Friedenberg Silverman (2011) Ch11 Social Cognition
Friedenberg Silverman (2011) Ch11 Social Cognition
EDITION
JAY FRIEDENBERG
GORDON SILVERMAN
Manhattan College
($)SAGE
Los Angeles I London I New Delhi
Singapore I Washington DC
11
THE SOCIAL APPROACH
Mind as Society
"A society can be no better than the men and women who compose it."
-Adlai E. Stevenson, 1952
SOCIAL COGNITION
Social cognition is the study of how people make sense of other people and of themselves.
Fiske and Taylor (2008) state that there are four assumptions that pervade the field. The
first is mentalism, which is that people use mental representations and processes. This is
nothing new, except in this case, the things that are being represented and computed on
are social in nature. They include people's faces and appearance, as well as the feelings,
thoughts, and explanations we have about them. As we shall see, we often think differently
about people than we do about inanimate objects, such as chairs, or abstract concepts,
such as justice. Table 11.1 summarizes some of the important differences in thinking about
people versus things.
The second basic assumption in research on social cognition is about cognitive pro-
cesses, or how representations develop, operate, and change over time. So, if we were
studying prejudice, we would want to know not just what prejudice is like but also how
it forms and how it might become more extreme. The third theme is cross-fertilization,
which is that the social cognitive approach is highly interdisciplinary in nature. In fact,
most of the work we report in this chapter occurs at the intersection of the cognitive,
social, and neuroscience perspectives. We discuss the intertwining of these three perspec-
tives in more depth later.
The fourth theme is the applicability of social cognitive research to the real world. Since
we live in a social society, we can use the findings we obtain to make the world a better place.
For example, our understanding of stereotypes may enable us to reduce the development of
324 Cognitive Science
Chapter 11 The Social Approach 325
Table 11.1 Differences in the way we think about people and things plays in social cognitive disorders such as autism. Finally, we sum up by covering some of
the basic social topics that explain how we think about others. These topics are impres-
sions, attributions, attitudes, stereotypes, and prejudice.
People Things
Principle (Example: your best friend) (Example: a rock)
l:'l
Intentionality People attempt to control the
SOCIAL COGNITIVE NEUROSCIENCE
Objects are not capable of !I:'
environment. controlling their environment.
A common theme throughout this book has been the interdisciplinary approach, where
Mutuality People think about each other (e.g ., Objects don't "think back." different perspectives and methodologies focus on a research topic. No area illustrates this
~'!
"I think of you, and you think of
me") . more clearly than the field of social cognitive neuroscience. As the name implies, this
endeavor attempts to describe phenomena in terms of interactions between three levels of
Self-concept Others judge us, provide information Objects cannot judge or provide analysis. At the broadest level, we have social psychology, which examines the role of
about us, and are more similar to us information about us and are human interaction, focusing on people's behavior in a group setting. At the intermediate
than are objects. very different from us.
level, there is the cognitive perspective, examining how we think about one another.
Observational People act differently when they know Objects do not act differently Ultimately, we arrive at the neuroscience level, which seeks to examine the location and
bias i'
they are being thought about. when thought about. type of neural processing that takes place in a social setting and that underpins and gives
ii,:
Nonobservable Traits that can't be observed are rise to our cognition and behavior.
Traits are more easily observed
attributes crucial to thinking about people . but less crucial. ,, Ochsner and Lieberman (2001) make several important points about this emerging
,,:: field. They note that emphasis needs to be placed on the cognitive approach, because
Change People change over time and in Objects change less over time f:l it is here where social psychologists and neuroscientists have the most in common and
different circumstances. and in different situations.
can most easily communicate. Cognition serves to "bridge the gap" and unites the
Accuracy of It is difficult to judge the traits or It is usually easier to judge :' fields of social cognition and cognitive neuroscience. They also note that each of these
cognition qualities of a person. object qualities. l' areas can enrich the other, with social psychologists using neuroscience findings to
Complexity People are complex; we need to
,, clarify and differentiate between competing psychological theories. On the other side
Objects are usually less
simplify in order to understand them . complex. of the divide, neuroscientists can use findings from the social literature to examine
Explanation
whether there are different neural mechanisms that underlie social versus nonsocial
We have to explain others' Explanations are not always II':
behaviors. stimuli.
necessary.
'I To illustrate the relationship between these three levels, Ochsner and Lieberman (2001)
present a prism model (shown in Figure 11.1). Each corner or vertex of the prism repre-
Source: Adapted from Fiske and Taylor (2008). sents one of four different disciplinary aspects. Each face is bordered by three corners and
stands for one of the three approaches. For instance, the base of the prism is formed by
the neural mechanisms, cognition/information-processing, and behavior and experience
negative stereotypes and enhance the formation of positive ones. Results from social cogni- vertices. This surface, which faces downward in Figure 11.1, corresponds to the domain
tive research have been applied to many areas, including healthcare, the legal system, adver- of inquiry for cognitive neuroscience but is lacking because it does not take into account
tising, and political campaigns.
the remaining vertex, personal and social context. Similarly, social psychology is the facet
There is a lot to say about social cognition. To organize this material and make it easier formed from the personal and social context, behavior and experience, and cognition/
to digest, we divide the chapter into three sections. In the first section, we talk about social information-processing vertices. This is the surface that faces toward the back in Figure 11.1.
cognitive neuroscience. In the second section, we examine perhaps the most important It, too, is lacking because it fails to account for the remaining vertex, neural mechanisms.
foundational aspect of social cognition. This is a theory of mind, which is how we come Only the field of social cognitive neuroscience takes all aspects into account and is, thus,
to know that other people have minds. In this section, we discuss the role theory of mind represented by the entire prism with all three faces.
326 Cognitive Science Chapter 11 The Social Approach 327
originally to allow us to remember key members of our social group. Once this was in
Figure 11.1 The prism model of social cognitive neuroscience
place, it could be applied to more abstract problems involving language, mathematics, and
spatial reasoning.
In contrast, the general intelligence hypothesis states that human intelligence did not
evolve in response to particular selection forces such as social organization. Instead, it
(D) Personal and
Social Context
emerged in a general form that could be suited to solving all kinds of problems, whether
they are social or not. One way to differentiate between these two notions is to examine
the physical and social cognitive abilities of children who have not yet been influenced by
cultural factors such as written language, symbolic mathematics, and formal education.
Physical cognition refers to reasoning about things such as space, quantities, and causality.
This could evolve in a wide variety of environments and is not specific to social settings.
(C) Behavior
Social cognition, on the other hand, is reasoning about those things directly linked to
and Experience (B) Cognition/ culture, such as social learning, communication, and theory of mind.
Information We can then compare the children's performance to that of our closest primate rela-
Processing tives. If the cultural intelligence hypothesis is correct, there should be no difference
between the children and primates on physical domain tests because both groups have
evolved under these conditions. There should, however, be a difference on tests of social
cognition. We would expect the children to score higher on these tests because humans
evolved in more demanding social environments.
This is exactly what Herrmann et al. (2009) did. They tested chimpanzees, orangutans,
(A) Neural
Mechanism
and 2 1/2-year-old human children. The results clearly supported the cultural hypothesis.
There was no difference between human and primate performance in the physical domain.
On tests designed to specifically measure social intelligence, the children's scores were
higher by a statistically significant amount. If humans had evolved a superior general form
Source: Ochsner and Lieberman (2001 ). of intelligence, they would have outscored the primates in both categories. It is not known
yet what specific social environments may have led to this divergence in our early ances-
tors, but it could have been complex types of collaborative activity involved in hunting
and gathering that would call on more sophisticated kinds of communicating and social
TOPICS IN SOCIAL COGNITIVE NEUROSCIENCE learning (Tomasello, Carpenter, Call, Behne, & Moll, 2005).
Additional evidence to support the evolution of specialized social intelligence comes from
Evolution studies that have examined brain size. There is a positive correlation between the relative
size of the neocortex and social factors such as group size and grooming clique size that are
We humans are a distinctly social race, what might be called "ultra-social" (Boyd & measures of social complexity (Dunbar, 1996). Hedgehogs are mammals that provide mater-
Richerson, 1996). More than any other species, we form cooperative groups such as fam- nal care to a small number of offspring, and they have comparatively small cortical regions.
ily, friendships, and teams as well as competing groups such as tribes, nations, and eth- Chimpanzees live in extended groups of a few dozen animals and show moderate cortical
nicities. Is it possible that this sort of social environment led to the development of our size. Humans, which interact in societies numbering in the millions, have the largest relative
intelligence? According to the cultural intelligence hypothesis , it has. In this view, complex cortices. In studies such as this, it is not absolute brain weight or volume that is measured
social interaction created a form of social intelligence that then laid the groundwork for but, rather, relative size expressed as a proportion of the animal's overall body size. It could
more general intelligence to emerge later (Herrmann, Call, Hernandez-Lloreda, Hare, & be that larger social groups cause greater competition for social skills that subsequently leads
Tomasello, 2009). For example, our working memory capacity may have evolved to the evolution of cognitive mechanisms for outsmarting others.
328 Cognitive Science Chapter 11 The Social Approach 329
Affention
Figure 11.2 The posterior and anterior attention systems underlying joint
attention (after Mundy et al. , 2000)
It has been known for some time that infants can follow somebody else's gaze (Scaife &
Bruner, 1975). This suggests that even at a young age, children can focus their attention
on what another person is attending. This ability to coordinate attention with a social
partner is called joint attention. Not only following but also initiating or joining shared Frontal eye fields and
Posterior parietal
attentional focus with others is a key part of social interaction. Without it, we would have association cortex
prefrontal association
a very difficult time understanding or making sense of what other people are saying cortex
(Mundy & Newell, 2007).
We can divide joint attention into two categories. In responding to joint attention
(RJA), we follow the gestures and direction of gaze in order to share a common point of Anterior Posterior
reference. This is the more passive form, because we are following others. In initiating Attentional Attentional
System System
joint attention (IJA), people use gestures and eye movements to direct attention to people, -Initiates -Responds
objects, and events. This is a more active form, because it involves the initiation of a joint to joint attention
attentional focus.
Joint attention plays a very important role in development. It can help disambiguate the
referent of a word during language acquisition. A mother pointing to a spoon and saying
the word gets a child to look at the object corresponding to the word and, so, aids in the Orbital prefrontal
growth of vocabulary (Baldwin, 1995). Unfortunately, children with autism show impair- association cortex
ments primarily in IJA rather than RJA. This may account for some of the developmental
delays experienced by autistic children. We cover autism in greater detail elsewhere in this
chapter. Interestingly, chimpanzees have the capacity for RJA but not IJA (Tomasello &
Carpenter, 2005).
Social cognition begins around the age of 9 to 12 months. At about this time, infants
should begin to be aware of their own goals and activities as well as those of others to pay attention to where people are looking or pointing and, so, is a more basic or
(Tomasello et al., 2005). Social cognitive skills allow the child to combine these. They thus perceptual level of joint attention.
realize that their intentions lead to actions and that this is the case for others as well. If I Secondarily, there is the anterior attention system. This is subserved by a collection of
want to eat, then I look at or point to the cereal box because I want to eat it or get mom different neural areas, including the frontal eye fields, which control eye movement; the
to help me eat it. If mom also points or looks at the cereal, then she, too, must be thinking orbital frontal cortex; the prefrontal association cortex; and the anterior cingulate. This
about it and plans to prepare it for me to eat. In this fashion, children learn that thought system underlies our capacity for IJA and emerges slightly later in development. It controls
is representational and directed toward objects and actions in the world, not just in them- voluntary, goal-directed attention and is what enables us to understand that our personal
selves but in others as well. This realization can then lead to the use of language and com- behavior is directed toward where we attend. This may be thought of as a more sophisti-
munication to convey the contents of one's mind to others. cated conceptual level of joint attention.
There are two attention-regulation systems underlying joint attention (Mundy, Figure 11.3 shows the attention-systems model of how these two systems interact to
Card, & Fox, 2000). These are depicted in Figure 11.2. The posterior orienting and determine joint attention. Early in development, at the age of 3 to 6 months, the anterior
perceptual system is used during the learning of RJA in infancy. It consists of the pari- and posterior systems interact with each other and with additional processes such as
etal and superior temporal cortex. This system is involuntary and is used to direct speed of processing, motivation, and the ability to divide attention. This dynamic leads
attention to biologically meaningful stimuli. This underlies our capacity to perceive to the development of integrated self-other attention processing at 7 to 9 months, which
the eye and head orientation of others as well as the spatial representation of our own in turn serves as the foundation for the emergence of social-cognitive joint attention at
bodies in relation to other people's bodies and objects in the environment. It allows us 10 to 18 months.
330 Cognitive Science
Chapter 11 The Social Approach jj I
Figure 11.3 How the two attention systems interact to determine joint attention and parietal areas, may explain autistic people's lack of social cognitive skills. In chimpan-
zees, the parietal attention system is represented, but the anterior system is not. This may
explain why their RJA ability is better than their IJA ability (Astafiev et al., 2003 ).
Other Cognitive
Mirror Neurons
Processes
Processi ng Speed
Splitting Atte ntion A mirror neuron is a neuron that fires when an animal performs some action and also
Motivation when the animal observes another animal performing that same action (Rizzolatti &
Craighero, 2004). So, for example, if a monkey were to reach out and grasp an object, we
might see activity in particular neurons. These neurons would also fire when that monkey
watches another monkey reach out to grasp the same object. In this sense, the neurons are
said to "mirror" the performance of others. These cells could play a role in imitative or
3- 6 Months
modeling behavior and help explain how some animals copy others as part of the learning
process. Although mirror neurons have been observed in primates, whether they exist in
humans is still a topic of debate.
In macaque monkeys, mirror neurons have been found in the inferior frontal gyrus and
IJA RAJ
Anterior Attention the inferior parietal lobe (Rizzolatti & Craighero, 2004). In one study, 41 neurons were
Posterior Attention
and Intentional and Reflexive studied in two macaque monkeys while they observed an experimenter either grasp an
Control of Attention Attention Control apple and bring it to his mouth or place it in a cup (Fogassi et al., 2005). They discovered
15 neurons that fired to the "grasp-to-eat" motion but did not fire to the "grasp-to-place"
condition. Conversely, there were 4 other neurons that showed increased activity to the
"grasp-to-place" motion but not to the "grasp-to-eat" motion. The cells fired in some
7- 9 Months
cases before the final grasp was performed (i.e., before the apple was brought to the mouth
or placed in the cup). These researchers speculate that these neurons, therefore, code for
the animal's understanding of another's intention. That is, they become active when the
monkey anticipates what somebody will do.
Several studies have examined whether mirror neurons exist in people. Iacoboni et al.
Social Attention
(1999), using fMRI, found activity in the inferior frontal cortex and the superior parietal
Executive Function
Integrated Self lobe when a person performs an action and also when the person sees another person
Additional Atte ntion perform that action. The specific pattern of brain activity in these areas also has been
Processes found to be similar during the perception of others' actions and when performing the act
oneself (Dinstein, Gardner, Jazayeri, & Heeger, 2008) . However, the problem with these
studies is that, although they may show activity in a mirror neuron "system" involving a
large population of cells, they fail to show activity in individual neurons. Data from other
Source : Mu ndy et al. (2000). studies suggest that, although there may be mirror neurons in the human brain, they con-
stitute only a tiny fraction of the cells that are active in these regions during observation
and execution of a movement (Lingnau, Gesierich, & Caramazza, 2009) .
The attention-systems model can account for two important observations
If mirror neurons are not used to model actions or behavioral intent, they may be used
mentione~ previously. Cherkassky, Kana, Keller, and Just (2006) propose that
instead as part of the experience of empathy (Decety & Jackson, 2004). There is some
problems m communication between brain areas, especially between the anterior
support for this .in imaging studies with human participants. Sections of the anterior
332 Cognitive Science
Chapter 11 The Social Approach 333
cingulate cortex that are thought to mediate the subjective experience of pain are also Mitchell (2008) strongly defends the field against these accusations. He points out that
active in individuals when they watch a video of somebody else about to experience bodily there are a small number of brain regions active during social cognition that are not active
harm-for instance, a person who is about to cut his or her finger with a kitchen knife during most other nonsocial cognitive tasks. For instance, the medial prefrontal cortex
(Singer et al., 2004). Smelling a foul odor was also found to produce activity in the ante- (mPFC) is activated when participants are asked to read stories that require mental attri-
rior insula, both when experiencing this directly and when watching somebody else bution, in comparison with those that involve physical causality (Fletcher et al., 1995).
express disgust to a smell (Wicker et al., 2003). However, these brain areas are not the Numerous other studies also have found differential mPFC activation in social cognition
same as those that mirror hand actions, and mirror neurons coding for empathy and emo- tasks under circumstances where the stimuli, type of mental state, and behavioral tasks
tional states have not yet been found in monkeys.
vary (Blakemore et al., 2004 ).
Additional evidence supports this rebuttal. Mitchell, Macrae, and Banaji (2004) per-
formed functional magnetic resonance imaging (fMRI) scans on individuals while they
Social Cognition as the Brain's Default State read descriptions of people and then were instructed either to form an impression based
on the passage or to memorize the information. The impression formation task produced
Mitchell (2008) proposes a speculative but fascinating new idea, which is that social cog- mPFC activity, while the memorization task did not. In addition, the mPFC activity was
nition may be so special that it is the default state for the brain. The areas implicated in correlated with the likelihood of remembering material, but only when it was associated
social thinking, such as the mPFC, rTPJ, and precuneus/posterior cingulate cortex, have with the impression-formation task.
an unusually high metabolic rate; they tend to burn through energy faster than other brain However, research also shows that there may be some functional overlap of brain areas
areas when at rest (Gusnard & Raichle, 2001). A higher resting rate means that these areas underlying social cognition and other more general cognitive operations. Saxe and
remain "on" and ready for action even though we may not be recruiting them to perform Kanwisher (2003) found right temporoparietal junction (rTPJ) activation when partici-
some task. This suggests that we remain in "people mode" whenever we are not actively pants processed stories about another person's erroneous beliefs. This activation was not
thinking about anything in particular.
seen for thinking about stories that referred to an erroneous physical representation. But
Social brain areas have also been shown to deactivate when people are thinking about to complicate things, it turns out that the rTPJ is activated during tasks that require shifts
nonsocial topics. So if you are pondering the answer to an exam question that requires seman- of attention (Mitchell, 2008). So it turns out this debate revolves around the specific type
tic and conceptual knowledge unrelated to people, you literally need to "turn off" the social of social cognition performed. Unique brain areas are called on when doing some types of
parts of your brain, perhaps so they don't interfere. This seems to be unique to social cognition. social thinking, but other forms of social mentalizing can recruit areas used in general
When thinking about a nonlinguistic problem, for example, we don't deactivate language cognitive processes.
areas. These regions simply remain at their lower resting metabolic rate. Mitchell (2008) con-
cludes that we are in a state of constant readiness to encounter other minds, and this normal
state must be shut down in order to reason about things that don't involve people. ADVANTAGES OF THE SOCIAL
COGNITIVE NEUROSCIENCE APPROACH
IS SOCIAL COGNITIVE NEUROSCIENCE SPECIAL? What are some of the potential benefits of this tripartite interdisciplinary venture? Ochsner
and Lieberman (2001) advance three possibilities. They argue that social cognitive neuro-
Social cognitive neuroscience is not without its critics. Some researchers question whether science could unify the field by discovering common brain mechanisms underlying what
it has been able to provide new insights into the nature of cognitive processes (Willingham were previously thought to be disparate phenomena. If the same parts of the brain are used
& Dunn, 2003 ). They question whether neuroimaging has yielded any novel information for attitudes and attributions, for example, then we may have to reconsider these not as
that could not be gleaned with more traditional methods such as accuracy and reaction two different processes but perhaps as two aspects of the same process. The amygdala, to
time. Another question that has been raised is whether the general cognitive processes used give another example, is implicated in many social cognitive processes, including attitudes,
i~ perception, attention, memory, language, and problem solving can explain social cogni- stereotyping, emotion, and person perception. An understanding of what the amygdala
tiOn or whether there are special cognitive mechanisms or neural processes that are specific does in each of these situations can unify these formally distinct research topics at least to
to social interaction (Blakemore, Winston, & Frith, 2004).
the extent that they call on similar neural systems.
Chapter 11 The Social Approach 335
334 Cognitive Science
Social cognitive neuroscience may also have the opposite effect. In some cases, it could Figure 11.4 The Sally-Anne task used to test theory of mind (after Gallagher,
separate what might have been thought to be unitary phenomena. Currently, stereotypes 2000)
are broadly categorized as implicit or explicit. However, it is likely that there are several
different subtypes that could include affective, semantic, declarative, and procedural ste-
reotypes. It remains to be seen whether different neural mechanisms underlie each of these
possibilities or whether there is a single or small number of distinct processes they share.
A third potential benefit of social cognitive neuroscience concerns bottom-up versus
top-down approaches. By bottom-up approaches, we mean investigating the basic neural
systems that service the perceptual, attentional, emotional, and behavioral processes of
social experience. Top-down approaches refer instead to the diverse types of social phe-
Anne has a box.
nomena, such as attitudes, attributions, stereotypes, and prejudice. It is generally agreed Sally has a basket.
that more can be learned by integrating these two approaches than by treating them sepa-
rately, as has been done most often in the past.
THEORY OF MIND
One of our most important social abilities, one that underlies all others, is the understand- Sally has a marble. She puts the marble into basket.
ing that other people have minds. When you speak to your roommate, you assume that he
is conscious and has feelings and thoughts in the same way that you do. You don't treat
him as though he is an inanimate object like a chair or table. This capacity is referred to
as a theory of mind (ToM) . It is the ability to understand others' mental states, such as
beliefs, desires, and intentions; to appreciate how these differ from our own; and to utilize
this understanding to guide our behavior when interacting with others.
Sally goes out for a walk.
A number of tests have been developed to diagnose ToM (Wimmer & Perner, 1983). The
use of these tests shows that children acquire a ToM by the age of 4 to 5 years. In the Sally-
Anne task, a cartoon-like drawing depicting two children is shown (Gallagher, 2000). Sally
puts her marble in a basket. When she leaves, Anne takes the marble out of the basket and
puts it into her box (Figure 11.4 ). Sally now returns. The question is, where will Sally look
for her toy? Children without disorders aged 4 years and older will answer that she first
looks for it in her pram. This requires an understanding of the situation from Sally's point Anne takes the marble out of the basket and puts it into the box.
of view-namely, that she holds the false belief that the toy is in the pram because that is She wants to play with her marble.
where she last left it. A child lacking ToM would predict that she would look for it in the Now Sally comes back.
box, even though she could not have experienced this since it was from Anne's perspective.