Journal of Applied Psychology © 2011 American Psychological Association

2012, Vol. 97, No. 2, 421– 434 0021-9010/12/$12.00 DOI: 10.1037/a0026701

Team Structure and Regulatory Focus:

The Impact of Regulatory Fit on Team Dynamic

Nikolaos Dimotakis Robert B. Davison and John R. Hollenbeck

Georgia State University Michigan State University

We report a within-teams experiment testing the effects of fit between team structure and regulatory task
demands on task performance and satisfaction through average team member positive affect and helping
behaviors. We used a completely crossed repeated-observations design in which 21 teams enacted 2 tasks
with different regulatory focus characteristics (prevention and promotion) in 2 organizational structures
(functional and divisional), resulting in 84 observations. Results suggested that salient regulatory
demands inherent in the task interacted with structure to determine objective and subjective team-level
outcomes, such that functional structures were best suited to (i.e., had best fit with) tasks with a
prevention regulatory focus and divisional structures were best suited to tasks with a promotion
regulatory focus. This contingency finding integrates regulatory focus and structural contingency
theories, and extends them to the team level with implications for models of performance, satisfaction,
and team dynamics.

Keywords: teams, regulatory fit, affect, helping, performance

It is a fundamental reality of human nature that people are of workplace functioning. The theory of regulatory fit posits that
motivated to approach pleasure and avoid pain. Known as the fit between goals and the means available to achieve these goals
hedonistic principle, this theoretical concept has underpinned influence perceptions, behaviors, and performance in meaningful
much of the scholarship in the area of motivation. In his theory of ways (Higgins, 2000, 2002). Understanding the ways in which
regulatory focus, Higgins (1997) argued that individuals are goals are selected and means provided is an important, yet under-
guided by two distinct motivational systems: prevention and pro- studied area.
motion. The needs that individuals seek to satisfy, the standards Although previous research on regulatory focus has provided
that individuals try to align themselves with, and the outcomes that important insights on the role of dispositions and states in influ-
are salient to them are three motivational factors that differentiate encing goals, the role of situational characteristics has been rela-
the two orientations (Brockner, Paruchuri, Idson, & Higgins, tively neglected, even though this is an area described as an
2002). Research has shown that each regulatory focus leads to important focus for future research (Brockner & Higgins, 2001;
unique psychological and behavioral consequences, and that these Wallace & Chen, 2006). One prominent factor that can help
processes are a function of dispositional and situational factors explain situation-induced regulatory focus in organizational set-
(see Förster, Higgins, & Bianco, 2003; Higgins, 2000; Wallace & tings is the demands placed upon individuals by task requirements.
Chen, 2006). For example, a sales team tasked with increasing penetration in a
Although regulatory focus relates to the goals to be pursued, an
specific region and another sales team responsible for maintaining
equally important consideration is the means available in their
existing accounts are subject to different regulatory situational
pursuit, as fit or misfit between the two can influence many aspects
demands, with the former being exposed to demands deriving from
a promotion-focused context and the latter being exposed to de-
mands deriving from a prevention-focused context. Research ex-
This article was published Online First December 19, 2011. amining the effects of work characteristics has demonstrated links
Nikolaos Dimotakis, Department of Managerial Sciences, J. Mack Rob-
between the motivational, social, and contextual features of the
inson College of Business, Georgia State University; Robert B. Davison
and John R. Hollenbeck, Department of Management, Eli Broad Graduate
work environment and outcomes such as performance and well-
College of Business, Michigan State University. being (Humphrey, Nahrgang, & Morgeson, 2007; Parker & Wall,
This research was supported by the U.S. Army via the Small Business 2001; Wallace, Johnson, & Frazier, 2009). Thus, conceptualizing
Innovation Research grant program. We would like to express thanks for regulatory orientation as inherent in the outcome demands of a task
this support, but note that the ideas and views presented in this article are enhances a number of literatures that are contingent upon features
the authors and do not necessarily reflect the views of the U.S. Army. We or characteristics of the task environment.
would also like to thank Daniel Serfaty, Jean MacMillan, Georgiy As mentioned above, regulatory fit investigations need to ex-
Levchuk, and the staff of Aptima Corporation for access to subject matter
amine means as well as goals, and the means available for goal
experts and programming support.
Correspondence concerning this article should be addressed to Nikolaos pursuit are influenced by characteristics of the organizations and
Dimotakis, Department of Managerial Sciences, J. Mack Robinson College groups that individuals work in. As work is increasingly structured
of Business, Georgia State University, Atlanta, GA 30303. E-mail: around teams (Ilgen, 1999), such considerations become progres-
[email protected] sively more important. Of the salient characteristics that can in-

421
422 DIMOTAKIS, DAVISON, AND HOLLENBECK

fluence available means, structure has long been identified as often associated with the pursuit of desired outcomes; thus, inves-
particularly important; its effects, however, are not universal but tigating their effects can inform frameworks that have situation at
rather context specific. Research in the organizational and team their core. Second, by integrating regulatory focus and structural
literatures has convincingly demonstrated that characteristics and contingency theories, this study expands the domain of both of
demands of the environment in which an organization is embedded these theoretical frameworks, extends regulatory fit approaches,
interact with its structure to determine the organization’s ultimate and contributes to an emerging stream of research in the area of
effectiveness (Beersma et al., 2003; Hollenbeck et al., 2002; regulatory focus within team settings (e.g., Chen, Kanfer, DeShon,
Thompson, 1967). Depending on the degree of fit between the Mathieu, & Kozlowski, 2009; Wallace & Chen, 2006). Third, we
characteristics of the structure and the demands of the situation or propose two basic and fundamental mediators of the effects of
environment, a particular organizational structure is advantageous regulatory fit (helping behaviors manifested and average levels of
in some task environments but a hindrance in others (Galunic & positive affect experienced), thus providing a look into the pro-
Eisenhardt, 1994). This suggests that contingency theories of or- cesses by which regulatory fit operates. Although prior work has
ganizational structure (Burns & Stalker, 1961; Katz & Kahn, 1966; investigated affect and citizenship behaviors at the level of the
Thompson, 1967) may hold the key to understanding the mean fit individual (e.g., Idson et al., 2000; Wallace & Chen, 2006), little
effects of situation-induced regulatory focus. extant research has considered these at the level of the group.
This article therefore aims to extend theories of regulation by
employing concepts of fit to integrate regulatory focus with struc- Theory and Hypotheses Development
tural contingency theory, investigating how context-derived task
goals interact with structure-provided means to influence team
Theories of Regulatory Focus and Fit
functioning and performance. Furthermore, to illuminate the pro-
cesses linking goals, means, and outcomes, we test two basic Regulatory focus theory (Higgins, 1997, 2000) argues that the
mediators of the relationship between regulatory fit and team-level nature of the outcomes (i.e., desired end-states) that are salient to
performance and satisfaction. The first—average level of team an individual is contingent upon their regulatory orientation, as is
member affect— captures the implicitly and explicitly shared, pos- the set of accessible and currently activated goals, because goals
itively valenced affective state of the team derived from a common are representational structures that guide pursuit of desired end-
context and purpose behavior (Kelly & Barsade, 2001). The sec- states (Markman & Brendl, 2000). The theory draws distinctions
ond—level of helping behavior— captures the rate of team mem- between the dominant means by which goals are strategically
bers assisting their teammates in the performance of their tasks by pursued as well, proposing that a natural fit exists between goal
providing verbal direction, feedback, or coaching, and thus is a means and regulatory orientation. Individuals in a promotion focus
subset of the team monitoring and backup behavior construct as eagerly seek to attain positive outcomes, whereas prevention-
defined by Marks, Mathieu, and Zaccaro (2001). focused individuals vigilantly seek to avert negative outcomes
Understanding helping behavior is important because it is one of (Förster, Higgins, & Idson, 1998). Importantly, strategic means
the most basic functions of teams (Kozlowski & Ilgen, 2006), and preference is the same regardless of whether the activated regula-
because little prior research has investigated this important behav- tory orientation is trait based or situation induced (Higgins &
ior in the regulatory domain (see Wallace et al., 2009, for an Spiegel, 2004).
exception). Positive affect, on the other hand, represents a basic When an individual pursues a goal in a manner that sustains his
psychological phenomenon with broad motivational and behav- or her regulatory orientation, the individual experiences regulatory
ioral implications, and has a recognized theoretical relationship fit (Higgins, Idson, Freitas, Spiegel, & Molden, 2003). Regulatory
with regulatory focus (Beal, Weiss, Barros, & MacDermid, 2005; fit influences judgments, decision making, feelings, emotions,
Brockner & Higgins, 2001; Idson, Liberman, & Higgins, 2000). attitude, behavior, and task performance (Higgins, 2005; Higgins,
Focusing on these two mediators allows us to provide an important Cesario, Hagiwara, Spiegel, & Pittman, 2010; Spiegel, Grant-
baseline investigation of the processes involved in linking team- Pillow, & Higgins, 2004). Friedman and Förster (2001) speculated
level regulatory fit to team outcomes. that environmental cues may automatically activate different pro-
Our level of analysis is the team, and though some of the cessing styles: one associated with a promotion orientation in
processes involved in our model are measured at the level of the response to nurturance cues and one associated with a prevention
individual, we posit that these represent individual-level psycho- orientation in response to threats to security cues. Several recent
logical and behavioral reactions to team-level structural stimuli studies have shown that regulatory fit plays a role in perception,
that emerge in the aggregate as meaningful team-level phenomena. information processing, persuasion, and decision confidence (e.g.,
Our model posits that situational demands interact with team Cesario & Higgins, 2008; Koenig, Cesario, Molden, Kosloff, &
structure to determine important team-level outcomes through Higgins, 2009). These studies suggest that salient cues in the
behavioral and affective mediating processes, and that both situa- environment, such as task demand characteristics (e.g., Chernev,
tional demands and team structure represent factors that the team 2009), activate different processing styles that induce a particular
experiences as a whole. regulatory orientation.
This study stands to contribute to the literature in three ways. By inducing a regulatory orientation, tasks stipulate the strategic
First, it responds to the call by Wallace and Chen (2006) for means by which they are preferentially accomplished (Förster et
research that examines situational factors affecting regulatory fo- al., 2003; Wallace et al., 2009). In addition, the degree of fit
cus and performance, underlining the importance of regulatory between preferred means and actual means effects behavioral and
demands as a task characteristic. Striving for accomplishment or affective responses when enacting these tasks. The effects of
striving to fulfill a responsibility are two motivational orientations different regulatory strategies and goal means fit have, however,
 REGULATORY FIT AND MEDIATING PROCESSES 423

prototypically been tested in tasks involving a single type of an attribute associated with a promotion orientation, whereas qual-
activity (e.g., Förster et al., 2003). Work teams, on the other hand, ity (avoiding errors of commission) is an attribute associated with
frequently face situations comprising much more complicated a prevention orientation (Förster et al., 2003).
tasks, involving activities that often need to be addressed simul-
taneously, and an assortment of resources that can be used to
achieve desired goals and outcomes. Although the dominant strat- Regulatory Focus, Structure, and Team Performance
egies to accomplish these tasks would remain the same, these
The way a team’s assets and responsibilities are structured
strategies are enacted in a complicated and often interrelated
directly relates to the ways in which a team will use its resources
context of different actions. An important challenge in such situ-
to successfully address the demands of the task environment.
ations, then, is to deploy available resources in a manner that
Members of divisional teams, where individual team members
optimally addresses the demands of the task environment.
have both a similar set of responsibilities and access to or control
over a wide variety of resources, have roles that are broad and
Structural Contingency Theory autonomous. Increased role breadth and independence of action
Contingency theories of structure (Burns & Stalker, 1961; Katz enable members of divisional teams to independently identify and
& Kahn, 1966; Thompson, 1967) hold that there is no one best way pursue a wide range of opportunities.
to structure an organization, but instead the appropriate structure On the other hand, structuring a team along functional lines,
for an organization is driven by the demands of the external task where individual team members are responsible for a specific
environment contingent on the resources (e.g., capabilities) avail- category of tasks and control specialized resources, tends to result
able. In environments typified by radical innovation, an organiza- in narrow, interdependent roles. Thus, individuals situated in a
tion can be more effective if it is structured to be less bureaucratic, functional structure often rely on the cooperation of others to
more informal and decentralized, with fewer levels of authority accomplish the task. Cooperation, by its nature, requires team
and a lesser degree of task specialization. Conversely, for rela- members to discuss and agree upon each action, which arguably
tively stable and predictable environments, an organization can be increases the quality of decisions but at the expense of quantity.
more effective when it realizes the efficiency gains that accrue Further, vigilance means are typified by a “be careful, be sure”
through the use of bureaucracy, centralization, and task special- attitude and achieve success through the absence of errors and
ization. Thus, the organization’s environment and its structural mistakes. Thus, they demand more task specialization, an attribute
characteristics interact to determine effectiveness. of a functional structure, than do eagerness means.
Recent work on structural contingency theory has translated this As discussed previously, individuals are more inclined toward
general proposition regarding abstract conceptualizations of fit goal means that represent a higher regulatory fit (Crowe & Hig-
into specific hypotheses involving concrete dimensions of task gins, 1997; Higgins, 2005), and their motivation is stronger during
structure in teams. For example, Hollenbeck et al. (2002) found goal pursuit when regulatory fit is higher (Förster et al., 1998;
that functional team structures that create roles that are narrow in Shah, Higgins, & Friedman, 1998). On the basis of these
scope but high in interdependence were superior in predictable individual-level results, as well as the team-level arguments, we
task environments, whereas divisional team structures that create expect that when teams experience fit between their structure and
roles that are broad in scope but low in interdependence demon- the regulatory focus orientation of the situation, they will exhibit
strated superiority in more unpredictable task environments. Sim- increased levels of performance (see Figure 1). Specifically, we
ilarly, Beersma et al. (2003) found that competitive structures argue that a natural fit exists between prevention-oriented task
promote performance in task contexts that placed an emphasis on demands and a functional form of team structure, and that a natural
performance quantity, whereas cooperative structures promoted fit exists between promotion-oriented task demands and a divi-
performance in contexts that emphasized quality of performance. It sional form of team structure. Further, we argue that compatibility
is important to note that quantity (avoiding errors of omission) is between team structure and regulatory focus will result in an

Figure 1. Conceptual model.

424 DIMOTAKIS, DAVISON, AND HOLLENBECK

experience of regulatory fit for the team, leading to increased Affective Outcomes of Regulatory Fit
performance. Therefore, we posit
In terms of the experience of working in a team structure that
Hypothesis 1: Team structure will interact with task regula- exhibits fit to the characteristics of the situation at hand versus one
tory focus orientation such that divisionally structured teams that does not, fit can be thought of as more likely to create frequent
will outperform functional teams in promotion task settings, events and circumstances that are generally positive in nature, such
whereas functionally structured teams will outperform divi- as being able to accomplish a part of a task quickly and with ease,
sional teams in prevention task settings. and making steady progress toward goal completion. According to
affective events theory (Weiss & Cropanzano, 1996), these events
have the capacity to influence individual affective levels. Affective
Regulatory Fit and Behavior events theory defines an event as “a change in circumstances, a
change in what one is currently experiencing” (Weiss & Cropan-
Configurations of resources and responsibilities that are optimal
zano, 1996, p. 31), and this definition is descriptive of the expe-
to the demands of the task at hand will not only enable team
rience when there is a fit between the structure of one’s team and
members to complete their tasks in an effective fashion, but also
the regulatory demands of the environment. Indeed, individuals’
enhance their capabilities to operate jointly to provide assistance to
feelings about a choice they might make is more positive for a
each other. In a divisional structure, this assistance is exhibited by
desirable choice and more negative for an undesirable choice
team members providing help to a teammate who is currently
(Idson et al., 2000; Idson, Liberman, & Higgins, 2004), and their
experiencing a higher workload, thus effectively using slack re-
retrospective evaluations of past decisions or goal pursuits are also
sources to tackle increased demands to the benefit of the team. In
more positive (Freitas & Higgins, 2002; Higgins et al., 2003; Idson
a functional structure, this assistance will be demonstrated in the
et al., 2000) when regulatory fit is higher. Therefore, we expect
form of team members engaging tasks within their own specialized
that regulatory fit will also result in increased average team mem-
function in an optimal manner, thus enabling fellow team members
ber positive affect, leading to a higher level of average positive
to focus on the tasks for which their resources are specialized,
affect for the team (see Figure 1).
regardless of the initial operational distribution of demands.
Conversely, this normal form of team functioning will be dis- Hypothesis 3: Team structure will interact with task regula-
rupted in cases of regulatory misfit, since the resources and capa- tory focus orientation such that members of divisionally
bilities available to team members will be ill-suited to accommo- structured teams will demonstrate higher levels of positive
date the necessary strategies for completing their tasks. The affect relative to members of functional teams in promotion
subsequent struggle to complete their tasks with a suboptimal task settings, whereas members of functionally structured
configuration of resources will leave team members unable to teams will demonstrate higher levels of positive affect rela-
provide the assistance necessary for optimal team functioning, tive to members of divisional teams in prevention settings.
reducing demonstrated instances of helping behavior. Thus, regu-
latory fit will be associated with the discretionary behaviors that Moreover, the same positive affective reaction to the expe-
team members will exhibit on average, such that teams experienc- rience of fit of team structure and regulatory focus within the
ing regulatory fit will display higher levels of helping behaviors task can be expected to relate to the average levels of job
relative to teams experiencing misfit. satisfaction within the team (Brief & Weiss, 2002). As we
articulated previously, the experience of regulatory fit in the
Hypothesis 2: Team structure will interact with task regula- task environment can result in the generation of affective events
tory focus orientation such that members of divisionally that can lead to positive affective reactions, thus indirectly
structured teams will demonstrate a higher level of helping heightening team members’ experience of satisfaction with
behaviors relative to members of functional teams in promo- their task. Experiencing regulatory fit can also result in cogni-
tion task settings, whereas functionally structured teams will tive judgments that result in increased satisfaction with one’s
demonstrate a higher level of helping behaviors relative to task, since team members performing the task in a context of
members of divisional teams in prevention settings. regulatory fit can be seen as operating within more agreeable
work conditions. This is in line with Motowidlo (1996), who
On the surface these predictions may seem somewhat in conflict defined reports of job satisfaction as “judgments about the
with the predictions and findings of Wallace et al. (2009), who favorability of the work environment” (p. 176). Thus, regula-
found that promotion-oriented individuals exhibited citizenship tory fit can result in increases in both components of satisfac-
behaviors, whereas prevention-oriented individuals did not. It is, tion with one’s work (defined in our case as satisfaction with
however, important to note that the latter prediction was based on the task).
the argument that prevention-oriented individuals are duty bound,
and since citizenship behavior is, by definition, an extrarole be- Hypothesis 4: Team structure will interact with task regula-
havior (Organ, 1988), these individuals feel no obligation to en- tory focus orientation such that members of divisionally
gage in them. The context of a team is quite different, however. In structured teams will demonstrate higher levels of task satis-
a team setting (as a member of a team pursuing a team-level task) faction relative to members of functional teams in promotion
helping one’s teammates is very much an in-role behavior, and settings, whereas members of functionally structured teams
thus we expect teams experiencing regulatory fit to exhibit in- will demonstrate higher levels of task satisfaction relative to
creased helping behaviors in both prevention and promotion tasks. members of divisional teams in prevention settings.
 REGULATORY FIT AND MEDIATING PROCESSES 425

Mediating Processes Method

With respect to mediating influences, affective experiences can Sample

relate to performance in two ways. First, increased levels of
positive affect indicate that an individual will not have to engage Our sample consisted of 110 undergraduates from a large uni-
in emotional regulation efforts (such as trying to control stress or versity, randomly assigned to 22 five-person teams. The average
frustration), freeing up resources that can then be focused on the participant age was 21 years (SD ⫽ 2.2), and 48% of the sample
task at hand (Beal et al., 2005; Schneider & Fisk, 1982). Second, was female. The sample consisted of 71% Caucasians, 8% African
increased positive affect has been associated with a broadening of Americans, 5% Hispanics, and 9% Asians, with the remainder
one’s scope of attention (Fredrickson & Branigan, 2005) and indicating other racial groups. Due to missing data, one team had
increased creativity (Isen, Daubman, & Nowicki, 1987). Com- to be excluded from the analyses; thus, our final sample consisted
bined these effects can help bring novel information and ideas to of 105 participants assigned to 21 teams. Each team completed
the team, further enhancing team performance. In general, positive four rounds of the simulation task (see below) for a total of 84
affect is associated with an increase in resource availability and team-level observations. Participants were compensated for their
resilience to adversity, and these effects can accumulate for teams efforts with a modest stipend ($30 for a 3-hr session).
in the same manner as individuals (Fredrickson & Losada, 2005).
Therefore, we expect that positive affect will mediate the relation- Procedure
ship between regulatory fit and performance.
Participants were randomly assigned to one of five computer
As we discussed above, increased positive affect as a result of
stations, each corresponding to a specific role in the simulation.
environmental events and features can influence judgments about
The simulation consisted of a new version of the dynamic distrib-
a task, which will in turn relate to an increase in task satisfaction
uted decision-making task specifically developed to explore tasks
(Weiss & Cropanzano, 1996). Although this effect relates more
differing in regulatory demands (prevention vs. promotion) that
strongly to the affective part of satisfaction with one’s task or job,
has not been previously used in any research setting. The basic task
we also expect that a positive mood will also relate to judgments
has been described in previous work (see Hollenbeck et al., 2002),
about the task of a more cognitive nature, through the informa-
but the differences from the basic dynamic distributed decision-
tional effects of experienced moods (Schwarz & Clore, 1983).
making task are described below.
Consequently, we expect that positive affect will mediate the
Each computer station was in control of a number of assets. In
relationship between regulatory fit and average team member
order to complete scenario objectives, these assets could be used to
satisfaction with the task.
engage a variety of hostile targets. Unlike the traditional dynamic
In terms of helping behaviors, we expect that they will relate to distributed decision-making task, all on-screen targets were sta-
both task performance and task satisfaction. In general, increased tionary and visible to all computer stations once they appeared. All
levels of helping behavior can benefit team performance by pro- targets required the simultaneous utilization of two to four assets to
viding aid to team members in need (Porter, 2005; Porter et al., be successfully engaged. These assets could come from any num-
2003), thus ensuring that all task responsibilities are completely ber of team members. Some targets could be engaged by a single
covered, as well as providing increased resources to teammates in team member using only his or her own assets, whereas others
need, resources that would potentially be idling otherwise. required members to collaborate in order to successfully engage a
Finally, providing and receiving help could also relate to an target. The required combinations of assets for each target were
increase in task satisfaction. Receiving help is a positive event that posted alongside each computer station for ease of reference.
helps the recipient fulfill his or her task responsibilities with
greater ease, and at the same time enhances the recipient’s per-
Research Design
ception of his or her work environment. Similarly, providing help
to others relates to significant psychological benefits for the indi- All teams completed four preprogrammed simulation scenarios
viduals providing the aid (Penner, Dovidio, Piliavin, & Schroeder, that lasted 25 min. Teams were assigned to a specific structure
2005), due in part to increased social integration and an increase in condition (divisional or functional) and to a task characterized by
positive moods (Midlarsky, 1991). Since an increased level of a specific regulatory focus condition (promotion or prevention) in
average helping behaviors in a team environment will translate to each scenario. Thus, each team completed every one of the four
a higher level of providing (for some team members) and receiving possible combinations of conditions; the order in which each game
help (for the team members being helped), we expect that helping was played was randomized across teams. After completing each
behaviors will also mediate the relationship between regulatory fit scenario, team members provided measures of positive affect
and average level of task satisfaction. See Figure 1 for a represen- experienced during the task, the helping behaviors they demon-
tation of these expected relationships. strated, and their level of task satisfaction. Scenarios were parallel
in terms of total and average workload, duration, and pacing; the
Hypothesis 5A: Positive affect and helping behaviors will graphical interface and starting locations for team member assets
mediate the effects of the interaction between structure and were also held constant across conditions.
regulatory focus on task performance.
Manipulations
Hypothesis 5B: Positive affect and helping behaviors will
mediate the effects of the interaction between structure and Team structure. The team structure, divisional or functional,
regulatory focus on task satisfaction. was manipulated by the assets under each team member’s control.
426 DIMOTAKIS, DAVISON, AND HOLLENBECK

In the divisional structure, team members were personally respon- entation for the prevention task and promotion orientation for the
sible for an entire geographic region of the map, and were able to promotion task). Furthermore, a majority of SMEs generally re-
engage most targets and objectives using only the assets under ported that the two tasks did not differ in terms of the other
their control. Thus, this structure was low in interdependence but characteristics examined, with 50% reporting no differences in
broad in scope, in the sense that each person was responsible for terms of complexity, 85% reporting no differences in teamwork
an average of 12 types of assets of varying capabilities (out of 17 requirements, 64% reporting no difference in cognitive ability
assets available; see Appendix). In the functional structure, team requirements, and 71% stating that speed and accuracy were
members controlled, on average, only six types of assets, and equally important for both tasks. The remaining responses were
hence this role was easier to learn and execute because of its split evenly among the two conditions (except for complexity, in
narrower scope. However, each person could engage only a rela- which 43% of SMEs found that the promotion task was more
tively limited number of targets by him- or herself, and hence each complex). Thus, our content validation analysis revealed that our
person had to coordinate the majority of his or her attacks with two scenarios were well characterized in terms of regulatory focus,
other team members. Thus, this structure was high in interdepen- and the data did not reveal any overwhelming differences on either
dence. The structures were similar in the average number of assets task.
controlled by each team member (see Appendix), in the position-
ing of these assets across the operational graphical interface, and in
Measures
relative workload.
Regulatory task characteristics. The regulatory character- Average team member positive affect. Average team mem-
istic of each task was manipulated by placing the participants in a ber positive affect was measured with the 10-item positive affect
scenario that involved either prevention or promotion objectives. component of the 20-item Positive and Negative Affect Schedule
These objectives were assigned by a description of the upcoming (Watson, Clark, & Tellegen, 1988). At the end of each scenario,
task requirements provided at the beginning of each scenario, as team members used Likert-type scale responses (ranging from 1 ⫽
well as by the actual simulation structure. In the promotion task not at all to 5 ⫽ extremely) to indicate how accurately the mood
condition, team members were told the following: “You have to descriptors described the way they felt during the scenario they
assume control of an area that is currently under enemy control. To had just completed; positive affect responses were then averaged to
do so, you have to engage enemy forces in order to take over the create a team-level variable reflecting the average affective level of a
maximum amount of territory” (a gain focus). Participants began team’s members for a given scenario. Across scenarios, the average
with their assets surrounding a marked territory on the game map reliability of this measure was .94. Before aggregating, we conducted
that was currently under the control of enemy forces. To succeed a variety of interrater agreement analyses to ensure there was enough
in this condition, participants had to use the assets they controlled agreement across team members to enable aggregation; these analyses
to advance on the enemy forces and successfully engage them in showed what LeBreton and Senter (2008) would characterize as
order to gain control over the target area. In other words, in this moderate to strong levels of agreement, adequate to proceed with
task the team members’ objective was to gain or acquire territory aggregation (Rwg(maximum variance) ⫽ .72; Rwg(uniform variance) ⫽ .61;
not currently held. ICC(1) ⫽ .20; ICC(2) ⫽ .84; F ⫽ 6.08, p ⬍ .01).
In the prevention task condition, team members were told the Task characteristics and team structure. Dummy coding
following: “In this scenario, an area you control is under attack, was used to capture team structure (0 ⫽ divisional, 1 ⫽ functional)
and you will have to be vigilant in defending it from enemy forces and task requirements (0 ⫽ promotion, 1 ⫽ prevention).
that will attempt to infiltrate and take it over” (a nonloss focus). Task performance. Task performance was operationalized
Participants began with their assets inside the same marked terri- as the number of targets that the team successfully engaged
tory on the map that they controlled. To succeed in this condition, during the scenario. Team performance scores could range from
participants had to use the assets they controlled to intercept and 0 to 74; observed scores in the current sample of teams ranged
stop enemy forces that were attempting to infiltrate the territory the from 8 to 74.
team was assigned to protect. Performance was measured as the Average team member task satisfaction. Average team
number of targets successfully engaged during the scenarios; thus member satisfaction with the task was assessed with a modified
the tasks had the same desired end-state (Idson et al., 2000). version of the five-item measure from Brayfield and Rothe (1951).
To ensure that our manipulation correctly represented situations At the end of each scenario, each team member was asked to use
with a promotion or prevention regulatory focus, we conducted a Likert-type scale responses (ranging from 1 ⫽ strongly disagree to
content validation analysis using a sample of 13 subject matter 5 ⫽ strongly agree) to indicate the degree of his or her agreement
experts (SMEs). These SMEs were provided with a description of with statements such as “I found real enjoyment in performing this
the two tasks (beginning with the common features of the situation task.” Responses were averaged to create a team-level variable
and proceeding to the two objectives and situational demands), and describing the mean level of task satisfaction for that scenario. We
were then asked to indicate which regulatory focus orientation best conducted interrater agreement analyses to ensure aggregation was
described each task. SMEs were also asked a series of questions appropriate; these analyses showed moderate levels of agreement,
related to their perception of the complexity, performance impor- allowing us to proceed with aggregation (Rwg(maximum variance) ⫽
tance (speed vs. accuracy), and the cognitive and teamwork re- .66; Rwg(uniform variance) ⫽ .68; ICC(1) ⫽ .17; ICC(2) ⫽ .81; F ⫽
quirements of each task; the SMEs were not informed in advance 5.22, p ⬍ .01). Across scenarios, the average reliability of this
of the purpose of this exercise. Results were supportive of the measure was .85.
intended characterization of each task; 100% of the SMEs char- Average team member helping behavior. Average team
acterized the tasks according to our expectations (prevention ori- member helping behavior was assessed with a modified five-item
 REGULATORY FIT AND MEDIATING PROCESSES 427

version of the Helping dimension of the Extra-Role Behaviors Results

Scale from Van Dyne and LePine (1998). At the end of each
scenario, each team member was asked to use Likert-type scale Table 1 presents the means, standard deviations, and intercor-
responses (ranging from 1 ⫽ strongly disagree to 5 ⫽ strongly relations among the variables included in the study. In terms of
agree) to indicate his or her agreement with statements such as “I main effects, being in a functional team structure, relative to being
assisted others in this group with their task for the benefit of the in a divisional structure, was associated with lower levels of task
group.” Responses were then averaged to create a team-level performance (standardized ␥ ⫽ ⫺.41, p ⬍ .01), lower levels of
variable reflecting the level of helping behavior exhibited for a task satisfaction (␥⬘ ⫽ ⫺.40, p ⬍ .01), and lower levels of average
given scenario. As the nature of the helping construct necessitates positive affect (␥⬘ ⫽ ⫺.22, p ⬍ .05). Being in a task characterized
that teams will have members who provide help and other team by a prevention regulatory focus, relative to being in a task
members who receive help, a high level of agreement is not characterized by a promotion regulatory focus, was associated with
expected. Accordingly, analyses did indicate mixed results in lower levels of average positive affect (␥⬘ ⫽ ⫺.22, p ⬍ .05).
terms of interrater agreement analyses with high Rwg but low ICC Neither team structure nor task characteristics had any main effects
values (Rwg(maximum variance) ⫽ .77; Rwg(uniform variance) ⫽ .73; relative to helping behaviors.
ICC(1) ⫽ .06; ICC(2) ⫽ .55; F ⫽ 2.22, p ⬍ .01). The average Before conducting hierarchical linear modeling analyses to test
reliability for this measure was .90. the study hypotheses, we conducted variance analyses to examine
To avoid having to rely exclusively on self-reported behavior, whether there was sufficient within-teams variance in the depen-
we also collected a validation measure at the end of each round. dent variables to proceed. These analyses are presented in Table 2;
This one-item measure required participants to indicate which specifically, the proportion of within-teams variance was 17.3%,
teammate was most helpful to their team in that particular scenario. 64.5%, 21.2%, and 47.5% for task performance, average helping
For each participant, others’ reports of helpfulness across scenarios behaviors, average team member positive affect, and task satisfac-
were summed to create an overall other-reported index of individ- tion, respectively. Therefore, there was sufficient within-teams
ual helping behaviors, which was then correlated with the partic- variance to enable estimations of hierarchical linear models.
ipants’ own average reports of helping. Results indicated that The results of the models testing Hypotheses 1– 4 are reported in
self-report measures strongly predicted others’ ratings of one’s Table 3. First, the interaction between team structure and task
own helpfulness (r ⫽ .49, p ⬍ .01). characteristics was significantly related to task performance (␥⬘ ⫽
.25, p ⬍ .05). Specifically, being in a divisional structure was
associated with higher performance levels in promotion-
Analyses
characterized tasks, whereas being in a functional structure was
To test the within-teams hypotheses, we used hierarchical linear associated with higher performance levels in prevention-
modeling (Bryk & Raudenbush, 1992), whereby the scores for the characterized tasks (see Figure 2).
relevant outcomes were regressed on the scores for the hypothe- Furthermore, the interaction between team structure and task
sized predictors across scenarios. In these analyses, team structure, characteristics was related to average helping behaviors (␥⬘ ⫽ .19,
task characteristics, and their interaction term were entered in p ⬍ .05). That is, being in a divisional team structure was asso-
models predicting task performance (Hypothesis 1), helping be- ciated with higher levels of helping behaviors in promotion-
haviors (Hypothesis 2), average team member positive affect (Hy- characterized tasks, whereas being in a functional team structure
pothesis 3), and task satisfaction (Hypothesis 4). To test the was associated with higher levels of helping behaviors in
mediation hypothesis (Hypothesis 5), we used the procedure de- prevention-characterized tasks. Conversely, functional teams in
scribed by Baron and Kenny (1986), as well as Sobel tests con- promotion tasks and divisional teams in prevention tasks evi-
ducted using the z⬘ unbiased estimator described by MacKinnon, denced lower levels of helping behaviors (see Figure 3).
Lockwood, Hoffman, West, and Sheets (2002), to test the signif- Average team member positive affect was also related to the
icance of the mediated effect. interaction between structure and task characteristics (␥⬘ ⫽ .29,

Table 1
Between- and Within-Individual Correlations Among Study Variables

SD

Variable M Between teams Within teams 1 2 3 4 5 6

ⴱⴱ ⴱ ⴱⴱ ⴱ
1. Positive affect 2.39 0.46 0.23 — .25 .27 .61 ⫺.22 ⫺.21ⴱ
2. Task performance 41.80 10.85 16.06 .57ⴱⴱ — .22ⴱ .43ⴱⴱ ⫺.41ⴱⴱ ⫺.04
3. Helping behaviors 3.70 0.25 0.28 .27 .12 — .45ⴱⴱ .05 .04
4. Task satisfaction 3.48 0.50 0.35 .79ⴱⴱ .55ⴱⴱ .59ⴱⴱ — ⫺.40ⴱⴱ ⫺.09
5. Structurea 0.50 0.50 0.50 — .00
6. Task characteristicsb 0.50 0.50 0.50 —

Note. The correlations below the diagonal represent between-teams associations (N ⫽ 21). The correlations above the diagonal represent within-teams
associations (over time) and were estimated from fixed-effects hierarchical linear models with single Level 1 predictors and no Level 2 predictors (N ⫽ 84).
a
Coded 0 ⫽ divisional, 1 ⫽ functional. b Coded 0 ⫽ promotion, 1 ⫽ prevention.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01 (two-tailed).
428 DIMOTAKIS, DAVISON, AND HOLLENBECK

Table 2
Variance Components for the Within-Individual Variables

Within-teams Between-teams % variability

Variable variance (␳2) variance (␶00) within teams

Task performance 53.32 257.90 17.3

Helping behaviors 0.078 0.041 64.5
Positive affect 0.053 0.197 21.2
Task satisfaction 0.123 0.136 47.5

Note. The variance components were obtained from null hierarchical

linear models. ␳2 ⫽ within-individual variance in dependent variable;
␶00 ⫽ between-individual variance in dependent variable. Percent variabil-
ity within individuals is computed as ␳2 ⫽ (␳2 ⫹ ␶00).

p ⬍ .01). Results indicated that being in a divisional structure and

a promotion-characterized task was associated with high levels of
positive affect; for this structure, being in a prevention-
characterized task was associated with low levels of positive affect Figure 2. Interaction of structure and regulatory focus in predicting
(see Figure 4). performance.
Finally, the interaction between team structure and task charac-
teristics was found to be significantly related to task satisfaction
(␥⬘ ⫽ .41, p ⬍ .01). Divisional team structures were associated in magnitude (becoming statistically nonsignificant in the case of
with higher levels of team-level task satisfaction in promotion the model predicting task performance), whereas average team
tasks, whereas functional structures were associated with higher member positive affect was significantly associated with higher
levels of team-level task satisfaction in prevention tasks. Divi- levels of performance (␥⬘ ⫽ .21, p ⬍ .01) as well as higher levels
sional structures, on the contrary, experienced a sharper decrease of task satisfaction (standardized ␥ ⫽ .46, p ⬍ .01). A similar
of task satisfaction in prevention tasks, relative to promotion tasks, pattern of results was found for helping behavior; when average
whereas functional structures experienced modestly higher task helping behavior was included in the models predicting perfor-
satisfaction in prevention tasks relative to promotion tasks (see mance and task satisfaction, the coefficient for the interaction term
Figure 5). between team structure and task characteristics decreased in mag-
In summary, these interactions showed a pattern of results that nitude and significance (albeit to a smaller extent relative to the
generally provided support for Hypotheses 1– 4. The specific in- positive affect models), and helping behaviors were significantly
stances of interactions, however, showed nuanced patterns, with associated with performance (␥⬘ ⫽ .22, p ⬍ .05) and satisfaction
each variable demonstrating the subtle effects of regulatory fit and (␥⬘ ⫽ .45, p ⬍ .01).
misfit for teams in different structures when predicting the objec- In a final set of exploratory models in which positive affect and
tive and subjective variables of interest. helping behaviors were both included, positive affect was found to
Hypothesis 5 was also supported (see Table 4). When average mediate the effect of the interaction between structure and task
team member positive affect was included in the models predicting characteristics when predicting both task performance and satis-
task performance and satisfaction, the coefficient for the interac- faction (␥⬘ ⫽ .18, p ⬍ .01, and ␥⬘ ⫽ .39, p ⬍ .01, respectively),
tion term between team structure and task characteristics decreased whereas helping behaviors was found to mediate the effect of the

Table 3
Hierarchical Linear Modeling Results for Models Examining Within-Individual Effects of Structure and Regulatory Focus (Hypotheses
1– 4)

Task performance Helping behaviors Positive affect Task satisfaction

Predictor ␥ t ␥ t ␥ t ␥ t

Model 1
Intercept 43.79 18.94ⴱⴱ 3.70 70.82ⴱⴱ 2.39 24.50ⴱⴱ 3.61 31.79ⴱⴱ
Team structure ⫺.41 ⫺5.22ⴱⴱ .05 0.44 ⫺.22 ⫺2.37ⴱ ⫺.30 ⫺3.86ⴱⴱ
Regulatory focus .09 1.04 .04 0.37 ⫺.22 ⫺2.32ⴱ ⫺.07 ⫺0.86
Model 2
Intercept 41.80 18.08ⴱⴱ 3.70 70.82ⴱⴱ 2.39 26.49ⴱⴱ 3.75 27.20ⴱⴱ
Team structure ⫺.41 ⫺5.22ⴱⴱ ⫺.13 ⫺0.70 ⫺.52 ⫺3.71ⴱⴱ ⫺.71 ⫺4.38ⴱⴱ
Regulatory focus ⫺.04 ⫺0.43 ⫺.14 ⫺1.02 ⫺.50 ⫺3.05ⴱⴱ ⫺.47 ⫺3.15ⴱⴱ
Interaction term .25 2.35ⴱ .19 1.99ⴱ .29 3.01ⴱⴱ .41 3.85ⴱⴱ

Note. N ⫽ 84 observations from 21 teams. All coefficients standardized.

ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01 (two-tailed).
 REGULATORY FIT AND MEDIATING PROCESSES 429

Figure 3. Interaction of structure and regulatory focus in predicting Figure 5. Interaction of structure and regulatory focus in predicting task
helping behaviors. satisfaction.

interaction between structure and task characteristics on satisfac-

tion only (␥⬘ ⫽ .37, p ⬍ .01). Sobel tests were supportive of the objective (task performance) and subjective (task satisfaction)
significance of the mediating effect of positive affect for both task nature. In other words, teams that operated under conditions of
performance (z⬘ ⫽ 2.34, p ⬍ .05) and satisfaction (z⬘ ⫽ 2.69, p ⬍ regulatory fit performed better at their assigned tasks, and team
.05). The mediating effect of helping behaviors for task satisfac- members also reported higher levels of task satisfaction, relative to
tion was also found to be significant (z⬘ ⫽ 3.03, p ⬍ .01). conditions of misfit. Importantly, we also found that these effects
were partially mediated through affective and behavioral pro-
Discussion cesses, specifically average levels of team member positive affect
and the helping behaviors enacted by team members.
Our results provide support for our assertion of an interaction
between team structure and the regulatory characteristics of the
task, underlining both the importance of conceptualizing self- Theoretical Implications
regulation (promotion/prevention) as a task characteristic and the Our findings have direct applications to models of regulatory fit
moderating role of team structure when assessing the fit between (Higgins, 2000, 2002) by demonstrating that situation-derived
the nature of the team and the nature of the task. We demonstrated goals interact with structure-determined means to influence objec-
that when team structure is aligned with the regulatory orientation tive and subjective outcomes, thus contributing to the growing
of the task, there is an increase in team-level outcomes of both an body of self-regulation research at the level of the team. We extend
this theory by documenting how team structure interacts with the
regulatory characteristics of a task to determine outcomes. This
illustrates how regulatory fit theory can operate at the level of the
group, affecting task performance and team viability through af-
fective and behavioral pathways. This complements previous work
that demonstrated the influence that situational context can have
on regulatory focus (e.g., Crowe & Higgins, 1997; Higgins, 1997,
1998, 2002) by demonstrating how situational context includes
specific demands that different team structures are better or worse
suited to address.
Second, we demonstrate that regulatory orientation is a task
characteristic that can serve as an influential situational cue or
demand. This finding has important implications for empirical and
theoretical studies of task characteristics (see Humphrey et al.,
2007; Morgeson & Campion, 2003), as regulatory focus represents
a widely applicable conceptualization that can be used as an
attribute to describe the motivational orientation induced by tasks
in the workplace. Our study is the first to conceptualize and
investigate regulatory focus orientation as a prescriptive charac-
Figure 4. Interaction of structure and regulatory focus in predicting teristic of the task itself (as opposed to an individual difference or
positive affect. an internally derived state) and to employ a within-teams research
430 DIMOTAKIS, DAVISON, AND HOLLENBECK

Table 4
Hierarchical Linear Modeling Results for Models Examining the Mediational Role of Positive
Affect and Helping Behaviors (Hypothesis 5)

Performance Task satisfaction

Predictor ␥ t ␥ t

Positive affect
Intercept 7.33 0.78 1.96 25.94ⴱⴱ
Team structure ⫺.36 ⫺4.51ⴱⴱ ⫺.47 ⫺2.98ⴱⴱ
Regulatory focus .01 0.09 ⫺.24 ⫺1.79
Interaction term .19 1.82 .27 2.68ⴱ
Positive affect .21 4.05ⴱⴱ .46 5.79ⴱⴱ
Helping behaviors
Intercept ⫺4.60 ⫺0.24 1.66 27.57ⴱⴱ
Team structure ⫺.42 ⫺5.29ⴱⴱ ⫺.66 ⫺4.44ⴱⴱ
Regulatory focus ⫺.04 ⫺0.54 ⫺.41 ⫺2.72ⴱⴱ
Interaction term .21 2.21ⴱ .33 3.20ⴱⴱ
Helping behaviors .22 2.50ⴱ .45 5.58ⴱⴱ
Dual mediator model
Intercept ⫺19.20 ⫺0.95 .49 1.24
Team structure ⫺.37 ⫺4.59ⴱⴱ ⫺.44 ⫺3.39ⴱⴱ
Regulatory focus .00 ⫺0.03 ⫺.21 ⫺1.70
Interaction term .17 1.82 .22 2.43ⴱ
Positive affect .18 3.51ⴱⴱ .39 5.62ⴱⴱ
Helping behaviors .14 1.43 .37 4.76ⴱⴱ

Note. N ⫽ 84 observations from 21 teams. All coefficients standardized.

ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01 (two-tailed).

design; we thereby extend the literatures of regulatory focus, task the regulatory characteristics of the task to impact both the per-
characteristics, and team structure. formance of the group and the typical subjective experience of
A comparison of regulatory focus theory and prospect theory team members. Furthermore, we demonstrated both affective and
(Kahneman & Tversky, 1979) provides valuable insight into the behavioral processes that can mediate these effects, thus providing
choice of the former over the latter for this study and may, thereby, a more comprehensive description of the phenomena of interest,
serve to inform the design of future research. Prospect theory and a link to wider literatures (Marks et al., 2001; Organ, 1988;
addresses situations where gain (presence of a positive outcome) is Weiss & Cropanzano, 1996). Importantly, the task characteristics
the desired end-state and loss (presence of a negative outcome) is we investigate represent an easily identifiable and widely applica-
the undesired end-state, and theorizes that a gain frame leads to ble contingency that is informed by a large body of available
conservatism, whereas a loss frame leads to riskiness. Regulatory literature.
focus theory addresses striving for goals that represent advance- Fourth, our findings can also extend affective events theory
ment, growth, and accomplishment versus striving for goals that (Weiss & Cropanzano, 1996), as well as other episodic conceptu-
represent protection, safety, and responsibility, and is concerned alizations on the effects of affect on attitudes, behavior, and
with both the presence and absence of outcomes, not just their performance (see Beal et al., 2005; Spector & Fox, 2002). Specif-
presence. It posits that comparisons of success and failure are ically, we demonstrate how the experience of fit between the
made relative to one reference point, whereas prospect theory structure of the team and the characteristics of the task can con-
employs two, and that the reference point for success (desired stitute an important nexus of affective events that can influence the
end-state) versus failure (undesired end-state) is different in the experienced states of team members and ultimately relate to im-
two regulatory focus orientations (Higgins, 1997, 1998; Idson et portant outcomes for the team. In general, our article can also be
al., 2000). The presence of positive outcomes (e.g., realized gains) seen as an initial effort to extend the individual-level processes
and the absence of positive outcomes (e.g., unrealized gains) discussed by affective events theory and other such models to the
define success and failure in a promotion focus. In a prevention team level, and outline conceptually the way these can tie into both
focus, on the other hand, success is defined by the absence of inputs (in this case, task characteristics) and outputs (task perfor-
negative outcomes (e.g., averted losses) and failure by the presence mance and team viability).
of negative outcomes (e.g., incurred losses). The regulatory focus Furthermore, by demonstrating how the average level of helping
theoretical lens is best suited to investigate the phenomena of behaviors within a team fluctuates based on regulatory fit and the
interest in this study because both conditions employed a single mediating role of such helping behaviors on the effects of fit on
reference point to delineate success from failure, and this point was team outcomes, our article provides a contribution to the literature
different in each condition. on helping and organizational citizenship behavior (Organ, 1988).
Third, our results inform and extend past and current approaches That is, we illustrate a widely applicable contingency that can
to structural contingency theory (Hollenbeck et al., 2002; Pen- affect these behaviors in a group setting and show that these
nings, 1975) by demonstrating how team structure interacts with behaviors can be associated with team viability (through the ex-
 REGULATORY FIT AND MEDIATING PROCESSES 431

perience of satisfaction with the task). As such, our results under- be thus framed, and this framing might not always be effective
line the importance of these discretionary behaviors for team depending on the specific task at hand.
functioning, and do so while embedded in a larger conceptual In the end, deciding on which policy to follow (matching teams
framework of team-level processes. to situations or changing either of the above to achieve fit) will
Finally, although our conceptual framework and analysis were depend on the specific context faced and resources possessed by
focused on the team level, the potential effects of task character- the organization. What these findings mean in more general terms,
istics that our results highlight can be of importance for any however, is that regulatory focus needs to be considered when an
number of frameworks that consider the effect of task character- organization creates or deploys teams, as achieving fit between
istics on employee performance, behaviors, and reactions (such as teams and the regulatory focus of their tasks can carry important
theories task design, task demands, and task context). Thus, our benefits for both the organization and its employees.
results can also contribute to these literatures by highlighting
regulatory focus as an important task characteristic in need of
empirical and theoretical attention across conceptual and theoret- Limitations
ical levels. That is, the team-level effects we discuss in this article
As with any research, our study has a number of limitations that
could potentially translate to (and inform research investigating)
should be acknowledged. First, we tested our model in a laboratory
both individual and organizational processes and outcomes.
setting using teams of undergraduate students, which can limit the
generalizability of our results. Nevertheless, our conceptual model
Practical Implications and results refer to basic processes (such as affect and helping
behaviors), as do the basic mechanisms underlying regulatory
Our results are also of interest to organizations and practitioners.
focus as a characteristic of the task, whose effects can be expected
With the emergence of teams as the fundamental building block of
to operate in naturalistic settings in a way similar to the one we
organizations (Mathieu, Maynard, Rupp, & Gilson, 2008), it is
report in this article. Although our laboratory setting provided a
critical to know how the characteristics of team tasks will interact
vehicle for theory testing and model development, and thus in-
with structure to influence important outcomes. Thus, our findings
forming future research, future research that attempts to replicate
have clear organizational design implications; our results indicate
our findings in work settings with established teams would be of
that achieving regulatory fit by aligning team structure with the
value in affirming these effects outside the laboratory.
task characteristics is operationally important. Therefore, we sug-
Moreover, our laboratory manipulation of regulatory focus was
gest that managers would be wise to consider matching team
aimed at objective task characteristics, rather than investigating the
structure to the regulatory focus of desired task outcomes.
psychological reactions involved in tasks with different regulatory
Managers could create and train teams that are optimally suited
requirements. Even though our SME validation process increased
to operate in divisional or functional structures and then deploy
our confidence at the appropriateness and validity of our manip-
these teams to address promotion- or prevention-oriented tasks,
ulation, future research could provide an important contribution to
respectively. For example, a sales department could have func-
the literature by integrating examinations of regulatory states with
tional teams assigned to customer retention (i.e., the loss or not of
examinations of situational or structural factors.
customers), and divisional teams could be assigned to increasing
In addition, our team size (N ⫽ 21) was fairly low; as such, this
market penetration (i.e., the gain or not of market share). Such an
sample size did not enable us to conduct any between-teams
effort aims to eliminate misfit between teams and their tasks, and
analyses beyond zero-order correlations, or to be highly confident
could prove of value to organizational performance, while also
of a lack of differences between teams due to random assignment.
enhancing the subjective experiences of the members of these
However, our within-teams (or Level 1) number of observations
teams.
was sufficient for our analyses (N ⫽ 84), and the consistent and
Another, more flexible approach requires that team structure be
strong pattern of results we found increases our confidence in our
changed to create regulatory fit. Utilizing teams that can change
findings. Moreover, the within-teams design we used alleviates
their structure to fit the demands of their task would seem to be an
concerns about team dissimilarity to some extent (see above),
optimal solution; however, research has shown that movement
assuaging worries about between-teams sample size.
between different structures does not always occur without issue
(Moon et al., 2004). Nevertheless, teams that can demonstrate this
level of dynamic adaptation will be able to perform optimally in a Future Research
wide range of tasks and, as such, can represent an important asset
for their organizations. Our findings invite a number of avenues of future research.
Alternatively, organizations can redesign the flow of work by First, even though we included affective processes in our model as
assigning tasks to fit the structure of existing teams. Task goals and an indicator of motivational processes, regulatory focus theory is
outcomes can be framed by managers so as to create fit; for rich with motivational processes to be investigated. For example,
example, implementing a safety policy can be done by emphasiz- future studies could examine the relationship of team regulatory fit
ing the losses that noncompliance can bring (accidents or injuries to intrinsic and extrinsic motivation through regulatory mode (see
in the workplace), or it can be done by emphasizing the gains to be Pierro, Kruglanski, & Higgins, 2006), or how goal congruent or
made by compliance (improved efficiency and employee well- incongruent team rewards or resources can affect motivational
being). By changing the framing of the task, managers can deploy strength levels within a team (see Spiegel et al., 2004). Similarly,
existing functional or divisional teams as needed to address spe- there are many potential avenues of research for a variety of
cific situations. The downside to this is that not all situations can processes linking regulatory fit to team- and individual-level out-
432 DIMOTAKIS, DAVISON, AND HOLLENBECK

comes (see, e.g., Kozlowski & Ilgen, 2006), thus illuminating the Toward a contingency approach. Academy of Management Journal, 46,
pathways by which regulatory fit operates. 572–590. doi:10.2307/30040650
Furthermore, our findings indicated that regulatory fit was re- Brayfield, A. H., & Rothe, H. F. (1951). An index of job satisfaction.
lated to both team processes and team-level outcomes. This can Journal of Applied Psychology, 35, 307–311. doi:10.1037/h0055617
Brief, A. P., & Weiss, H. M. (2002). Organizational behavior: Affect in the
inform frameworks such as Woolley’s (2009) process– outcome
workplace. Annual Review of Psychology, 53, 279 –307. doi:10.1146/
focus by investigating the types of regulatory fit on whether teams
annurev.psych.53.100901.135156
focus on outcomes or processes by examining the relative impor- Brockner, J., & Higgins, E. T. (2001). Regulatory focus theory: Implica-
tance that the situational demands and structural capabilities place tions for the study of emotions at work. Organizational Behavior and
on means versus ends and how that relates to the ways teams Human Decision Processes, 86, 35– 66. doi:10.1006/obhd.2001.2972
function. Brockner, J., Paruchuri, S., Idson, L. C., & Higgins, E. T. (2002). Regu-
Different aspects of performance can also be valuable avenues latory focus and the probability estimates of conjunctive and disjunctive
of research, by examining the way in which regulatory fit can have events. Organizational Behavior and Human Decision Processes, 87,
differential effects on various aspects of performance. One poten- 5–24. doi:10.1006/obhd.2000.2938
tial issue is the differential effects of regulatory fit in predicting Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models:
speed versus accuracy performance, thus helping create more Applications and data analysis methods. Newbury Park, CA: Sage.
Burns, T., & Stalker, G. M. (1961). The management of innovation.
nuanced predictions of how structure affects performance across
London, England: Tavistock.
different contexts. Similarly, investigations of various aspects of
Cesario, J., & Higgins, E. T. (2008). Making message recipients “feel
team viability can provide a more complete picture of team func- right”: How nonverbal cues can increase persuasion. Psychological
tioning. Science, 19, 415– 420. doi:10.1111/j.1467-9280.2008.02102.x
Moreover, although our article sought to expand regulatory Chen, G., Kanfer, R., DeShon, R. P., Mathieu, J. E., & Kozlowski, S. W. J.
focus beyond dispositional considerations to situational character- (2009). The motivating potential of teams: Test and extension of Chen
istics, integrating these two approaches can provide important and Kanfer’s (2006) cross-level model of motivation in teams. Organi-
insights. For example, future studies could investigate how the zational Behavior and Human Decision Processes, 110, 45–55. doi:
dispositional regulatory focus of a team member can influence his 10.1016/j.obhdp.2009.06.006
or her performance and satisfaction based on team tasks and/or Chernev, A. (2009). Choosing versus rejecting: The impact of goal–task
structure. At the team level, future studies can examine how compatibility on decision confidence. Social Cognition, 27, 249 –260.
doi:10.1521/soco.2009.27.2.249
regulatory focus dispositions within the team can influence how
Crowe, E., & Higgins, E. T. (1997). Regulatory focus and strategic incli-
well team members adapt to different situations or structures.
nations: Promotion and prevention in decision-making. Organizational
Another important consideration would be the way the team Behavior and Human Decision Processes, 69, 117–132. doi:10.1006/
receives implicit or explicit feedback on its performance and obhd.1996.2675
whether this feedback is provided to (or perceived by) the team in Förster, J., Higgins, E. T., & Bianco, A. T. (2003). Speed/accuracy deci-
a way that is congruent or incongruent with the regulatory focus of sions in task performance: Built-in trade-off or separate strategic con-
its situation (see Idson & Higgins, 2000). For example, do pre- cerns? Organizational Behavior and Human Decision Processes, 90,
vention teams react better to loss-focused feedback because they 148 –164. doi:10.1016/S0749-5978(02)00509-5
are better suited to vigilance means? Such research can provide an Förster, J., Higgins, E. T., & Idson, L. C. (1998). Approach and avoidance
integration of team situation–structure fit with various motivation strength during goal attainment: Regulatory focus and the “goal looms
theories to provide an important contribution. larger” effect. Journal of Personality and Social Psychology, 75, 1115–
1131. doi:10.1037/0022-3514.75.5.1115
Finally, future research needs to consider how teams develop as
Fredrickson, B. L., & Branigan, C. A. (2005). Positive emotions broaden
they continue to function over time within and across specific
the scope of attention and thought–action repertoires. Cognition and
situations. Although past work has investigated some issues in- Emotion, 19, 313–332. doi:10.1080/02699930441000238
volved with teams moving across structures (see Johnson et al., Fredrickson, B. L., & Losada, M. F. (2005). Positive affect and the
2006), an investigation of how the regulatory characteristics of the complex dynamics of human flourishing. American Psychologist, 60,
task affect changes in structure can provide an important contri- 678 – 686. doi:10.1037/0003-066X.60.7.678
bution. Likewise, investigations of how regulatory fit affects teams Freitas, A. L., & Higgins, E. T. (2002). Enjoying goal-directed action: The
over time and across stages of team development can shed addi- role of regulatory fit. Psychological Science, 13, 1– 6. doi:10.1111/1467-
tional light on the temporal effects of regulatory fit. 9280.00401
Friedman, R. S., & Förster, J. (2001). The effects of promotion and
prevention cues on creativity. Journal of Personality and Social Psy-
References chology, 81, 1001–1013. doi:10.1037/0022-3514.81.6.1001
Galunic, D. C., & Eisenhardt, K. M. (1994). Renewing the strategy–
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable structure–performance paradigm. In B. M. Staw & L. L. Cummings
distinction in social psychological research: Conceptual, strategic and (Eds.), Research in organizational behavior (Vol. 16, pp. 215–255).
statistical considerations. Journal of Personality and Social Psychology, Greenwich, CT: JAI Press.
51, 1173–1182. doi:10.1037/0022-3514.51.6.1173 Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist,
Beal, D. J., Weiss, H. M., Barros, E., & MacDermid, S. M. (2005). An 52, 1280 –1300. doi:10.1037/0003-066X.52.12.1280
episodic process model of affective influences on performance. Journal Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a
of Applied Psychology, 90, 1054 –1068. doi:10.1037/0021- motivational principle. In M. P. Zanna (Ed.), Advances in experimental
9010.90.6.1054 social psychology (Vol. 30, pp. 1– 46). New York, NY: Academic Press.
Beersma, B., Hollenbeck, J. R., Humphrey, S. E., Moon, H., Conlon, D. E., Higgins, E. T. (2000). Making a good decision: Value from fit. American
& Ilgen, D. R. (2003). Cooperation, competition, and team performance: Psychologist, 55, 1217–1230. doi:10.1037/0003-066X.55.11.1217
 REGULATORY FIT AND MEDIATING PROCESSES 433

Higgins, E. T. (2002). How self-regulation creates distinct values: The case intervening variable effects. Psychological Methods, 7, 83–104. doi:
of promotion and prevention decision making. Journal of Consumer 10.1037/1082-989X.7.1.83
Psychology, 12, 177–191. doi:10.1207/S15327663JCP1203_01 Markman, A. B., & Brendl, C. M. (2000). The influence of goals on value
Higgins, E. T. (2005). Value from regulatory fit. Current Directions in and choice. In D. L. Medin (Ed.), The psychology of learning and
Psychological Science, 14, 209 –213. doi:10.1111/j.0963-7214.2005 motivation (Vol. 39, pp. 97–129) San Diego, CA: Academic Press.
.00366.x Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based
Higgins, E. T., Cesario, J., Hagiwara, N., Spiegel, S., & Pittman, T. (2010). framework and taxonomy of team processes. Academy of Management
Increasing or decreasing interest in activities: The role of regulatory fit. Review, 26, 356 –376.
Journal of Personality and Social Psychology, 98, 559 –572. doi: Mathieu, J. E., Maynard, M. T., Rapp, T., & Gilson, L. (2008). Team
10.1037/a0018833 effectiveness 1997–2007: A review of recent advancements and a
Higgins, E. T., Idson, L. C., Freitas, A. L., Spiegel, S., & Molden, D. C. glimpse into the future. Journal of Management, 34, 410 – 476. doi:
(2003). Transfer of value from fit. Journal of Personality and Social 10.1177/0149206308316061
Psychology, 84, 1140 –1153. doi:10.1037/0022-3514.84.6.1140 Midlarsky, E. (1991). Helping as coping. In M. C. Clark (Ed.), Review of
Higgins, E. T., & Spiegel, S. (2004). Promotion and prevention strategies Personality and Social Psychology: Vol. 12. Prosocial behavior (Vol.
for self-regulation: A motivated cognition perspective. In R. F. Baumeis- 12, pp. 238 –264). Newbury Park, CA: Sage.
ter & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, Moon, H., Hollenbeck, J. R., Humphrey, S. E., Ilgen, D. R., West, B., Ellis,
and applications (pp. 171–187). New York, NY: Guilford Press. A. P. J., & Porter, C. O. L. H. (2004). Asymmetric adaptability: Dy-
Hollenbeck, J. R., Moon, H., Ellis, A. P. J., West, B. J., Ilgen, D. R., namic team structures as one-way streets. Academy of Management
Sheppard, L., . . . Wagner, J. A., III. (2002). Structural contingency Journal, 47, 681– 695. doi:10.2307/20159611
theory and individual differences: Examination of external and internal Morgeson, F. P., & Campion, M. A. (2003). Work design. In W. C.
person–team fit. Journal of Applied Psychology, 87, 599 – 606. doi: Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of Psychology:
10.1037/0021-9010.87.3.599 Vol. 12. Industrial and organizational psychology (pp. 423– 452). Hobo-
Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). Integrating ken, NJ: Wiley.
motivational, social, and contextual work design features: A meta- Motowidlo, S. J. (1996). Orientation toward the job and organization. In
K. R. Murphy (Ed.), Individual differences and behavior in organiza-
analytic summary and theoretical extension of the work design literature.
tions (pp. 175–208). San Francisco, CA: Jossey-Bass.
Journal of Applied Psychology, 92, 1332–1356. doi:10.1037/0021-
Organ, D. W. (1988). Organizational citizenship behavior: The good
9010.92.5.1332
soldier syndrome. Lexington, MA: Lexington Books.
Idson, L. C., Liberman, N., & Higgins, E. T. (2000). Distinguishing gains
Parker, S. K., & Wall, T. D. (2001). Work design: Learning from the past
from nonlosses and losses from nongains: A regulatory focus perspec-
and mapping a new terrain. In N. Anderson, D. S. Ones, H. K. Sinangil,
tive on hedonic intensity. Journal of Experimental Social Psychology,
& C. Viswesvaran (Eds.), Handbook of Industrial, Work and Organiza-
36, 252–274. doi:10.1006/jesp.1999.1402
tional Psychology: Vol. 1. Personnel psychology (pp. 90 –109). London,
Idson, L. C., Liberman, N., & Higgins, E. T. (2004). Imagining how you’d
England: Sage.
feel: The role of motivational experiences from regulatory fit. Person-
Penner, L. A., Dovidio, J. F., Piliavin, J. A., & Schroeder, D. A. (2005).
ality and Social Psychology Bulletin, 30, 926 –937. doi:10.1177/
Prosocial behavior: Multilevel perspectives. Annual Review of Psychol-
0146167204264334
ogy, 56, 365–392. doi:10.1146/annurev.psych.56.091103.070141
Ilgen, D. R. (1999). Teams embedded in organizations: Some implications.
Pennings, J. M. (1975). Relevance of structural-contingency model for
American Psychologist, 54, 129 –139. doi:10.1037/0003-066X.54.2.129
organizational effectiveness. Administrative Science Quarterly, 20, 393–
Isen, A. M., Daubman, K. A., & Nowicki, G. P. (1987). Positive affect
410. doi:10.2307/2391999
facilitates creative problem solving. Journal of Personality and Social Pierro, A., Kruglanski, A. W., & Higgins, E. T. (2006). Regulatory mode
Psychology, 52, 1122–1131. doi:10.1037/0022-3514.52.6.1122 and the joys of doing: Effects of “locomotion” and “assessment” on
Johnson, M. D., Hollenbeck, J. R., Humphrey, S. E., Ilgen, D. R., Jundt, D., intrinsic and extrinsic task-motivation. European Journal of Personality,
& Meyer, C. J. (2006). Cutthroat cooperation, Asymmetrical adaptation 20, 355–375. doi:10.1002/per.600
to changes in team reward structures. Academy of Management Journal, Porter, C. O. L. H. (2005). Goal orientation: Effects on backing up
49, 103–119. doi:10.5465/AMJ.2006.20785533 behavior, performance, efficacy, and commitment in teams. Journal of
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of Applied Psychology, 90, 811– 818. doi:10.1037/0021-9010.90.4.811
decision under risk. Econometrica, 47, 263–291. doi:10.2307/1914185 Porter, C. O. L. H., Hollenbeck, J. R., Ilgen, D. R., Ellis, A. P. J., West,
Katz, D., & Kahn, R. L. (1966). The social psychology of organizations. B. J., & Moon, H. (2003). Backing up behaviors in teams: The role of
New York, NY: Wiley. personality and legitimacy of need. Journal of Applied Psychology, 88,
Kelly, J. R., & Barsade, S. G. (2001). Mood and emotions in small groups 391– 403. doi:10.1037/0021-9010.88.3.391
and work teams. Organizational Behavior and Human Decision Pro- Schneider, W., & Fisk, A. D. (1982). Concurrent automatic and control
cesses, 86, 99 –130. doi:10.1006/obhd.2001.2974 visual search, Can it be done without cost? Journal of Experimental
Koenig, A. M., Cesario, J., Molden, D. C., Kosloff, S., & Higgins, E. T. Psychology: Learning, Memory, and Cognition, 8, 261–278. doi:
(2009). Incidental experiences of regulatory fit and the processing of 10.1037/0278-7393.8.4.261
persuasive appeals. Personality and Social Psychology Bulletin, 35, Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments
1342–1355. doi:10.1177/0146167209339076 of well-being: Informative and directive functions of affective states.
Kozlowski, S. W. J., & Ilgen, D. R. (2006). Enhancing the effectiveness of Journal of Personality and Social Psychology, 45, 513–523. doi:
work groups and teams. Psychological Science in the Public Interest, 7, 10.1037/0022-3514.45.3.513
77–124. Shah, J., Higgins, E. T., & Friedman, R. S. (1998). Performance incentives
LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about and means: How regulatory focus influences goal attainment. Journal of
interrater reliability and interrater agreement. Organizational Research Personality and Social Psychology, 74, 285–293. doi:10.1037/0022-
Methods, 11, 815– 852. doi:10.1177/1094428106296642 3514.74.2.285
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Spector, P. E., & Fox, S. (2002). An emotion-centered model of voluntary
Sheets, V. (2002). A comparison of methods to test mediation and other work behavior: Some parallels between counterproductive work behav-
434 DIMOTAKIS, DAVISON, AND HOLLENBECK

ior and organizational citizenship behavior. Human Resource Manage- at Work Scale. Journal of Organizational Behavior, 30, 805– 831. doi:
ment Review, 12, 269 –292. doi:10.1016/S1053-4822(02)00049-9 10.1002/job.572
Spiegel, S., Grant-Pillow, H., & Higgins, E. T. (2004). How regulatory fit Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
enhances motivational strength during goal pursuit. European Journal of dation of brief measures of positive and negative affect: The PANAS
Social Psychology, 34, 39 –54. doi:10.1002/ejsp.180 scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Thompson, J. D. (1967). Organizations in action. New York, NY: McGraw-Hill. doi:10.1037/0022-3514.54.6.1063
Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A
behaviors: Evidence of construct and predictive validity. Academy of theoretical discussion of the structure, causes and consequences of
Management Journal, 41, 108 –119. doi:10.2307/256902 affective experiences at work. In B. M. Staw & L. L. Cummings (Eds.),
Wallace, J. C., & Chen, G. (2006). A multilevel integration of personality, Research in organizational behavior (Vol. 18, pp. 1–74). Greenwich,
climate, self-regulation, and performance. Personnel Psychology, 59, CT: JAI Press
529 –557. doi:10.1111/j.1744-6570.2006.00046.x Woolley, A. W. (2009). Means vs. ends: Implications of process and
Wallace, J. C., Johnson, P. D., & Frazier, M. L. (2009). An examination of outcome focus for team adaptation and performance. Organization Sci-
the factorial, construct, and predictive validity of the Regulatory Focus ence, 20, 500 –515. doi:10.1287/orsc.1080.0382

Appendix

Distribution of Assets Across Team Structures

Functional structure Divisional structure

Asset 1 2 3 4 5 1 2 3 4 5

Rifle company 11 11 7 6 8 7 7 7 6
Weapons company 2 2 2 2
Motorized troops 2 2 1 3
Dismount troops 1 1 1 1
Reconnaissance troops 10 8 4 6 3 3 2
Missile battery 2 2 1 1 2
Tank company 2 2 4 1 3 1 1 2
Field artillery battery 3 1 1
Mechanical engineer company 3 1 1 1
Combat engineer company 2 2 2 1 1 1 1 2
Military police company 5 2 2 1
Military intelligence company 5 1 1 1 1 1
Recon helicopter 7 1 2 2 1 1
Attack helicopter 6 1 1 2 2
Transport helicopter 12 2 3 2 1 4
Heavy transport helicopter 3 2 1
Civil affairs division 6 2 1 2 1
Total assets controlled 22 22 28 42 17 28 28 25 23 26
Number of different assets 7 7 5 6 4 14 12 13 12 11

Received October 11, 2010

Revision received September 6, 2011
Accepted September 12, 2011 䡲