beef labeled “75% lean” (positive
Other forms of linguistic framing can shape beliefs framing) compared with “25% fat” (negative framing).
as well (e.g., Chestnut et al., 2021; Chestnut & Markman, They report it to be less greasy, higher quality, and
2018; Holmes et al., 2022; Nelson & Oxley, 1999; better tasting (Levin, 1987). This effect is observed even
Schaffner & Atkinson, 2009). In one experiment, par- when people take a bite of the beef after receiving the
ticipants read a press release about a real study that linguistic frame (Levin & Gaeth, 1988; see also Anderson
analyzed 3 years of standardized math test scores, & Barrett, 2016). In a study by Schmidt et al. (2017),
revealing that boys and girls performed equally well participants tasted an array of wines while their brain
(Chestnut & Markman, 2018). For some participants, activity was measured through neuroimaging. All the
the report began with the statement “A recent study has wines cost the same amount of money, but participants
shown that girls do just as well as boys at math.” This saw different price labels for each one during the
places “boys” in the complement position of the sen- experiment. The results showed they preferred wines
tence (i.e., the reference point) and “girls” in the subject that had been framed as more expensive, and this was
position. For others, the positions were reversed (“boys mediated by activity in the brain’s valuation system (see
do just as well as girls at math”). Participants were then also Werner et al., 2021). In other words, from the per-
asked which group was naturally more skilled at math spective of their brains, the pricier-framed wines actu-
(or must work harder to be good at math). The results ally tasted better.
showed that the group in the complement position was Framing manipulations can also shape attitudes
viewed as more innately math-inclined. Notably, using toward important social issues. In one set of experi-
a different sentence structure (“girls and boys do ments, researchers examined how different group labels
equally well at math”) eliminated the bias to believe influenced American participants’ attitudes toward peo-
that one group was better than the other. We discuss ple living in the United States without authorization
this further in the “Social-Pragmatic Mechanisms” (Rucker et al., 2019; see also Ommundsen et al., 2014).
section. For now, we simply highlight that even An initial study revealed that the labels “illegal aliens,”
exceedingly subtle differences in language can shape “illegal immigrants,” and “undocumented aliens” were
beliefs. perceived as more negatively valenced than the labels
“undocumented immigrants” and “noncitizens.” In a
follow-up framing experiment, participants received
Attitudes
one of those five group labels in the instructions
Some of us love anchovies but loathe raisins, enjoy prompt: “The following questions deal with your
musical theater but dislike opera, and adore puppies thoughts about the term ‘[group label].’ As you know,
but have an aversion to snakes (you may feel otherwise, the issue of [group label] in the United States is hotly
of course). The previous sentence describes a set of debated right now.” Next, they used a “feelings ther-
attitudes: “a psychological tendency that is expressed mometer” to register their attitudes toward the group
by evaluating a particular entity with some degree of on a scale from extremely cold to extremely warm.
favor or disfavor” (Eagly & Chaiken, 1993, p. 1). Attitudes Participants who received the “illegal aliens,” “illegal
play an important role in our daily lives and influence immigrants,” or “undocumented aliens” labels expressed
how we navigate our physical and social environments significantly more prejudiced (colder) attitudes than
(although the relationship between attitudes and behav- those who received the “undocumented immigrants” or
ior can be complicated; Ajzen et al., 2018; Fazio, 1986). “noncitizens” labels. This shows how negative associa-
As a result, a central goal of many persuasive appeals tions conjured by a simple word or phrase can subtly
is to change how people feel about a given issue. Much shape attitudes.
Psychological Science in the Public Interest 25(3) 113

Memory it (McCloskey & Zaragoza, 1985). That is, participants

who do not remember the accident very well may sim-
Some words and phrases are more memorable than ply respond in a way that is compatible with the leading
others, such as those that evoke negative feelings (Aka question—reporting a relatively high speed and broken
et al., 2021, 2023). This may boost their influence and glass when questioned with “smashed,” for example.
facilitate accurate retrieval of a message or idea later On this account, participants’ responses reflect the
on. But language can also bias how you respond to intensity of the verb not because the question distorted
questions about the past. For example, try to recall the their memory but because it is the only information
last movie you saw in a theater. How long was the film they have to rely on. On a practical level, whether fram-
(in minutes)? Research suggests that your estimate ing alters memory per se may not matter much. So long
would be reduced if we instead asked, “How short was as different words or phrases lead people to give dif-
it?” (Harris, 1973; Lipscomb, Bregman, & McAllister, ferent accounts of their experiences, this could have
1985; Lipscomb, McAllister, & Bregman, 1985; see also important consequences for eyewitness identification
Inbar & Evers, 2022; Stephensen et al., 2021). and other real-world decisions (Loftus, 1979). More
Psychologist Elizabeth Loftus conducted many pio- problematic is that Loftus and Palmer’s (1974) findings
neering studies that suggested leading questions can have not been consistently replicated (e.g., Goldschmied
distort eyewitness memory (e.g., Loftus, 1975; Loftus & et al., 2017; Lipscomb, Bregman, & McAllister, 1985;
Palmer, 1974; Loftus & Zanni, 1975). In two famous McAllister et al., 1988; J. D. Read & Bruce, 1984; J. D.
experiments, Loftus and Palmer (1974) investigated the Read et al., 1978), which may preclude any practical
impact of verb intensity on memory for vehicular acci- applications of the findings. We have more to say about
dents. Participants watched a series of short video clips the importance of replication in studies of linguistic
depicting traffic accidents. After each one, they wrote framing in the “Opportunities for Future Research”
an account of the accident and answered questions section.
about what they had seen. The critical question asked However, many similar studies have shown that lan-
participants to estimate the speed of the cars during guage can interfere with tasks in which people must
the collision, and the intensity of the verb in this ques- recall what they have seen (e.g., Alogna et al., 2014;
tion varied across conditions: “About how fast were the Carmichael et al., 1932; Lupyan, 2008; Meissner &
cars going when they [contacted vs. hit vs. bumped vs. Brigham, 2001; Melcher & Schooler, 1996; Schooler &
collided vs. smashed] each other?” The results showed Engstler-Schooler, 1990; Wang & Gennari, 2019). An early
that more intense verbs led to greater speed estimates. demonstration of this phenomenon comes from a study
For example, “smashed” resulted in a mean estimate of by Leonard Carmichael and his colleagues (1932).
40.8 miles per hour, whereas “contacted” led to an Participants viewed a set of simple line drawings (e.g., a
estimate of only 31.8 miles per hour. shape that looked roughly like O–O) and were tasked
In a follow-up study, participants watched a single with reproducing them. Some participants received a par-
video of a multicar accident and then answered the ticular set of labels alongside the images (e.g., “eye-
same questions as before. This time, one third of par- glasses”), whereas others received a different set of labels
ticipants received the verb “smashed” in the critical that could also apply (e.g., “dumbbells”). When partici-
probe question, another third received “hit,” and the pants later redrew the images from memory, they unwit-
remaining third watched the video but received no tingly introduced changes to make the image better
communicative frame. Once again, the more intense match the label accompanying it. For example, the “eye-
verb led to greater speed estimates. The innovative glasses” group drew something like O^O, whereas the
feature of this study was that participants were asked “dumbbells” group drew something more like O=O (apol-
to come back to the lab a week later and indicate ogies for our primitive ASCII art skills). These findings,
whether they had seen any broken glass in the video like Loftus and Palmer’s (1974), may reflect the influence
they watched the previous week. Those in the “smashed” of language in our recollections rather than the direct
condition were more than twice as likely as those in impact of language on visual memory (McCloskey &
the “hit” and no-framing conditions to report having Zaragoza, 1985). Someone who remembers only the label
seen broken glass. Yet no broken glass had been visible “eyeglasses” will tend to produce a drawing that looks
in the crash film. This suggests that the critical verb not more like O^O than O=O (Hanawalt & Demarest, 1939),
only skewed participants’ memory for the accident but but this does not mean their memory of the drawing was
also led them to remember details that did not exist. somehow fused with the label. Nevertheless, research on
There is another explanation for these classic find- language and memory demonstrates how elements of
ings, however. The verb may have only affected par- our cognitive and semantic frames, cued by language,
ticipants’ reports of the accident, not their memory of can subtly distort our recollections.
114 Flusberg et al.

Reasoning, judgment, Thibodeau & Boroditsky, 2011). In other experiments,

and decision-making participants must choose between two or more response
options, as in the risky-choice framing paradigm. In
One of the most seminal framing experiments in cogni- one such study, participants read a series of vignettes
tive psychology goes something like this (adapted from about various everyday issues (e.g., crime, politics, bil-
Tversky & Kahneman, 1981): Imagine you’re the mayor liards, medical research), each one framed using one
of a small city preparing for the outbreak of an unusual of two metaphors (Thibodeau, 2016). After reading each
disease, which is expected to kill 600 people. Your vignette, participants answered a question by selecting
health minister presents you with two alternative pro- between two response options, each conceptually con-
grams to address the outbreak. You must now decide gruent with one of the metaphor frames. For example,
which one to implement: a cancer researcher was described as imagining herself
either “scaling a mountain” or “working on a puzzle”
A. If Program A is adopted, 200 people will be saved. while conducting her research. Participants were asked
B. If Program B is adopted, there is a one-third prob- to make a prediction about the researcher’s process.
ability that all 600 people will be saved and a two- Does she “look for connections by testing completely
thirds probability that nobody will be saved. novel theories” (congruent with the puzzle metaphor)
or “gain ground by using methods that are simple to
This is a study of judgment and decision-making. follow” (congruent with the mountain metaphor)?
Your task is to think about this dire situation and make Across all target issues, participants were about 10%
the most optimal choice. In the original version of the more likely to choose the metaphor-consistent response
experiment, 72% of participants selected the sure option. This illustrates how cognitive frames evoked by
option, A, rather than the uncertain option, B. Now metaphor can guide people’s reasoning.
imagine that your health minister had provided the fol-
lowing two options instead: Behavior
Most of the studies we have considered so far elicited
C. If Program C is adopted, 400 people will die. effects of language in the laboratory. That’s not a bad
D. If Program D is adopted, there is a one-third prob- thing. Lab studies give researchers control over many
ability that nobody will die and a two-thirds prob- extraneous variables that impact human behavior,
ability that 600 people will die. enabling them to better establish causal relationships
of interest. Still, most of us want to know whether the
In this case, 78% of participants in the original study effects we carefully isolate in the lab generalize to the
selected Option D, the risky choice. Notice, however, real world. This is difficult to ascertain for several rea-
that Options A and C are logically equivalent: If 200 sons, including ethical considerations, time, and
people are saved, that means 400 will die, because 600 money—three things that researchers struggle with.
people were initially expected to perish in the out- One stepping stone to reality is to assess behavioral
break. A parallel equivalence holds for Options B and intentions. In many framing studies, participants are
D. Framing choices in a positive manner, as lives saved asked to indicate whether they plan to take some action
(a “gain” frame), leads people to select the sure option, in the future or to rate their likelihood of engaging in
whereas framing them in a negative manner, in terms a particular behavior (e.g., Flusberg et al., 2017; Gerend
of deaths (a “loss” frame), leads people to select the & Shepherd, 2007; D. J. Hauser & Schwarz, 2015, 2020;
riskier option. The “risky-choice” framing paradigm has Thibodeau & Flusberg, 2017). In one study, for exam-
been quite influential in fields such as health commu- ple, female college students rated their intentions to
nication and behavioral economics, and variations of obtain the human papillomavirus vaccine after reading
this basic study design have been used thousands of a booklet that used either a gain frame (i.e., discussing
times across these fields (for meta-analyses, see the benefits of receiving the vaccine) or a loss frame
Kühberger, 1998; Steiger & Kühberger, 2018). (i.e., discussing the costs of not receiving it; Gerend &
Many other studies ask participants to solve prob- Shepherd, 2007; for additional context on this para-
lems, make decisions, or otherwise engage in reasoning digm, see our discussion of goal framing in the “Valence
about an issue that has been framed in a particular way Framing” section). The loss frame increased intentions
(Flusberg et al., 2020; Thibodeau, 2016; Thibodeau & to get the vaccine, but only for participants who
Boroditsky, 2011, 2013). In some cases, participants reported multiple sexual partners and rarely used con-
freely generate their own solution to a problem, which traception. The relationship between intentions and
is then coded by the experimenters or naive raters (e.g., behavior is complicated, however, and people do not
Psychological Science in the Public Interest 25(3) 115

always follow through on what they say they will do (Gerber et al., 2016, 2018, 2023). This may be because
(Conner & Norman, 2022; Sheeran & Webb, 2016; for it is difficult to conduct exact replications in the field
a meta-analysis of the intention-to-behavior literature, (Bryan et al., 2019), and heterogeneous samples and
see Webb & Sheeran, 2006). Therefore, there are limits methods will yield noisy results for any behavioral
to what we can conclude from studies of behavioral intervention, including framing studies (Bryan et al.,
intentions. 2021). We explore some of this heterogeneity later on
One way to get at behavior more directly is to ana- when discussing moderators of linguistic framing
lyze naturally occurring data sets that involve interac- effects. For now, we turn to general mechanisms that
tions between people using systematically different give rise to these effects.
communicative frames. For instance, one study exam-
ined nearly 200 medical emergency call transcripts from Mechanics of Communication
the Perth, Australia, metropolitan area over 2 years (Riou
et al., 2017). Protocol dictates that emergency operators
and Linguistic Framing
use a scripted prompt to elicit information from the When scholars write about framing, they tend to focus
caller: “Okay, tell me exactly what happened.” About on the framing manipulation (“we emphasized gains
60% of the time, however, the operators spontaneously versus losses”) and the framing effect (“the loss frame
shifted from the past tense to the present perfect tense, led to increased behavioral intentions”). Emphasizing
saying instead, “Tell me what’s happened.” This subtle the outcome, rather than the process, may contribute
change was associated with substantially different to the perception that the framing literature is “frac-
response patterns from the callers. Whereas the prompt tured” (Entman, 1993), consisting of hundreds of iso-
in the past tense tended to elicit longer, narrative lated “effects” without any organizing framework (or
accounts of what led them to place the call (18 s on cognitive frame). Viewing framing through the lens of
average), the prompt in the present perfect elicited human communication helps to address this concern.
shorter, more direct reports (9 s on average). This trans- This perspective connects research on the cognitive
lated into a nearly 15% faster ambulance dispatch time science of language and reasoning with insights from
in the latter case (from 58 to 50 s). A similar study from the study of persuasion and social influence. It also
the United Kingdom found that people in crisis were reveals how a few general mechanisms can account for
more likely to reject a proposal for dialogue when a broad range of framing effects. We begin this section
police negotiators used the verb “talk” than when they by briefly discussing language comprehension. We then
used the verb “speak” (Sikveland & Stokoe, 2020). unpack the cognitive, social-pragmatic, and emotional
Nothing beats a randomized experiment for estab- processes that give rise to framing effects in everyday
lishing causal relationships between variables, however. communication. We conclude by relating this work to
As a result, some of the most eye-catching framing a leading theory of persuasion and questioning the
studies have used a field study design to measure real- popular view that framing effects reflect the irrational
world behaviors (e.g., Bryan et al., 2011; Chou & side of human cognition.
Murnighan, 2013; Heritage et al., 2007; Hershfield et al.,
2020). In one study, conducted on the eve of the 2008
Making sense of language
U.S. presidential election, registered California voters
recruited on social media completed a 10-item election Many species use sophisticated communication sys-
survey (Bryan et al., 2011). For half of the participants, tems, but human language differs in several key
the survey questions used a noun form to refer to the respects. For one, language is far more complex and
act of voting (e.g., “How important is it to you to be a flexible: The typical person knows many thousands of
voter in tomorrow’s election?”). For the other half, the (mostly) arbitrary symbols or words, which can be com-
survey questions used a verb form (e.g., “How impor- bined and reformatted to express an infinite number of
tant is it to you to vote in tomorrow’s election?”). After ideas. This makes language a productive or “generative”
the election, an analysis of voting records revealed that system. We can also use our words to communicate
those exposed to the noun frame were almost 14% about people, places, objects, events, and ideas beyond
more likely to vote in the election than those exposed the here and now (e.g., “The stingy extraterrestrials
to the verb frame (95.5% vs. 81.8% voting rate). Similar insisted on buying the cheapest spaceship, which led
results were later found in a New Jersey gubernatorial to the crash”). This property, called “displacement,”
race, and related work has extended this methodology exists only in very limited forms in the rest of the ani-
to nonvoting contexts (Bryan et al., 2011, 2013, 2014). mal kingdom (e.g., in the honeybee’s “waggle dance,”
More recently, however, other researchers have failed used to communicate the location of its most recently
to replicate the original effect on voting behavior visited flowers; von Frisch, 1967).
116 Flusberg et al.

Put together, these unique characteristics enable a message in a particular way encourages the reader or
what linguist Daniel Dor (2015) describes as the central listener to construct a corresponding mental model,
function of language: the “instruction of imagination.” which may have downstream consequences for attitudes,
To an unprecedented degree, we can recreate our judgments, and behavior. In the next few sections, we
thoughts, intentions, and experiences in the minds of unpack the cognitive, social-pragmatic, and emotional
our interlocutors. This is not a direct transfer of infor- mechanisms that support language comprehension and
mation, such as sending an email attachment that some- reasoning and thus underlie many framing effects. This
one downloads onto their neural hardware (cf. Reddy, provides a general framework for understanding how
1979). Rather, understanding language is an active, framing works. For a summary, see Table 2.
reconstructive process. As Dor (2015) puts it:
Cognitive mechanisms
The speaker provides the receiver with a code, a
plan, a skeletal list of the basic coordinates of the Schematic structuring. Earlier we defined cognitive
experience—which the receiver is then expected frames as a form of memory comprising relatively abstract
to use as a scaffold for experiential imagination. knowledge about the world that we use to make sense of
Following the code, the interlocutor raises past our experiences. This type of prior knowledge is critical
experiences from memory, and then reconstructs for forming rich situation models from language. Con-
and recombines them to produce novel, imagined sider, for example, the following piece of gastronomic
experiences. (p. 2)1 fiction (inspired by Bower et al., 1979):

The “code” consists of the specific sequence of Isabel went to a restaurant.

words that a speaker uses to deliver their message. In
She ordered a burger but sent it back because it was
spoken or signed conversations, as opposed to writing,
overcooked.
the code might also include other factors such as our
tone of voice, gestures, and facial expressions (Kita & A while later, she left in a hurry.
Emmorey, 2023). The reader or listener must then
“decode” these communicative signals. This requires This story never explicitly mentions that Isabel sat
generating a mental model of the situation or events down at the restaurant, looked at a menu, spoke with a
described in the text—also called a mental “simulation” server, or received a bill for her meal. And yet you likely
or “situation model” (Bergen, 2012; Graesser et al., assumed that those events took place—they were pres-
1997; Johnson-Laird, 1983; Zwaan & Radvansky, 1998). ent in your situation model (Bower et al., 1979). This
If the situation model aligns with the speaker’s intended is because words such as “restaurant” and “ordered a
message, then successful language comprehension has burger” evoke a particular cognitive or semantic frame,
occurred. the “restaurant” frame, which includes these elements.
Decoding a linguistic message isn’t like translating Other knowledge stored in memory informed your
Morse code into English, however. As Dor (2015) notes, interpretation of the story as well. For instance, you
linguistic messages are “skeletal”—they are underspeci- probably inferred that Isabel doesn’t like well-done
fied, and the receiver must fill in details based on con- beef—otherwise, why would she send the burger back?
text, shared background knowledge, and inference. As to why she left in a hurry, you can’t know for sure.
This process is dynamic, unfolding over time as we Situation models can be schematic or fuzzy. But several
process a message and activate relevant semantic and plausible reasons come to mind that are consistent with
cognitive frames. We anticipate what information will this event. Maybe Isabel left without paying because
come next and use our working mental model to inter- she didn’t like the food and wanted to get away quickly.
pret the communicative signals as they arrive Or maybe she just had somewhere else to be. Regardless,
(Christiansen & Chater, 2022; Pickering & Gambi, 2018). notice what happens when we change the organizing
All this takes place in a social context, between speaker frame in the story:
and listener, writer and reader, in which certain norms
and rules constrain how we respond and react. It also Isabel went to a friend’s barbecue.
takes place in a biological context: Reasoning is not a She ordered a burger but sent it back because it was
purely logical, symbolic process but a deeply embodied overcooked.
one that is suffused with imagery, emotion, and physi-
A while later, she left in a hurry.
cal movement (Bergen, 2012; Lakoff, 1987).
Ultimately, our situation models are the basis for fur-
ther reasoning about the topic of conversation, guiding Now your interpretation of the story—the content of
how we answer questions and make decisions. Framing your situation model—is quite different. You no longer
Psychological Science in the Public Interest 25(3) 117

Table 2. Psychological Mechanisms That Contribute to Linguistic Framing Effects

Mechanism Explanation Example

Cognitive
Schematic structuring Cognitive and semantic frames evoked Participants were more likely to recommend
by a message influence how people enforcement-related solutions to a crime problem—
organize their mental model of a target such as hiring more police and building more jails—
issue, licensing certain patterns of when crime was metaphorically framed as a “beast” as
reasoning. opposed to a “virus” (Thibodeau & Boroditsky, 2011).
Enforcement is consistent with the beast frame.
Value reweighting Issue-relevant values and beliefs invoked Liberal Americans expressed less support for Democratic
by a message seem more important presidential candidate Hillary Clinton after reading an
and therefore figure more prominently anti-Clinton message that described her as violating
in reasoning, especially when they the important liberal value of fairness as opposed to
resonate with the audience. the more conservative value of loyalty (Voelkel &
Feinberg, 2018).
Priming Exposure to a message makes the Participants rated ground beef as less greasy, higher
ideas communicated in it (and quality, and better tasting when it was labeled “75%
related concepts) temporarily more lean,” which primes positive associations, than when
cognitively accessible and therefore it was labeled “25% fat,” which primes negative
more prominent in evaluations and associations (Levin, 1987; but see discussion in text
reasoning. on how emotional and pragmatic factors may explain
related effects).
Processing fluency Some prose is easier to process than Participants were more persuaded by audio narratives
other prose, making a message seem about the health risks of caffeine compared with
more persuasive. nonnarratives, and this was mediated by greater
processing fluency for narratives (Bullock et al., 2021).
Social-pragmatic
Pragmatic inference People read between the lines of a When exposed to generic statements about a social
message and infer what the speaker group (e.g., “Zarpies are good at baking pizzas”),
intended to communicate, going children and adults inferred that a member of an
beyond literal meaning. unmentioned group (“Gorps”) was bad at this skill
(Moty & Rhodes, 2021). This did not occur for
nongeneric statements (e.g., “This Zarpie is good at
baking pizzas”).
Speaker inference People draw inferences about the Conservatives were less likely to oppose a
speaker’s attributes from the language proenvironmental policy when it appealed to a need
they use (e.g., their identity, values, for closure—a trait conservatives exhibit more than
and character traits), which can bias liberals. This resulted from the fact that participants
their response to the message. inferred the policy was proposed by a fellow
conservative (Lammers et al., 2023).
Emotional
Affective valence Language evokes positive and negative Participants expressed more negative attitudes
feelings, which can influence attitudes toward immigration when it was discussed using
and decision-making. Effects may be negatively valenced group labels such as “illegal
stronger (or weaker, in some cases) aliens” compared with more neutral labels such as
when the feelings are more intense or “noncitizens” (Rucker et al., 2019).
arousing.
Discrete emotions Language evokes specific emotions (e.g., Dutch participants expressed more support for
anger, fear, sadness) that figure into cooperation among European Union nations when
evaluations and reasoning. economic investments were framed positively than
negatively, and this was mediated specifically by
enthusiasm and anger, not other emotions (Lecheler
et al., 2013).

assume Isabel looked at a menu, talked to a server, or perhaps it was her partner. You certainly haven’t
received a bill. Instead, she probably just asked for a inferred that Isabel left in a hurry because she didn’t
burger from whomever was working the grill at the pay for her food (unless that sort of transaction is typi-
party. Maybe it was her friend who was hosting, or cal at the barbecues you frequent). Maybe she was
118 Flusberg et al.

embarrassed for sending food back at a friend’s gather- coping with cancer. The “journey” semantic frame
ing because it was a socially awkward thing to do. includes elements such as moving down a path toward
What these examples illustrate is that communicative a destination and encountering and overcoming obsta-
frames guide the aspects of prior knowledge—our cog- cles. When this frame is evoked in the context of cancer,
nitive and semantic frames—that we use to organize these elements call up thoughts about the path of
and interpret the events described in a text or in speech. receiving treatment over time, the destination of remis-
Different communicative frames lead people to con- sion and recovery, and obstacles such as pain and
struct different situation models, which then inform nausea.
judgments and predictions about the situation. This is The “battle” semantic frame, on the other hand,
one of the central cognitive mechanisms that has been includes elements such as two opposing groups fighting,
proposed for how linguistic framing works (Lakoff, allies and enemies, and winners and losers. When this
1996, 2008; Lakoff & Johnson, 1980; Thibodeau, frame is evoked in the context of cancer, these elements
Hendricks, & Boroditsky, 2017; Thibodeau et al., 2019). summon thoughts about fighting the enemy cancer cells
However, we want to flag that this is simply part of the in the body, the goal of winning the battle by eliminating
normal process for making meaning from language. 2 the threat, and the all-too-real possibility of losing the
To demonstrate this effect using a real-world example, war by succumbing to the disease. As D. J. Hauser and
let’s examine the following viral Twitter (now X) Schwarz (2020) note: “One way to approach a battle is
exchange between The Economist magazine and user to yield, give up control, and surrender to the attacker”
@ZachBoomG (can we call you Zach?), which took (p. 1699). They suggest that this element of the battle
place on August 8, 2020. After you read the first post, frame—the notion of surrender—was also evoked by
ask yourself how you feel about the U.S. Postal Service the story, leading participants to entertain the idea that
(USPS). Then read Zach’s reply and ask yourself if your people with cancer may lose the war by surrendering.
attitude has shifted: Consequently, they expressed a more fatalistic outlook
toward the disease.

@TheEconomist: The USPS is viewed favourably by Reweighting of issue-relevant beliefs and values. In
91% of Americans despite billions in losses (3.9K some cases, communicative frames do not exert influ-
likes, 1.0K retweets) ence by restructuring our mental models. Rather, they
@ZachBoomG: It’s a service. It doesn’t lose money. work by making issue-relevant beliefs or values evoked
It costs money. No one says the military loses $750b by the frame seem more important (Chong & Druckman,
a year (162.3K likes, 42.5K retweets) 2007; Nelson, Clawson, & Oxley, 1997; Nelson & Oxley,
1999; Nelson, Oxley, & Clawson, 1997). As Nelson, Oxley,
The Economist frames the USPS as a business, in and Clawson (1997) explain in an article on the psychol-
which monetary losses symbolize failure. Zach counters ogy of news framing effects: “Frames tell people how to
by reframing the USPS as a service, such as the military; weight the often conflicting considerations that enter into
losses don’t fit the “service” frame—costs do. Our taxes everyday political deliberations. Frames may supply no
pay for many vital services, including the military, new information about an issue, yet their influence on
police, firefighters, clean water, and, on this construal, our opinions may be decisive through their effect on the
the USPS. Your beliefs and attitudes toward the USPS perceived relevance of alternative considerations” (p. 226).
might shift depending on which frame is organizing In one experiment that illustrates this mechanism,
your current mental model (although it doesn’t hurt college students read a news report about a proposed
Zach’s case that “service” is in the name). Ku Klux Klan (KKK) rally at their university (Nelson,
Many framing effects in the experimental literature Clawson, & Oxley, 1997). For half of them, the story
can be similarly explained as a result of schematic was framed as a free-speech issue, with a headline
(re)structuring of our mental model for the target issue. reading “Ku Klux Klan Tests OSU’s Commitment to Free
For example, in the “Beliefs” section, we described a Speech,” along with a few other sentences that rein-
study showing that relative to framing cancer as a jour- forced the frame. For the other half, the story was
ney framing cancer as a battle increased fatalistic think- framed as a public-order issue, with the following head-
ing about the disease (D. J. Hauser & Schwarz, 2020). line instead: “Possible Ku Klux Klan Rally Raises Safety
Even though this study used a figurative framing device, Concerns.” Participants exposed to the “free-speech”
the structuring process works the same as the previous frame expressed greater tolerance for KKK rallies and
examples we discussed. That is, participants used the speeches. Critically, however, this effect was mediated
different semantic frames evoked by the two commu- by the importance participants placed on the values of
nicative frames to organize their mental model of freedom of speech and public order. Those exposed to
Psychological Science in the Public Interest 25(3) 119

the “public-order” frame assigned significantly more discount, even when used to argue for a position
weight to this value, which reduced their tolerance for one would typically oppose. As a result, when
the proposed rally. individuals face a morally reframed argument that
Value reweighting helps explain the potency of com- resonates with their fundamental moral convic-
municative frames that evoke moral values and other tions, they are more likely to evaluate the argu-
attributes tied to a sense of personal identity. This is ment positively and revise their relevant attitudes
the explanation provided by Bryan and colleagues as a result. (p. 4)3
(2011) for why the noun form “being a voter” increased
voter turnout compared with the verb form “vote” in Priming. After reading the word “table,” you are quicker
the study we discussed in the “Behavior” section. Nouns to recognize the word “chair” (but not “suitcase”). You
communicate more stable, essential trait information are also more likely to answer a question such as “What’s
than verbs do (Gelman & Heyman, 1999; Walton & an important piece of furniture?” with “table” or “chair”
Banaji, 2004). Because voting is a socially desirable than with “bookshelf.” This is known as priming, in
activity in the United States, people may be motivated which exposure to one stimulus—in this case, a word—
to see themselves (and be seen) as someone who temporarily increases the accessibility and processing of
votes—a voter. Exposure to the noun frame in the elec- related stimuli. Priming is a well-established, basic cogni-
tion survey may have caused people to assign more tive process (Schacter & Buckner, 1998), and it has been
importance to this self-identity trait, increasing their used to explain a range of framing effects. For example,
likelihood of voting in the election. Although this effect a negative/loss frame such as “25% fat” may prime associ-
has not been consistently replicated (Gerber et al., ated negative concepts (e.g., “unhealthy,” “bad”). This
2016, 2018, 2023), other studies have shown how com- would lead you to evaluate a product more negatively
municative frames that invoke moral values relevant to than a logically equivalent positive/gain frame such as
the receiver may be especially effective at shaping atti- “75% lean,” which primes more positive concepts (Levin
tudes and beliefs (e.g., Bloemraad et al., 2016; Bryan et al., 1998).
et al., 2013, 2014; Feinberg & Willer, 2013, 2015, 2019; Communication scholars have argued that priming
Feygina et al., 2010; Franks & Scherr, 2019; Hurst & helps explain the influence of political and news-media
Stern, 2020; Kidwell et al., 2013; Nath et al., 2022; frames (Hoewe, 2020; Roskos-Ewoldsen & Roskos-
Voelkel et al., 2022, 2023; Voelkel & Feinberg, 2018; Ewoldsen, 2009). Stories and messages in the media
Wolsko et al., 2016). become more accessible (“primed”) in the minds of the
For example, one experiment found that conserva- public, which can then impact attitudes. When you are
tive Americans expressed increased support for same- asked how you feel toward a political candidate run-
sex marriage when it was framed in terms of loyalty to ning for office, any negatively (or positively) framed
the nation (“same-sex couples are proud and patriotic recent news stories will likely come to mind. Information
Americans”), an important conservative value (Feinberg that is easily accessible in this way is often assumed to
& Willer, 2015). In the same set of studies, liberal be more common or diagnostic (the “availability heu-
Americans expressed increased support for military ristic”; Tversky & Kahneman, 1973), so priming a nega-
spending when it was framed in terms of fairness and tive story may bias your evaluation against the candidate.
equality (“through the military, the disadvantaged can As a result, what the mass media choose to focus on
achieve equal standing and overcome the challenges in their coverage can powerfully shape what the public
of poverty and inequality”), which are important liberal views as relevant and important (this is known as
values. Similar effects are evident in various sociopoliti- “agenda setting” in communication studies; Weaver
cal domains, including support for environmental et al., 2004).
causes (Feinberg & Willer, 2013; Feygina et al., 2010; That said, priming alone cannot fully explain certain
Hurst & Stern, 2020), immigration (Nath et al., 2022; media framing effects (Hendricks et al., 2018; Nelson,
Voelkel et al., 2022), and political figures (Voelkel et al., Clawson, & Oxley, 1997; Thibodeau & Boroditsky, 2011;
2023; Voelkel & Feinberg, 2018). Aligning the moral Thibodeau, Crow, & Flusberg, 2017). For example, in
values in a communicative frame with the moral values the KKK rally study described in the previous section
held by the target audience is known as “moral refram- (Nelson, Clawson, & Oxley, 1997), participants com-
ing.” Feinberg and Willer (2019) provide a compelling pleted a “lexical-decision” task after reading their
explanation for the potency of this rhetorical strategy, assigned news article. They were shown a series of
especially in the context of political persuasion: letter strings in a randomized order and had to indicate
as quickly as possible whether each string was a real
Because moral convictions are so strongly held, word or not. Some of the real words were associated
arguments that appeal to them are difficult to with free speech (e.g., “liberty”), some were associated
120 Flusberg et al.

with public order (e.g., “danger”), and the rest were Mr. Colquhoun; Laham et al., 2012), viewing them as
neutral fillers (e.g., “planet”). If participants were more trustworthy (Silva et al., 2017) and truthful (Newman
primed by reading the message frames, you would et al., 2014).
expect their reaction times to depend on which frame Consequently, several studies have found that pro-
they received. Specifically, those exposed to the free- cessing fluency can increase the persuasive power of a
speech frame should be faster to respond to the free- message frame (Bullock et al., 2021; Kidwell et al., 2013;
speech-related words, and those exposed to the H. J. Kim & Jang, 2018; A. Y. Lee & Aaker, 2004; Mayer
public-order frame should be faster to respond to the & Tormala, 2010; Okuhara et al., 2017). This may be
public-order-related words. No such pattern was another reason why it is so effective to align the moral
observed. As we described earlier, however, the degree values in a communicative frame with the moral values
of importance participants placed on free speech and held by the target audience (Feinberg & Willer, 2019).
public order did predict their support for the KKK rally, In one study, for example, participants were randomly
suggesting that value reweighting—and not priming—is assigned to view one of two messages aimed at increas-
the central mechanism for eliciting these effects. ing support for a recycling program (Kidwell et al.,
Other studies indicate that priming cannot fully 2013). One appeal was framed in terms of “individual-
account for the framing effects of metaphors (Hendricks izing” moral foundations such as fairness and harm
et al., 2018; Thibodeau & Boroditsky, 2011; Thibodeau, reduction, which resonate more with liberal values. The
Crow, & Flusberg, 2017). For example, in one experi- other was framed in terms of “binding” moral founda-
ment, participants read a brief report about a city expe- tions such as being part of a group and duty to authority,
riencing a surge in violent crime. The first sentence of which resonate more with conservative values. Recycling
the article framed crime as either a “beast” or a “virus” intentions were higher when the moral values of the
that was ravaging the city. Compared with participants message aligned with the moral values of the partici-
in the “virus” condition, those who read that crime was pant. However, this was mediated by processing fluency:
a “beast” were more likely to generate enforcement- Liberal participants found the “individualizing” message
related solutions to the crime problem (e.g., build more clearer and easier to follow, whereas conservatives felt
jails)—conceptually congruent with the “beast” frame. that way about the “binding” message. Although these
In another version of the study, participants were sim- results are intriguing, processing fluency is a relatively
ply asked to list a synonym for the word “beast” or understudied mechanism in the linguistic framing litera-
“virus” before reading a nonmetaphorically framed ver- ture. More research is needed to better understand its
sion of the news report. In this case, there was no dif- role in different types of framing.
ference in the solutions generated across conditions.
This suggests that simply being primed with a word is
Social-pragmatic mechanisms
not sufficient to generate a metaphor framing effect.
Rather, the word has to be used as a metaphor—a kind Pragmatic inference. Language comprehension involves
of schematic structure—to influence people’s mental “reading between the lines” and interpreting the meaning
model of the target issue. of words in context. This often requires going beyond
the literal meaning of a statement to figure out what the
Processing fluency. Some messages feel easier to read person really means. As we read or listen to language,
and understand than others. Consider how much more we draw pragmatic inferences about the speaker’s or
arduous it is to trudge through a legal contract than to writer’s communicative intentions, assuming that they
float through breezy prose in a romance novel. This sub- have chosen their words for good reason—because they
jective feeling of how easy or difficult it is to process new are informative and relevant (Christiansen & Chater,
information is known as processing fluency—a kind of 2022; Goodman & Frank, 2016; Grice, 1975; Sperber &
metacognitive judgment. Research has shown that peo- Wilson, 1986). For example, imagine you’re out with a
ple often use fluency as a heuristic in reasoning, favoring friend and you ask about the blind date she went on the
information that is easier to process and giving it greater night before. She pauses for a moment before replying,
weight in their evaluations (Alter & Oppenheimer, 2006; “He was nice.” How would you interpret this response?
Bullock et al., 2021; Claypool et al., 2015; Schwarz et al., “Nice” is a positive attribute, so you might conclude she
2021; Shah & Oppenheimer, 2007; for evidence that the liked her date. But if she really liked him, wouldn’t she
relationship between fluency and reasoning is not so say so more directly? By saying he was “nice”—and not
simple, however, see Markowitz & Shulman, 2021; saying he was “amazing, charming, and sexy”—your
Oppenheimer, 2008). All else being equal, for example, friend might be politely communicating that she found
people form more positive impressions of others with him boring, unattractive, and unworthy of a second date.
names that are easier to pronounce (e.g., Mr. Smith vs. This example illustrates that our situation models are
Psychological Science in the Public Interest 25(3) 121

based not only on the literal meaning of the words we directions framed using the present perfect tense (“Tell
hear but also on our assumptions about what the speaker me what’s happened”) compared with the past tense
intends to communicate. (“Tell me what happened”; Riou et al., 2017; see
Evidence suggests that pragmatic inferences contrib- “Behavior” section). This may be because the past tense
ute to a broad range of framing effects, including sev- implies that the speaker is interested in hearing about
eral we have introduced already (e.g., Flusberg, everything leading up to the current situation, whereas
Thibodeau, & Holmes, 2022; Flusberg, van der Vord, the present perfect tense implies an interest in the here
et al., 2022; Frisch, 1993; Holmes et al., 2022; Kühberger, and now. Similarly, people in crisis may be less likely
1995; Kühberger & Tanner, 2010; Leong et al., 2017; to engage with police negotiators who propose that
Mandel, 2001, 2014; McKenzie & Nelson, 2003; Rook & they “talk” rather than “speak” because “talk” implies a
Holmes, 2023; Sher & McKenzie, 2006; Wu et al., 2021). more confrontational exchange (Sikveland & Stokoe,
For instance, multiple scholars have promoted a prag- 2020; for additional discussion of these cases, see
matics-based account of the “unusual disease” risky- Enfield, 2022). These examples serve as a lesson to
choice framing effect we described in the “Reasoning, researchers on the importance of paying careful atten-
Judgment, and Decision-Making” section (e.g., Frisch, tion to the wording of questionnaires and other linguis-
1993; Gigerenzer, 2018; Kühberger, 1995; Kühberger & tic measures, which will inevitably be interpreted
Tanner, 2010; Mandel, 2001, 2014; Pinker, 2007). Mandel pragmatically (Schuman & Presser, 1996; Schwarz &
(2014) hypothesized that people assume communicated Oyserman, 2001; for similar recommendations in the
quantities are “lower-bounded” estimates rather than domain of law, see Kellermann, 2007).
precise values unless otherwise stated (for a similar
claim, see Pinker, 2007). In other words, when you are Inferences about the speaker. In addition to provid-
presented with the response option “If Program A is ing clues about a speaker’s intended meaning, language
adopted, 200 people will be saved,” you’ll tend to inter- provides clues about the speaker. People draw many
pret it as “at least 200 people will be saved” rather than inferences about others based on their choice of words
“exactly 200 people will be saved.” The same is true for or their “linguistic style,” including judgments of intelli-
the loss-framed option, “If Program C is adopted, 400 gence, group membership, social power, familiarity with
people will die.” Here you would probably interpret it the topic, and credibility (Areni & Sparks, 2005; Blanken-
as “at least 400 people will die.” Mandel argued that ship & Craig, 2011; Gibbons et al., 1991; Holtgraves &
this is why people are more likely to choose Program Lasky, 1999; Hosman & Siltanen, 2006). For example, one
A than Program C in the two versions of the experi- study found that participants judged a writer to be less
ment: C implies more deaths than A, pragmatically intelligent when they used more complex and esoteric
speaking. To support this claim, he showed that the language (Oppenheimer, 2006). This effect was mediated
framing effect is eliminated if you add the word by the reduced processing fluency associated with com-
“exactly” to the language of the response options but plicated prose.
not if you add “at least” (for conflicting evidence, how- Our impressions of people can affect how persuasive
ever, see Chick et al., 2016; Claus, 2022). we find them (Blankenship & Craig, 2011). For exam-
A similar account may explain why subject-comple- ple, some people communicate in a seemingly “power-
ment statements of equality (e.g., “girls do just as well less” style, using more hesitations (“um . . .”), hedges
as boys at math”; see “Beliefs” section) lead people to (“I sort of think that . . .”), and tag questions (“Right?
conclude that the group occupying the complement You know what I mean?”). People form more negative
position (“boys”) is superior (Holmes et al., 2022). impressions of powerless speakers and are less per-
People tend to infer that the speaker framed this group suaded by their messages (Areni & Sparks, 2005;
as the standard or “reference point” to communicate that Gibbons et al., 1991; Holtgraves & Lasky, 1999). In
it truly is superior. They then use this insight to update contrast, speakers who use many intensifiers (e.g.,
their own beliefs. Consistent with this explanation, a “really,” “extremely”) are often perceived as more com-
recent set of studies showed that subject-complement petent and in control (Hosman & Siltanen, 2006),
statements elicited stronger framing effects for partici- although their persuasiveness can differ as a function
pants who were more sensitive to their pragmatic impli- of other situational factors (Blankenship & Craig, 2011;
cations (Holmes, Wu, et al., 2024; Wu et al., 2021). Hamilton et al., 1990). Consider former President
Finally, the pragmatics of everyday conversation can Donald Trump, known for frequently using intensifiers
help explain many framing effects related to how peo- such as “very” and “tremendously” in his remarks. For
ple respond to questions and directions. For example, those supporters who view him as a credible source of
we noted earlier that people who called an emergency information, this might increase Trump’s persuasive
medical hotline provided shorter responses to power. For those who view him as untrustworthy and
122 Flusberg et al.

deceitful, however, such language might reinforce that “cognitive” regions such as the prefrontal cortex was
negative opinion. A study by Hamilton and colleagues associated with reduced framing effects (De Martino
(1990) found empirical support for this possibility: et al., 2006). In a recent behavioral experiment, risky-
Language intensity enhanced the persuasiveness of a choice framing was mediated by the valence of people’s
message delivered by a high-credibility source but self-reported evaluations of gain- and loss-framed
inhibited persuasiveness for a message delivered by a response options (Stark et al., 2017b). Finally, a meta-
low-credibility source. analysis of 30 years of communication research found
We are also quick to form impressions about the that whereas gain frames tend to induce positive emo-
social identities of communicators based on subtle lin- tions, loss frames tend to induce negative emotions, and
guistic details. Such inferences can lead to framing the presence of these emotions increases the influence of
effects with significant implications for society (e.g., the frames (Nabi et al., 2020).
Gaucher et al., 2011; Lammers et al., 2023). For exam- Language itself is valenced as well, as individual
ple, one set of studies examined the impact of gender- words and phrases can evoke positive or negative feel-
stereotypical wording in job advertisements (Gaucher ings. This can be considered a form of emotional prim-
et al., 2011). Certain words and phrases are more ing and helps explain certain framing effects. In the
strongly associated with masculine stereotypes (e.g., “Attitudes” section, for example, we discussed how
determined, strong, competitive, superior), whereas people form more negative attitudes toward immigra-
others are more strongly associated with feminine ste- tion when it is framed using negatively valenced group
reotypes (e.g., supportive, committed, sensitive, nurtur- labels such as “illegal aliens” compared with more neu-
ing). When a job advertisement was constructed to tral labels such as “noncitizens” (Rucker et al., 2019).
include more masculine than feminine wording, par- The affective associations of these terms are the prin-
ticipants inferred that there were more men working in cipal drivers of people’s reactions. This is why emo-
that occupation. As a result, women expressed less tional language is ubiquitous in political communication.
interest in the job, and this was mediated by their feel- Partisans on both sides of the aisle routinely use words
ing that they wouldn’t belong at the company advertis- and phrases that elicit positive feelings for positions
ing the position. they favor and negative feelings for positions they
oppose. This is reflected in how Republican politicians
use the phrase “death tax” to describe a tax on large
Emotional mechanisms
inheritances because they are averse to taxing wealthy
Aristotle argued that appeals to the emotions (pathos) citizens (Lakoff, 2008, 2014; Luntz, 2007).
form a central pillar of rhetoric. This view is echoed in Even the most subtle differences in valence can
contemporary scholarship on attitudes, reasoning, and shape attitudes. Some words seem neutral on their own
persuasion, in which emotion is understood to play a but have a tendency to appear alongside mostly posi-
critical role (Dillard & Seo, 2013; Lerner et al., 2015; tive or negative words in speech and writing. For
Nabi, 2002b; Slovic et al., 2007). Feelings and emotions instance, the word “cause” is often followed by nega-
elicited by language infuse our situation models and tively valenced words such as “death,” “problems,”
guide how we respond to a message. Here we consider “damage,” and “pain.” The word “produce” is nearly
the role of affective valence and discrete emotional synonymous with “cause,” yet it is more likely to be
experiences as mechanisms in this process. followed by positively valenced words such as “results,”
“effects,” and “goods.” As a result, “cause” is said to
Affective valence. We often use our current affective have more negative “semantic prosody” than “produce”
state as a heuristic for decision-making, gravitating toward (D. J. Hauser & Schwarz, 2016). Several studies have
options that evoke positive feelings and leaning away shown that semantic prosody can elicit framing effects
from options that evoke negative feelings (Slovic et al., (for review, see D. J. Hauser & Schwarz, 2023). In one
2007).4 This can help explain a variety of framing effects. experiment, participants read a sentence about a man
For example, several scholars have argued that affective and then evaluated him on several traits (D. J. Hauser
responses play a key role in risky-choice and other types & Schwarz, 2018). Across conditions, the sentences
of gain/loss or “valence” framing (De Martino et al., 2006; were identical but for one word that was either positive
Druckman & McDermott, 2008; Loewenstein et al., 2001; (“totally”) or negative (“utterly”) in semantic prosody:
Nabi et al., 2020; Stark et al., 2017b; Young et al., 2019; “As his siblings discovered, Daniel was a[n] totally
but see H. Cheng et al., 2022). One functional neuroim- [utterly] changed man when he returned.” Those who
aging study found that risky-choice framing effects were read the sentence framed with “utterly” rated Daniel as
associated with increased activity in “emotional” brain less competent and colder than those who read the
regions such as the amygdala, whereas activity in more sentence framed with “totally.”
Psychological Science in the Public Interest 25(3) 123

Discrete emotions. When we communicate about feel- framing effects. Before moving on, we make two final
ings, we typically use discrete emotion labels such as points about these mechanisms and raise an important
“happy,” “sad,” “guilty,” “scared,” and “angry.” The latter related question.
four are all negatively valenced but used to express very First, the mechanisms described earlier are not mutu-
different feelings, which may generate different reactions ally exclusive. A single, seemingly basic framing effect
to a message. This may influence the persuasive power of may result from two or more mechanisms operating
a communicative frame. One meta-analysis found that simultaneously (or separately in different individuals).
fear-based appeals reliably affect attitudes and increase For example, we have reviewed evidence that prag-
behavioral intentions and behaviors because people are matic inference and affective valence both contribute
motivated to avoid sources of fear described in a message to risky-choice framing, and schematic structuring may
(Tannenbaum et al., 2015). Another meta-analysis found play a role as well (see ”Equivalency Framing” section).
that anger-based appeals have a somewhat inconsistent Our broader framework suggests that, as in all forms
persuasive influence that depends on many other factors of human communication, framing arises as people use
such as argument quality (Walter et al., 2019). These con- whatever clues are available to generate a situation
trasting effects may be related to differences in approach- model of the topic at hand. That means they will readily
avoidance motivation associated with fear and anger integrate semantic, social-pragmatic, and affective infor-
(Adams et al., 2006; A. J. Elliot et al., 2013). This is one mation in the course of constructing a mental model of
reason scholars have argued that it is necessary to con- the target issue. From there, they will tend to answer
sider the role of discrete emotions in framing, over and whatever question has been posed to them by drawing
above a focus on affective valence (Lerner & Keltner, on the cognitive and emotional contours of the model.
2000; Nabi, 1999, 2002a, 2002b, 2010). Second, framing research is deeply connected to other
Many studies in the communication literature concur, psychological theories of persuasion and cognitive pro-
showing that different emotions can generate different cessing. This has implications for how we think about
reactions to persuasive messages, even when those the nature of linguistic framing. In the 20th century, the
emotions have the same valence or arousal levels (e.g., scientific study of social influence and persuasion
Dillard & Nabi, 2006; Lecheler et al., 2013; Nabi, 2002a; emerged as a central topic for researchers in social psy-
Yang & Chu, 2018). In this way, discrete emotions func- chology and communication, coinciding with the rise of
tion as cognitive frames—evoked by language—that mass-media technologies (Cialdini, 2001; Crano & Prislin,
structure how people conceptualize an issue (Nabi, 2006; Perloff, 2017; Pratkanis & Aronson, 1991). Like
2003). Once activated, these emotions shape what infor- Aristotle, these scholars asked what makes people more
mation is accessible and guide how people reason or less likely to change their attitudes, beliefs, and behav-
about the content of a message. In one experiment, for ior in response to messages with persuasive appeal.
example, Dutch participants read a news report about Much of this work focused on how people process and
economic investments in Bulgaria and Romania after engage with such information. One highly influential
those countries had joined the European Union theory of persuasion is the elaboration likelihood model
(Lecheler et al., 2013). The article framed this situation (ELM; Petty & Cacioppo, 1986; for alternative models,
either in a positive way, emphasizing the excitement of however, see, e.g., Chaiken, 1987; Kruglanski &
Dutch investors, or in a negative way, emphasizing the Thompson, 1999). The ELM is a “dual-process” model.
investors’ outrage. Participants then rated their support This popular characterization of the human mind posits
for the view that cooperation between the European two general modes of cognitive processing: one fast,
Union, Bulgaria, and Romania would be profitable. automatic, and intuitive and the other slow, reflective,
They also rated their feelings of enthusiasm, content- and deliberate ( J. S. B. Evans, 2008; J. S. B. Evans &
ment, fear, and anger in response to this economic Stanovich, 2013; Kahneman, 2011).
agreement. The framing manipulation had the predicted According to the ELM, both modes of thinking can
effect, with the positive frame eliciting greater support. lead to persuasion. The “central route” to persuasion
However, this effect was mediated by two specific emo- engages the rational, reflective mode (think Aristotle’s
tions: enthusiasm (but not contentment) and anger (but logos). In this case, you would carefully assess the
not fear). This suggests that measuring discrete emo- quality of the persuasive appeal, generate and evaluate
tions can provide more nuanced mechanistic accounts counterarguments, and otherwise “elaborate” on the
than measuring affective valence alone. content of the message. For example, after reading an
article recommending a new vaccine, you might analyze
the risks on the basis of medical data and your personal
Are framing effects irrational?
health situation, consult your doctor for a second opin-
In this section, we have described several mechanisms ion, and then decide to get vaccinated. The “peripheral
that explain the provenance of common linguistic route” to persuasion, on the other hand, engages the
124 Flusberg et al.

intuitive, reflexive mode of thinking (think ethos and McKenzie, 2006). As our discussion of pragmatics illus-
pathos). In this mode, you would rely on heuristics, or trated, the “meaning” of a message goes beyond the
mental shortcuts, to come to a decision, such as your surface-level, literal content of a proposition. Even
immediate emotional reaction to the message. For seemingly equivalent message frames can communicate
example, maybe you decide to get vaccinated because very different—and relevant—information about a tar-
the tone of the article makes you anxious, or because get issue. It is perfectly sensible to use such information
you trust your experienced doctor who sent it to you. in the course of your decision-making, even if you are
The ELM further specifies who is likely to elaborate on not fully aware of all the ways the message frame has
a persuasive message through the central route (hence shaped your reasoning. Moreover, and importantly, it
“elaboration” and “likelihood” in the name of the is not the case that all framing effects result from auto-
model)—namely, those who are sufficiently motivated matic or reflexive processing. Sometimes communica-
and able to interrogate the persuasive appeal. For tive frames are effective precisely because they motivate
example, someone who cares about their health, has the audience to process a message more deeply (e.g.,
mixed feelings about pharmaceutical interventions, has Ottati et al., 1999). Ultimately, it is overly simplistic to
some spare time, and enjoys analyzing scientific data view framing as a form of subliminal manipulation that
would be especially likely to consciously reflect on the exposes irrational flaws in human reasoning. Rather,
article recommending the new vaccine. framing is a complex but natural consequence of the
At first glance, all the framing effects we have way people communicate.
described—and the mechanisms that explain them—
seem to fall under the umbrella of “fast and automatic”
processing. That is, most framing manipulations appear
A Taxonomy of Framing Devices
to operate via the peripheral route to persuasion. Their So far, we have covered a wide range of effects that
effects seem almost subliminal. This is one reason they result from linguistic framing and multiple mechanisms
are so interesting and attention-grabbing, raising the that underlie these effects. In this section, we turn to
specter of Orwellian social control. Although people the communicative frames themselves: How can we alter
can engage in deliberate reasoning about, say, the prag- the structure or content of a message to induce a fram-
matic implications of a message (“Hmmm, I wonder ing effect? This question gets at the heart of applying
what she meant by saying her date was ‘nice.’ Maybe framing to issues in the public interest. We have already
she doesn’t really like him . . .”), most of the experi- introduced many of these framing devices, such as meta-
ments we have described do not encourage such elabo- phor framing and risky-choice framing, but here we
ration. Participants are usually expected to work organize these subtypes into a broader taxonomy and
through a problem quickly, without much reflection, highlight some of the prominent findings for each one.
so any impact of a message frame is assumed to be This list is admittedly partial and idiosyncratic
largely automatic. When researchers require people to because no formal classification system for framing
consciously articulate their reasoning before making a devices currently exists. Researchers interested in fram-
decision, framing effects are often reduced or elimi- ing tend to coin their own terms, only some of which
nated (e.g., F. F. Cheng et al., 2014; Hodgkinson et al., take hold and are picked up by other scholars.
1999; Sieck & Yates, 1997; Takemura, 1994; but see Igou Therefore, it is challenging to develop a coherent tax-
& Bless, 2007; LeBoeuf & Shafir, 2003). onomy for different types of communicative frames.
The dominance of peripheral-route explanations in One popular approach distinguishes between two
the framing literature raises an important question: Are broad categories of framing: equivalency framing and
framing effects irrational? Consider that people often emphasis framing (Bullock & Shulman, 2020; Druckman,
respond quite differently when presented with subtly 2001). Equivalency framing is when the same informa-
different communicative frames, even when the “logi- tion is presented in two different forms that are “logi-
cal” content of the frames is identical (e.g., in terms of cally equivalent”—having the same literal meaning (or
numbers of lives saved). This has led many researchers “truth value,” as philosophers put it). We have consid-
to conclude that such effects reveal the irrationality of ered several kinds of equivalency framing already,
human judgment, which deviates from idealized models including risky-choice framing and attribute framing
of rational choice (Ariely, 2008; Kahneman, 2011). (e.g., describing beef as 25% fat vs. 75% lean). Framing
However, we side with critics who have argued that that uses subject-complement syntax (e.g., “girls are just
it is rational to respond differently when the same situ- as skilled as boys” vs. “boys are just as skilled as girls”)
ation is framed differently (Bermúdez, 2020; Flusberg, is another example. This is a useful category, and we
Thibodeau, & Holmes, 2022; Gigerenzer, 2018; McKenzie address the many subtypes of equivalency framing in
& Nelson, 2003; Pinker, 2007; Sher et al., 2022; Sher & the next section.
Psychological Science in the Public Interest 25(3) 125

Fig. 1. A humorous illustration of emphasis framing. The six headlines at the top are reworked to emphasize a pro-dog
or pro-raccoon perspective. Any similarities to our current partisan media landscape are likely intentional. Reprinted with
permission from Wait But Why (Urban, 2023).

Emphasis framing, on the other hand, is when the into two categories does little to illuminate meaningful
communicator emphasizes one logically distinct inter- differences. In addition, there are types of linguistic fram-
pretation over another, deliberately or otherwise. ing that do not fit neatly into the equivalency/emphasis
Consider the differences between the pro-dog and pro- binary. Some of the framing effects we have reviewed,
raccoon headlines at the bottom of Figure 1, which such as the difference between “what happened” and
promote a partisan interpretation of the “neutral” head- “what’s happened,” are so subtle that a social scientist
lines at the top. We have encountered many examples might balk at placing them in the same category as the
of emphasis framing already, including the use of dif- contrast between “free speech” and “public order.”
ferent metaphors (e.g., “battle” vs. “journey”), labels Thus, we take a different approach to classifying lin-
(e.g., “illegal aliens” vs. “noncitizens”), and values (e.g., guistic framing techniques. We focus on those aspects of
“free speech” vs. “public order”). Scholars sometimes linguistic form or content that are manipulated in the
delineate different subtypes of emphasis framing in frame (for a summary of our taxonomy, see Table 3; for
discussions of how politicians and the news media alternative typologies, see Hallahan, 1999; O’Keefe,
frame important issues. For example, episodic frames 2017b). This approach dovetails with our contention that
focus on individual people and events, whereas the- framing should be viewed through the lens of human
matic frames zoom out to consider broader historical communication, and it provides a more practical organi-
and contextual factors (Iyengar, 1991; see also our dis- zational schema for these purposes. Please note, how-
cussion of psychological-distance framing below). ever, that the framing devices we describe below comprise
Although useful, the distinction between equivalency fuzzy categories and subcategories that are not mutually
and emphasis framing is somewhat limiting. For one exclusive. For example, equivalency frames and gram-
thing, it is too broad. There are dozens of framing manip- matical frames can both evoke different degrees of psy-
ulations discussed in the literature, and lumping them all chological distance, and some metaphors might be
126 Flusberg et al.

Table 3. Taxonomy of Linguistic Framing Techniques

Type of framing Definition Example of framing effect

Equivalency When information is presented in different
forms that are logically equivalent/have the
same literal meaning; can involve a contrast
between positive/negative frames (valence
framing) or more neutral frames (nonvalenced
framing); traditionally distinguished from
emphasis framing, in which a communicator
emphasizes one logically distinct
interpretation over another
Valence
  Risky choice Describing the response options in an In response to a scenario describing an outbreak
uncertainty-related problem using positive of a new disease expected to kill 600 people,
(e.g. “saved”) or negative (e.g., “killed”) participants were more likely to select a sure
language (as opposed to risky) health program when
it was framed positively (“200 people will be
saved”) than when it was framed negatively
(“400 people will die”; Tversky & Kahneman,
1981).
  Attribute Describing an object or event in terms of a Participants assigned greater fines to a company
positive or desirable attribute (e.g., “85% whose deceptive advertising practices were
survival rate”) or a negative or undesirable described using a negative frame (80% chance
attribute (e.g., “15% mortality rate”) they knew the ad was deceptive) than a
positive frame (20% chance they did not know
the ad was deceptive; Dunegan, 1996).
  Goal Describing the consequences of an action (or College women expressed greater intentions to
inaction) in terms of positive or negative engage in breast self-examination (BSE) after
outcomes reading a loss-framed pamphlet (“women who
do not do BSE have a decreased chance of
finding a tumor . . .”) than after reading a gain-
framed pamphlet (“women who do BSE have
an increased chance of finding a tumor . . .”;
Meyerowitz & Chaiken, 1987).
  Comparison Comparing items in terms of one being “better” Participants rated two items (e.g., sodas) as lower
than the other (positive frame) versus one quality when they were exposed to the negative
being “worse” than the other (negative frame) “worse” frame compared with the positive
“better” frame. The perceived difference
between the items was also larger following the
negative frame (Inbar & Evers, 2022).
Nonvalenced
  Subject-complementa Describing an equivalence between groups Participants who read that “Christians are just
using subject-complement syntax (e.g., “Xs as likely as Muslims to commit terrorist acts”
are just as good as Ys” places Xs in the rated Muslims as more likely to be terrorists
subject position and Ys in the complement compared with when the positions of the
position); also an example of grammatical groups were reversed (but only if they did not
framing cite this statement as figuring prominently in
their evaluations; Holmes et al., 2022).
  Unit Describing a quantitative value using different Four times as many people enrolled in a savings
units; often involves presenting a smaller program when deposits were framed in terms
annuity (e.g., “$1 per day”) as opposed to of smaller daily amounts than larger monthly
a larger aggregate amount (e.g., “$365 per amounts (Hershfield et al., 2020).
year”)
  Order Changing the order of a set of items in a list When presented with a set of traits about another
(e.g., “A, B, C” vs. “C, B, A”); the first item person, participants who heard a positive trait
often anchors people’s evaluation of the set first endorsed more positive descriptors of the
person than those who heard a negative trait
first (J. Sullivan, 2019).

(continued)
Psychological Science in the Public Interest 25(3) 127

Table 3. (continued)

Type of framing Definition Example of framing effect

Figurative Use of figurative language—including
hyperbole, irony, or metaphor—to frame
discussion of an issue
Metaphora Describing a target issue metaphorically, often Participants were more likely to endorse
in terms of a concrete or experiential source biological causes of addiction and depression
domain; also an example of lexical framing, when these conditions were framed as
especially in the case of nominal metaphors infectious “brain diseases” than as “demons”
(e.g., “Police officers are warriors” vs. “Police preying on people (Flusberg et al., 2023).
officers are guardians”)

Psychological distance Describing the same issue at different levels Participants had more favorable views of
of abstraction that evoke different degrees a fictional senate candidate when the
of psychological distance—the sense of psychological distance of their campaign
separation between us and what is being matched the level of abstraction in their
described; may be especially effective when campaign statement: When the campaign was
there is a match between the abstractness of set to begin in a week, concrete language led
a message and the psychological distance of to greater favorability; when it was set to begin
the events being described in 6 months, abstract language led to greater
favorability (Kim et al., 2009).
Narrative Describing an issue using a personal story, Chinese university students expressed greater
often from the perspective of an individual, in intentions to get vaccinated against COVID-19
contrast to a more abstract, statistical, or facts- after reading a narrative account of the benefits
based account of vaccination compared with a nonnarrative
account (Ye et al., 2021).
Pronoun Describing a situation using different pronouns Participants who used “you” statements to
that communicate different levels of psych themselves up (e.g., “You can do it!”)
psychological distance performed better on a subsequent anagram task
than those who used psychologically closer
“I” statements (e.g., “I can do it!”; Dolcos &
Albarracín, 2014).

Lexical Describing a situation using different labels or Participants evaluated meat more negatively—and
descriptions that have different connotations consumed less of it—when the animal was
described as having been raised on a “factory
farm” as opposed to a “humane family farm”
(Anderson & Barrett, 2016).
Moral reframing Describing a situation by invoking moral values Liberals expressed greater support for immigration
relevant to the target audience (especially for when an appeal was framed in terms of
issues the audience may not be inclined to “compassion” compared with “patriotism,”
support) whereas conservatives showed the opposite
pattern (Nath et al., 2022).
Victim Common rhetorical technique that involves Participants who read a report about sexual-
labeling the alleged perpetrator of a crime assault allegations that framed the alleged
as the “real” victim in an attempt to mitigate assailant as a victim (of false accusations)
blame and punishment expressed more support for him and less
support for his accuser compared with a report
that framed the accuser or neither character as
the victim (Flusberg, van der Vord, et al., 2022).

Grammatical Describing a situation by manipulating the

grammatical form or structure of a message
rather than the content
Agentive Describing an action by including the causal Participants attributed more blame and financial
agent (“Carlos opened the umbrella”; agentive liability to a person who accidentally started a
frame) or omitting the causal agent (“the fire at a restaurant when the report was framed
umbrella opened”; nonagentive frame) agentively (“she flopped the napkin. . .”) than
nonagentively (“the napkin flopped. . .”; Fausey
& Boroditsky, 2010).
(continued)
128 Flusberg et al.

Table 3. (continued)

Type of framing Definition Example of framing effect

Aspectual Using grammatical markers to emphasize either Participants evaluated a political candidate more
the completion of an event (“swerved”; negatively when prior negative actions were
perfective aspect) or the ongoing nature framed imperfectively (“was having an affair”)
of the event (“was swerving”; imperfective than perfectively (“had an affair”; Fausey &
aspect); imperfective aspect seems to elicit a Matlock, 2011).
more dynamic situation model
Generic Describing an attribute using generic language Generic statements about Democrats or
(e.g. “Dogs are great swimmers”), which Republicans (e.g., “Democrats support House
promotes more essentialist reasoning than Bill 858”) compared with quantified statements
nongeneric language (e.g., “That dog is a (e.g., “Many Democrats support House Bill
great swimmer”) 858”) led participants to infer that the two
parties held more polarized views (Novoa et al.,
2023).
Dynamic norma Describing normative behavior in terms of Participants were more willing to reduce meat
change over time as opposed to more stable consumption when presented with a dynamic
information; often involves both grammatical norm message (“in the last 5 years, 30% of
and lexical changes to a message Americans have now started to make an effort
to limit their meat consumption. . .”) than a
static norm message (“30% of Americans make
an effort to limit their meat consumption. . .”;
Sparkman & Walton, 2017).

Note: This list is not exhaustive, and the categories that are provided are fuzzy. aMay belong in multiple categories.

classified as lexical frames. This list is therefore not meant “75% lean” or “85% survival rate”) or a negative or unde-
to be exhaustive or final; there are likely many variants sirable attribute (e.g., “25% fat” or “15% mortality rate”).
of linguistic framing that we have inadvertently left out. Attribute framing affects the evaluation of a target item
We encourage you to read on with these caveats in mind. as indexed by some measure of attitudes (e.g., product
ratings) or behavior (e.g., willingness to purchase a prod-
uct). In general, people show more favorable attitudes
Equivalency framing
toward items and situations that are framed positively.
Valence framing. As we discussed, equivalency fram- Goal framing is when the consequences of an action
ing involves presenting people with seemingly equiva- (or inaction) are framed in terms of positive or negative
lent information in different ways. This often entails outcomes. This type of framing is often used to influ-
contrasting a positive (“gain”) frame and a negative ence people’s behavior or behavioral intentions in a
(“loss”) frame, leading some scholars to prefer the term positive way—one reason it is prevalent in health and
valence framing. Levin and colleagues (1998) distin- environmental communication. It is slightly more com-
guished between three types of valence framing based plicated than other forms of valence framing, however.
on differences in what is framed, what is affected, and A gain frame might describe how engaging in the action
how this is measured. will lead to a desirable outcome (“if you use sunscreen,
Risky-choice framing should be familiar by now. you will be protected from the sun’s harmful rays”) or
Here, response options in a problem-solving task with help you avoid an undesirable outcome (“if you use
different levels of associated risk or uncertainty are sunscreen, you will lower your risk of getting cancer”).
framed in terms of gains or losses. This influences risk Similarly, a loss frame might describe how not doing
preferences and decision-making, as indexed by which the action will lead you to miss out on the desirable
option participants select to solve the problem. This outcome (“if you don’t use sunscreen, you won’t be
type of valence framing originated with Tversky and protected from the sun’s harmful rays”) or experience
Kahneman’s (1981) unusual disease study, and has been the undesirable outcome (“if you don’t use sunscreen,
a major topic of interest ever since, especially in psy- you will increase your risk of getting cancer”). This
chology and behavioral economics. differs from risky-choice and attribute framing because
Attribute framing is when an object or situation is the target action is framed as something “good” in all
framed in terms of a positive or desirable attribute (e.g., circumstances.
Psychological Science in the Public Interest 25(3) 129

In a seminal study of goal framing, college women (Freling et al., 2014; McDonald et al., 2021; Piñon &
were presented with a short pamphlet on breast self- Gambara, 2005; Steiger & Kühberger, 2018). However,
examination (BSE) that included either the following there are differences across different subtypes of
underlined (gain frame) or bracketed (loss frame) infor- valence framing. One recent meta-analysis in the
mation (Meyerowitz & Chaiken, 1987): domain of moral reasoning found an overall medium-
sized effect (Cohen’s d = 0.50; d = 0.22 when correcting
By doing [not doing] BSE now, you can [will not] for publication bias; McDonald et al., 2021). An analysis
learn what your normal, healthy breasts feel like of studies published between 1997 and 2003 found
so that you will be better prepared [ill prepared] reliable, moderate-to-medium effect sizes for risky-
to notice any small, abnormal changes that might choice framing (d = 0.44), attribute framing (d = 0.26),
occur as you get older. Research shows that and goal framing (d = 0.44; Piñon & Gambara, 2005).
women who do [do not do] BSE have an increased Recent meta-analyses have found similar effect sizes for
[a decreased] chance of finding a tumor in the risky-choice framing (d = 0.52; Steiger & Kühberger,
early, more treatable stage of the disease. . . . You 2018) and attribute framing (Pearson’s r = .25 for atti-
can gain [lose] several potential health benefits by tude outcome measures and r = .21 for behavioral out-
spending [failing to spend] only 5 minutes each come measures; Freling et al., 2014).
month doing BSE. Take [Don’t fail to take] advan- However, several meta-analyses by O’Keefe and col-
tage of this opportunity. leagues suggest that the impact of goal framing in
health communication is smaller and less consistent
The results revealed that the loss frame was more (O’Keefe & Jensen, 2006, 2007, 2009, 2015; O’Keefe &
impactful, as women in this condition reported greater Wu, 2012). This is not surprising. Goal framing often
intentions to engage in BSE than women who received targets people’s real-world health behaviors, which are
the gain-framed pamphlet. A 4-month follow-up showed complicated and influenced by many other factors. For
that the loss-frame participants had followed through messages encouraging disease detection behaviors,
on their intentions, performing BSE more often than loss-framed appeals were found to be more persuasive
the other participants in the intervening months. than gain-framed appeals (O’Keefe & Jensen, 2009).
However, loss frames are not necessarily more impact- This effect was very small overall (r = −.04), and only
ful than gain frames in studies involving other kinds of statistically significant for breast cancer detection
target actions (O’Keefe & Jensen, 2007, 2009, 2015; behaviors (r = −.06). Loss framing had no measurable
O’Keefe & Wu, 2012). impact on detection behaviors for other conditions such
A fourth type of valence framing, comparison fram- as skin cancer or dental problems. For messages
ing, was recently investigated by Inbar and Evers (2022). encouraging disease prevention behaviors, however,
When comparing two items (e.g., two toasters, teams, or gain-framed appeals were slightly more persuasive than
sodas), you can describe Item A as better than Item B or loss-framed appeals (O’Keefe & Jensen, 2007). Again,
Item B as worse than Item A. Across multiple studies, this effect was small (r = .03) and driven entirely by
participants rated both items lower in quality when they messages encouraging dental hygiene (r = .15). Other
had been exposed to the negative “worse” frame com- meta-analyses support this slight gain-frame advantage
pared with the positive “better” frame. The perceived for promoting prevention behaviors (e.g., Gallagher &
difference between A and B was also larger following Updegraff, 2012). For example, gain-framed appeals
the negative frame. Inbar and Evers suggest that this is were more effective than loss-framed appeals for
because of linguistic “markedness.” For any pair of ant- encouraging physical activity to address obesity (r =
onyms, one word will be more dominant or “unmarked” .17; O’Keefe & Jensen, 2015). However, goal framing
(e.g., “better,” “longer,” “heavier”), whereas the other will appears to have no impact on healthy eating (O’Keefe
be less dominant or “marked” (e.g., “worse,” “shorter,” & Jensen, 2015) or skin-cancer prevention behaviors
“lighter”). Unmarked terms are generally easier to pro- (O’Keefe & Wu, 2012).
cess and remember, and they communicate a more neu- As we discussed in the “Mechanics of Communication
tral position on the underlying scale. Marked terms, on and Linguistic Framing” section, multiple mechanisms
the other hand, require more effort to understand, and contribute to valence framing effects, including prag-
they communicate a specific anchoring on one end of matic inferences and emotional reactions associated
the scale (often the negative end; see also Clark, 1969; with the message frames (De Martino et al., 2006; Inbar
Harris, 1973; Lipscomb, Bregman, & McAllister, 1985; & Evers, 2022; Leong et al., 2017; Mandel, 2014; Nabi
Lipscomb, McAllister, & Bregman, 1985). et al., 2020; Stark et al., 2017; see also Kreiner &
The results of several meta-analyses suggest that Gamliel, 2018). Researchers have also devoted consider-
valence framing reliably yields moderately sized effects able attention to the cognitive mechanisms that underlie
130 Flusberg et al.

risky-choice framing, although a full treatment of the Order framing concerns how the order in which a
various models is beyond the scope of this article (for set of items is presented can shape the way people
a recent attempt to compare formal models, see respond (e.g., Mantonakis et al., 2009; J. Sullivan, 2019).
Huizenga et al., 2023). Much of this work describes the For example, in a classic study on impression forma-
mental representations people construct in response to tion, Solomon Asch (1946) discovered a primacy effect:
different frames and the internal (unconscious) calcula- Participants tended to form a more favorable impres-
tions they run before making a choice. sion of someone described as “compassionate, stub-
born, and envious” than “envious, stubborn, and
Nonvalenced equivalency framing. There are sev- compassionate.” A recent large-scale replication of this
eral other types of equivalency framing effects that do work found that hearing a positive trait first leads peo-
not involve contrasting explicitly positively versus nega- ple to endorse about 7% more positive descriptors of
tively valenced message frames. a person than hearing a negative trait first ( J. Sullivan,
Subject-complement framing is when this particular 2019). Interestingly, this effect does not seem to be
grammatical construction is used to express that two explained by the pragmatic inference that the most
groups are equivalent on some trait, as we have dis- informative trait was intentionally mentioned first:
cussed previously (e.g., “girls are just as skilled as Participants still showed a primacy effect when they
boys”; see “Beliefs” section; Chestnut & Markman, 2018; were told that a computer randomly put the list of traits
Holmes et al., 2022). People tend to surmise that the together. This suggests that schematic structuring may
group in the complement position is superior, setting be the main contributing factor, with the first trait scaf-
the standard for the group in the subject position. This folding the situation model of the person being
judgment appears to be driven by the pragmatic impli- described. Pollsters and survey makers are well aware
cations of the syntactic construction (Holmes, Wu, of order framing because it affects how questions and
et al., 2024; Wu et al., 2021). In comparative statements response options are worded and organized
more generally, the group framed as the standard (e.g., (Kellermann, 2007; Schuman & Presser, 1996).
“dogs” in “cats use different hunting methods than
dogs”) is viewed as having more agency and power
than the other group (Bruckmüller & Abele, 2010;
Figurative framing
Bruckmüller et al., 2012). People often express themselves figuratively rather than
Unit framing—a term we just coined—refers to the literally. This includes using hyperbole (“Everyone
act of describing a quantitative value using different knows this is the worst policy in American history”),
units. This typically involves presenting a smaller annu- irony (“Yeah, this policy is amazing, sure”), and meta-
ity as opposed to a larger aggregate amount. For exam- phor (“This policy is toxic trash”). Many scholars have
ple, an animal shelter might advertise that you can save explored the framing effects of figurative language, ask-
a sick puppy for 35¢ per day rather than $127.75 per ing when, how, and why it affects reasoning and per-
year, a “pennies-a-day” strategy that Gourville (1998) suasion (e.g., Boeynaems et al., 2021; Brugman et al.,
called “temporal reframing.” This can impact people’s 2019; Burgers et al., 2016, 2018; Flusberg et al., 2018;
decision-making and behavior, which is why it is Lakoff, 1996, 2008, 2014; Lakoff & Johnson, 1980;
favored by marketers. For example, a recent field study Sopory & Dillard, 2002; Thibodeau et al., 2019). Most
found that four times as many people used a new of the related empirical research has focused on meta-
finance app to enroll in a savings program when depos- phor framing (but see Boeynaems et al., 2021; Burgers
its were framed in terms of smaller daily amounts than et al., 2015, 2016, 2018).
larger monthly amounts (Hershfield et al., 2020). People Metaphors play an essential role in language and
struggle to construct quantitatively precise mental mod- thought and are pervasive in everyday conversations,
els on the fly, so the smaller daily amount may simply written texts, and public discourse (Gibbs, 2017; Holmes
feel less costly. This effect depends on the amount of et al., 2018; Lakoff & Johnson, 1980; Thibodeau et al.,
money we’re talking about, however. Another study 2019; Thibodeau, Hendricks, & Boroditsky, 2017).
examined the extent to which people perceive that a Metaphors enable us to communicate about complex,
certain amount of wealth is adequate for retirement abstract, and unfamiliar “target domains” (e.g., a new
(Goldstein et al., 2016). At lower amounts (e.g., governmental policy, cancer) by borrowing language
$100,000), a lump sum was rated as more adequate than from a more familiar or concrete “source domain” (e.g.,
a comparable monthly annuity, but at higher amounts trash, battles and journeys). Many studies have demon-
(e.g., $2,000,000) this effect was reversed. More research strated the effects of metaphor framing on attitudes,
is needed on this type of framing to assess the reliability beliefs, and reasoning about everything from crime and
of this technique and fully unpack all of the underlying law enforcement (Christmann & Göhring, 2016; Thibodeau
mechanisms at play. & Boroditsky, 2011, 2013; Thibodeau, Crow, & Flusberg,
Psychological Science in the Public Interest 25(3) 131

2017), to climate change and genius ideas (Elmore & mentioned earlier, metaphors tend to be more effective
Luna-Lucero, 2017; Flusberg et al., 2017; Flusberg & when they appear early in a message (Sopory & Dillard,
Thibodeau, 2023), to immigration and the federal budget 2002; Thibodeau & Boroditsky, 2011). That way, they
(Brown et al., 2019; Thibodeau & Flusberg, 2017), to can scaffold how people interpret and integrate the rest
cancer and mental illness (Flusberg et al., 2023; D. J. of the message into their situation model. If a metaphor
Hauser & Schwarz, 2020; Hendricks et al., 2018; Magaña appears toward the end of a message, people may have
& Matlock, 2018; Schroder et al., 2023). We have dis- already formed an opinion on the target issue, and
cussed some of these examples already. there may be little opportunity for the metaphor to
Several meta-analyses indicate that metaphors are structure their thinking.
(slightly) more persuasive than comparable literal lan- Second, metaphors are more impactful when they
guage. For example, Sopory and Dillard (2002) found are extended throughout a message and into the
a small but significant overall effect size across a range response options in a decision-making scenario
of studies of metaphor framing (r = .07). This effect was (Flusberg et al., 2020; Keefer et al., 2014; Thibodeau
substantially larger in studies that used a single, novel et al., 2016). Extended metaphors reinforce the struc-
metaphor that appeared early in a message to describe ture of a situation model and help people map the
a familiar target domain (r = .42). A more recent meta- metaphor’s “entailments”—the inferences licensed by
analysis of studies conducted between 2001 and 2015 the metaphor—onto their choices in the task. For exam-
found a similarly small but reliable metaphor framing ple, we asked earlier whether it would make any dif-
effect (r = .09; Van Stee, 2018). A third meta-analysis ference to describe political polarization using “war”
on metaphor framing in the context of political dis- versus “theater” metaphors. Studies suggest that the
course also found a small but reliable overall effect (d = answer is “yes,” and that extending the metaphor
0.11; Brugman et al., 2019). This effect was larger for enhances the effect (Flusberg et al., 2020; Thibodeau
measures of belief (d = 0.29) than attitudes (d = 0.10). et al., 2016). This example comes from an extension of
Whereas many studies contrast metaphorically versus the study discussed in the “Reasoning, Judgment, and
nonmetaphorically framed messages, other studies Decision-Making” section in which participants
compare the effects of different metaphors in discus- responded to multiple metaphorically framed reasoning
sions of a single target issue (e.g., cancer as a “battle” problems (Thibodeau et al., 2016). In the follow-up
vs. “journey,” crime as a “beast” vs. “virus,” politics as study (Flusberg et al., 2020), participants completed
“war” vs. “theater,” mental illness as “an infectious dis- four different metaphor framing trials, including one
ease” vs. “a demon”). Together, this research has pro- that described legislative stagnation using either a the-
vided valuable insights into the cognitive, pragmatic, ater or war metaphor:
and emotional mechanisms that drive metaphor framing
effects (Lakoff, 2008, 2014; Thibodeau & Flusberg, 2022; The Democrats and Republicans have been [play-
Thibodeau, Hendricks, & Boroditsky, 2017; Thibodeau ing political theater/fighting a battle] with each
et al., 2019). As discussed in the “Schematic Structuring” other in which both parties seem more interested
section, metaphors encourage people to map their in [performing dramatic monologues/attacking
schematic knowledge of the source domain (e.g., bat- their opponent] than engaging with difficult policy
tles) onto the target domain (e.g., cancer). This provides questions. Congress has passed roughly 80% fewer
an organizational structure—a semantic or cognitive bills in recent terms than it did in the ’70s and ’80s.
frame—for what might otherwise be a vague or ambigu-
ous mental model. By leveraging knowledge of the Participants were then asked: “Which of the follow-
source domain to construct an understanding of the ing do you think would be more likely to change the
target domain, metaphors are especially useful for guid- culture in Washington?” They had to select between two
ing reasoning about complex, abstract issues such as response options, each conceptually congruent with
cancer, crime, and climate change, which do not evoke one of the metaphors. In the first version of the experi-
well-delineated cognitive frames on their own. Thus, ment, the response options extended the metaphor
metaphors are most effective when (a) they involve from the vignette: (a) “Close the curtain on the saga by
source domains that are familiar, accessible, and have ending the 24-hour media coverage of politicians” (con-
a well-defined schematic structure and (b) the source gruent with theater) or (b) “Bring a truce to the war by
domain structure can be clearly mapped onto the target forcing politicians to acknowledge their common obli-
domain (i.e., the metaphor is “apt”; Thibodeau, gations” (congruent with battle).
Hendricks, & Boroditsky, 2017; Thibodeau et al., 2019). In the second version of the experiment, the response
Several other findings support the idea that meta- options did not extend the metaphor, although they
phors stimulate schematic (re)structuring. First, as we were still conceptually congruent with each metaphor
132 Flusberg et al.

according to a norming study (Thibodeau, 2016): (a) whatever it is we are considering at the moment
“End the 24-hour media coverage of politicians” (con- (Liberman et al., 2007). More abstract “construals” feel
gruent with theater) or (b) “Force politicians to acknowl- more psychologically distant, whereas more concrete
edge their common obligations” (congruent with construals feel psychologically closer and evoke more
battle). of an emotional response. This has downstream con-
In both experiments, participants selected the met- sequences for reasoning and decision-making (S. J. Lee,
aphor-congruent response option at above-chance lev- 2019; Trope & Liberman, 2012).
els. However, this metaphor framing effect was For example, people may feel more comfortable and
significantly larger when the metaphor was extended motivated to engage with an issue that is framed more
(13.4% advantage for the metaphor-congruent option) concretely, and thus psychologically closer. In one
than when it was not (7.8% advantage). study, graduate-level business students indicated their
The emotional and pragmatic effects of metaphor willingness to invest in a new company that was about
are also important. The emotional dimension has to go public (W. B. Elliott et al., 2015). Participants were
received limited attention (Holyoak & Stamenković, more willing to invest when the prospectus they read
2018), but research suggests metaphors tend to be more highlighted concrete (as opposed to abstract) informa-
emotionally engaging than comparable literal language tion about the company. On the other hand, abstract
(Citron & Goldberg, 2014; Mon et al., 2021). It is these language may be effective for communicating with a
affective associations that drive the effects of some larger and more diverse audience, where people may
metaphors (e.g., war metaphors promote a sense of vary in their concrete circumstances but share abstract
urgency; Flusberg et al., 2017, 2018). But emotionally ideals ( Joshi & Wakslak, 2014). This is one reason why
charged metaphors can also backfire. One study found political slogans are often pitched at a high level of
that a far-right, anti-immigration message framed using abstraction (“Make America Great Again,” “Yes We Can”).
violent and dehumanizing metaphors (e.g., “a plague Some research suggests that messages are especially
of immigrants swarming into the country”) increased effective when there is a match between the abstractness
pro-immigration attitudes (Hart, 2021). of the message and the psychological distance of the
Metaphors can also influence persuasion more indi- events being described. In one experiment, participants
rectly by impacting speaker judgments, although here read a statement from a fictional candidate running for
the evidence is mixed (Ottati & Renstrom, 2010; S. J. the U.S. Senate (H. Kim et al., 2009). The statement was
Read et al., 1990). For example, Ottati and Renstrom framed at either an abstract level (“a refocus on why we
(2010) reviewed some studies suggesting that metaphors do things”) or a more concrete level (“a refocus on how
enhance persuasion by eliciting more positive speaker we do things”). The researchers further manipulated
impressions, perhaps because metaphor use is inter- whether the campaign was set to begin next week (psy-
preted as a sign of intelligence. Yet they also reviewed chologically close) or in 6 months (psychologically dis-
other studies suggesting that metaphors sometimes make tant). Participants had more favorable views of the
the speaker seem less credible. They conclude that cer- candidate when the psychological distance and level of
tain properties of the metaphor may moderate these abstraction matched: When the campaign was set to
effects, such as the relevance or aptness of the metaphor begin in a week, concrete language led to greater favor-
in context. Additional research is needed to tease apart ability, but when it was set to begin in 6 months, abstract
the pragmatic functions of metaphors in framing. language led to greater favorability.
Several specific framing techniques involve manipu-
lating psychological distance:
Psychological distance framing
Narrative framing describes the use of concrete sto-
The same event, issue, or action can be framed at dif- ries about individual characters and their motivations.
ferent levels of abstraction. For example, you can talk Evidence suggests that narratives are (slightly) more
about a pandemic at a high level of abstraction (a persuasive than other messages about the same issue.
“thematic” frame) by discussing policies and statistics, One meta-analysis found that narrative messages sig-
such as the number of people getting ill each day or nificantly predicted narrative-consistent beliefs (r = .17),
the financial losses associated with a lockdown. attitudes (r = .19), intentions (r = .17), and behaviors
Alternatively, you can frame a pandemic in very con- (r = .23; Braddock & Dillard, 2016; see also Oschatz &
crete terms by using a personal narrative (an “episodic” Marker, 2020). A meta-analysis of the health-communi-
frame), describing the experiences of a single individual cation literature found that narratives were more per-
who became sick or lost their job. Abstraction is associ- suasive than nonnarratives overall (r = .06; Shen et al.,
ated with psychological distance—the sense of separa- 2015; see also De Graaf et al., 2016; Xu, 2023). This
tion in time, space, or experience between us and was only statistically significant for messages
Psychological Science in the Public Interest 25(3) 133

advocating disease detection and prevention behaviors, and I” or “we” to refer to the two friends (Fitzsimons &
however, and not for messages advocating cessation Kay, 2004). The friendship was perceived as closer and
behaviors. higher quality—and the friends as forming a more
Narratives are effective for several reasons (for coherent unit—when described with “we.” A similar
review, see Walsh et al., 2022). For one, they are more effect was observed when participants were asked to
emotionally engaging than nonnarratives, and they help vary their pronoun use while reflecting on their own
people connect with the experiences of others (Alam friendships.
& So, 2020; Nabi & Green, 2015). As a result, narratives The pronoun “you” may be especially effective as a
can reduce audience resistance to a message (Moyer- framing device for making sense of the social world.
Gusé & Nabi, 2010; Ratcliff & Sun, 2020). A sense of Although most often used when addressing a specific
psychological closeness to the central characters is individual (“How are you doing?”), “you” can also be
important in narrative framing. Health narratives have used to reframe personal challenges as shared, univer-
been found to be more effective when delivered from sal experiences (“You have to recognize you can’t
a first-person point of view than from a third-person change people”; Orvell et al., 2017). This use of “you”—
point of view, especially when the protagonist is known as generic-you—seems to foster a sense of con-
depicted as similar to the audience (M. Chen & Bell, nection with the ideas expressed. For example, in one
2022; De Graaf et al., 2016). Relatedly, narratives can study, participants indicated that generic-you state-
reduce stigma toward different social groups, especially ments resonated more strongly with them than other-
when they are delivered from the first-person perspec- wise identical “I” statements (e.g., “Sometimes, [you/I]
tive (Zhuang & Guidry, 2022). Finally, narratives are have to take a step back before [you/I] can take a step
generally easier to understand and processed more flu- forward”; Orvell et al., 2020). People may regard such
ently than nonnarratives, which has been shown to statements as insightful because they perceive the
increase their persuasive power (Bullock et al., 2021). speaker as expressing a healthy degree of psychological
However, narratives are not always more effective distance from the events that inspired them (Orvell
than nonnarratives. As we mentioned before, message et al., 2017).
frames are often most persuasive when the level of Another function of “you” is to address ourselves
psychological distance they evoke matches the target from an outsider’s perspective. For example, right
issue. One recent meta-analysis compared the persua- before you deliver a big presentation, you might say to
sive effects of anecdotal versus statistical information yourself: “You can do this!” Unlike most of the examples
in different communicative contexts (Freling et al., of framing we have reviewed so far, this use of “you”
2020). Concrete personal anecdotes were generally is intended to regulate our own thoughts and feelings
found to be more persuasive than abstract statistics in rather than influence others (cf. Bermúdez, 2020).
situations involving high emotional engagement Several studies have shown that using “you” statements
(Hedges’s g = −0.06). This includes health issues and of this sort in the midst of a stressful situation increases
other personally-relevant threats, which are psychologi- psychological distance, reducing negative emotions and
cally close. On the other hand, statistical information enhancing task performance (Dolcos & Albarracín,
was found to be more persuasive in situations involving 2014; Kross et al., 2014, 2017). Other linguistic shifts in
low emotional engagement, like non-health issues response to aversive stimuli (e.g., avoiding present-
affecting other people, which are more psychologically tense verbs, using longer words) also support effective
distant (g = −0.16). Another meta-analysis found that in emotion regulation (Holmes, Kassin, et al., 2024; Kassin
health communication, narratives have a greater impact et al., 2023; Nook et al., 2017, 2020). Like “you,” these
than statistics on behavioral intentions (d = 0.10), but words may help us process stressful situations by fram-
statistics have a greater impact on beliefs (d = −0.16) ing them (to ourselves) from a more abstract, distanced
and attitudes (d = −0.11; Zebregs et al., 2015). perspective.
Pronoun framing is when we use words such as “I,”
“we,” and “you” to communicate psychological distance
Lexical framing
in the social domain (Nook et al., 2017; Orvell et al.,
2022). Unlike the first-person singular “I,” the plural One of the most basic framing techniques involves
“we” and second-person “you” signal that the speaker’s manipulating the label or description applied to an
perspective extends beyond the self. Research has issue, individual, or group. This is known as lexical
shown that subtle differences in the use of these pro- framing. For example, one set of studies found that
nouns can affect how interpersonal relationships are people spend more and save less when an unantici-
evaluated. In one study, for example, participants read pated financial windfall is labeled a “bonus” than when
a description of a friendship that used either “Valerie it’s labeled a “rebate” (Epley et al., 2006). The “bonus”
134 Flusberg et al.

frame leads people to conceptualize the money as heterogeneous, context-specific effects. The same is
something extra, over and above their normal income. true of grammatical framing, which we discuss next.
This licenses freer spending habits because it is not
included in their mental model of their budget. The
Grammatical framing
“rebate” frame, on the other hand, leads people to
conceptualize the money as something that was already Some of the subtlest framing effects result from manipu-
theirs and is only now being returned to them. As a lating grammatical form, including sentence structure,
result, they see the money as part of their normal bud- and grammatical category, such as tense and aspect.
get, which likely includes limits on spending. Grammatical features are often chosen without any
We have seen many other examples of lexical fram- awareness of the implications they carry. For example,
ing throughout this article, including the contrast someone might attempt to combat Islamophobia by
between different labels for people who enter the saying “Christians are just as likely as Muslims to com-
United States without authorization (Rucker et al., mit terrorist acts,” not realizing that this statement tacitly
2019), as well as work that emphasizes different values reinforces anti-Muslim stereotypes by placing Muslims
in a persuasive message or news report (e.g., free in the complement position—implying they are more
speech vs. public order; Nelson, Clawson, & Oxley, likely to commit terrorist acts (Holmes et al., 2022). In
1997; for a review of the literature on moral reframing, addition to such subject-complement statements (which
see Feinberg & Willer, 2019). This reflects the promi- are also a form of equivalency framing; see “Nonvalenced
nence of lexical framing in politics, in which groups Equivalency Framing” section), we have already
with competing interests work to influence public senti- reviewed several other examples of grammatical fram-
ment by labeling and relabeling programs, policies, and ing, including messages that imply different levels of
political opponents (for an extended discussion of this speaker interest by varying tense (“Tell me what hap-
issue from the perspective of a political pollster, see pened” vs. “Tell me what’s happened”; Riou et al., 2017)
Luntz, 2007; see also Lakoff, 2008, 2014). and different personal values by varying whether
Sometimes, lexical framing involves selectively actions are described using nouns or verbs (“be a voter”
applying a single label to one person or situation versus vs. “vote”; Bryan et al., 2011, 2013, 2014).
another. For example, one series of studies investigated Many other grammatical contrasts are known to
victim framing, a common rhetorical technique that influence how people reason about everyday events
casts the alleged perpetrator of a crime as the “real” and situations. Suppose you witness a traffic accident
victim in an attempt to mitigate blame and punishment in which a car swerves off the road and crashes into a
(Flusberg, van der Vord, et al., 2022). Participants read fence. Later, when recounting what you saw, you might
a brief news report about an alleged sexual assault on use an “agentive” expression such as “The driver
a college campus. The report framed either the female crashed the car.” Alternatively, you could omit the
accuser as a victim (of sexual assault) or the male causal agent by saying “The car crashed”—a construc-
alleged assailant as a victim (of false allegations). tion that seems evasive in English but is commonly used
Relative to a baseline condition in which no one was to describe accidents in other languages (Fausey &
framed as a victim, participants generally expressed Boroditsky, 2011). Changes in agentive framing can
more support for the victim-framed character and less affect how people construe what happened. In one
support for the other character. A key mechanism driv- study, participants attributed more blame and financial
ing this effect is pragmatic inference. Only those par- liability to the agent of an accidental event after reading
ticipants who explicitly cited the victim-related language an agentive description (e.g., “she toppled the candle”)
in the report as influencing their evaluations showed than a comparable nonagentive description (e.g., “the
significant framing effects. This suggests that partici- candle toppled”; Fausey & Boroditsky, 2010). This may
pants inferred the victim label was chosen for a rea- be because the agent figured more prominently in the
son—to indicate which character was deserving of situation model triggered by the agentive description,
support. shaping subsequent memory and reasoning (Fausey &
By their nature, lexical framing effects are incredibly Boroditsky, 2011).
diverse. The linguistic properties of generativity and In your description of the car accident you saw, you
displacement we discussed in the “Making Sense of might say “The car swerved” or “The car was swerving.”
Language” section mean that common words and These statements differ in aspectual framing. One
phrases can express a virtually limitless array of mean- emphasizes the completion of the event (“swerved”;
ings, and researchers have only scratched the surface perfective aspect), whereas the other emphasizes its
of their potential effects. Thus, we view lexical framing ongoing nature (“was swerving”; imperfective aspect).
as a loose category that encompasses a wide variety of Research suggests that your choice of aspectual frame
Psychological Science in the Public Interest 25(3) 135

will impact how others conceptualize the accident. In bad drivers under certain circumstances”). A large body
descriptions of motion events, imperfective aspect of research suggests that the use of generic language—
tends to elicit a more dynamic situation model—with generic framing—can promote psychological essential-
motion that is more extensive, vivid, and, in the case ism, the belief that members of a category share a deep,
of accidents, reckless—than perfective aspect (Huette inherent nature or “essence” that makes them funda-
et al., 2014; Matlock, 2011; Matlock et al., 2012). mentally similar (e.g., Benitez et al., 2022; Cimpian &
Aspectual framing can even shape the interpretation of Markman, 2011; Rhodes et al., 2012; but see Noyes &
political messages. In one study, participants were less Keil, 2019; Vasilyeva & Lombrozo, 2020).
confident that a senator would be reelected after read- For example, children are more likely to believe that
ing that he “was taking hush money”—implying that a property associated with a novel social category is
the misdeed was long-lasting—than that he “took hush heritable after hearing relevant generic language (e.g.,
money” (Fausey & Matlock, 2011). “Zarpies are scared of ladybugs”) than nongeneric lan-
The effects of aspectual framing dovetail with recent guage (e.g., “This Zarpie is scared of ladybugs”; Rhodes
work on dynamic norm framing, in which people are et al., 2012), even when the property is said to have a
presented with information about how normative behav- cultural origin (e.g., “because they are taught that lady-
ior is changing over time (Loschelder et al., 2019; bugs are dangerous”; Benitez et al., 2022). This suggests
Sparkman & Walton, 2017, 2019). This type of framing that the form of the generic—Xs are Y—communicates
has been found to be more persuasive than comparable the speaker’s belief that Xs are a “natural kind,” and
“static norm” information (but for evidence it can back- that even young listeners are sensitive to this implicit
fire, see Boenke et al., 2022). In one study, for example, message. For this reason, saying “Asian women are bad
some participants were presented with the following drivers” imparts the presumption that Asian women
message about meat consumption that emphasized a have a distinctive nature—that they’re a homogeneous
stable demographic norm (Sparkman & Walton, 2017): group to whom many stereotypes are likely to apply.
“Recent research has shown that 30% of Americans make This harmful idea may be transmitted by the statement
an effort to limit their meat consumption. That means regardless of whether the listener buys the claim that
that 3 in 10 people eat less meat than they otherwise they’re bad drivers (Gelman, 2021; Wodak et al., 2015).
would.” Other participants were presented with a dynamic Generic framing is common in scientific communica-
norm version of the same message: “Recent research has tion, which favors broad conclusions delivered in a
shown that, in the last 5 years, 30% of Americans have concise, accessible manner. One study found that read-
now started to make an effort to limit their meat con- ers judged scientific results expressed with generics
sumption. That means that, in recent years, 3 in 10 people (e.g., “Group discussion improves lie detection”) as
have changed their behavior and begun to eat less meat more important and generalizable than the same results
than they otherwise would.” Those in the dynamic norm expressed nongenerically (e.g., as past-tense state-
condition reported more interest in reducing their own ments: “Group discussion improved lie detection”;
meat consumption in the future. Follow-up studies sug- DeJesus et al., 2019). This converges with evidence that
gested that this framing led people to assume that more the present tense—a grammatical form that distin-
and more people would reduce their meat consumption guishes generics from nongenerics—implies objective
over time, and so they “preconformed” to this anticipated truths and makes the speaker seem more certain of their
future norm. In other words, their situation model message relative to the past tense (Packard et al., 2023;
included a trajectory of continued change, similar to how for a discussion of how noun labels such as “voter”
imperfective aspectual framing leads people to represent invite similar essentialist inferences, see Gelman &
actions as ongoing. In this case, dynamic norm framing Roberts, 2017). In scientific communication, such
also led people to infer that the framed behavior is gener- generic framing can be problematic because broad gen-
ally considered valuable, which caused them to update eralizations are often unwarranted from complex, vari-
their own weighting of the issue. able data, especially when derived from studies with
Let’s return to the car accident you witnessed. small, unrepresentative samples (DeJesus et al., 2019;
Suppose you learn that the driver was an Asian woman. for other “persuasive communication devices” to watch
In this case, some people might assume that the acci- out for in scientific writing, see Corneille et al., 2023).
dent confirms a well-worn stereotype and assert that Generic framing is also pervasive in political commu-
“Asian women are bad drivers.” This is a generic state- nication (e.g., “Democrats favor affirmative action”) and
ment because it expresses a generalization about an poses similar dangers, exaggerating perceived party
entire category of people, not just the individual in lines (Novoa et al., 2023).
question (“Susan is a bad driver”), and it does not As the writers of this article, we are hardly immune
quantify or qualify the claim (“Some Asian women are to the communicative allure of generics. In presenting
136 Flusberg et al.

a wide variety of findings and conclusions from the affected by the metaphors. Republicans generally have
framing literature, we have used many generic state- stronger views on crime and therefore endorsed
ments. At the same time, we have been careful to enforcement-related solutions regardless of framing.
acknowledge nuances and limitations of the evidence A similar pattern may occur even for ideologically
(for further discussion in this vein, see “Moderators of benign stimuli. For example, effects of framing on recall
Framing Effects” and “Opportunities for Future Research” like those we discussed in the “Memory” section (e.g.,
sections). Nevertheless, if you’re wary of our (generic) reproducing O–O as O^O when it is paired with the
framing of the literature, we invite you to evaluate the label “eyeglasses”; Carmichael et al., 1932) may be
evidence yourself and draw your own conclusions. weaker when memory for the original perceptual stim-
uli is strong (McCloskey & Zaragoza, 1985). Having a
strong memory in these studies is analogous to having
Moderators of Framing Effects strong prior beliefs about crime: In both cases, people
Throughout this article, we have emphasized the general are less susceptible to framing.
impact of communicative frames and the basic mecha-
nisms that underlie these effects. However, we have also
Knowledge and interest
pointed out that there is substantial heterogeneity across
different studies and framing manipulations, and some Research on emphasis framing in news media suggests
effects do not always replicate across samples or experi- that framing effects are larger when the audience has
mental contexts (Bryan et al., 2021; Krefeld-Schwalb more knowledge about the target issue—as long as this
et al., 2024). The presence of moderating factors is one is not confounded with strong prior attitudes and beliefs
reason global effect size estimates in the framing litera- (Chong & Druckman, 2007; Druckman & Nelson, 2003).
ture tend to be small.5 In this section, we discuss a This is because knowledgeable media consumers can
number of well-known moderating factors that can more easily connect the content of a message frame
increase or decrease the magnitude of various framing with their mental model of the target issue.
effects. This list is not meant to be exhaustive. Other research suggests that metaphor frames are
more effective when the audience is interested in and
knowledgeable about the metaphorical source domain
Strength of prior attitudes and beliefs (Ottati et al., 1999; Ottati & Renstrom, 2010; Thibodeau
Framing effects can weaken, disappear, or even backfire et al., 2019). In one experiment, for example, college
when people have strong prior attitudes or beliefs students listened to an audio recording of a message
about the target issue (e.g., Boenke et al., 2022; Chong arguing for a new senior-thesis requirement. The mes-
& Druckman, 2007; Flusberg, van der Vord, et al., 2022; sage included either a sports metaphor or a comparable
Hardisty et al., 2010; Holmes et al., 2022; Landau et al., literal filler sentence, and the arguments in the message
2014; Lecheler et al., 2009; Thibodeau & Boroditsky, made either a strong or weak case for the requirement.
2011, 2013; Thibodeau & Flusberg, 2017). For example, Stronger arguments have been shown to motivate
we described a metaphor framing study in the “Priming” deeper message processing (Petty & Cacioppo, 1986).
section in which a city’s growing crime problem was In this experiment, however, only students who reported
framed as either a “beast” or a “virus” (Thibodeau & liking sports were persuaded by argument strength
Boroditsky, 2011, 2013). American participants exposed when the message included a sports metaphor.
to the beast frame tended to support enforcement- In the case of equivalency framing, knowledge of a
related solutions to the crime problem, such as building domain is sometimes associated with reduced framing
more jails and hiring more police officers. This type of effects. In an experiment on attribute framing, for exam-
solution is conceptually congruent with the beast frame ple, people with little knowledge of NBA basketball
because an issue with an actual beast would be man- judged a target player more valuable when he was
aged by capturing and caging it. Those exposed to the described as “making 60%” of his free throws than when
virus frame, on the other hand, tended to endorse he was described as “missing 40%” (Leong et al., 2017).
reform-related solutions, such as addressing economic However, NBA fans didn’t show this framing effect. The
and educational issues in the city. These solutions are fans knew the player’s performance was poor—the
conceptually congruent with the virus frame because average free throw percentage is much higher—and
real epidemics require targeting root causes and inocu- didn’t need to rely on the frame to draw pragmatic
lating people against the problem. Across multiple ver- inferences about typical free throw rates. That said,
sions of this experiment, however, only participants even experts in a given domain are susceptible to fram-
who identified as Democrats or Independents showed ing when evaluating decontextualized scenarios, such
this framing effect; Republican participants were not as when physicians or public-policy professionals are
Psychological Science in the Public Interest 25(3) 137

asked to use aggregated data to choose between medi- models of persuasion such as the ELM (Petty &
cal treatments framed in terms of “survival” or “death” Cacioppo, 1986). One critical judgment is speaker cred-
(Banuri et al., 2019; McNeil et al., 1982). In such cases, ibility: People are more likely to resist a message frame
experts may find it difficult to apply their specialized when they view the speaker as having low credibility
knowledge, so they rely on the frame instead (Leong (Carpenter, 2012; Chong & Druckman, 2007; Druckman,
et al., 2017). 2001). Conversely, message frames have a greater influ-
More generally, heterogeneity in linguistic framing ence when the speaker or writer is presented as more
effects may result from misalignments in people’s seman- similar to the message recipient (e.g., in political beliefs,
tic frames. The knowledge that comes to mind in response gender, etc.), which enhances their credibility (Balietti
to a given word or phrase is not the same for everyone et al., 2021; Lammers et al., 2023).
(Martí et al., 2023), which means the same message may
structure people’s mental models in different ways. More
Nonnative language
research on individual differences in semantic knowledge
is needed to better address this possibility. So far, we have taken for granted that people are usually
exposed to messages in their native tongue. Some schol-
ars believe, however, that the majority of the world’s
Other individual differences population speaks more than one language (e.g.,
Several studies have investigated the moderating impact Grosjean, 2010). This means people often encounter
of individual differences in cognitive and personality message frames in a nonnative language. It is therefore
traits on valence framing, with somewhat mixed results noteworthy that researchers have observed a “foreign-
(Best & Charness, 2015; Druckman & Nelson, 2003; language effect,” in which the influence of a communi-
Krefeld-Schwalb et al., 2024; LeBoeuf & Shafir, 2003; cative frame is attenuated or even eliminated when a
Levin et al., 2002; Mahoney et al., 2011; Mandel & Kapler, message is presented in a nonnative language (Del
2018; McElroy & Seta, 2003; Stark et al., 2017a). For Maschio et al., 2022; Hayakawa et al., 2016; Keysar et al.,
example, people who score higher on measures of “intui- 2012; Liu et al., 2022). In the first demonstration of this
tive” or “experiential” thinking styles—relying more phenomenon, native speakers of English and Korean
heavily on fast, automatic processing—tend to exhibit completed a risky-choice framing task in either their
stronger risky-choice framing effects (Levin et al., 2002; native language or their second language ( Japanese,
Mahoney et al., 2011; Stark et al., 2017a). Scoring higher English, or French; Keysar et al., 2012). Participants
on measures of “analytic” or “rational” thinking styles, showed the typical framing effect when tested in their
however, does not consistently moderate risky-choice native language in that they were more likely to select
framing (Levin et al., 2002; Mandel & Kapler, 2018; Stark the sure option under the gain frame than the loss
et al., 2017a). Additionally, a meta-analysis indicated that frame. But this framing effect disappeared when par-
younger adults are more likely than older adults to ticipants were tested in their nonnative language.
choose the risky option when presented with a positive/ One explanation for the foreign-language effect is
gain frame, but age does not predict decision-making that people react less emotionally to words and ideas
under negative/loss framing (Best & Charness, 2015). expressed in their nonnative language (Hayakawa et al.,
Some research suggests that individual differences in 2016; Keysar et al., 2012; but see Oganian et al., 2016).
pragmatic sensitivity also matter: Subject-complement This may be because new languages are often learned
framing effects are stronger for those who can explicitly in a formal academic setting, not in the context of our
identify the pragmatic implications of the message frame emotionally rich everyday experiences. Given that
(Holmes et al., 2022; Holmes, Wu, et al., 2024; Wu et al., affective responses contribute to valence framing
2021). Finally, people approach social interactions with effects, the increased emotional distance associated
varying levels of receptiveness to opposing views with using a nonnative language may blunt the effects
(Minson & Chen, 2022). Although receptiveness has not of these message frames. Consistent with this possibil-
been directly measured in any study of linguistic fram- ity, the foreign-language effect has been shown to be
ing that we are aware of, it may be an important inter- less pronounced (i.e., people show similar framing
personal variable for any persuasive communication. effects in both languages) when participants acquired
their second language from an early age (Flexas et al.,
2023) or are presented with less emotionally charged
Judgments of the speaker
valence framing vignettes (Costa et al., 2014).
Since the time of Aristotle, scholars have known that Few studies have examined the foreign-language
audience judgments of a speaker moderate the speak- effect using other types of framing manipulations
er’s persuasive power. This is a core component in besides valence framing. One exception is a study on
138 Flusberg et al.

metaphor framing with Indian participants using a task and invest in larger field studies that measure factors
presented in English (Thibodeau et al., 2016). likely to modulate behavior in the wild.
Participants who reported using English in more infor-
mal, social contexts (e.g., with friends and family)
Reliability and open science
showed the expected metaphor framing effect, whereas
those who reported using English in more formal con- How reliable are framing effects? Whenever possible,
texts (e.g., school or work) showed no framing effect. we have presented the results of meta-analyses and
More research is needed to fully unpack the moderating cited multiple studies of the same topic to indicate the
impact of nonnative languages in the broader framing general robustness of a given finding. Quite deliber-
literature. ately, we have also highlighted several instances in
which high-profile studies have failed to replicate, indi-
cating certain effects may be more brittle than they
Opportunities for Future Research appear in the popular press. Studies showing that a
There are a number of open questions, exciting new very subtle framing device dramatically changes real-
avenues to pursue, and other golden opportunities for world behaviors or strongly held attitudes seem espe-
future research on linguistic framing. This is our posi- cially difficult to replicate. This is not surprising.
tive framing of the problems, limitations, and method- Framing manipulations are often designed to be incon-
ological concerns with the current literature. Here we spicuous, and human reasoning is complex and sensi-
discuss five important issues: ecological validity; reli- tive to many factors. In addition, framing research has
ability and open science; the need for more precise historically suffered from many of the same method-
models; the narrow focus on Western, educated, indus- ological deficiencies as the rest of the social and behav-
trialized, rich, and democratic (WEIRD) societies; and ioral sciences (Shrout & Rodgers, 2018).
AI. We discuss each issue briefly in turn. Therefore, we support the emerging “credibility revo-
lution” that was sparked by the “replication crisis” in
psychology and other sciences (Shrout & Rodgers, 2018;
Ecological validity Vazire, 2018). This movement embraces principles of
In the “Behavior” section, we noted that most framing open science, which include greater transparency about
research is conducted in the lab but that researchers research practices, preregistration of studies, more
are ultimately interested in real-world behavior. In other direct replications of key findings, and improved standards
words, researchers want to know whether framing for evaluating the quality of evidence. Simultaneously,
effects have ecological validity. Several field studies we think researchers and science communicators
have shown promising results, but some studies fail to should recognize that heterogeneity in behavioral inter-
replicate across different contexts, as we have dis- vention effects is to be expected (Bryan et al., 2019,
cussed. One concern is that we don’t yet know how 2021). We encourage people to carefully consider con-
long most framing effects last. Some studies suggest text effects, individual differences, and other moderat-
certain framing effects are short-lived (Druckman & ing factors.
Nelson, 2003) whereas others are “surprisingly persis-
tent,” lasting several weeks (Lecheler & de Vreese,
More precise models
2011). To date, however, no systematic work has
assessed the endurance of different framing effects and Many discussions of framing rely on overly broad,
what this means for real-world impact (but see Baden descriptive theories or general process models of the
& Lecheler, 2012). phenomenon of interest, such as the ELM (see also
Another issue is that some framing interventions with “Mechanics of Communication and Linguistic Framing”
a mixed record of success have not been evaluated with section). Researchers in cognitive psychology and
respect to the moderating factors we discussed in the behavioral economics have formalized models of cer-
previous section, such as knowledge or interest in the tain valence framing effects (Huizenga et al., 2023), but
target issue or the ability to draw pragmatic inferences these do not generalize beyond a very restricted domain
from the framing language. For example, teaching stu- (see also Schwartzstein & Sunderam, 2021). This leaves
dents that their brain can “get stronger like a muscle”— researchers and those who wish to use framing tech-
a “mindset” intervention designed to improve academic niques with only a vague idea of where to start. In most
achievement—shows promise, but the strength and cases, we can’t even specify whether a language manip-
reliability of its effects are still under debate (Macnamara ulation provides the right “dose” to elicit a framing
& Burgoyne, 2023; Yeager et al., 2019). We recommend effect without a great deal of testing (Rothman et al.,
that future work address these gaps in the literature 2020). Therefore, we recommend that researchers
Psychological Science in the Public Interest 25(3) 139

develop more precise models of linguistic framing join us on a spontaneous adventure, sell a new product,
effects, integrating insights from across the various dis- create a health-messaging campaign, or solicit a dona-
ciplines that contribute to this field. tion to our favorite nonprofit. But, as we have seen,
some frames are more influential than others. In this
section, we offer step-by-step guidelines for effective
WEIRD problems
framing, emphasizing applications in the public interest.
Like many areas of psychology and communication, Although our recommendations are intended to be
research on framing suffers from a number of WEIRD applicable in many different contexts, our examples
problems (de Oliveira & Baggs, 2023; Henrich et al., focus on the promotion of health-related policies and
2010). Most researchers, participant samples, and lan- initiatives:
guages used in this work originated from WEIRD societ-
ies, largely in the Anglophone world. To the extent that 1. Who’s your audience? First, you should think
we are interested in understanding and motivating about your target audience. What is its makeup,
people in these societies, this narrow focus is not an and what are their values, needs, goals, and
issue. However, if our ultimate goal is to develop gen- knowledge base? How receptive are they to
eral theories of language, framing, and persuasion, then opposing views (Minson & Chen, 2022)? What is
this is a significant problem. A majority of humans are the cultural context of the communication?
not from WEIRD societies, and any research that fails Research suggests that matching features of your
to account for the full range of human variation is message frame to the characteristics of your
inherently limited. We know that languages and cul- audience can enhance your persuasive power
tures differ in a number of important ways that can (Balietti et al., 2021; Feinberg & Willer, 2019;
have downstream effects on cognitive processing (see Joyal-Desmarais et al., 2022; Tappin et al., 2023;
Box 1; see also Blasi et al., 2022). Therefore, we recom- but for guidance on when and why matching a
mend that framing researchers form more multidisci- message to a recipient may backfire, see Teeny
plinary, cross-cultural, and cross-linguistic collaborations et al., 2021). We recommend leveraging this fact
in future work. as you craft your message frame. For example,
if you are speaking to a conservative audience
in the United States about a new health-care
Artificial Intelligence
policy, it might help to frame the discussion in
AI has advanced rapidly in recent years. The develop- terms of values such as loyalty or patriotism
ment of large language models (LLMs) such as ChatGPT rather than fairness or harm. You could say, for
means we can now have complex, interactive conversa- instance, that we have a patriotic duty to keep
tions with systems that communicate in surprisingly Americans healthy and that loyal citizens must
sophisticated—and sometimes perplexing—ways. Users ensure that our children and elders have access
frequently share tips for eliciting better responses from to lifesaving medicine.
these chatbots, making suggestions for how to elaborate 2. What is your goal? Next you must determine your
a message and follow up in a dialogue to get the best goal. Are you aiming to change attitudes, beliefs,
results. Such “prompt engineering” is really just framing or behaviors? If you are targeting behaviors, what
under another guise (or lexical frame). Researchers have is it you want your audience to do? Is the behav-
recently shown that LLMs can be used to generate per- ior easy or hard? Is it a one-off, short-term action
suasive, microtargeted political ads at scale, which raises (such as making a donation) or a long-term,
concerns about the use of these systems by bad actors habitual one (such as increasing weekly exer-
(Simchon et al., 2024). However, such advances also cise)? Your choice of communicative frame will
present an opportunity for researchers. The systematic depend greatly on how you intend your message
study of framing in LLMs may help us better understand recipients to respond. If you are aiming to shift
both these systems and ourselves. attitudes toward a new health-care initiative, for
example, more valenced language could help. If
you want to improve people’s understanding of
Recommendations for Effective Framing
a complex issue such as vaccination or govern-
From the beginning of this article, we have emphasized ment spending on health care, on the other
that framing effects are ubiquitous—a natural and hand, then using an extended metaphor may be
unavoidable consequence of human communication. most effective.
We must always select one communicative frame or 3. Consult existing research. As we have surely dem-
another, whether we are trying to convince a friend to onstrated by now, the literature on framing is
140 Flusberg et al.

Box 1. How Framing Informs Linguistic Relativity

Approximately 7,000 languages are spoken on Earth, and their diversity is astonishing (N. Evans & Levinson, 2009;
Malt & Majid, 2013). As N. Evans and Levinson (2009) observed: “Languages differ so fundamentally from one
another at every level of description (sound, grammar, lexicon, meaning) that it is very hard to find any single
structural property they share” (p. 429). As a result, the semantic systems of different languages—their repertoire
of communicative frames—differ in systematic ways. This includes not only idiosyncratic expressions tied to local
cultural concerns but also category labels, conventional metaphors, and grammatical devices that communicate
seemingly core aspects of human experience. Unlike English, for example, Russian and Greek distinguish
categorically between light and dark shades of blue (Thierry et al., 2009; Winawer et al., 2007); Turkish and
Farsi describe high and low tones using words for thin and thick (Dolscheid et al., 2013, 2020); and German
and Swedish lack aspectual markers for distinguishing ongoing from completed events (e.g., “was swerving” vs.
“swerved”; see “Grammatical Framing” section; Athanasopoulos & Bylund, 2013).
The amateur linguist Benjamin Lee Whorf famously argued that differences of this sort lead speakers of
different languages to form different conceptions of reality (Whorf, 1940/2012). Writing in the early 20th century,
Whorf wasn’t aware of the myriad examples of linguistic variation that scientists have since documented. Instead,
he relied on his own informal (and often flawed) observations, many of which were inspired by his work as an
inspector for the Hartford Fire Insurance Company. In one much-cited anecdote, an explosion occurred after
a factory worker carelessly tossed a burning cigarette into a drum labeled “empty” that was actually filled with
hazardous vapor. This, it would seem, is a framing effect: The communicative frame “empty” led to a behavior
with disastrous consequences, perhaps because it prompted the worker to construct a situation model of the
apparently “null and void, negative, inert” contents of the drum (Whorf, 1940/2012, p. 175).
Then again, the explosion may have occurred simply because the vapor-filled drum appeared empty to the
worker, not because it was labeled that way (Lenneberg, 1953; Pinker, 1994). For good reason, many scholars
roundly dismissed this example and most of Whorf’s other linguistic observations as misleading or erroneous
in implicating language as the source of apparent cognitive and behavioral patterns. This delayed serious
consideration of the claim that speakers of different languages think differently—the so-called Whorfian, Sapir-
Whorf, or linguistic relativity hypothesis—until many decades later, when the science finally caught up with the
hype (Gentner & Goldin-Meadow, 2003).
Today, linguistic relativity enjoys a more positive reception in the literature, and research on framing can help
explain why. The Whorfian hypothesis essentially has three parts (Wolff & Holmes, 2011). First, languages differ
in how their lexicons and grammatical structures partition the world. This assertion—Whorf’s shoddy linguistics
notwithstanding—is now indisputable. Second, the lexicon and grammatical structure of a given language
influence how its speakers think and reason about the world. This claim is supported by the research we’ve
discussed in this article, which provides many striking demonstrations of language’s impact on cognition, albeit
mostly in English (which may be why classic and contemporary framing research is rarely cited in the linguistic
relativity literature; cf. Blasi et al., 2022).
Finally, if the first two propositions are true, it seems to follow that speakers of different languages will think
and reason about the world in ways that mirror the lexical and grammatical differences in their languages. Much
recent research supports this conclusion through cross-linguistic comparisons of all the aspects of cognition we
surveyed in the “What Does Language Do?” section, among others (for review, see Wolff & Holmes, 2011). For
example, Russian and Greek speakers display enhanced discrimination of colors at their language’s boundary
between light and dark blue (Thierry et al., 2009; Winawer et al., 2007); Turkish and Farsi speakers match low
(“thick”) tones to thick, high lines rather than thin, low lines (Dolscheid et al., 2013, 2020); and German and
Swedish speakers give more weight to endpoints than midpoints when judging the similarity of motion events
(Athanasopoulos & Bylund, 2013; Athanasopoulos et al., 2015). English speakers show none of these patterns.
Importantly, none of these are “framing” effects in the sense we have been using this term: The stimuli
in these studies (and the format of participants’ responses) involve little to no language. This is a strength in
linguistic relativity research because it suggests that speakers of different languages think differently even when
they’re not “thinking for speaking” (Boroditsky, 2001; Slobin, 1996). Nevertheless, the mechanisms of framing
we’ve discussed in this article can help make sense of these cross-linguistic differences. As we suggested earlier,
languages have systematically different communicative frames: different color words, pitch metaphors, aspectual
markers, and so on. And as we discussed in the “Mechanics of Communication and Linguistic Framing” section,
communicative frames evoke situation models in the minds of readers and listeners that have downstream
(continued)
Psychological Science in the Public Interest 25(3) 141

Box 1. (continued)

consequences for thought and behavior, mediated by a host of cognitive, social-pragmatic, and emotional
mechanisms. This means speakers of different languages—who habitually use different communicative frames—
will habitually construct different situation models. As a result, we would expect them to perform differently on
tasks in which these mental models are used to answer questions or solve problems. This is exactly what the
linguistic relativity literature indicates. From this framing-informed perspective, cross-linguistic differences in
cognition are the consequences of a lifetime of exposure to the communicative frames of one’s native language.
Of course, just because speakers of different languages exhibit different patterns of thought or behavior doesn’t
mean language was the cause of them. That’s because linguistic differences are often confounded with differences
in culture, environment, and life experience (Casasanto, 2005; Li & Gleitman, 2002). For example, speakers of
languages that grammatically distinguish present from future events (e.g., English: “It’s raining” vs. “It’s going to
rain”) tend to save less money than speakers of “futureless” languages that conflate the two tenses (M. K. Chen,
2013), but this correlation disappears when accounting for historical and geographical relationships between
languages that track shared cultural values (Roberts et al., 2015; but see Ayres et al., 2023).
In many studies, however, “extra-linguistic” factors such as these can be ruled out through experimental
manipulations that establish a causal role for language. One approach is to have participants complete a “verbal
interference” task that blocks their mental access to language, such as shadowing speech or mentally rehearsing
a string of digits. Cross-linguistic differences often disappear under verbal interference, suggesting that they
were driven by language and not some other factor (for a systematic review of verbal interference studies, see
Nedergaard et al., 2023).
Another approach is to expose speakers of one language to a “concentrated dose” of the lexical or grammatical
patterns of another language (e.g., Boroditsky, 2001; Casasanto, 2008; Dolscheid et al., 2013; Fausey et al.,
2010). This is essentially a framing manipulation, in which a new set of communicative frames is introduced and
their impact is assessed. After exposure to the “new language,” participants often perform like speakers of that
language typically do on tasks previously shown to yield a cross-linguistic difference. This suggests that language
alone can create the difference. It also reinforces the point that habitual use of certain communicative frames—
presumably the sum of many short-term exposures—drives many linguistic relativity effects.
For some scholars, the fact that a short-term linguistic manipulation can temporarily override habitual
thought patterns is evidence that the effects of language don’t run very deep (Gleitman & Papafragou, 2012;
January & Kako, 2007). For framing researchers, however, short-term effects are par for the course. Most of the
findings reviewed in this article concern the influence of language on in-the-moment decision-making, guided
by our current (but subject-to-change) mental model of the target issue or situation. Although longer-lasting
effects might bolster the real-world impact of framing, even temporary effects can be powerful if they result in
important decisions or behaviors. Several recent accounts of linguistic relativity dovetail with framing research in
emphasizing the dynamic, context-sensitive nature of human thought—and the potential for language to meddle
with it (Athanasopoulos et al., 2015; Casasanto, 2016; Lupyan et al., 2020; Ünal & Papafragou, 2016).

vast. This is both a blessing and a curse. It cer- example, we reviewed much research on the
tainly makes locating and synthesizing relevant impact of goal framing in health communication,
findings a daunting prospect, although hopefully showing that gain and loss frames may be better
we have done some of this work for you. The suited to different target behaviors (disease pre-
plus side is that researchers have examined hun- vention vs. detection, respectively).
dreds if not thousands of individual framing tech- 4. Engage the emotions wisely. The power of emo-
niques across a wide range of messaging contexts. tional appeals has been recognized for as long
We suggest consulting this literature for guidance as people have been thinking about persuasive
as you construct your message frame. There is language. And for good reason. Emotions can
no need to reinvent the wheel when hundreds motivate interest, engagement, concern, and
of scholars have been prototyping and testing action. But they can also motivate violent out-
different tires, hubcaps, and axles for decades. rage, polarization, helplessness, and withdrawal.
The taxonomy we have laid out in this article We recommend designing communicative frames
(see Table 3) is a good place to start. For carefully to leverage the positive power of
142 Flusberg et al.

emotions without slipping into extreme partisan message frame, you could invite them to get
rhetoric or negative doomsaying (for examples screened at their earliest convenience.
related to climate change communication, see 7. Test your message. After reading through this
Box 2). For example, to inspire a sense of article, you have hopefully improved your intu-
urgency and concern during a health-care emer- itions for which message frames are likely to be
gency, such as when a new virus is circulating effective. That’s great, but intuitions are only a
in the population, using negatively valenced lan- starting point. Just because several studies have
guage (e.g., war metaphors) may be a productive found that a particular framing device is effective
first step—but only if you follow it with a more in one domain, with one group of people, does
optimistic message about how to address the not mean it will work in another domain, with
issue. Giving people hope that a problem can another group of people. Many framing effects
be solved is critical to getting people to engage are subtle and context-dependent. We recom-
and not tune out. mend, therefore, that you use some of the meth-
5. Generate your frame. Only at this stage should odological techniques we have described to test,
you draft your message frame. We recommend evaluate, and improve your messages. Many
thinking about the mental model you would like researchers, companies, advertisers, and political
your audience to adopt and working backward, campaigns do just that. For example, a recent
assessing how different linguistic cues (e.g., study examined the impact of different explana-
labels, metaphors, grammatical constructions) tory metaphors in vaccine communication and
might or might not help instantiate this model. compared them to briefer, more “literal” explana-
Once again, the taxonomy in Table 3 may be use- tions (Flusberg et al., 2024). The two types of
ful, and we have provided many other relevant explanations improved vaccine attitudes to a
examples throughout this article. For example, similar degree. However, metaphorical explana-
imagine your goal is to increase the number of tions led people to generate more elaborate
people who participate in a regular screening for explanations of their own, suggesting they have
skin cancer. The mental model you’d like to com- an advantage in facilitating further social com-
municate is that more and more people are get- munication about this issue.
ting this screening because it can help detect the 8. Be open to reframing. We also recommend that
disease when it is still very treatable. Working you remain flexible and are willing to change
backward, then, you could combine dynamic and update your message frames over time as
norm framing with goal (loss) framing, generating circumstances change. Although repetition can
a message such as “Recent research has shown increase the saliency of a message, people may
that, in the last 5 years, 30% of Americans have come to resist popular framings. For example,
now started to get screened for skin cancer every using a violent battle metaphor may be a good
year. If you do not engage in regular screenings, first step to attract attention and impart a sense
you will have a decreased chance of finding skin of urgency for an issue such as a pandemic, but
cancer in the early, more treatable stage of the the effects of such language could backfire over
disease. Talk to your dermatologist today!” time as the audience realizes there is no way to
6. Provide opportunities for action. The effects of a truly “win” the war (Flusberg et al., 2018). A
message frame will likely be short-lived, quickly more apt metaphor in this case might be fire-
overshadowed by the passage of time and other fighting, in which a contagion is likened to a fire
conflicting messages. We recommend providing spreading in a forest, and the more tractable goal
an opportunity for action as quickly as possible is to control and reduce the spread rather than
(cf. Lakoff, 2010). This can be small, such as “kill” the entire fire (Semino, 2021).
acquiring a signature or email address, or it can 9. Temper your expectations. Language is powerful,
be larger, such as securing a donation or purchase but it’s not a magic bullet. Study after study (and
right away. Engaging a behavioral commitment meta-analysis after meta-analysis) shows that
early on makes it more likely that the impact of framing effects are relatively small, so changing
your message will make a difference. Following your message frame will likely have only a mod-
up on the example provided in the previous step, est impact on your audience (O’Keefe & Hoeken,
for example, you could partner with a mobile 2021). Of course, if a message spreads through
clinic that offers free or low-cost skin-cancer a large enough population, even marginal returns
screenings and meets people where they are. As could yield a significant real-world impact. But
soon as your target audience is exposed to the most of us don’t have that reach. Real,
Psychological Science in the Public Interest 25(3) 143

Box 2. Framing in Environmental Communication

Anthropogenic climate change poses an existential threat to humanity. Even if you are skeptical of such claims,
you probably appreciate clean air and water and a healthy natural environment. Scientists, activists, and political
leaders have deployed a wide range of messaging strategies to advance several interrelated, proenvironmental
goals. Some messages are used to drum up support for new policies, programs, or regulations (attitude change).
Others are aimed at increasing public understanding of complex ideas, such as why excess carbon emissions
cause global warming (belief change). And still others are used to elicit donations to environmental causes or
votes for proenvironmental candidates (behavior change). Researchers in this space have tested the efficacy of
many of the framing techniques we described in the “A Taxonomy of Framing Devices” section.
Homar and Cvelbar (2021) provided a systematic review of valence framing effects on environmental decision-
making. On the whole, negative/loss frames were found to be more effective than positive/gain frames for
increasing proenvironmental intentions and behaviors. The authors recommend that environmentally relevant
decisions be framed in terms of a potential loss or negative consequence that must be prevented. Gain framing
can be effective in some contexts, but mostly for improving attitudes toward an issue rather than eliciting
behavior change.
Metaphor framing is pervasive in environmental messaging, from the “war” on coal and the “race” to net-zero
emissions to carbon “footprints” and “greenhouse” gases. Flusberg and Thibodeau (2023) recently conducted
a broad assessment of the literature on English metaphors in environmental discourse, highlighting several
promising message frames. For example, (a) describing the earth as our “common home” may help people
feel more connected to nature, which has been shown to increase proenvironmental attitudes (see Thibodeau,
Frantz, & Berretta, 2017); (b) explaining how the atmosphere functions as a “giant bathtub” can help people
understand why carbon emissions continue to accumulate and drive global warming (see Guy et al., 2013); and
(c) war metaphors can enhance the sense of urgency and risk surrounding climate change and increase intentions
to behave in ways that would mitigate those risks (see Flusberg et al., 2017). However, there have not been
many large, replicable studies of metaphor framing in environmental messaging, so much of this work remains
speculative.
One compelling idea is that people struggle to engage with the issue of climate change because it seems
so psychologically distant (Van Lange & Huckelba, 2021). It can feel like an abstract problem that affects other
people who live far away and will only affect us in the distant future. This suggests that psychological distance
framing may be effective if it can be used to make climate change feel more proximal and concrete. The
research on this topic is promising but mixed. A recent large-scale study conducted in more than 60 countries
found that framing climate change as psychologically close increased belief in climate change more than any
other intervention, but the magnitude of this effect was modest (Vlasceanu et al., 2024). And neither this nor any
of 10 other expert-backed interventions increased people’s motivation to engage in effortful climate-mitigation
behaviors such as tree planting. Another set of experiments found that narrative stories about climate change
were more effective than informational texts for eliciting proenvironmental behaviors (Morris et al., 2019). This
was due to the fact that the narratives were more emotionally engaging. However, communicating statistics
about the scientific consensus on climate change can also increase belief in climate change and support for
environmental policies (van der Linden, 2021). Other studies that have directly manipulated the perceived
psychological distance of climate change have yielded inconsistent results, suggesting that moderators and other
contextual factors are relevant for this type of messaging (Maiella et al., 2020; Yang et al., 2021). For example,
greater psychological closeness may increase the impact of a loss-framed environmental message, whereas greater
psychological distance may increase the impact of a gain-framed message (Homar & Cvelbar, 2021).
Research on lexical framing and moral reframing highlights the importance of individual differences in
environmental communication. Studies suggest that people who are more ideologically conservative, culturally
individualistic, and supportive of free-market principles are more skeptical of climate change and resistant to
standard environmental messaging (Hornsey & Fielding, 2020). Therefore, messages that appeal to these values
should be more effective. In one experiment, American participants had to choose which of two products they
would purchase (e.g., airline flights; Hardisty et al., 2010). The products were identical, but one was slightly more
expensive so that the extra profits could be used to fund carbon-reduction programs. The extra cost was labeled
either a carbon “tax” or a carbon “offset.” Republican and Independent participants were significantly more likely
to choose the pricier product when it was labeled an “offset” rather than a “tax,” presumably because they are
ideologically opposed to new taxes. Many other experiments have shown that invoking conservative values such
(continued)
144 Flusberg et al.

Box 2. (continued)

as loyalty, patriotism, and purity can enhance proenvironmental attitudes and behavior among conservative
participants (e.g., Feinberg & Willer, 2013; Feygina et al., 2010; Hurst & Stern, 2020; Kidwell et al., 2013; Wolsko
et al., 2016).
There are several additional factors to consider in environmental messaging. First, emotions play a critical
but nuanced role. Negative feelings, including a sense of urgency, fear, and concern about the future, are
especially motivating. But too much negativity is counterproductive. Dire messaging about climate change has
been shown to backfire, reducing belief in global warming by threatening people’s sense that the world is a just
place (Feinberg & Willer, 2011). Strong negative emotions can also undermine feelings of efficacy and promote
the sense that nothing can be done to address environmental devastation. Therefore, positive feelings, such as
a sense of hope, are important (Geiger et al., 2023; Nabi et al., 2018), although too much optimism can make
the potential impact of climate change on our own lives seem insubstantial (Hornsey & Fielding, 2016). Taken
together, this work suggests that environmental messages should be carefully crafted to elicit the right balance of
emotional response.
Second, messages may be more effective if they target certain dimensions of climate change over others. A
meta-analysis by Li and Su (2018) found that message frames were largely ineffective when they focused on
geographical identity or public health but had small-to-medium-sized effects on engagement when they focused
on the moral, economic, and environmental aspects of climate change. It may therefore be a good idea to
emphasize these dimensions in climate communication.
Finally, environmental issues are enormously complex and require massive, cooperative, and systemic
solutions. Leveraging the power of social norms to encourage behavior change is one promising avenue to
pursue (Constantino et al., 2022), and dynamic norm framing may be helpful in this regard (Loschelder et al.,
2019; Sparkman & Walton, 2017). However, such interventions are “no panacea” (Constantino et al., 2022).
Message framing effects are typically small and transient. To maximize their impact, they must be coupled with
immediate opportunities for action and embedded within broader social and political movements aimed at
enacting meaningful structural changes across the globe (Constantino et al., 2022; Flusberg & Thibodeau, 2023;
Lakoff, 2010).

long-lasting change in attitudes, beliefs, and emotional reactions that influence our behavior. A
behavior will take more than a single slogan, deeper understanding of the psychology of framing can
article, advertisement, or conversation. Investing enhance not only our communication skills and ability
in deeper structural changes will be the only way to navigate our linguistic environments but also our
to address complex issues such as health-care prospects for getting things done. The research we have
policy, climate change, and social inequality. reviewed serves as a reminder to be mindful of both
the words and phrases we use ourselves and the ones
Conclusions: Language in the Public that grab our attention or generate a strong emotional
response. We don’t like to be cynical about people’s
Interest intentions, but we think it’s wise to consider the goals
Language is the fuel that powers social and political and incentives behind the loud voices that intrude on
life. It’s through language that coalitions and move- daily life in the media, in the public square, and online,
ments are formed and reformed, new policies and ideas including sometimes our own.
are generated and spread, and leaders inspire hope, But remember that Newspeak is science fiction, not
excitement, and fear. As a result, it’s in the public’s fact. Language is powerful, but it doesn’t transform us
interest to better understand how and why language into mindless marionettes. Most framing effects are a
shapes the way we think, feel, and act—how commu- sensible response to the information that is communi-
nicative frames can shape and restructure our mental cated by a linguistic message, and most are quite mod-
models of the world. We began this article with a medi- est in size. We always have the power to take control
tation on speculative fiction, in which language is often of the narrative, so to speak. We can seek out alterna-
presented as an irresistible force that undermines our tive opinions and arguments, reframe issues and events
free will. There is a kernel of truth to this idea: Linguistic to ourselves and to others, and engage in careful reflec-
messages can trigger a cascade of cognitive and tion and thoughtful dialogue to develop a richer
Psychological Science in the Public Interest 25(3) 145

understanding of the world around us. What’s more, a new way, through the lens of a moral value they hold dear.
we can leverage these insights to promote positive 4. Our moods, feelings, and emotions vary along two dimensions
policy reforms, environmental and health-care initia- that comprise what psychologists call “core affect”: valence, or
tives, and other vital causes in the public interest. the degree of positivity or negativity, and arousal, or the degree
of intensity (Russell, 2003). Most researchers who discuss affect
in the context of framing and persuasive messaging focus on the
Transparency
valence dimension (e.g., Slovic et al., 2007), but it’s likely that
Action Editor: Nora Newcombe arousal plays a role as well (e.g., Flusberg et al., 2018).
Editor: Nora Newcombe 5. O’Keefe (2017b) argued that people often misunderstand
Declaration of Conflicting Interests effect size estimates in message effects research. The effect
The author(s) declared that there were no conflicts of size indicates the relative efficacy of one message frame ver-
interest with respect to the authorship or the publication sus another, not the absolute magnitude of a particular effect.
of this article. This type of research typically compares two or more message
frames to each other (e.g., narratives vs. statistics, or Metaphor
Notes A vs. Metaphor B). So, a small effect size does not necessarily
mean that a message frame has little impact on its own, because
1. Cognitive scientist David Rumelhart used a different metaphor
it is rarely compared to a pretest score or baseline condition
to make a similar point (as cited in Elman, 2009): “My approach
with no message. Rather, a small effect size indicates that a
suggests that comprehension, like perception, should be likened
message frame is only slightly more impactful than the other
to [a] paleontologist, who uses his beliefs and knowledge about
message frame it is being compared to.
dinosaurs in conjunction with the clues provided by the bone
fragments available to construct a full-fledged model of the origi-
nal. In this case the words spoken and the actions taken by the References
speaker are likened to the clues of the paleontologist, and the
Adams, R. B., Ambady, N., Macrae, C. N., & Kleck, R. E.
dinosaur, to the meaning conveyed through these clues” (p. 548).
(2006). Emotional expressions forecast approach-avoid-
2. One consequence of needing prior knowledge to “decode”
ance behavior. Motivation and Emotion, 30, 177–186.
language is that stories are difficult to understand if we can’t
https://doi.org/10.1007/s11031-006-9020-2
activate the right frames. Words are more abstract than many
Ajzen, I., Fishbein, M., Lohmann, S., & Albarracín, D. (2018).
people presume (Lupyan & Winter, 2018), so the same words
The influence of attitudes on behavior. In D. Albarracín
mean different things in different contexts. Without the proper
& B. T. Johnson (Eds.), The handbook of attitudes. Vol.
frame in mind, generating a coherent situation model may be
1: Basic principles (pp. 197–255). Routledge.
impossible. For example, consider the following description:
Aka, A., Bhatia, S., & McCoy, J. (2023). Semantic determinants
The procedure is actually quite simple. First you arrange of memorability. Cognition, 239, Article 105497. https://
things into different groups depending on their makeup. doi.org/10.1016/j.cognition.2023.105497
Of course, one pile may be sufficient depending on how Aka, A., Phan, T. D., & Kahana, M. J. (2021). Predicting recall
much there is to do. If you have to go somewhere else of words and lists. Journal of Experimental Psychology:
due to lack of facilities, that is the next step; otherwise you Learning, Memory, and Cognition, 47(5), 765–784.
are pretty well set. It is better to do too few things at once https://doi.org/10.1037/xlm0000964
than too many. Remember mistakes can be expensive. At Alam, N., & So, J. (2020). Contributions of emotional flow
first the whole procedure will seem quite complicated. in narrative persuasion: An empirical test of the emo-
Soon, however, it will become just another fact of life. It tional flow framework. Communication Quarterly, 68(2),
is difficult to foresee any end to the necessity for this task 161–182. https://doi.org/10.1080/01463373.2020.1725079
in the immediate future, but then one never knows. Albarracín, D., Sunderrajan, A., Lohmann, S., Chan, M. P. S.,
& Jiang, D. (2018). The psychology of attitudes, motiva-
If you’re anything like us—or the participants in the origi- tion, and persuasion. In D. Albarracín & B. T. Johnson
nal experiment that used this vignette (Bransford & Johnson, (Eds.), The handbook of attitudes. Vol. 1: Basic principles
1972)—you are probably struggling to make heads or tails of (pp. 3–44). Routledge.
this paragraph. This is the case even though you are familiar Alogna, V. K., Attaya, M. K., Aucoin, P., Bahník, Š., Birch, S., Birt,
with all of the individual words. Things would be different if A. R., Bornstein, B. H., Bouwmeester, S., Brandimonte, M.
we had provided a title that evoked the proper frame, such A., Brown, C., Buswell, K., Carlson, C., Carlson, M., Chu,
as “How to do laundry.” If you reread the paragraph with this S., Cislak, A., Colarusso, M., Colloff, M. F., Dellapaolera,
frame in mind, you should be able to generate a coherent situ- K. S., Delvenne, J.-F., . . . Zwaan, R. A. (2014). Registered
ation model with the meaning of each word falling into place. replication report: Schooler and Engstler-Schooler (1990).
The laundry frame is metacommunicative, to use Bateson’s Perspectives on Psychological Science, 9(5), 556–578.
(1972/1987) term: it tells you how to interpret the rest of the https://doi.org/10.1177/1745691614545653
message so you can successfully decode what, for example, Alter, A. L., & Oppenheimer, D. M. (2006). Predicting short-term
“things” and “facilities” mean in this context. stock fluctuations by using processing fluency. Proceedings
3. That said, moral reframing may also rely on schematic restruc- of the National Academy of Sciences, USA, 103(24), 9369–
turing because the audience is led to view the target issue in 9372. https://doi.org/10.1073/pnas.0601071103
146 Flusberg et al.

Anderson, E. C., & Barrett, L. F. (2016). Affective beliefs Personality Psychology Compass, 5(4), 194–205. https://
influence the experience of eating meat. PLOS ONE, doi.org/10.1111/j.1751-9004.2011.00344.x
11(8), Article e0160424. https://doi.org/10.1371/journal Blasi, D. E., Henrich, J., Adamou, E., Kemmerer, D., & Majid,
.pone.0160424 A. (2022). Over-reliance on English hinders cognitive sci-
Areni, C. S., & Sparks, J. R. (2005). Language, power, and per- ence. Trends in Cognitive Sciences, 26(12), 1153–1170.
suasion. Psychology & Marketing, 22(6), 507–525. https:// https://doi.org/10.1016/j.tics.2022.09.015
doi.org/10.1002/mar.20071 Bloemraad, I., Silva, F., & Voss, K. (2016). Rights, economics,
Ariely, D. (2008). Predictably irrational. HarperCollins. or family? Frame resonance, political ideology, and the
Asch, S. (1946). Forming impressions of personality. Journal immigrant rights movement. Social Forces, 94(4), 1647–
of Abnormal Social Psychology, 41(3), 258–290. https:// 1674. https://doi.org/10.1093/sf/sov123
doi.org/10.1037/h0055756 Boenke, L., Panning, M., Thurow, A., Hörisch, J., & Loschelder,
Athanasopoulos, P., & Bylund, E. (2013). Does grammatical D. D. (2022). Who can nudge for sustainable develop-
aspect affect motion event cognition? A cross-linguistic ment? How nudge source renders dynamic norms (in-)
comparison of English and Swedish speakers. Cognitive effective in eliciting sustainable behavior. Journal of
Science, 37(2), 286–309. https://doi.org/10.1111/ Cleaner Production, 368, Article 133246. https://doi.org/
cogs.12006 10.1016/j.jclepro.2022.133246
Athanasopoulos, P., Bylund, E., Montero-Melis, G., Boeynaems, A., Burgers, C., & Konijn, E. A. (2021). When
Damjanovic, L., Schartner, A., Kibbe, A., Riches, N., & figurative frames decrease political persuasion: The
Thierry, G. (2015). Two languages, two minds: Flexible case of right-wing anti-immigration rhetoric. Discourse
cognitive processing driven by language of operation. Processes, 58(3), 193–212. https://doi.org/10.1080/0163
Psychological Science, 26(4), 518–526. https://doi.org/ 853X.2020.1851121
10.1177/0956797614567509 Boroditsky, L. (2001). Does language shape thought? Mandarin
Ayres, I., Katz, T. K., & Regev, T. (2023). Languages and and English speakers’ conceptions of time. Cognitive
future-oriented economic behavior—Experimental evi- Psychology, 43(1), 1–22. https://doi.org/10.1006/
dence for causal effects. Proceedings of the National cogp.2001.0748
Academy of Sciences, USA, 120(7), Article e2208871120. Bower, G. H., Black, J. B., & Turner, T. J. (1979). Scripts in
https://doi.org/10.1073/pnas.2208871120 memory for text. Cognitive Psychology, 11(2), 177–220.
Baden, C., & Lecheler, S. (2012). Fleeting, fading, or far- https://doi.org/10.1016/0010-0285(79)90009-4
reaching? A knowledge-based model of the persistence of Braddock, K., & Dillard, J. P. (2016). Meta-analytic evidence
framing effects. Communication Theory, 22(4), 359–382. for the persuasive effect of narratives on beliefs, attitudes,
https://doi.org/10.1111/j.1468-2885.2012.01413.x intentions, and behaviors. Communication Monographs,
Balietti, S., Getoor, L., Goldstein, D. G., & Watts, D. J. (2021). 83(4), 446–467. https://doi.org/10.1080/03637751.2015.
Reducing opinion polarization: Effects of exposure to 1128555
similar people with differing political views. Proceedings Brown, M., Keefer, L. A., Sacco, D. F., & Bermond, A.
of the National Academy of Sciences, USA, 118(52), Article (2019). Is the cure a wall? Behavioral immune sys-
e2112552118. https://doi.org/10.1073/pnas.2112552118 tem responses to a disease metaphor for immigration.
Banuri, S., Dercon, S., & Gauri, V. (2019). Biased policy Evolutionary Psychological Science, 5, 343–356. https://
professionals. The World Bank Economic Review, 33(2), doi.org/10.1007/s40806-019-00191-3
310–327. https://doi.org/10.1093/wber/lhy033 Bruckmüller, S., & Abele, A. E. (2010). Comparison focus in
Bartlett, F. C. (1932). Remembering: A study in experimental intergroup comparisons: Who we compare to whom influ-
and social psychology. Cambridge University Press. ences who we see as powerful and agentic. Personality
Bateson, G. (1987). Steps to an ecology of mind: Collected essays and Social Psychology Bulletin, 36(10), 1424–1435.
in anthropology, psychiatry, evolution, and epistemology. https://doi.org/10.1177/0146167210383581
Jason Aronson Inc. (Original work published 1972) Bruckmüller, S., Hegarty, P., & Abele, A. E. (2012). Framing
Benitez, J., Leshin, R. A., & Rhodes, M. (2022). The influence gender differences: Linguistic normativity affects percep-
of linguistic form and causal explanations on the devel- tions of power and gender stereotypes. European Journal
opment of social essentialism. Cognition, 229, Article of Social Psychology, 42(2), 210–218. https://doi.org/
105246. https://doi.org/10.1016/j.cognition.2022.105246 10.1002/ejsp.858
Bergen, B. K. (2012). Louder than words: The new science of Brugman, B. C., Burgers, C., & Vis, B. (2019). Metaphorical
how the mind makes meaning. Basic Books. framing in political discourse through words vs. concepts:
Berger, J. (2023). Magic words: What to say to get your way. A meta-analysis. Language and Cognition, 11(1), 41–65.
Harper Business. https://doi.org/10.1111/comt.12096
Bermúdez, J. L. (2020). Frame it again: New tools for rational Bryan, C. J., Adams, G. S., & Monin, B. (2013). When cheating
decision-making. Cambridge University Press. would make you a cheater: Implicating the self prevents
Best, R., & Charness, N. (2015). Age differences in the effect of unethical behavior. Journal of Experimental Psychology:
framing on risky choice: A meta-analysis. Psychology and General, 142(4), 1001–1005. https://doi.org/10.1037/
Aging, 30(3), 688–698. https://doi.org/10.1037/a0039447 a0030655
Blankenship, K. L., & Craig, T. Y. (2011). Language use and Bryan, C. J., Master, A., & Walton, G. M. (2014). “Helping” ver-
persuasion: Multiple roles for linguistic styles. Social and sus “being a helper”: Invoking the self to increase helping
Psychological Science in the Public Interest 25(3) 147

in young children. Child Development, 85(5), 1836–1842. Chaiken, S. (1987). The heuristic model of persuasion. In M.
https://doi.org/10.1111/cdev.12244 P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social influ-
Bryan, C. J., Tipton, E., & Yeager, D. S. (2021). Behavioural ence: The Ontario symposium (Vol. 5, pp. 3–39). Lawrence
science is unlikely to change the world without a het- Erlbaum Associates.
erogeneity revolution. Nature Human Behaviour, 5(8), Chatruc, M. R., Stein, E., & Vlaicu, R. (2021). How issue
980–989. https://doi.org/10.1038/s41562-021-01143-3 framing shapes trade attitudes: Evidence from a multi-
Bryan, C. J., Walton, G. M., Rogers, T., & Dweck, C. S. (2011). country survey experiment. Journal of International
Motivating voter turnout by invoking the self. Proceedings Economics, 129, Article 103428. https://doi.org/10.1016/
of the National Academy of Sciences, USA, 108(31), 12653– j.jinteco.2021.103428
12656. https://doi.org/10.1073/pnas.1103343108 Chen, M., & Bell, R. A. (2022). A meta-analysis of the impact
Bryan, C. J., Yeager, D. S., & O’Brien, J. M. (2019). Replicator of point of view on narrative processing and persuasion
degrees of freedom allow publication of misleading fail- in health messaging. Psychology & Health, 37(5), 545–562.
ures to replicate. Proceedings of the National Academy https://doi.org/10.1080/08870446.2021.1894331
of Sciences, USA, 116(51), 25535–25545. https://doi.org/ Chen, M. K. (2013). The effect of language on economic
10.1038/s41562-021-01143-3 behavior: Evidence from savings rates, health behav-
Bullock, O. M., & Shulman, H. C. (2020). Framing. In J. iors, and retirement assets. American Economic Review,
Bulck (Ed.), The international encyclopedia of media 103(2), 690–731. https://doi.org/10.1257/aer.103.2.690
psychology. John Wiley & Sons. https://doi.org/ Cheng, F. F., Wu, C. S., & Lin, H. H. (2014). Reducing the
10.1002/9781119011071.iemp0268 influence of framing on internet consumers’ decisions:
Bullock, O. M., Shulman, H. C., & Huskey, R. (2021). Narratives The role of elaboration. Computers in Human Behavior,
are persuasive because they are easier to understand: 37, 56–63. https://doi.org/10.1016/j.chb.2014.04.015
Examining processing fluency as a mechanism of nar- Cheng, H., Burns, C., & Revie, M. (2022). Risky-choice fram-
rative persuasion. Frontiers in Communication, 188. ing and its null effect on integral emotions. Journal of
https://doi.org/10.3389/fcomm.2021.719615 Risk Research, 25(4), 453–467. https://doi.org/10.1080/1
Burgers, C., de Lavalette, K. Y. R., & Steen, G. J. (2018). 3669877.2021.1936611
Metaphor, hyperbole, and irony: Uses in isolation and in Chestnut, E. K., & Markman, E. M. (2018). “Girls are as good
combination in written discourse. Journal of Pragmatics, as boys at math” implies that boys are probably better: A
127, 71–83. https://doi.org/10.1016/j.pragma.2018.01.009 study of expressions of gender equality. Cognitive Science,
Burgers, C., Konijn, E. A., & Steen, G. J. (2016). Figurative 42(7), 2229–2249. https://doi.org/10.1111/cogs.12637
framing: Shaping public discourse through metaphor, Chestnut, E. K., Zhang, M. Y., & Markman, E. M. (2021). “Just
hyperbole, and irony. Communication Theory, 26(4), as good”: Learning gender stereotypes from attempts to
410–430. https://doi.org/10.1111/comt.12096 counteract them. Developmental Psychology, 57(1), 114–
Burgers, C., Konijn, E. A., Steen, G. J., & Iepsma, M. A. (2015). 125. https://doi.org/10.1037/dev0001143
Making ads less complex, yet more creative and persuasive: Chick, C. F., Reyna, V. F., & Corbin, J. C. (2016). Framing
The effects of conventional metaphors and irony in print effects are robust to linguistic disambiguation: A criti-
advertising. International Journal of Advertising, 34(3), cal test of contemporary theory. Journal of Experimental
515–532. https://doi.org/10.1080/02650487.2014.996200 Psychology: Learning, Memory, and Cognition, 42(2),
Cacciatore, M. A., Scheufele, D. A., & Iyengar, S. (2016). The 238–256. https://doi.org/10.1037/xlm0000158
end of framing as we know it . . . and the future of media Chong, D., & Druckman, J. N. (2007). A theory of framing
effects. Mass Communication and Society, 19(1), 7–23. and opinion formation in competitive elite environments.
https://doi.org/10.1080/15205436.2015.1068811 Journal of Communication, 57(1), 99–118. https://doi
Carmichael, L., Hogan, H. P., & Walter, A. A. (1932). An .org/10.1111/j.1460-2466.2006.00331_3.x
experimental study of the effect of language on the Chou, E. Y., & Murnighan, J. K. (2013). Life or death deci-
reproduction of visually perceived form. Journal of sions: Framing the call for help. PLOS ONE, 8(3), Article
Experimental Psychology, 15(1), 73–86. https://doi.org/ e57351. https://doi.org/10.1371/journal.pone.0057351
10.1037/h0072671 Christiansen, M. H., & Chater, N. (2022). The language game:
Carpenter, C. J. (2012). A meta-analysis and an experiment How improvisation created language and changed the
investigating the effects of speaker disfluency on persua- world. Basic Books.
sion. Western Journal of Communication, 76(5), 552–569. Christmann, U., & Göhring, A. L. (2016). A German-language
https://doi.org/10.1080/10570314.2012.662307 replication study analysing the role of figurative speech
Casasanto, D. (2005). Crying “Whorf.” Science, 307(5716), in reasoning. Scientific Data, 3, Article 160098. https://
1721–1722. https://doi.org/10.1126/science.307.5716.1721 doi.org/10.1038/sdata.2016.98
Casasanto, D. (2008). Who’s afraid of the big bad Whorf? Cialdini, R. B. (2001). The science of persuasion. Scientific
Crosslinguistic differences in temporal language and American, 284(2), 76–81. https://www.jstor.org/stable/
thought. Language Learning, 58(Suppl. 1), 63–79. https:// 26059056
doi.org/10.1111/j.1467-9922.2008.00462.x Cimpian, A., & Markman, E. M. (2011). The generic/nongeneric
Casasanto, D. (2016). A shared mechanism of linguistic, cul- distinction influences how children interpret new informa-
tural, and bodily relativity. Language Learning, 66(3), tion about social others. Child Development, 82(2), 471–
714–730. https://doi.org/10.1111/lang.12192 492. https://doi.org/10.1111/j.1467-8624.2010.01525.x
148 Flusberg et al.

Citron, F. M., & Goldberg, A. E. (2014). Metaphorical sen- Bilingualism: Language and Cognition, 25(4), 617–630.
tences are more emotionally engaging than their literal https://doi.org/10.1017/S1366728921001012
counterparts. Journal of Cognitive Neuroscience, 26(11), De Martino, B., Kumaran, D., Seymour, B., & Dolan, R. J.
2585–2595. https://doi.org/10.1162/jocn_a_00654 (2006). Frames, biases, and rational decision-making in
Clark, H. H. (1969). Influence of language on solving three- the human brain. Science, 313(5787), 684–687. https://
term series problems. Journal of Experimental Psychology, doi.org/10.1126/science.1128356
82(2), 205–215. https://doi.org/10.1037/h0028153 de Oliveira, G. S., & Baggs, E. (2023). Psychology’s WEIRD
Clark, H. H. (1996). Using language. Cambridge University problems. Cambridge University Press.
Press. Dillard, J. P., & Nabi, R. L. (2006). The persuasive influence
Claus, B. (2022). Framing effects: Three linguistic accounts and of emotion in cancer prevention and detection messages.
experimental evidence. In R. Hörnig, S. von Wietersheim, Journal of Communication, 56, S123–S139. https://doi
A. Konietzko, & S. Featherston (Eds.), Linguistic theory .org/10.1111/j.1460-2466.2006.00286.x
enriched by experimental data (pp. 421–442). University Dillard, J. P., & Seo, K. (2013). Affect and persuasion. In J. P.
of Tübingen. Dillard & L. Shen (Eds.), The SAGE handbook of persuasion:
Claypool, H. M., Mackie, D. M., & Garcia-Marques, T. (2015). Developments in theory and practice (pp. 150–166). Sage.
Fluency and attitudes. Social and Personality Psychology DiMaggio, P. (1997). Culture and cognition. Annual Review
Compass, 9(7), 370–382. https://doi.org/10.1111/ of Sociology, 23(1), 263–287. https://doi.org/10.1146/
spc3.12179 annurev.soc.23.1.263
Conner, M., & Norman, P. (2022). Understanding the inten- Dolcos, S., & Albarracín, D. (2014). The inner speech of
tion-behavior gap: The role of intention strength. Frontiers behavioral regulation: Intentions and task performance
in Psychology, 13, Article 923464. https://doi.org/10.3389/ strengthen when you talk to yourself as a you. European
fpsyg.2022.923464 Journal of Social Psychology, 44(6), 636–642. https://doi
Constantino, S. M., Sparkman, G., Kraft-Todd, G. T., Bicchieri, .org/10.1002/ejsp.2048
C., Centola, D., Shell-Duncan, B., Vogt, S., & Weber, E. Dolscheid, S., Çelik, S., Erkan, H., Küntay, A., & Majid, A.
U. (2022). Scaling up change: A critical review and practi- (2020). Space-pitch associations differ in their susceptibil-
cal guide to harnessing social norms for climate action. ity to language. Cognition, 196, Article 104073. https://
Psychological Science in the Public Interest, 23(2), 50–97. doi.org/10.1016/j.cognition.2019.104073
https://doi.org/10.1177/15291006221105279 Dolscheid, S., Shayan, S., Majid, A., & Casasanto, D. (2013).
Corneille, O., Havemann, J., Henderson, E. L., IJzerman, H., The thickness of musical pitch: Psychophysical evidence
Hussey, I., Orban de Xivry, J. J., Jussim, L., Holmes, N. P., for linguistic relativity. Psychological Science, 24(5), 613–
Pilacinski, A., Beffara, B., Carroll, H., Outa, N. O., Lush, 621. https://doi.org/10.1177/0956797612457374
P., & Lotter, L. D. (2023). Point of view: Beware “persua- Dor, D. (2015). The instruction of imagination: Language as a
sive communication devices” when writing and reading social communication technology. Oxford University Press.
scientific articles. eLife, 12, Article e88654. https://doi Druckman, J. N. (2001). On the limits of framing effects:
.org/10.7554/eLife.88654 Who can frame? The Journal of Politics, 63(4), 1041–1066.
Costa, A., Foucart, A., Arnon, I., Aparici, M., & Apesteguia, https://doi.org/10.1111/0022-3816.00100
J. (2014). “Piensa” twice: On the foreign language effect Druckman, J. N., & McDermott, R. (2008). Emotion and the
in decision making. Cognition, 130(2), 236–254. https:// framing of risky choice. Political Behavior, 30, 297–321.
doi.org/10.1016/j.cognition.2013.11.010 https://doi.org/10.1007/s11109-008-9056-y
Crano, W. D., & Prislin, R. (2006). Attitudes and persuasion. Druckman, J. N., & Nelson, K. R. (2003). Framing and delib-
Annual Review of Psychology, 57, 345–374. https://doi eration: How citizens’ conversations limit elite influence.
.org/10.1146/annurev.psych.57.102904.190034 American Journal of Political Science, 47(4), 729–745.
de Bruijn, H. (2019). The art of political framing: How https://doi.org/10.1111/1540-5907.00051
politicians convince us that they are right. Amsterdam Dunegan, K. J. (1996). Fines, frames, and images:
University Press. Examining formulation effects on punishment decisions.
De Graaf, A., Sanders, J., & Hoeken, H. (2016). Characteristics Organizational Behavior and Human Decision Processes,
of narrative interventions and health effects: A review 68(1), 58–67. https://doi.org/10.1006/obhd.1996.0089
of the content, form, and context of narratives in Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes.
health-related narrative persuasion research. Review of Harcourt Brace Jovanovich College Publishers.
Communication Research, 4, 88–131. https://doi.org/ Elliot, A. J., Eder, A. B., & Harmon-Jones, E. (2013). Approach–
10.12840/issn.2255-4165.2016.04.01.011 avoidance motivation and emotion: Convergence and
DeJesus, J. M., Callanan, M. A., Solis, G., & Gelman, S. divergence. Emotion Review, 5(3), 308–311. https://doi
A. (2019). Generic language in scientific communica- .org/10.1177/1754073913477517
tion. Proceedings of the National Academy of Sciences, Elliott, W. B., Rennekamp, K. M., & White, B. J. (2015). Does
USA, 116(37), 18370–18377. https://doi.org/10.1073/ concrete language in disclosures increase willingness to
pnas.1817706116 invest? Review of Accounting Studies, 20, 839–865. https://
Del Maschio, N., Crespi, F., Peressotti, F., Abutalebi, J., & doi.org/10.1007/s11142-014-9315-6
Sulpizio, S. (2022). Decision-making depends on lan- Elman, J. L. (2009). On the meaning of words and dinosaur
guage: A meta-analysis of the Foreign Language Effect. bones: Lexical knowledge without a lexicon. Cognitive
Psychological Science in the Public Interest 25(3) 149

Science, 33(4), 547–582. https://doi.org/10.1111/j.1551- Personality and Social Psychology Bulletin, 41(12), 1665–
6709.2009.01023.x 1681. https://doi.org/10.1177/0146167215607842
Elmore, K. C., & Luna-Lucero, M. (2017). Light bulbs or seeds? Feinberg, M., & Willer, R. (2019). Moral reframing: A tech-
How metaphors for ideas influence judgments about nique for effective and persuasive communication across
genius. Social Psychological and Personality Science, 8(2), political divides. Social and Personality Psychology
200–208. https://doi.org/10.1177/1948550616667611 Compass, 13(12), Article e12501. https://doi.org/10.1111/
Enfield, N. J. (2022). Language vs. reality: Why language is spc3.12501
good for lawyers and bad for scientists. MIT Press. Feygina, I., Jost, J. T., & Goldsmith, R. E. (2010). System
Entman, R. M. (1993). Framing: Toward clarification of a justification, the denial of global warming, and the pos-
fractured paradigm. Journal of Communication, 43(4), sibility of “system-sanctioned change.” Personality and
51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x Social Psychology Bulletin, 36(3), 326–338. https://doi
Epley, N., Mak, D., & Idson, L. C. (2006). Bonus of rebate? .org/10.1177/0146167209351435
The impact of income framing on spending and saving. Fillmore, C. J. (1982). Frame semantics. In Linguistic Society
Journal of Behavioral Decision Making, 19(3), 213–227. of Korea (Ed.), Linguistics in the morning calm (pp. 111–
https://doi.org/10.1002/bdm.519 137). Hanshin Publishing Co.
Evans, J. S. B. (2008). Dual-processing accounts of reason- Fillmore, C. J. (2008). The merging of “frames.” In R. Rossini
ing, judgment, and social cognition. Annual Review of Favretti (Ed.), Frames, corpora, and knowledge represen-
Psychology, 59, 255–278. https://doi.org/10.1146/annurev. tation (pp. 1–12). Bononia University Press.
psych.59.103006.093629 Fillmore, C. J., & Baker, C. (2009). A frames approach to
Evans, J. S. B., & Stanovich, K. E. (2013). Dual-process theories semantic analysis. In B. Heine & H. Narrog (Eds.), The
of higher cognition: Advancing the debate. Perspectives Oxford handbook of linguistic analysis (pp. 313–340).
on Psychological Science, 8(3), 223–241. https://doi Oxford University Press.
.org/10.1177/1745691612460685 Fitzsimons, G. M., & Kay, A. C. (2004). Language and inter-
Evans, N., & Levinson, S. C. (2009). The myth of language personal cognition: Causal effects of variations in pro-
universals: Language diversity and its importance for noun usage on perceptions of closeness. Personality and
cognitive science. Behavioral and Brain Sciences, 32(5), Social Psychology Bulletin, 30(5), 547–557. https://doi
429–448. https://doi.org/10.1017/S0140525X0999094X .org/10.1177/0146167203262852
Farrow, K., Grolleau, G., & Mzoughi, N. (2018). What in Flexas, A., López-Penadés, R., Aguilar-Mediavilla, E., &
the word! The scope for the effect of word choice on Adrover-Roig, D. (2023). Meanness trumps language:
economic behavior. Kyklos, 71(4), 557–580. https://doi Lack of foreign language effect in early bilinguals’ moral
.org/10.1111/kykl.12186 choices. PLOS ONE, 18(11), Article e0294523. https://doi
Fausey, C. M., & Boroditsky, L. (2010). Subtle linguistic .org/10.1371/journal.pone.0294523
cues influence perceived blame and financial liability. Flusberg, S. J., DellaValle, M., & Thibodeau, P. H. (2023).
Psychonomic Bulletin & Review, 17(5), 644–650. https:// How metaphor and political ideology shape lay theo-
doi.org/10.3758/PBR.17.5.644 ries of mental disorders. Journal of Social & Clinical
Fausey, C. M., & Boroditsky, L. (2011). Who dunnit? Cross- Psychology, 42(4), 293–322. https://doi.org/10.1521/
linguistic differences in eye-witness memory. Psychonomic jscp.2023.42.4.293
Bulletin & Review, 18, 150–157. https://doi.org/10.3758/ Flusberg, S. J., Lauria, M., Balko, S., & Thibodeau, P. H.
s13423-010-0021-5 (2020). Effects of communication modality and speaker
Fausey, C. M., Long, B. L., Inamori, A., & Boroditsky, L. (2010). identity on metaphor framing. Metaphor & Symbol,
Constructing agency: The role of language. Frontiers 35(2), 136–152. https://doi.org/10.1080/10926488.2020
in Psychology, 1, Article 162. https://doi.org/10.3389/ .1767336
fpsyg.2010.00162 Flusberg, S. J., Mackey, A., & Semino, E. (2024). Seatbelts and
Fausey, C. M., & Matlock, T. (2011). Can grammar win elec- raincoats, or banks and castles: Investigating the impact
tions? Political Psychology, 32(4), 563–574. https://doi of vaccine metaphors. PLOS ONE, 19(1), Article e0294739.
.org/10.1111/j.1467-9221.2010.00802.x https://doi.org/10.1371/journal.pone.0294739
Fazio, R. H. (1986). How do attitudes guide behavior? In R. Flusberg, S. J., Matlock, T., & Thibodeau, P. H. (2017).
M. Sorrentino & E. T. Higgins (Eds.), Handbook of motiva- Metaphors for the war (or race) against climate change.
tion and cognition: Foundations of social behavior (pp. Environmental Communication, 11(6), 769–783. https://
204–243). Guilford Press. doi.org/10.1080/17524032.2017.1289111
Feinberg, M., & Willer, R. (2011). Apocalypse soon? Dire mes- Flusberg, S. J., Matlock, T., & Thibodeau, P. H. (2018). War
sages reduce belief in global warming by contradicting metaphors in public discourse. Metaphor & Symbol, 33(1),
just-world beliefs. Psychological Science, 22(1), 34–38. 1–18. https://doi.org/10.1080/10926488.2018.1407992
https://doi.org/10.1177/0956797610391911 Flusberg, S. J., & Thibodeau, P. H. (2023). Why is mother earth
Feinberg, M., & Willer, R. (2013). The moral roots of envi- on life support? Metaphors in environmental discourse.
ronmental attitudes. Psychological Science, 24(1), 56–62. Topics in Cognitive Science, 15(3), 522–545. https://doi
https://doi.org/10.1177/0956797612449177 .org/10.1111/tops.12651
Feinberg, M., & Willer, R. (2015). From gulf to bridge: Flusberg, S. J., Thibodeau, P. H., & Holmes, K. J. (2022).
When do moral arguments facilitate political influence? Even simple framing effects are rational. Commentary on
150 Flusberg et al.

Bermúdez. Behavioral & Brain Sciences, 45, Article e228. cues might not affect voter behavior. Proceedings of the
https://doi.org/10.1017/S0140525X22000942 National Academy of Sciences, USA, 113(26), 7112–7117.
Flusberg, S. J., van der Vord, J., Husney, S. Q., & Holmes, https://doi.org/10.1073/pnas.1513727113
K. J. (2022). Who’s the “real” victim? How vic- Gerber, A. S., Huber, G. A., & Fang, A. (2018). Do subtle lin-
tim framing shapes attitudes toward sexual assault. guistic interventions priming a social identity as a voter
Psychological Science, 33(4), 524–537. https://doi.org/ have outsized effects on voter turnout? Evidence from a
10.1177/09567976211045935 new replication experiment. Political Psychology, 39(4),
Franks, A. S., & Scherr, K. C. (2019). Economic issues are 925–938. https://doi.org/10.1111/pops.l2446
moral issues: The moral underpinnings of the desire Gerber, A. S., Huber, G. A., & Fang, A. H. (2023). Voting
to reduce wealth inequality. Social Psychological and behavior is unaffected by subtle linguistic cues: Evidence
Personality Science, 10(4), 553–562. https://doi.org/ from a psychologically authentic replication. Behavioural
10.1177/194855061877282 Public Policy, 7(2), 380–394. https://doi.org/10.1017/
Freling, T. H., Vincent, L. H., & Henard, D. H. (2014). When bpp.2020.57
not to accentuate the positive: Re-examining valence Gerend, M. A., & Shepherd, J. E. (2007). Using message fram-
effects in attribute framing. Organizational Behavior and ing to promote acceptance of the human papillomavirus
Human Decision Processes, 124(2), 95–109. https://doi vaccine. Health Psychology, 26(6), 745–752. https://doi
.org/10.1016/j.obhdp.2013.12.007 .org/10.1037/0278-6133.26.6.745
Freling, T. H., Yang, Z., Saini, R., Itani, O. S., & Abualsamh, R. Gibbons, P., Busch, J., & Bradac, J. J. (1991). Powerful ver-
R. (2020). When poignant stories outweigh cold hard facts: sus powerless language: Consequences for persuasion,
A meta-analysis of the anecdotal bias. Organizational impression formation, and cognitive response. Journal of
Behavior and Human Decision Processes, 160, 51–67. Language and Social Psychology, 10(2), 115–133. https://
https://doi.org/10.1016/j.obhdp.2020.01.006 doi.org/10.1177/0261927X91102003
Frisch, D. (1993). Reasons for framing effects. Organizational Gibbs, R. W. (2017). Metaphor wars. Cambridge University
Behavior and Human Decision Processes, 54(3), 399–429. Press.
https://doi.org/10.1006/obhd.1993.1017 Gigerenzer, G. (2018). The bias bias in behavioral economics.
Gallagher, K. M., & Updegraff, J. A. (2012). Health message Review of Behavioral Economics, 5(3–4), 303–336. https://
framing effects on attitudes, intentions, and behavior: doi.org/10.1561/105.00000092
A meta-analytic review. Annals of Behavioral Medicine, Gleitman, L., & Papafragou, A. (2012). New perspectives on
43(1), 101–116. https://doi.org/10.1007/s12160-012-9446-6 language and thought. In K. J. Holyoak & R. G. Morrison
Gaucher, D., Friesen, J., & Kay, A. C. (2011). Evidence that (Eds.), The Oxford handbook of thinking and reason-
gendered wording in job advertisements exists and sus- ing (pp. 543–568). Oxford University Press. https://doi.
tains gender inequality. Journal of Personality and Social org/10.1093/oxfordhb/9780199734689.013.0028
Psychology, 101(1), 109–128. https://doi.org/10.1037/ Goffman, E. (1974). Frame analysis: An essay on the organiza-
a0022530 tion of experience. Harvard University Press.
Geiger, N., Dwyer, T., & Swim, J. K. (2023). Hopium or Goldschmied, N., Sheptock, M., Kim, K., & Galily, Y. (2017).
empowering hope? A meta-analysis of hope and climate Appraising Loftus and Palmer (2017) post-event infor-
engagement. Frontiers in Psychology, 14, Article 1139427. mation versus concurrent commentary in the context
https://doi.org/10.3389/fpsyg.2023.1139427 of sport. Quarterly Journal of Experimental Psychology,
Gelman, S. A. (2021). Generics in society. Language 70(11), 2347–2356. https://doi.org/10.1080/17470218.20
in Society, 50(4), 517–532. https://doi.org/10.1017/ 16.1237980
S0047404521000282 Goldstein, D. G., Hershfield, H. E., & Benartzi, S. (2016). The
Gelman, S. A., & Heyman, G. D. (1999). Carrot-eaters and illusion of wealth and its reversal. Journal of Marketing
creature-believers: The effects of lexicalization on chil- Research, 53(5), 804–813. https://doi.org/10.1509/
dren’s inferences about social categories. Psychological jmr.14.0652
Science, 10(6), 489–493. https://doi.org/10.1111/1467- Goodman, N. D., & Frank, M. C. (2016). Pragmatic lan-
9280.00194 guage interpretation as probabilistic inference. Trends
Gelman, S. A., & Roberts, S. O. (2017). How language shapes in Cognitive Sciences, 20(11), 818–829. https://doi.org/
the cultural inheritance of categories. Proceedings of the 10.1016/j.tics.2016.08.005
National Academy of Sciences, USA, 114(30), 7900–7907. Gourville, J. T. (1998). Pennies-a-day: The effect of temporal
https://doi.org/10.1073/pnas.1621073114 reframing on transaction evaluation. Journal of Consumer
Gentner, D. (1983). Structure-mapping: A theoretical frame- Research, 24(4), 395–408. https://doi.org/10.1086/209517
work for analogy. Cognitive Science, 7(2), 155–170. Graesser, A. C., Millis, K. K., & Zwaan, R. A. (1997). Discourse
https://doi.org/10.1016/S0364-0213(83)80009-3 comprehension. Annual Review of Psychology, 48(1),
Gentner, D., & Goldin-Meadow, S. (2003). Whither Whorf. 163–189. https://doi.org/10.1146/annurev.psych.48.1.163
In D. Gentner & S. Goldin-Meadow (Eds.), Language in Grice, H. P. (1975). Logic and conversation. In P. Cole & J.
mind: Advances in the study of language and thought Morgan (Eds.), Syntax and semantics (Vol. 3, pp. 41–58).
(pp. 3–14). MIT Press. Academic Press.
Gerber, A. S., Huber, G. A., Biggers, D. R., & Hendry, D. J. Grosjean, F. (2010). Bilingual: Life and reality. Harvard
(2016). A field experiment shows that subtle linguistic University Press.
Psychological Science in the Public Interest 25(3) 151

Guy, S., Kashima, Y., Walker, I., & O’Neill, S. (2013). in Cognitive Sciences, 20(11), 791–793. https://doi.org/
Comparing the atmosphere to a bathtub: Effectiveness 10.1016/j.tics.2016.08.004
of analogy for reasoning about accumulation. Climatic Hendricks, R. K., Demjén, Z., Semino, E., & Boroditsky, L.
Change, 121(4), 579–594. https://doi.org/10.1007/s10584- (2018). Emotional implications of metaphor: Consequences
013-0949-3 of metaphor framing for mindset about cancer. Metaphor
Hallahan, K. (1999). Seven models of framing: Implications and Symbol, 33(4), 267–279. https://doi.org/10.1080/109
for public relations. Journal of Public Relations 26488.2018.1549835
Research, 11(3), 205–242. https://doi.org/10.1207/ Henrich, J., Heine, S., & Norenzayan, A. (2010). The weirdest
s1532754xjprr1103_02 people in the world? Behavioral and Brain Sciences, 33(2–
Hamilton, M. A., Hunter, J. E., & Burgoon, M. (1990). An 3), 61–83. https://doi.org/10.1017/S0140525X0999152X
empirical test of an axiomatic model of the relationship Heritage, J., Robinson, J. D., Elliott, M. N., Beckett, M., &
between language intensity and persuasion. Journal of Wilkes, M. (2007). Reducing patients’ unmet concerns in
Language and Social Psychology, 9(4), 235–255. https:// primary care: The difference one word can make. Journal
doi.org/10.1177/0261927X9094002 of General Internal Medicine, 22, 1429–1433. https://doi
Hanawalt, N. G., & Demarest, I. H. (1939). The effect of .org/10.1007/s11606-007-0279-0
verbal suggestion in the recall period upon the reproduc- Hershfield, H. E., Shu, S., & Benartzi, S. (2020). Temporal
tion of visually perceived forms. Journal of Experimental reframing and participation in a savings program: A field
Psychology, 25(2), 159–174. https://doi.org/10.1037/ experiment. Marketing Science, 39(6), 1039–1051. https://
h0057682 doi.org/10.1287/mksc.2019.1177
Hardisty, D. J., Johnson, E. J., & Weber, E. U. (2010). A Hodgkinson, G. P., Bown, N. J., Maule, A. J., Glaister, K. W., &
dirty word or a dirty world? Attribute framing, political Pearman, A. D. (1999). Breaking the frame: An analysis of
affiliation, and query theory. Psychological Science, 21(1), strategic cognition and decision making under uncertainty.
86–92. https://doi.org/10.1177/0956797609355572 Strategic Management Journal, 20(10), 977–985. https://
Harris, R. J. (1973). Answering questions containing marked doi.org/10.1002/(SICI)1097-0266(199910)20:10<977::AID-
and unmarked adjectives and adverbs. Journal of SMJ58>3.0.CO;2-X
Experimental Psychology, 97(3), 399–401. https://doi Hoewe, J. (2020). Toward a theory of media priming. Annals
.org/10.1037/h0034165 of the International Communication Association, 44(4),
Hart, C. (2021). Animals vs. armies: Resistance to extreme 312–321. https://doi.org/10.1080/23808985.2020.1815232
metaphors in anti-immigration discourse. Journal of Holmes, K. J., Doherty, E., & Flusberg, S. J. (2022). How
Language and Politics, 20(2), 226–253. https://doi.org/ and when does syntax perpetuate stereotypes? Probing
10.1075/jlp.20032.har the framing effects of subject-complement statements of
Hauser, D. J., & Schwarz, N. (2015). The war on prevention: equality. Thinking & Reasoning, 28(2), 226–260. https://
Bellicose cancer metaphors hurt (some) prevention inten- doi.org/10.1080/13546783.2021.1963841
tions. Personality and Social Psychology Bulletin, 41(1), Holmes, K. J., Flusberg, S. J., & Thibodeau, P. H. (2018).
66–77. https://doi.org/10.1177/0146167214557006 Compound words reflect cross-culturally shared bodily
Hauser, D. J., & Schwarz, N. (2016). Semantic prosody and metaphors. Cognitive Science, 42(8), 3071–3082. https://
judgment. Journal of Experimental Psychology: General, doi.org/10.1111/cogs.12671
145(7), 882–896. https://doi.org/10.1037/xge0000178 Holmes, K. J., Kassin, L., Buchillon-Almeida, D., & Canseco-
Hauser, D. J., & Schwarz, N. (2018). How seemingly innocuous Gonzalez, E. (2024). Linguistic distancing in emotion
words can bias judgment: Semantic prosody and impres- regulation: Evidence for cross-linguistically shared and
sion formation. Journal of Experimental Social Psychology, language-specific signatures. [Manuscript submitted for
75, 11–18. https://doi.org/10.1016/j.jesp.2017.10.012 publication]. Department of Psychology, Reed College.
Hauser, D. J., & Schwarz, N. (2020). The war on preven- Holmes, K. J., & Wolff, P. (2013). Spatial language and the psy-
tion II: Battle metaphors undermine cancer treatment chological reality of schematization. Cognitive Processing,
and prevention and do not increase vigilance. Health 14, 205–208. https://doi.org/10.1007/s10339-013-0545-5
Communication, 35(13), 1698–1704. https://doi.org/10 Holmes, K. J., Wu, S. H., Elpers, N., Doherty, E. M., & Flusberg,
.1080/10410236.2019.1663465 S. J. (2024). How syntax promotes stereotypes: Assessing
Hauser, D. J., & Schwarz, N. (2023). Semantic prosody: the role of pragmatic inference [Manuscript submitted for
How neutral words with collocational positivity/nega- publication]. Department of Psychology, Reed College.
tivity color evaluative judgments. Current Directions in Holtgraves, T., & Lasky, B. (1999). Linguistic power and per-
Psychological Science, 32(2), 98–104. https://doi.org/ suasion. Journal of Language and Social Psychology, 18(2),
10.1177/09637214221127978 196–205. https://doi.org/10.1177/0261927X99018002004
Hauser, M. D., Chomsky, N., & Fitch, W. T. (2002). The faculty Holyoak, K. J., & Stamenković, D. (2018). Metaphor com-
of language: What is it, who has it, and how did it evolve? prehension: A critical review of theories and evidence.
Science, 298(5598), 1569–1579. https://doi.org/10.1126/ Psychological Bulletin, 144(6), 641–671. https://psycnet
science.298.5598.1569 .apa.org/doi/10.1037/bul0000145
Hayakawa, S., Costa, A., Foucart, A., & Keysar, B. (2016). Homar, A. R., & Cvelbar, L. K. (2021). The effects of fram-
Using a foreign language changes our choices. Trends ing on environmental decisions: A systematic literature
152 Flusberg et al.

review. Ecological Economics, 183, Article 106950. https:// Joyal-Desmarais, K., Scharmer, A. K., Madzelan, M. K., See,
doi.org/10.1016/j.ecolecon.2021.106950 J. V., Rothman, A. J., & Snyder, M. (2022). Appealing to
Hommel, B., Pratt, J., Colzato, L., & Godijn, R. (2001). Symbolic motivation to change attitudes, intentions, and behavior:
control of visual attention. Psychological Science, 12(5), A systematic review and meta-analysis of 702 experimen-
360–365. https://doi.org/10.1111/1467-9280.00367 tal tests of the effects of motivational message matching
Hornsey, M. J., & Fielding, K. S. (2016). A cautionary note on persuasion. Psychological Bulletin, 148(7–8), 465–517.
about messages of hope: Focusing on progress in reduc- https://doi.org/10.1037/bul0000377
ing carbon emissions weakens mitigation motivation. Kahneman, D. (2011). Thinking, fast and slow. Macmillan.
Global Environmental Change, 39, 26–34. https://doi Kassin, L., Castellano, A., Canseco-Gonzalez, E., & Holmes, K.
.org/10.1016/j.gloenvcha.2016.04.003 J. (2023, July 26–29). Linguistic distancing and emotion
Hornsey, M. J., & Fielding, K. S. (2020). Understanding (and regulation in English and Spanish [Paper presentation].
reducing) inaction on climate change. Social Issues and 45th Annual Meeting of the Cognitive Science Society,
Policy Review, 14(1), 3–35. https://doi.org/10.1111/ Sydney, Australia.
sipr.12058 Keefer, L. A., Landau, M. J., Sullivan, D., & Rothschild, Z. K.
Hosman, L. A., & Siltanen, S. A. (2006). Powerful and pow- (2014). Embodied metaphor and abstract problem solv-
erless language forms: Their consequences for impres- ing: Testing a metaphoric fit hypothesis in the health
sion formation, attributions of control of self and control domain. Journal of Experimental Social Psychology, 55,
of others, cognitive responses, and message memory. 12–20. https://doi.org/10.1016/j.jesp.2014.05.012
Journal of Language and Social Psychology, 25(1), 33–46. Kellermann, K. (2007, September 27–30). Persuasive question
https://doi.org/10.1177/0261927X05284477 asking: How question wording influences answers [Paper
Howard, D. J., & Kerin, R. A. (2011). The effects of name presentation]. Annual Meeting of the State Bar Association
similarity on message processing and persuasion. Journal of California, Anaheim, CA, United States.
of Experimental Social Psychology, 47(1), 63–71. https:// Keysar, B., Hayakawa, S. L., & An, S. G. (2012). The foreign-
doi.org/10.1016/j.jesp.2010.08.008 language effect: Thinking in a foreign tongue reduces
Huette, S., Winter, B., Matlock, T., Ardell, D. H., & Spivey, decision biases. Psychological Science, 23(6), 661–668.
M. (2014). Eye movements during listening reveal https://doi.org/10.1177/0956797611432178
spontaneous grammatical processing. Frontiers in Psychology, Kidwell, B., Farmer, A., & Hardesty, D. M. (2013). Getting
5, Article 410. https://doi.org/10.3389/fpsyg.2014.00410 liberals and conservatives to go green: Political ideology
Huizenga, H. M., Zadelaar, J. N., Jansen, B. R. J., Olthof, M. C., and congruent appeals. Journal of Consumer Research,
Steingroever, H., Dekkers, L. M. S., van Duijvenvoorde, 40(2), 350–367. https://doi.org/10.1086/670610
A. C. K., Figner, B., & Agelink van Rentergem, J. (2023). Kim, H., Rao, A. R., & Lee, A. Y. (2009). It’s time to vote:
Formal models of differential framing effects in decision The effect of matching message orientation and tempo-
making under risk. Decision, 10(3), 197–234. https://doi ral frame on political persuasion. Journal of Consumer
.org/10.1037/dec0000201 Research, 35(6), 877–889. https://doi.org/10.1086/593700
Hurst, K., & Stern, M. J. (2020). Messaging for environmental Kim, H. J., & Jang, J. M. (2018). The easier the better: How
action: The role of moral framing and message source. processing fluency influences self-efficacy and behav-
Journal of Environmental Psychology, 68, Article 101394. ioral intention in pro-social campaign advertising.
https://doi.org/10.1016/j.jenvp.2020.101394 Sustainability, 10(12), 4777. https://doi.org/10.3390/
Igou, E. R., & Bless, H. (2007). On undesirable consequences su10124777
of thinking: Framing effects as a function of substantive Kintsch, W., & van Dijk, T. A. (1978). Toward a model of
processing. Journal of Behavioral Decision Making, 20(2), text comprehension and production. Psychological
125–142. https://doi.org/10.1002/bdm.543 Review, 85(5), 363–394. https://doi.org/10.1037/0033-
Inbar, Y., & Evers, E. (2022). Worse is bad: Divergent infer- 295X.85.5.363
ences from logically equivalent comparisons. Journal Kita, S., & Emmorey, K. (2023). Gesture links language and
of Experimental Psychology: General, 151(3), 665–675. cognition for spoken and signed languages. Nature
https://doi.org/10.1037/xge0000804 Reviews Psychology, 2, 407–420. https://doi.org/10.1038/
Iyengar, S. (1991). Is anyone responsible? How television s44159-023-00186-9
frames political issues. University of Chicago Press. Knowlton, T., Pietroski, P., Halberda, J., & Lidz, J. (2022). The
January, D., & Kako, E. (2007). Re-evaluating evidence mental representation of universal quantifiers. Linguistics
for linguistic relativity: Reply to Boroditsky (2001). and Philosophy, 45(4), 911–941. https://doi.org/10.1007/
Cognition, 104(2), 417–426. https://doi.org/10.1016/ s10988-021-09337-8
j.cognition.2006.07.008 Krefeld-Schwalb, A., Sugerman, E. R., & Johnson, E. J. (2024).
Johnson-Laird, P. N. (1983). Mental models: Towards a cog- Exposing omitted moderators: Explaining why effect sizes
nitive science of language, inference, and consciousness. differ in the social sciences. Proceedings of the National
Harvard University Press. Academy of Sciences, USA, 121(12), Article e2306281121.
Joshi, P. D., & Wakslak, C. J. (2014). Communicating with the https://doi.org/10.1073/pnas.2306281121
crowd: Speakers use abstract messages when address- Kreiner, H., & Gamliel, E. (2018). The role of attention in
ing larger audiences. Journal of Experimental Psychology: attribute framing. Journal of Behavioral Decision Making,
General, 143(1), 351–362. https://doi.org/10.1037/a0032413 31(3), 392–401. https://doi.org/10.1002/bdm.2067
Psychological Science in the Public Interest 25(3) 153

Kross, E., Bruehlman-Senecal, E., Park, J., Burson, A., Psychology, 53, 125–138. https://doi.org/10.1016/j.jesp
Dougherty, A., Shablack, H., Bremner, R., Moser, J., & .2014.03.009
Ayduk, O. (2014). Self-talk as a regulatory mechanism: Landau, M. J., Sullivan, D., & Greenberg, J. (2009). Evidence
How you do it matters. Journal of Personality and Social that self-relevant motives and metaphoric framing interact
Psychology, 106(2), 304–324. https://doi.org/10.1037/ to influence political and social attitudes. Psychological
a0035173 Science, 20(11), 1421–1427. https://doi.org/10.1111/
Kross, E., Vickers, B. D., Orvell, A., Gainsburg, I., Moran, j.1467-9280.2009.02462.x
T. P., Boyer, M., Jonides, J., Moser, J., & Ayduk, O. LeBoeuf, R. A., & Shafir, E. (2003). Deep thoughts and shallow
(2017). Third-person self-talk reduces Ebola worry and frames: On the susceptibility to framing effects. Journal
risk perception by enhancing rational thinking. Applied of Behavioral Decision Making, 16(2), 77–92. https://doi
Psychology: Health and Well-Being, 9(3), 387–409. https:// .org/10.1002/bdm.433
doi.org/10.1111/aphw.12103 Lecheler, S., & de Vreese, C. H. (2011). Getting real: The dura-
Kruglanski, A. W., & Thompson, E. P. (1999). Persuasion by tion of framing effects. Journal of Communication, 61(5),
a single route: A view from the unimodel. Psychological 959–983. https://doi.org/10.1111/j.1460-2466.2011.01580.x
Inquiry, 10(2), 83–109. https://doi.org/10.1207/ Lecheler, S., & de Vreese, C. H. (2019). News Framing effects:
S15327965PL100201 Theory and Practice. Routledge.
Kühberger, A. (1995). The framing of decisions: A new look Lecheler, S., de Vreese, C. H., & Slothuus, R. (2009).
at old problems. Organizational Behavior and Human Issue importance as a moderator of framing effects.
Decision Processes, 62(2), 230–240. https://doi.org/ Communication Research, 36(3), 400–425. https://doi
10.1006/obhd.1995.1046 .org/10.1177/0093650209333028
Kühberger, A. (1998). The influence of framing on risky deci- Lecheler, S., Schuck, A. R., & De Vreese, C. H. (2013). Dealing
sions: A meta-analysis. Organizational Behavior and with feelings: Positive and negative discrete emotions as
Human Decision Processes, 75(1), 23–55. https://doi mediators of news framing effects. Communications, 38(2),
.org/10.1006/obhd.1998.2781 189–209. https://doi.org/10.1515/commun-2013-0011
Kühberger, A., & Tanner, C. (2010). Risky choice framing: Lee, A. Y., & Aaker, J. L. (2004). Bringing the frame into
Task versions and a comparison of prospect theory focus: The influence of regulatory fit on processing flu-
and fuzzy-trace theory. Journal of Behavioral Decision ency and persuasion. Journal of Personality and Social
Making, 23(3), 314–329. https://doi.org/10.1002/bdm.656 Psychology, 86(2), 205–218. https://doi.org/10.1037/0022-
Laham, S. M., Koval, P., & Alter, A. L. (2012). The name- 3514.86.2.205
pronunciation effect: Why people like Mr. Smith more Lee, S. J. (2019). The role of construal level in message effects
than Mr. Colquhoun. Journal of Experimental Social research: A review and future directions. Communication
Psychology, 48(3), 752–756. https://doi.org/10.1016/ Theory, 29(3), 319–338. https://doi.org/10.1093/ct/qty030
j.jesp.2011.12.002 Lenneberg, E. H. (1953). Cognition in ethnolinguistics.
Lakoff, G. (1987). Women, fire, and dangerous things. Language, 29(4), 463–471. https://doi.org/10.2307/409956
University of Chicago Press. Leong, L. M., McKenzie, C. R., Sher, S., & Müller-Trede, J.
Lakoff, G. (1996). Moral politics: What conservatives know that (2017). The role of inference in attribute framing effects.
liberals don’t. University of Chicago Press. Journal of Behavioral Decision Making, 30(5), 1147–1156.
Lakoff, G. (2008). The political mind: A cognitive scientist’s https://doi.org/10.1002/bdm.2030
guide to your brain and its politics. Penguin. Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward
Lakoff, G. (2010). Why it matters how we frame the envi- a model of emotion-specific influences on judgment and
ronment. Environmental Communication, 4(1), 70–81. choice. Cognition & Emotion, 14(4), 473–493. https://doi
https://doi.org/10.1080/17524030903529749 .org/10.1080/026999300402763
Lakoff, G. (2014). The all new don’t think of an elephant! Lerner, J. S., Li, Y., Valdesolo, P., & Kassam, K. S. (2015).
Know your values and frame the debate. Chelsea Green Emotion and decision making. Annual Review of
Publishing. Psychology, 66(1), 799–823. https://doi.org/10.1146/
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. annurev-psych-010213-115043
University of Chicago Press. Levin, I. P. (1987). Associative effects of information framing.
Lammers, J., Schulte, A., & Baldwin, M. (2023). Does framing Bulletin of the Psychonomic Society, 25(2), 85–86. https://
climate change policies to fit with epistemic needs for doi.org/10.3758/BF03330291
predictability reduce conservatives’ opposition? Analyses Levin, I. P., & Gaeth, G. J. (1988). How consumers are affected
of Social Issues and Public Policy, 23(3), 571–591. https:// by the framing of attribute information before and after
doi.org/10.1111/asap.12362 consuming the product. Journal of Consumer Research,
Landau, M. J., & Keefer, L. A. (2014). This is like that: 15(3), 374–378. https://doi.org/10.1086/209174
Metaphors in public discourse shape attitudes. Social and Levin, I. P., Gaeth, G. J., Schreiber, J., & Lauriola, M. (2002).
Personality Psychology Compass, 8(8), 463–473. https:// A new look at framing effects: Distribution of effect sizes,
doi.org/10.1111/spc3.12125 individual differences, and independence of types of
Landau, M. J., Keefer, L. A., & Rothschild, Z. K. (2014). effects. Organizational Behavior and Human Decision
Epistemic motives moderate the effect of metaphoric Processes, 88(1), 411–429. https://doi.org/10.1006/
framing on attitudes. Journal of Experimental Social obhd.2001.2983
154 Flusberg et al.

Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames Lupyan, G. (2008). From chair to “chair”: A representational
are not created equal: A typology and critical analysis shift account of object labeling effects on memory.
of framing effects. Organizational Behavior and Human Journal of Experimental Psychology: General, 137(2),
Decision Processes, 76(2), 149–188. https://doi.org/ 348–369. https://doi.org/10.1037/0096-3445.137.2.348
10.1006/obhd.1998.2804 Lupyan, G., Rahman, R. A., Boroditsky, L., & Clark, A.
Li, N., & Su, L. Y. (2018). Message framing and climate (2020). Effects of language on visual perception. Trends
change communication: A meta-analytical review. Journal in Cognitive Sciences, 24(11), 930–944. https://doi.org/
of Applied Communications, 102(3). https://doi.org/ 10.1016/j.tics.2020.08.005
10.4148/1051-0834.2189 Lupyan, G., & Spivey, M. J. (2010a). Making the invisible vis-
Li, P., & Gleitman, L. (2002). Turning the tables: Language ible: Verbal but not visual cues enhance visual detection.
and spatial reasoning. Cognition, 83(3), 265–294. https:// PLOS ONE, 5(7), Article e11452. https://doi.org/10.1371/
doi.org/10.1016/S0010-0277(02)00009-4 journal.pone.0011452
Liberman, N., Trope, Y., & Stephan, E. (2007). Psychological Lupyan, G., & Spivey, M. J. (2010b). Redundant spoken
distance. In A. W. Kruglanski & E. T. Higgins (Eds.), labels facilitate perception of multiple items. Attention,
Social psychology: Handbook of basic principles (pp. 353– Perception, & Psychophysics, 72(8), 2236–2253. https://
381). Guilford Press. doi.org/10.3758/BF03196698
Lipscomb, T. J., Bregman, N. J., & McAllister, H. A. (1985). A Lupyan, G., & Ward, E. J. (2013). Language can boost other-
developmental inquiry into the effects of postevent infor- wise unseen objects into visual awareness. Proceedings of
mation on eyewitness accounts. The Journal of Genetic the National Academy of Sciences, USA, 110(35), 14196–
Psychology, 146(4), 551–556. https://doi.org/10.1080/00 14201. https://doi.org/10.1073/pnas.1303312110
221325.1985.10532475 Lupyan, G., & Winter, B. (2018). Language is more abstract
Lipscomb, T. J., McAllister, H. A., & Bregman, N. J. (1985). than you think, or, why aren’t languages more iconic?
Bias in eyewitness accounts: The effects of question Philosophical Transactions of the Royal Society B:
format, delay interval, and stimulus presentation. The Biological Sciences, 373(1752), 20170137. https://doi.
Journal of Psychology, 119(3), 207–212. https://doi.org org/10.1098/rstb.2017.0137
/10.1080/00223980.1985.10542888 Macnamara, B. N., & Burgoyne, A. P. (2023). Do growth mind-
Liu, C., Wang, H., Timmer, K., & Jiao, L. (2022). The foreign set interventions impact students’ academic achievement?
language effect on altruistic decision making: Insights A systematic review and meta-analysis with recommenda-
from the framing effect. Bilingualism: Language and tions for best practices. Psychological Bulletin, 149(3–4),
Cognition, 25(5), 890–898. https://doi.org/10.1017/ 133–173. https://doi.org/10.1037/bul0000352
S1366728922000128 Magaña, D., & Matlock, T. (2018). How Spanish speakers
Loewenstein, G. F., Weber, E. U., Hsee, C. K., & Welch, N. use metaphor to describe their experiences with cancer.
(2001). Risk as feelings. Psychological Bulletin, 127(2), Discourse & Communication, 12(6), 627–644. https://doi
267–286. https://doi.org/10.1037/0033-2909.127.2.267 .org/10.1177/1750481318771446
Loftus, E. F. (1975). Leading questions and the eyewitness Mahoney, K. T., Buboltz, W., Levin, I. P., Doverspike, D.,
report. Cognitive Psychology, 7(4), 560–572. https://doi & Svyantek, D. J. (2011). Individual differences in a
.org/10.1016/0010-0285(75)90023-7 within-subjects risky-choice framing study. Personality
Loftus, E. F. (1979). Eyewitness testimony. Harvard University and Individual Differences, 51(3), 248–257. https://doi
Press. .org/10.1016/j.paid.2010.03.035
Loftus, E. F., & Palmer, J. C. (1974). Reconstruction of automo- Maiella, R., La Malva, P., Marchetti, D., Pomarico, E., Di
bile destruction: An example of the interaction between Crosta, A., Palumbo, R., Cetara, L., Di Domenico, A., &
language and memory. Journal of Verbal Learning and Verrocchio, M. C. (2020). The psychological distance and
Verbal Behavior, 13(5), 585–589. https://doi.org/10.1016/ climate change: A systematic review on the mitigation and
S0022-5371(74)80011-3 adaptation behaviors. Frontiers in Psychology, 11, Article
Loftus, E. F., & Zanni, G. (1975). Eyewitness testimony: The 568899. https://doi.org/10.3389/fpsyg.2020.568899
influence of the wording of a question. Bulletin of the Malik-Moraleda, S., Ayyash, D., Gallée, J., Affourtit, J.,
Psychonomic Society, 5(1), 86–88. https://doi.org/10.3758/ Hoffmann, M., Mineroff, Z., Jouravlev, O., & Fedorenko,
BF03336715 E. (2022). An investigation across 45 languages and 12
Logan, G. D. (1995). Linguistic and conceptual control of language families reveals a universal language network.
visual spatial attention. Cognitive Psychology, 28(2), 103– Nature Neuroscience, 25(8), 1014–1019. https://doi.org/
174. https://doi.org/10.1006/cogp.1995.1004 10.1038/s41593-022-01114-5
Loschelder, D. D., Siepelmeyer, H., Fischer, D., & Rubel, J. A. Malt, B. C., & Majid, A. (2013). How thought is mapped into
(2019). Dynamic norms drive sustainable consumption: words. Wiley Interdisciplinary Reviews: Cognitive Science,
Norm-based nudging helps café customers to avoid dis- 4(6), 583–597. https://doi.org/10.1002/wcs.1251
posable to-go-cups. Journal of Economic Psychology, 75, Mandel, D. R. (2001). Gain-loss framing and choice: Separating
102146. https://doi.org/10.1016/j.joep.2019.02.002 outcome formulations from descriptor formulations.
Luntz, F. (2007). Words that work: It’s not what you say, it’s Organizational Behavior and Human Decision Processes,
what people hear. Hachette UK. 85(1), 56–76. https://doi.org/10.1006/obhd.2000.2932
Psychological Science in the Public Interest 25(3) 155

Mandel, D. R. (2014). Do framing effects reveal irrational Meissner, C. A., & Brigham, J. C. (2001). A meta-analysis of
choice? Journal of Experimental Psychology: General, the verbal overshadowing effect in face identification.
143(3), 1185–1198. https://doi.org/10.1037/a0034207 Applied Cognitive Psychology, 15(6), 603–616. https://doi
Mandel, D. R., & Kapler, I. V. (2018). Cognitive style and frame .org/10.1002/acp.728
susceptibility in decision-making. Frontiers in Psychology, Melcher, J. M., & Schooler, J. W. (1996). The misremembrance
9, Article 1461. https://doi.org/10.3389/fpsyg.2018.01461 of wines past: Verbal and perceptual expertise differen-
Mandler, J. M. (2014). Stories, scripts, and scenes: Aspects of tially mediate verbal overshadowing of taste memory.
schema theory. Psychology Press. Journal of Memory and Language, 35(2), 231–245. https://
Mantonakis, A., Rodero, P., Lesschaeve, I., & Hastie, R. (2009). doi.org/10.1006/jmla.1996.0013
Order in choice: Effects of serial position on preferences. Meyerowitz, B. E., & Chaiken, S. (1987). The effect of mes-
Psychological Science, 20(11), 1309–1312. https://doi sage framing on breast self-examination attitudes, inten-
.org/10.1111/j.1467-9280.2009.02453.x tions, and behavior. Journal of Personality and Social
Markowitz, D. M., & Shulman, H. C. (2021). The predic- Psychology, 52(3), 500–510. https://doi.org/10.1037/0022-
tive utility of word familiarity for online engagements 3514.52.3.500
and funding. Proceedings of the National Academy of Minsky, M. (1974). A framework for representing knowledge
Sciences, USA, 118(18), Article e2026045118. https://doi (MIT-AI Laboratory Memo 306). MIT Artificial Intelligence
.org/10.1073/pnas.2026045118 Laboratory.
Martí, L., Wu, S., Piantadosi, S. T., & Kidd, C. (2023). Latent Minson, J. A., & Chen, F. S. (2022). Receptiveness to oppos-
diversity in human concepts. Open Mind, 7, 79–92. ing views: Conceptualization and integrative review.
https://doi.org/10.1162/opmi_a_00072 Personality and Social Psychology Review, 26(2), 93–111.
Matlock, T. (2011). The conceptual motivation of aspect. In https://doi.org/10.1177/10888683211061037
K. Panther & G. Radden (Eds.), Motivation in grammar Mon, S. K., Nencheva, M., Citron, F. M., Lew-Williams, C.,
and the lexicon (pp. 133–147). John Benjamins. https:// & Goldberg, A. E. (2021). Conventional metaphors elicit
doi.org/10.1075/hcp.27.09mat greater real-time engagement than literal paraphrases or
Matlock, T. (2012). Framing political messages with grammar concrete sentences. Journal of Memory and Language, 121,
and metaphor. American Scientist, 100(6), 478–483. Article 104285. https://doi.org/10.1016/j.jml.2021.104285
Matlock, T., Sparks, D., Matthews, J. L., Hunter, J., & Huette, Moray, N. (1959). Attention in dichotic listening: Affective
S. (2012). Smashing new results on aspectual framing: cues and the influence of instructions. Quarterly Journal
How people describe car accidents. Studies in Language, of Experimental Psychology, 11, 56–60. https://doi
36(3), 700–721. https://doi.org/10.1075/sl.36.3.09mat .org/10.1080/17470215908416289
Mayer, N. D., & Tormala, Z. L. (2010). “Think” versus “feel” Morris, B. S., Chrysochou, P., Christensen, J. D., Orquin, J.
framing effects in persuasion. Personality and Social L., Barraza, J., Zak, P. J., & Mitkidis, P. (2019). Stories
Psychology Bulletin, 36(4), 443–454. https://doi.org/ vs. facts: Triggering emotion and action-taking on cli-
10.1177/0146167210362981 mate change. Climatic Change, 154, 19–36. https://doi
McAllister, H. A., Bregman, N. J., & Lipscomb, T. J. (1988). .org/10.1007/s10584-019-02425-6
Speed estimates by eyewitnesses and earwitnesses: How Moty, K., & Rhodes, M. (2021). The unintended conse-
vulnerable to postevent information? The Journal of quences of the things we say: What generic statements
General Psychology, 115(1), 25–35. https://doi.org/10.10 communicate to children about unmentioned catego-
80/00221309.1988.9711085 ries. Psychological Science, 32(2), 189–203. https://doi
McCloskey, M., & Zaragoza, M. (1985). Misleading postevent .org/10.1177/0956797620953132
information and memory for events: Arguments and evi- Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the
dence against memory impairment hypotheses. Journal of effects of narrative in an entertainment television pro-
Experimental Psychology: General, 114(1), 1–16. https:// gram: Overcoming resistance to persuasion. Human
doi.org/10.1037/0096-3445.114.1.1 Communication Research, 36(1), 26–52. https://doi
McDonald, K., Graves, R., Yin, S., Weese, T., & Sinnott- .org/10.1111/j.1468-2958.2009.01367.x
Armstrong, W. (2021). Valence framing effects on moral Nabi, R. L. (1999). A cognitive-functional model for the effects
judgments: A meta-analysis. Cognition, 212, Article of discrete negative emotions on information process-
104703. https://doi.org/10.1016/j.cognition.2021.104703 ing, attitude change, and recall. Communication Theory,
McElroy, T., & Seta, J. J. (2003). Framing effects: An ana- 9(3), 292–320. https://doi.org/10.1111/j.1468-2885.1999.
lytic-holistic perspective. Journal of Experimental Social tb00172.x
Psychology, 39(6), 610–617. https://doi.org/10.1016/ Nabi, R. L. (2002a). Anger, fear, uncertainty, and atti-
S0022-1031(03)00036-2 tudes: A test of the cognitive-functional model.
McKenzie, C. R. M., & Nelson, J. (2003). What a speaker’s Communication Monographs, 69(3), 204–216. https://
choice of frame reveals: Reference points, frame selec- doi.org/10.1080/03637750216541
tion, and framing effects. Psychonomic Bulletin & Review, Nabi, R. L. (2002b). Discrete emotions and persuasion. In J.
10(3), 596–602. https://doi.org/10.3758/BF03196520 P. Dillard & M. Pfau (Eds.), The persuasion handbook:
McNeil, B. J., Pauker, S. G., Sox, H. C., & Tversky, A. (1982). Developments in theory and practice (pp. 289–308). Sage.
On the elicitation of preferences for alternative therapies. Nabi, R. L. (2003). Exploring the framing effects of emotion:
New England Journal of Medicine, 306(21), 1259–1262. Do discrete emotions differentially influence information
https://doi.org/10.1056/NEJM198205273062103 accessibility, information seeking, and policy preference?
156 Flusberg et al.

Communication Research, 30(2), 224–247. https://doi Noyes, A., & Keil, F. C. (2019). Generics designate kinds
.org/10.1177/0093650202250881 but not always essences. Proceedings of the National
Nabi, R. L. (2010). The case for emphasizing discrete emo- Academy of Sciences, USA, 116(41), 20354–20359. https://
tions in communication research. Communication doi.org/10.1073/pnas.190010511
Monographs, 77(2), 153–159. https://doi.org/10.1080/ Oganian, Y., Korn, C. W., & Heekeren, H. R. (2016). Language
03637751003790444 switching—but not foreign language use per se—reduces
Nabi, R. L., & Green, M. C. (2015). The role of a narrative’s the framing effect. Journal of Experimental Psychology:
emotional flow in promoting persuasive outcomes. Media Learning, Memory, and Cognition, 42(1), 140–148.
Psychology, 18(2), 137–162. https://doi.org/10.1080/152 https://doi.org/10.1037/xlm0000161
13269.2014.912585 O’Keefe, D. J. (2017a). Message framing variations in health
Nabi, R. L., Gustafson, A., & Jensen, R. (2018). Framing cli- and risk messaging. In M. Powers (Ed.), Oxford research
mate change: Exploring the role of emotion in generat- encyclopedia of communication. Oxford University Press.
ing advocacy behavior. Science Communication, 40(4), https://doi.org/10.1093/acrefore/9780190228613.013.308
442–468. https://doi.org/10.1177/1075547018776019 O’Keefe, D. J. (2017b). Misunderstandings of effect sizes in
Nabi, R. L., Walter, N., Oshidary, N., Endacott, C. G., Love- message effects research. Communication Methods and
Nichols, J., Lew, Z. J., & Aune, A. (2020). Can emotions Measures, 11(3), 210–219. https://doi.org/10.1080/19312
capture the elusive gain-loss framing effect? A meta-anal- 458.2017.1343812
ysis. Communication Research, 47(8), 1107–1130. https:// O’Keefe, D. J., & Hoeken, H. (2021). Message design choices
doi.org/10.1177/0093650219861256 don’t make much difference to persuasiveness and can’t
Nath, L., Pedriana, N., Gifford, C., McAuley, J. W., & Fülöp, be counted on—not even when moderating conditions
M. (2022). Examining moral foundations theory through are specified. Frontiers in Psychology, 12, Article 664160.
immigration attitudes. Athens Journal of Social Science, https://doi.org/10.3389/fpsyg.2021.664160
9(1), 9–30. https://doi.org/10.30958/ajss.9-1-1 O’Keefe, D. J., & Jensen, J. D. (2006). The advantages of
Nedergaard, J. S., Wallentin, M., & Lupyan, G. (2023). Verbal compliance or the disadvantages of noncompliance? A
interference paradigms: A systematic review investigating meta-analytic review of the relative persuasive effective-
the role of language in cognition. Psychonomic Bulletin & ness of gain-framed and loss-framed messages. Annals
Review, 30(2), 464–488. https://doi.org/10.3758/s13423- of the International Communication Association, 30(1),
022-02144-7 1–43. https://doi.org/10.1080/23808985.2006.11679054
Nelson, T. E., Clawson, R. A., & Oxley, Z. M. (1997). Media O’Keefe, D. J., & Jensen, J. D. (2007). The relative persuasive-
framing of a civil liberties conflict and its effect on toler- ness of gain-framed loss-framed messages for encourag-
ance. American Political Science Review, 91(3), 567–583. ing disease prevention behaviors: A meta-analytic review.
https://doi.org/10.2307/2952075 Journal of Health Communication, 12(7), 623–644.
Nelson, T. E., & Kinder, D. R. (1996). Issue frames and group- https://doi.org/10.1080/10810730701615198
centrism in American public opinion. The Journal of Politics, O’Keefe, D. J., & Jensen, J. D. (2009). The relative persua-
58(4), 1055–1078. https://doi.org/10.2307/2960149 siveness of gain-framed and loss-framed messages for
Nelson, T. E., & Oxley, Z. M. (1999). Issue framing effects encouraging disease detection behaviors: A meta-ana-
on belief importance and opinion. The Journal of Politics, lytic review. Journal of Communication, 59(2), 296–316.
61(4), 1040–1067. https://doi.org/10.2307/2647553 https://doi.org/10.1111/j.1460-2466.2009.01417.x
Nelson, T. E., Oxley, Z. M., & Clawson, R. A. (1997). Toward O’Keefe, D. J., & Jensen, J. D. (2015). The relative effective-
a psychology of framing effects. Political Behavior, 19, ness of gain-framed and loss-framed persuasive appeals
221–246. https://doi.org/10.1023/A:1024834831093 concerning obesity-related behaviors: Meta-analytic evi-
Newman, E. J., Sanson, M., Miller, E. K., Quigley-McBride, dence and implications. In R. Batra, V. J. Strecher, &
A., Foster, J. L., Bernstein, D. M., & Garry, M. (2014). P. A. Keller (Eds.), Leveraging consumer psychology for
People with easier to pronounce names promote truthi- effective health communications: The obesity challenge
ness of claims. PLOS ONE, 9(2), Article e88671. https:// (pp. 171–185). Routledge.
doi.org/10.1371/journal.pone.0088671 O’Keefe, D. J., & Wu, D. (2012). Gain-framed messages do
Nook, E. C., Schleider, J. L., & Somerville, L. H. (2017). A not motivate sun protection: A meta-analytic review
linguistic signature of psychological distancing in emotion of randomized trials comparing gain-framed and loss-
regulation. Journal of Experimental Psychology: General, framed appeals for promoting skin cancer prevention.
146(3), 337–346. https://doi.org/10.1037/xge0000263 International Journal of Environmental Research and
Nook, E. C., Vidal Bustamante, C. M., Cho, H. Y., & Somerville, Public Health, 9(6), 2121–2133. https://doi.org/10.3390/
L. H. (2020). Use of linguistic distancing and cognitive ijerph9062121
reappraisal strategies during emotion regulation in chil- Okuhara, T., Ishikawa, H., Okada, M., Kato, M., & Kiuchi, T.
dren, adolescents, and young adults. Emotion, 20(4), (2017). Designing persuasive health materials using process-
525–540. https://doi.org/10.1037/emo0000570 ing fluency: A literature review. BMC Research Notes, 10(1),
Novoa, G., Echelbarger, M., Gelman, A., & Gelman, S. A. Article 198. https://doi.org/10.1186/s13104-017-2524-x
(2023). Generically partisan: Polarization in political Ommundsen, R., Larsen, K. S., van der Veer, K., & Eilertsen, D.
communication. Proceedings of the National Academy of E. (2014). Framing unauthorized immigrants: The effects
Sciences, USA, 120(47), Article e2309361120. https://doi of labels on evaluations. Psychological Reports, 114(2),
.org/10.1073/pnas.2309361120 461–478. https://doi.org/10.2466/17.PR0.114k20w0
Psychological Science in the Public Interest 25(3) 157

Oppenheimer, D. M. (2006). Consequences of erudite vernac- Stanford University. https://plato.stanford.edu/archives/

ular utilized irrespective of necessity: Problems with using spr2022/entries/aristotle-rhetoric
long words needlessly. Applied Cognitive Psychology, Ratcliff, C. L., & Sun, Y. (2020). Overcoming resistance through
20(2), 139–156. https://doi.org/10.1002/acp.1178 narratives: Findings from a meta-analytic review. Human
Oppenheimer, D. M. (2008). The secret life of fluency. Communication Research, 46(4), 412–443. https://doi
Trends in Cognitive Sciences, 12(6), 237–241. https://doi .org/10.1093/hcr/hqz017
.org/10.1016/j.tics.2008.02.014 Read, J. D., Barnsley, R. H., Ankers, K., & Whishaw, I. Q.
Orvell, A., Gelman, S. A., & Kross, E. (2022). What “you” (1978). Variations in severity of verbs and eyewitnesses’
and “we” say about me: How small shifts in language testimony: An alternative interpretation. Perceptual and
reveal and empower fundamental shifts in perspective. Motor Skills, 46(3), 795–800. https://doi.org/10.2466/
Social and Personality Psychology Compass, 16(5), Article pms.1978.46.3.79
e12665. https://doi.org/10.1111/spc3.12665 Read, J. D., & Bruce, D. (1984). On the external validity
Orvell, A., Kross, E., & Gelman, S. A. (2017). How “you” of questioning effects in eyewitness testimony. Applied
makes meaning. Science, 355(6331), 1299–1302. https:// Psychology, 33(1), 33–49. https://doi.org/10.1111/j.1464
doi.org/10.1126/science.aaj2014 -0597.1984.tb01415.x
Orvell, A., Kross, E., & Gelman, S. A. (2020). “You” speaks to Read, S. J., Cesa, I. L., Jones, D. K., & Collins, N. L. (1990).
me: Effects of generic-you in creating resonance between When is the federal budget like a baby? Metaphor in
people and ideas. Proceedings of the National Academy political rhetoric. Metaphor and Symbol, 5(3), 125–149.
of Sciences, USA, 117(49), 31038–31045. https://doi.org/ https://doi.org/10.1207/s15327868ms0503_1
10.1073/pnas.2010939117 Reali, F., Spivey, M. J., Tyler, M. J., & Terranova, J. (2006).
Oschatz, C., & Marker, C. (2020). Long-term persuasive effects Inefficient conjunction search made efficient by con-
in narrative communication research: A meta-analysis. current spoken delivery of target identity. Perception
Journal of Communication, 70(4), 473–496. https://doi & Psychophysics, 68, 959–974. https://doi.org/10.3758/
.org/10.1093/joc/jqaa017 BF03193358
Ottati, V. C., & Renstrom, R. A. (2010). Metaphor and persua- Reddy, M. (1979). The conduit metaphor. In A. Ortony
sive communication: A multifunctional approach. Social (Ed.), Metaphor and thought (pp. 285–324). Cambridge
and Personality Psychology Compass, 4(9), 783–794. University Press.
https://doi.org/10.1111/j.1751-9004.2010.00292.x Rhodes, M., Leslie, S. J., & Tworek, C. M. (2012). Cultural
Ottati, V. C., Rhoads, S., & Graesser, A. (1999). The effect transmission of social essentialism. Proceedings of the
of metaphor on processing style in a persuasion task: National Academy of Sciences, USA, 109(34), 13526–
A motivational resonance model. Journal of Personality 13531. https://doi.org/10.1073/pnas.1208951109
and Social Psychology, 77(4), 688–697. https://doi.org/ Riou, M., Ball, S., Williams, T. A., Whiteside, A., O’halloran,
10.1037/0022-3514.77.4.688 K. L., Bray, J., Perkins, G. D., Smith, K., Cameron, P.,
Packard, G., Berger, J., & Boghrati, R. (2023). How verb tense Fatovich, D. M., Inoue, M., Bailey, P., Brink, D., & Finn,
shapes persuasion. Journal of Consumer Research, 50(3), J. (2017). ‘Tell me exactly what’s happened’: When lin-
645–660. https://doi.org/10.1093/jcr/ucad006 guistic choices affect the efficiency of emergency calls
Perloff, R. M. (2017). Dynamics of persuasion: Communication for cardiac arrest. Resuscitation, 117, 58–65. https://doi
and attitudes in the twenty-first century. Routledge. .org/10.1016/j.resuscitation.2017.06.002
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likeli- Roberts, S. G., Winters, J., & Chen, K. (2015). Future tense
hood model of persuasion. In Communication and per- and economic decisions: Controlling for cultural evolu-
suasion (pp. 1–24). Springer. tion. PLOS ONE, 10(7), Article e0132145. https://doi.org/
Pickering, M. J., & Gambi, C. (2018). Predicting while com- 10.1371/journal.pone.0132145
prehending language: A theory and review. Psychological Rogers, T. T., & McClelland, J. L. (2004). Semantic cognition:
Bulletin, 144(10), 1002–1044. https://doi.org/10.1037/ A parallel distributed processing approach. MIT Press.
bul0000158 Rook, E. D., & Holmes, K. J. (2023). How language shapes
Pinker, S. (1994). The language instinct: How the mind creates anti-fat bias: Comparing the effects of disease and fat-
language. HarperCollins. rights framing. Frontiers in Communication, 8, Article
Pinker, S. (2007). The stuff of thought: Language as a window 1284074. https://doi.org/10.3389/fcomm.2023.1284074
into human nature. Penguin. Rosch, E. (1983). Prototype classification and logical classi-
Piñon, A., & Gambara, H. (2005). A meta-analytic review fication: The two systems. In E. F. Scholnick (Ed.), New
of framing effect: Risky, attribute and goal framing. trends in conceptual representation: Challenges to Piaget’s
Psicothema, 17(2), 325–331. theory (pp. 73–86). Lawrence Erlbaum Associates.
Powell, T. E., Boomgaarden, H. G., De Swert, K., & De Vreese, Roskos-Ewoldsen, D. R., & Roskos-Ewoldsen, B. (2009).
C. H. (2015). A clearer picture: The contribution of visuals Media priming: An updated synthesis. In J. Bryant & M.
and text to framing effects. Journal of Communication, B. Oliver (Eds.), Media effects (pp. 90–109). Routledge.
65(6), 997–1017. https://doi.org/10.1111/jcom.12184 Rothman, A. J., Desmarais, K. J., & Lenne, R. L. (2020). Moving
Pratkanis, A. R., & Aronson, E. (1991). Age of propaganda: from research on message framing to principles of message
The everyday use and abuse of persuasion. Macmillan. matching: The use of gain-and loss-framed messages to
Rapp, C. (2022). Aristotle’s rhetoric. In E. N. Zalta (Ed.), The promote healthy behavior. Advances in Motivation Science,
Stanford encyclopedia of philosophy (Spring 2022 ed.). 7, 43–73. https://doi.org/10.1016/bs.adms.2019.03.001
158 Flusberg et al.

Rucker, J. M., Murphy, M. C., & Quintanilla, V. D. (2019). The Semino, E. (2021). “Not soldiers but fire-fighters”—Metaphors
immigrant labeling effect: The role of immigrant group and Covid-19. Health Communication, 36(1), 50–58.
labels in prejudice against noncitizens. Group Processes https://doi.org/10.1080/10410236.2020.1844989
& Intergroup Relations, 22(8), 1139–1160. https://doi.org/ Semino, E., Demjén, Z., Demmen, J., Koller, V., Payne,
10.1177/1368430218818744 S., Hardie, H., & Rayson, P. (2017). The online use of
Rumelhart, D. E. (1975). Notes on a schema for stories. In Violence and Journey metaphors by cancer patients, as
D. G. Bobrow & A. Collins (Eds.), Representation and compared with health professionals: A mixed methods
understanding: Studies in cognitive science (pp. 211–236). study. BMJ Supportive and Palliative Care, 7(1), 60–66.
Morgan Kaufmann. https://doi.org/10.1016/B978-0-12- https://doi.org/10.1136/bmjspcare-2014-000785
108550-6.50013-6 Semino, E., Demjén, Z., Hardie, A., Payne, S., & Rayson, P.
Ruppenhofer, J., Ellsworth, M., Petruck, M. R., Johnson, C. (2018). Metaphor, cancer and the end of life: A corpus-
R., & Scheffczyk, J. (2016). FrameNet II: Extended theory based study. Routledge.
and practice. International Computer Science Institute. Shah, A. K., & Oppenheimer, D. M. (2007). Easy does it:
Russell, J. A. (2003). Core affect and the psychological con- The role of fluency in cue weighting. Judgment and
struction of emotion. Psychological Review, 110(1), 145– Decision Making, 2(6), 371–379. https://doi.org/10.1017/
172. https://doi.org/10.1037/0033-295X.110.1.145 S1930297500000516
Sahni, N. S., Wheeler, S. C., & Chintagunta, P. (2018). Shapiro, K. L., Caldwell, J., & Sorensen, R. E. (1997). Personal
Personalization in email marketing: The role of nonin- names and the attentional blink: A visual “cocktail party”
formative advertising content. Marketing Science, 37(2), effect. Journal of Experimental Psychology: Human
236–258. https://doi.org/10.1287/mksc.2017.1066 Perception and Performance, 23(2), 504–514. https://doi
Schacter, D. L., & Buckner, R. L. (1998). Priming and the .org/10.1037/0096-1523.23.2.504
brain. Neuron, 20(2), 185–195. https://doi.org/10.1016/ Sheeran, P., & Webb, T. L. (2016). The intention-behavior
s0896-6273(00)80448-1 gap. Social and Personality Psychology Compass, 10(9),
Schaffner, B. F., & Atkinson, M. L. (2009). Taxing death or 503–518. https://doi.org/10.1111/spc3.12265
estates? When frames influence citizens’ issue beliefs. In Shen, F., Sheer, V. C., & Li, R. (2015). Impact of narratives
B. F. Schaffner & P. J. Sellers (Eds.), Winning with words: on persuasion in health communication: A meta-analysis.
The origins & impact of political framing (pp. 121–135). Journal of Advertising, 44(2), 105–113. https://doi.org/10
Routledge. .1080/00913367.2015.1018467
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals, Sher, S., & McKenzie, C. (2006). Information leakage from
and understanding: An inquiry into human knowledge logically equivalent frames. Cognition, 101(3), 467–494.
structures. Psychology Press. https://doi.org/10.1016/j.cognition.2005.11.001
Schmidt, L., Skvortsova, V., Kullen, C., Weber, B., & Plassmann, Sher, S., McKenzie, C. R., Müller-Trede, J., & Leong, L.
H. (2017). How context alters value: The brain’s valuation (2022). Rational choice in context. Current Directions in
and affective regulation system link price cues to expe- Psychological Science, 31(6), 518–525. https://doi.org/
rienced taste pleasantness. Scientific Reports, 7, Article 10.1177/09637214221120387
8098. https://doi.org/10.1038/s41598-017-08080-0 Shrout, P. E., & Rodgers, J. L. (2018). Psychology, sci-
Schooler, J. W., & Engstler-Schooler, T. Y. (1990). Verbal ence, and knowledge construction: Broadening per-
overshadowing of visual memories: Some things are bet- spectives from the replication crisis. Annual Review
ter left unsaid. Cognitive Psychology, 22(1), 36–71. https:// of Psychology, 69, 487–510. https://doi.org/10.1146/
doi.org/10.1016/0010-0285(90)90003-M annurev-psych-122216-011845
Schroder, H. S., Devendorf, A., & Zikmund-Fisher, B. J. Sieck, W., & Yates, J. F. (1997). Exposition effects on decision
(2023). Framing depression as a functional signal, not a making: Choice and confidence in choice. Organizational
disease: Rationale and initial randomized controlled trial. Behavior and Human Decision Processes, 70(3), 207–219.
Social Science & Medicine, 328, Article 115995. https:// https://doi.org/10.1006/obhd.1997.2706
doi.org/10.1016/j.socscimed.2023.115995 Sikveland, R. O., & Stokoe, E. (2020). Should police negotia-
Schuman, H., & Presser, S. (1996). Questions and answers in tors ask to “talk” or “speak” to persons in crisis? Word
attitude surveys: Experiments on question form, wording, selection and overcoming resistance to dialogue propos-
and context. Sage. als. Research on Language and Social Interaction, 53(3),
Schwartzstein, J., & Sunderam, A. (2021). Using models to 324–340. https://doi.org/10.1080/08351813.2020.1785770
persuade. American Economic Review, 111(1), 276–323. Silva, R. R., Chrobot, N., Newman, E., Schwarz, N., &
https://doi.org/10.1257/aer.20191074 Topolinski, S. (2017). Make it short and easy: Username
Schwarz, N., Jalbert, M., Noah, T., & Zhang, L. (2021). complexity determines trustworthiness above and beyond
Metacognitive experiences as information: Processing objective reputation. Frontiers in Psychology, 8, Article
fluency in consumer judgment and decision making. 2200. https://doi.org/10.3389/fpsyg.2017.02200
Consumer Psychology Review, 4(1), 4–25. https://doi.org/ Simchon, A., Edwards, M., & Lewandowsky, S. (2024). The
10.1002/arcp.1067 persuasive effects of political microtargeting in the age of
Schwarz, N., & Oyserman, D. (2001). Asking questions about generative artificial intelligence. PNAS Nexus, 3(2), Article
behavior: Cognition, communication, and questionnaire pgae035. https://doi.org/10.1093/pnasnexus/pgae035
construction. American Journal of Evaluation, 22(2), 127– Slobin, D. I. (1996). From “thought and language” to “thinking
160. https://doi.org/10.1177/10982140010220020 for speaking.” In J. J. Gumperz & S. C. Levinson (Eds.),
Psychological Science in the Public Interest 25(3) 159

Rethinking linguistic relativity: Studies in the social and to fear: A meta-analysis of fear appeal effectiveness and
cultural foundations of language (pp. 70–96). Cambridge theories. Psychological Bulletin, 141(6), 1178–1204.
University Press. https://doi.org/10.1037/a0039729
Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. Tappin, B. M., Wittenberg, C., Hewitt, L. B., Berinsky, A. J.,
G. (2007). The affect heuristic. European Journal of & Rand, D. G. (2023). Quantifying the potential persua-
Operational Research, 177(3), 1333–1352. https://doi sive returns to political microtargeting. Proceedings of
.org/10.1016/j.ejor.2005.04.006 the National Academy of Sciences, USA, 120(25), Article
Sopory, P., & Dillard, J. P. (2002). The persuasive effects of met- e2216261120. https://doi.org/10.1073/pnas.2216261120
aphor: A meta-analysis. Human Communication Research, Teeny, J. D., Siev, J. J., Briñol, P., & Petty, R. E. (2021).
28(3), 382–419. https://doi.org/10.1111/j.1468-2958.2002 A review and conceptual framework for understanding
.tb00813.x personalized matching effects in persuasion. Journal of
Sparkman, G., & Walton, G. M. (2017). Dynamic norms pro- Consumer Psychology, 31(2), 382–414. https://doi.org/
mote sustainable behavior, even if it is counternorma- 10.1002/jcpy.1198
tive. Psychological Science, 28(11), 1663–1674. https:// Thibodeau, P. H. (2016). Extended metaphors are the home
doi.org/10.1177/0956797617719950 runs of persuasion: Don’t fumble the phrase. Metaphor
Sparkman, G., & Walton, G. M. (2019). Witnessing change: and Symbol, 31(2), 53–72. https://doi.org/10.1080/1092
Dynamic norms help resolve diverse barriers to personal 6488.2016.1150756
change. Journal of Experimental Social Psychology, 82, Thibodeau, P. H., & Boroditsky, L. (2011). Metaphors we
238–252. https://doi.org/10.1016/j.jesp.2019.01.007 think with: The role of metaphors in reasoning. PLOS
Sperber, D., & Wilson, D. (1986). Relevance: Communication ONE, 6(2), Article e16782. https://doi.org/10.1371/journal
and cognition. Harvard University Press. .pone.0016782
Stark, E., Baldwin, A. S., Hertel, A. W., & Rothman, A. J. Thibodeau, P. H., & Boroditsky, L. (2013). Natural language
(2017a). The role of rational and experiential processing metaphors covertly influence reasoning. PLOS ONE,
in influencing the framing effect. The Journal of Social 8(1), Article e52961. https://doi.org/10.1371/journal
Psychology, 157(3), 308–321. https://doi.org/10.1080/00 .pone.0052961
224545.2016.1198301 Thibodeau, P. H., Crow, L., & Flusberg, S. J. (2017). The
Stark, E., Baldwin, A. S., Hertel, A. W., & Rothman, A. J. metaphor police: A case study of the role of metaphor
(2017b). Understanding the framing effect: Do affective in explanation. Psychonomic Bulletin & Review, 24(5),
responses to decision options mediate the influence of 1375–1386. https://doi.org/10.3758/s13423-016-1192-5
frame on choice? Journal of Risk Research, 20(12), 1585– Thibodeau, P. H., & Flusberg, S. J. (2017). Metaphorical
1597. https://doi.org/10.1080/13669877.2016.1200654 accounting: How framing the federal budget like a house-
Steiger, A., & Kühberger, A. (2018). A meta-analytic re- hold’s affects voting intentions. Cognitive Science, 41(S5),
appraisal of the framing effect. Zeitschrift für Psychologie, 1168–1182. https://doi.org/10.1111/cogs.12475
226(1), 45–55. https://doi.org/10.1027/2151-2604/a000321 Thibodeau, P. H., & Flusberg, S. J. (2022). Metaphor and
Stephensen, M. B., Schulze, C., Landrø, M., Hendrikx, J., elaboration in context. In H. L. Colston, T. Matlock, &
& Hetland, A. (2021). Should I judge safety or danger? G. J. Steen (Eds.), Dynamism in metaphor and beyond
Perceived risk depends on the question frame. Journal of (pp. 223–240). John Benjamins. https://doi.org/10.1075/
Experimental Psychology: Applied, 27(3), 485–502. https:// milcc.9.12thi
doi.org/10.1037/xap0000354 Thibodeau, P. H., Frantz, C. M., & Berretta, M. (2017). The
Strauss, C., & Quinn, N. (1997). A cognitive theory of cultural earth is our home: Systemic metaphors to redefine our
meaning. Cambridge University Press. relationship with nature. Climatic Change, 142(1), 287–
Sullivan, J. (2019). The primacy effect in impression formation: 300. https://doi.org/10.1007/s10584-017-1926-z
Some replications and extensions. Social Psychological Thibodeau, P. H., Hendricks, R. K., & Boroditsky, L. (2017).
and Personality Science, 10, 432–439. https://doi.org/ How linguistic metaphor scaffolds reasoning. Trends
10.1177/1948550618771003 in Cognitive Sciences, 21(11), 852–863. https://doi.org/
Sullivan, K. (2023). Three levels of framing. Wiley 10.1016/j.tics.2017.07.001
Interdisciplinary Reviews: Cognitive Science, Article e1651. Thibodeau, P. H., Lee, D., & Flusberg, S. J. (2016). Context,
https://doi.org/10.1002/wcs.1651 but not proficiency, moderates the effects of metaphor
Swidler, A. (1986). Culture in action: Symbols and strategies. framing: A case study in India. In A. Papafragou, D. J.
American Sociological Review, 51(2), 273–286. https:// Grodner, D. Mirman, & J. Trueswell (Eds.), Proceedings
doi.org/10.2307/2095521 of the 38th Annual Conference of the Cognitive Science
Takemura, K. (1994). Influence of elaboration on the framing Society (pp. 1571–1576). Cognitive Science Society.
of decision. The Journal of Psychology, 128(1), 33–39. Thibodeau, P. H., Matlock, T., & Flusberg, S. J. (2019).
https://doi.org/10.1080/00223980.1994.9712709 The role of metaphor in communication and thought.
Tannen, D. (Ed.). (1993). What’s in a frame? Surface evidence Language & Linguistics Compass, 13(5), Article e12327.
for underlying expectations. Framing in discourse (pp. https://doi.org/10.1111/lnc3.12327
14–56). Oxford University Press. Thierry, G., Athanasopoulos, P., Wiggett, A., Dering, B., &
Tannenbaum, M. B., Hepler, J., Zimmerman, R. S., Saul, L., Kuipers, J. R. (2009). Unconscious effects of language-specific
Jacobs, S., Wilson, K., & Albarracín, D. (2015). Appealing terminology on preattentive color perception. Proceedings
160 Flusberg et al.

of the National Academy of Sciences, USA, 106(11), 4567– American Academy of Political and Social Science, 700(1),
4570. https://doi.org/10.1073/pnas.0811155106 73–85. https://doi.org/10.1177/00027162221083877
Trope, Y., & Liberman, N. (2012). Construal level theory. In Voelkel, J. G., Mernyk, J. S., & Willer, R. (2023). Moral refram-
P. A. M. Van Lange, A. W. Kruglanski, & E. T. Higgins ing increases support for economically progressive can-
(Eds.), Handbook of theories of social psychology: Vol. 1 didates. PNAS Nexus, 2(6), Article pgad154. https://doi
(pp. 118–134). Sage. .org/10.1093/pnasnexus/pgad154
Tversky, A., & Kahneman, D. (1973). Availability: A heu- von Frisch, K. (1967). The dance language and orientation of
ristic for judging frequency and probability. Cognitive bees. Harvard University Press.
Psychology, 5(2), 207–232. https://doi.org/10.1016/0010- Walsh, J., Vaida, N., Coman, A., & Fiske, S. T. (2022). Stories in
0285(73)90033-9 action. Psychological Science in the Public Interest, 23(3),
Tversky, A., & Kahneman, D. (1981). The framing of deci- 99–141. https://doi.org/10.1177/15291006231161337
sions and the psychology of choice. Science, 211(4481), Walter, N., Tukachinsky, R., Pelled, A., & Nabi, R. (2019).
453–458. https://doi.org/10.1126/science.7455683 Meta-analysis of anger and persuasion: An empirical inte-
Ünal, E., & Papafragou, A. (2016). Interactions between lan- gration of four models. Journal of Communication, 69(1),
guage and mental representations. Language Learning, 73–93. https://doi.org/10.1093/joc/jqy054
66(3), 554–580. https://doi.org/10.1111/lang.12188 Walton, G. M., & Banaji, M. R. (2004). Being what you say:
Urban, T. (2023). What’s our problem? A self-help book for The effect of essentialist linguistic labels on preferences.
societies. Wait But Why. Social Cognition, 22(2), 193–213. https://doi.org/10.1521/
van der Linden, S. (2021). The Gateway Belief Model (GBM): soco.22.2.193.35463
A review and research agenda for communicating the Wang, Y., & Gennari, S. P. (2019). How language and event
scientific consensus on climate change. Current Opinion recall can shape memory for time. Cognitive Psychology,
in Psychology, 42, 7–12. https://doi.org/10.1016/j.co 108, 1–21. https://doi.org/10.1016/j.cogpsych.2018.10.003
psyc.2021.01.005 Weaver, D., McCombs, M., & Shaw, D. L. (2004). Agenda-
van Dijk, T. A. (2023). Analyzing frame analysis: A critical setting research: Issues, attributes, and influences. In
review of framing studies in social movement research. L. L. Kaid (Ed.), Handbook of political communication
Discourse Studies, 25(2), 153–178. https://doi.org/ research (pp. 257–280). Lawrence Erlbaum Associates.
10.1177/14614456231155080 Webb, T. L., & Sheeran, P. (2006). Does changing behavioral
Van Lange, P. A., & Huckelba, A. L. (2021). Psychological intentions engender behavior change? A meta-analysis of
distance: How to make climate change less abstract and the experimental evidence. Psychological Bulletin, 132(2),
closer to the self. Current Opinion in Psychology, 42, 249–268. https://doi.org/10.1037/0033-2909.132.2.249
49–53. https://doi.org/10.1016/j.copsyc.2021.03.011 Werner, C. P., Birkhaeuer, J., Locher, C., Gerger, H.,
Van Stee, S. K. (2018). Meta-analysis of the persuasive effects Heimgartner, N., Colagiuri, B., & Gaab, J. (2021). Price
of metaphorical vs. literal messages. Communication information influences the subjective experience of
Studies, 69(5), 545–566. https://doi.org/10.1080/105109 wine: A framed field experiment. Food Quality and
74.2018.1457553 Preference, 92, Article 104223. https://doi.org/10.1016/
Vasilyeva, N., & Lombrozo, T. (2020). Structural thinking j.foodqual.2021.104223
about social categories: Evidence from formal explana- Whorf, B. L. (2012). Language, thought, and reality: Selected
tions, generics, and generalization. Cognition, 204, Article writings of Benjamin Lee Whorf (J. B. Carroll, S. C. Levinson,
104383. https://doi.org/10.1016/j.cognition.2020.104383 & P. Lee, Eds.). MIT Press. (Original work published 1940)
Vazire, S. (2018). Implications of the credibility revolution Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A.
for productivity, creativity, and progress. Perspectives on R., & Boroditsky, L. (2007). Russian blues reveal effects
Psychological Science, 13(4), 411–417. https://doi.org/ of language on color discrimination. Proceedings of the
10.1177/1745691617751884 National Academy of Sciences, USA, 104(19), 7780–7785.
Vlasceanu, M., Doell, K. C., Bak-Coleman, J. B., Todorova, https://doi.org/10.1073/pnas.0701644104
B., Berkebile-Weinberg, M. M., Grayson, S. J., Patel, Y., Wodak, D., Leslie, S. J., & Rhodes, M. (2015). What a loaded
Goldwert, D., Pei, Y., Chakroff, A., Pronizius, E., van den generalization: Generics and social cognition. Philosophy
Broek, K. L., Vlasceanu, D., Constantino, S., Morais, M. Compass, 10(9), 625–635. https://doi.org/10.1111/phc3.12250
J., Schumann, P., Rathje, S., Fang, K., Aglioti, S. M., . . . Wolff, P., & Holmes, K. J. (2011). Linguistic relativity. Wiley
Van Bavel, J. J. (2024). Addressing climate change with Interdisciplinary Reviews: Cognitive Science, 2(3), 253–
behavioral science: A global intervention tournament in 265. https://doi.org/10.1002/wcs.104
63 countries. Science Advances, 10(6), Article eadj5778. Wolsko, C., Ariceaga, H., & Seiden, J. (2016). Red, white,
https://doi.org/10.1126/sciadv.adj5778 and blue enough to be green: Effects of moral framing
Voelkel, J. G., & Feinberg, M. (2018). Morally reframed argu- on climate change attitudes and conservation behaviors.
ments can affect support for political candidates. Social Journal of Experimental Social Psychology, 65, 7–19.
Psychological and Personality Science, 9(8), 917–924. https://doi.org/10.1016/j.jesp.2016.02.005
https://doi.org/10.1177/1948550617729408 Wood, N. L., & Cowan, N. (1995). The cocktail party phe-
Voelkel, J. G., Malik, M., Redekopp, C., & Willer, R. (2022). nomenon revisited: Attention and memory in the clas-
Changing Americans’ attitudes about immigration: Using sic selective listening procedure of Cherry (1953).
moral framing to bolster factual arguments. Annals of the Journal of Experimental Psychology: Learning, Memory,
Psychological Science in the Public Interest 25(3) 161

& Cognition, 21, 255–260. https://doi.org/10.1037//0096- Yeager, D. S., Hanselman, P., Walton, G. M., Murray, J.
3445.124.3.243 S., Crosnoe, R., Muller, C., Tipton, E., Schneider, B.,
Wu, S. H., Elpers, N., Doherty, E. M., Flusberg, S. J., & Hulleman, C. S., Hinojosa, C. P., Paunesku, D., Romero,
Holmes, K. J. (2021, July 26–29). Pragmatic reasoning C., Flint, K., Roberts, A., Trott, J., Iachan, R., Buontempo,
ability predicts syntactic framing effects on social judg- J., Yang, S. M., Carvalho, C. M., . . . Dweck, C. S. (2019).
ments [Paper presentation]. 43rd Annual Meeting of the A national experiment reveals where a growth mind-
Cognitive Science Society, Virtual. set improves achievement. Nature, 573(7774), 364–369.
Xu, J. (2023). A meta-analysis comparing the effectiveness of https://doi.org/10.1038/s41586-019-1466-y
narrative vs. statistical evidence: Health vs. non-health Young, N. A., Shuster, M. M., & Mikels, J. A. (2019). The sure
contexts. Health Communication, 38(14), 3113–3123. thing: The role of integral affect in risky choice fram-
https://doi.org/10.1080/10410236.2022.2137750 ing. Emotion, 19(6), 1035–1043. https://doi.org/10.1037/
Yang, J. Z., & Chu, H. (2018). Who is afraid of the Ebola emo0000505
outbreak? The influence of discrete emotions on risk per- Zebregs, S., van den Putte, B., Neijens, P., & de Graaf, A.
ception. Journal of Risk Research, 21(7), 834–853. https:// (2015). The differential impact of statistical and narra-
doi.org/10.1080/13669877.2016.1247378 tive evidence on beliefs, attitude, and intention: A meta-
Yang, J. Z., Rickard, L. N., Liu, Z., & Boze, T. (2021). Too analysis. Health Communication, 30(3), 282–289. https://
close to care? A replication study to re-examine the doi.org/10.1080/10410236.2013.842528
effect of cued distance on climate change engagement. Zhuang, J., & Guidry, A. (2022). Does storytelling reduce
Environmental Communication, 15(1), 1–11. https://doi stigma? A meta-analytic view of narrative persuasion on
.org/10.1080/17524032.2020.1777181 stigma reduction. Basic and Applied Social Psychology,
Ye, W., Li, Q., & Yu, S. (2021). Persuasive effects of mes- 44(1), 25–37. https://doi.org/10.1080/01973533.2022.203
sage framing and narrative format on promoting COVID- 9657
19 vaccination: A study on Chinese college students. Zwaan, R. A., & Radvansky, G. A. (1998). Situation models
International Journal of Environmental Research and in language comprehension and memory. Psychological
Public Health, 18(18), Article 9485. https://doi.org/ Bulletin, 123(2), 162–185. https://doi.org/10.1037/0033-
10.3390/ijerph18189485 2909.123.2.162