Blazing the trail versus trailing the group: Culture and perceptions of the leader’s position

Organizational Behavior and Human Decision Processes 113 (2010) 51–61

Contents lists available at ScienceDirect

Organizational Behavior and Human Decision Processes

journal homepage: www.elsevier .com/ locate/obhdp

Blazing the trail versus trailing the group: Culture and perceptionsof the leader’s position

Tanya Menon a,*, Jessica Sim b,1, Jeanne Ho-Ying Fu c,1, Chi-yue Chiu d, Ying-yi Hong d

a Booth School of Business, University of Chicago, 5807 South Woodlawn Avenue, Chicago, Illinois 60637, USAb Department of Psychology, University of Chicago, 5848 S. University Avenue, Chicago, IL 60637, USAc Department of Management, City University of Hong Kong, Kowloon Tong, Hong Kongd Division of Strategy, Management and Organization Nanyang Business, School Nanyang Technological University, S3-01C-81 Nanyang Avenue, Singapore 639798, Singapore

a r t i c l e i n f o

Article history:Received 3 February 2008Accepted 2 April 2010Available online 5 May 2010

Accepted by Dan van Knippenberg

Keywords:CultureLeadershipPosition

0749-5978/$ - see front matter � 2010 Elsevier Inc. Adoi:10.1016/j.obhdp.2010.04.001

* Corresponding author. Address after June 30: IN138676 Singapore, Singapore. Fax: +1 (773)702 0458

E-mail addresses: [email protected], tany(T. Menon).

1 These authors contributed equally to this paper.

a b s t r a c t

Research suggests that power triggers assertive action. However, people from different cultures mightexpect different types of action from powerful individuals such as leaders. In comparing cultural differ-ences in leadership imagery, we find that Americans represent leaders standing ahead of groups, whereasAsians also represent leaders behind groups. We propose that front versus back positions embody twofaces of leader action: individual assertion versus group-focused action. Studies 1a and 1b respectivelyemployed etic and emic methods to demonstrate that Singaporeans were more likely than Americansto represent leaders behind groups. In Study 2, Singaporeans evaluated back leaders more favorably thanAmericans did, and group focus mediated cultural differences. Simulating the conditions under whichcultural differences arise, Study 3 demonstrates that a primarily Western managerial sample primed withthreat (versus opportunity) preferred back leaders. By describing cultural variations in imagery, we revealmore nuanced implicit theories of leader action.

� 2010 Elsevier Inc. All rights reserved.

Introduction

‘‘Leadership involves finding a parade and getting in front of it”(Naisbitt, 1982, p. 162).

‘‘To lead the people, walk behind them” (Lao Tzu, cited in Gro-the, 2004, p. 5–6).

People often assume that leaders head the pack, lead the charge,and blaze the trail (Grint, 2005; Northouse, 2001). The Oxford Eng-lish Dictionary defines a leader as ‘‘the first man in a file, one in thefront rank, one of the foremost in a moving body” (Simpson & Wei-ner, 1989). Pilots direct planes from the cockpit, and when a pres-ident enters a room, advisors often trail behind. Although militaryformations sometimes situate leaders at the rear (see Andrzejew-ski, 1954/2003), it seems overwhelmingly obvious and indeed nat-ural that leaders would and should stand in front of the group.

However, consider an opposite image, also associated withleadership. Just as Lao Tzu advised leaders to walk behind follow-ers, another classical Chinese philosopher, Sun Tzu (1963), warnedthat when an army’s weight ‘‘is at the front not at the rear-

ll rights reserved.

SEAD, 1 Ayer Rajah Avenue,[email protected]

although [it may be] massive, it is not firm” (p. 155). These quota-tions suggest that the association between social and physicalposition might reflect a cultural schema (e.g., McArthur & Post,1977).

The present research explores cross-cultural variations in theimagery people associate with leadership to develop a morenuanced view of leader action. The image of the leader in frontaligns with recent social psychological research demonstratingthat powerful people such as leaders activate the behavioralapproach system (Galinsky, Gruenfeld, & Magee, 2003; Higgins;1997; Keltner, Gruenfeld, & Anderson, 2003). Powerful people ini-tiate behavior (e.g., leaders as ‘‘agents of change,” Bass, 1981; firstoffers in negotiations, Magee, Galinsky, & Gruenfeld, 2007); seekrewards in the environment (French & Raven, 1959), feel uninhib-ited, and often fail to see others’ perspectives (Galinsky, Magee,Inesi, & Gruenfeld, 2006), because they care less about what othersthink (Fiske, 1993). Although the leader in the front of the group ispoorly positioned to monitor group members, he or she is wellpositioned to act assertively upon the environment. By contrast,we argue that the image of the leader behind the group aligns withAsian expectations about leader action. Specifically, the back leadercan readily watch over the group to protect it from threats, disrup-tion, and failure, consistent with Asian notions of paternalisticleadership (Cheng, Chou, Wu, Huang, & Fahr, 2004).

By describing how preferences for these representations differcross-culturally, we redefine the types of action people associate

http://dx.doi.org/10.1016/j.obhdp.2010.04.001

mailto:[email protected]

mailto:[email protected]

http://www.sciencedirect.com/science/journal/07495978

http://www.elsevier.com/locate/obhdp

52 T. Menon et al. / Organizational Behavior and Human Decision Processes 113 (2010) 51–61

with leaders. Rather than assuming that power triggers assertiveaction only (e.g., Galinsky et al., 2003; Keltner et al., 2003), wespecify a dual-function model of leader action, with front versusback positions embodying assertive versus group-focused orienta-tions (Kark & Van Dijk; 2007; Zhong, McGee, Maddux, & Galinsky,2006). We hypothesize that when cultures emphasize individualassertion (e.g., the United States, Lee, Aaker, & Gardner, 2000), theirmembers perceptually represent leaders in front, breaking awayfrom groups to exert their will on the environment (Kark & VanDijk, 2007). In contrast, when cultures prioritize responsibilitiesto the group (e.g., Singapore; Chen, Ng, & Rao, 2005; Lee et al.,2000), their members also perceptually represent leaders behindgroups to watch over and protect them. By describing how theseperceptions differ cross-culturally, we reveal more nuanced expec-tations about leader action.

Leadership imagery across cultures: simulating assertion versusgroup responsibility

Researchers define power as control over resources, and hence,over others’ outcomes (Fiske, 1993; French & Raven, 1959; Keltneret al., 2003). Leadership, on the other hand, is the process of influ-encing others to accomplish tasks (Chemers, 2001). Noting thesedefinitions, Giessner and Schubert (2007) contend that, althoughpowerful people might not always be leaders, leaders always re-quire power.

Rather than defining what power and leadership are, or describ-ing how leaders might act, we draw from implicit leadership the-ory and assess people’s lay understandings of leaders. Inparticular, we focus on how perceivers mentally represent the lea-der’s physical position (Giessner & Schubert, 2007; Raghubir & Val-enzuela, 2006; Schubert, 2005). Implicit theories are categorizationschemas that help observers to simplify information (Phillips &Lord, 1986). Followers’ reactions to leadership are more stronglyshaped by their own, often automatic, construction than by the lea-der’s objective traits and behaviors (Ritter & Lord, 2007; Schyns,Meindl, & Croon, 2007). As a result, followers afford leaders discre-tion and power based on the match between the leader’s actualcharacteristics and their expectations of the leader, as capturedby their leader prototype (Giessner & van Knippenberg, 2008; Rit-ter & Lord, 2007; van Knippenberg & Hogg, 2003).

The present investigation builds on implicit leadership theory intwo ways. First, by focusing on one of the most basic elements of aleader prototype, the leader’s physical position, we move beyondprevious research that directly taps people’s beliefs about leaderbehaviors (e.g., House, Hanges, Javidan, Dorfman, & Gupta, 2004),and reveal the subtle ways in which people mentally simulatethose behaviors. Second, we explore how the content of these the-ories might differ cross-culturally (House et al., 2004), showingthat, in addition to the common Western association betweenpower and assertive action, Asian contexts might feature other lea-der representations (Zhong et al., 2006), because people are morelikely to expect leaders to protect the group.

Representing leader action through physical position

Implicit theories have largely been assessed by directly tappingpeople’s judgments about leaders. For instance, the GLOBE studies(House et al., 2004) explored cultural differences in people’s expec-tations about leadership by asking participants from 62 countriesto assess whether particular attributes impeded or facilitated lead-ership. The researchers confirmed that cultural values around lead-ership differed. Whereas 22 leader descriptors (e.g., trustworthy)were universal, another 35 (e.g., sensitive) were culturallycontingent.

The present research similarly focuses on cross-cultural differ-ences in implicit leadership theories, but we study a more subtlecue that shapes people’s perceptions of leaders. Specifically, westudy the leader’s physical position, defined as the leaders’ spatialorientation to the group, i.e., where leaders stand with respect to fol-lowers, how close they stand, whether they face followers or turnaway from them (Taylor & Fiske, 1975), and more generally, howleaders occupy space (see Tiedens & Fragale, 2003). Well before peo-ple begin to perceive others’ traits and behaviors, they infer powerthrough physical position (e.g., being seated at the head of the table,Bass & Klubeck, 1952). Likewise, people associate high vertical posi-tions with power (Giessner & Schubert, 2007; Schubert, 2005) andbelieve that important or prominent people occupy central positions(Raghubir & Valenzuela, 2006). In the game show The Weakest Link,contestants who were randomly assigned to central locations withinthe group won more often than those in peripheral locations, be-cause observers frequently applied the heuristic that ‘‘importantpeople sit in the middle,” which substituted for individuating perfor-mance information (Raghubir & Valenzuela, 2006). These associa-tions seem to emerge over time as people form correlationsbetween power and particular spatial locations in their environ-ments (Giessner & Schubert, 2007; Lakoff & Johnson, 1980).

Extant research on spatial orientation has explored hierarchy(up–down position, Giessner & Schubert, 2007; Schubert, 2005)and centrality (center–periphery position, Raghubir & Valenzuela,2006; Taylor & Fiske, 1975). In spite of the prominence of recentconceptualizations of power as assertive action (Galinsky et al.,2003; Keltner et al., 2003), physical representations of action havenot yet been studied. The present research identifies two ways tophysically represent a leader’s action: positioning leaders in frontof groups (trailblazers) versus behind groups (trailing behind,Fiske, 2004; Lakoff & Johnson, 1980). Whereas vertical and centralpositions depict stable power hierarchies (Giessner & Schubert,2007; Raghubir & Valenzuela, 2006; Schubert, 2005), front-backpositions represent leadership on a horizontal plane, dynamicallysimulating a group’s forward progress towards specific goals(Boroditsky, 2000).

Physical position is crucial to leadership theory because it reflectsa relatively primitive form of implicit theory. The front-back posi-tion in particular captures people’s mental simulations of leader ac-tion, offering a more implicit and less formalized view of leadershipperceptions (Barsalou, 1999). Physical position does not merelyindicate literal spatial arrangements; it symbolizes social relation-ships, and reveals the rich perceptual apparatus that accompaniesimplicit theories of leadership (Barsalou, 1999; Lakoff & Johnson,1980). Position influences both how people talk about leadership(e.g., ‘‘trailblazer,” ‘‘ahead of the pack,” versus ‘‘core team member”),and how people think about leaders (Lakoff & Johnson, 1980). Posi-tion can determine whom people treat as leaders (Bass & Klubeck,1952; Raghubir & Valenzuela, 2006). Further, if leaders deviate fromexpected positions, this could reduce their conformity with proto-types and hence, perceptions of their effectiveness (Giessner & vanKnippenberg, 2008; van Knippenberg & Hogg, 2003).

These perceptions particularly matter to symbolic leadership the-ories, which argue that leaders use symbols to shape followers’ sense-making (Weick, 1987). Because an organization’s environment con-strains the leader’s control over substantive activities (e.g., strategicdecisions such as mergers/acquisitions), Pfeffer (1981) argued thatleaders primarily engage in symbolic action. By initiating symbolicactivities (such as ceremonies) and evoking imagery (e.g., throughphysical space), leaders develop shared paradigms among followersand convey their power and influence (Pfeffer, 1981).

Symbolic theories tend to be ‘‘leader-centric,” focusing on howleaders can and should use symbols, rather than showing howfollowers interpret leader symbols. While Pfeffer noted theconstraints that surround a leader’s substantive decisions, we

a

b

Fig. 1. Trailblazing and trailing-behind imagery.

T. Menon et al. / Organizational Behavior and Human Decision Processes 113 (2010) 51–61 53

suggest that a leader’s symbolic repertoire might similarly be con-strained. Although past research recognizes that different spatialpositions and symbols can communicate leadership and power(Giessner & Schubert, 2007; Raghubir & Valenzuela, 2006), weshow that these position effects can also vary according to culturalnorms. Leaders’ socio-cultural context endows them with a sym-bolic repertoire that also colors the ways that followers interprettheir actions. We consider this cultural context next.

Representing the two faces of leader action

In addition to developing the perceptual route through whichpeople internalize implicit theories about leaders, we also explorethe content of these theories. We begin by reviewing recent re-search that shows how power triggers assertive action (Galinskyet al., 2003; Keltner et al., 2003). We then consider how suchexpectations might be represented through physical position, andwhether these forms of action might generalize cross-culturally.

Power as assertive action.When people have power, they activate a unique psychological

state, characterized by two interrelated aspects. Specifically, theyenter into an assertive mode whereby they: (a) act on the environ-ment to change and shape it according to their will and (b) fail torecognize the perspectives of others. First, according to the ap-proach/inhibition theory of power (Keltner et al., 2003), power isassociated with reward-rich environments and freedom, whichactivate the approach system. By contrast, lacking power is associ-ated with threats, punishment, and constraint, which activate inhi-bition and avoidance.

Recent research argues that, by triggering the approach system,power leads directly to action, channeling specific goals intobehavior (Galinsky et al., 2003). To illustrate how people primedwith power exerted their will upon their immediate environments,Galinsky et al. (2003) placed participants in a room with an annoy-ing fan, priming half with power and the other half with lowpower. They found that 69% of people primed with power turnedoff the annoying fan, while only 42% of those primed with lowpower did. Power encouraged people to change the environmentaccording to their will.

Second, past research typically contends that the powerful areinsensitive to how their assertive behavior affects others (Keltneret al., 2003; Kipnis, 1972). In support of this contention, Galinskyet al. (2006) found that power diminished the ability to take others’perspectives. In one experiment, participants primed with highpower were three times more likely to draw letters on their fore-head that were only readable from their own perspective, whilepeople in the low power condition spontaneously took the perspec-tive of another person, drawing the letters so that they were read-able to others. Follow-up experiments revealed that high powerparticipants anchored more on their own knowledge than on otherpeople’s vantage points, and that they also failed to recognize oth-ers’ emotions. One explanation for these effects is that powerfulpeople control resources, so they are not dependent on others anddo not need to pay attention to them (Fiske, 1993). On the otherhand, low-power individuals have to be more vigilant about othersto survive threatening environments (Ellyson & Dovidio, 1985).

Representing assertive power: trailblazing imagery. Just as powerprimes assertive behavior among Westerners, Western followersmay likewise expect assertion from leaders (Eaton, Visser,Krosnick, & Anand, 2009). Indeed, implicit theories about powerare replete with assumptions that leaders are assertive and action-oriented and challenge the status quo (Anderson & Galinsky, 2006;Bass, 1981; Meindl, Ehrlich, & Dukerich, 1985; Weber, 1947;Zaleznik, 1989).

Although it is often assumed that assertion is inextricably boundto the concept of power, it is possible that it particularly features inWestern leader representations. For instance, recent research(Kitayama, Ishii, Imada, Takemura, & Ramaswamy, 2006) suggeststhat such expectations might have their antecedents with the fron-tier narrative, which has been propagated through history lessons,literature, and Western movies, leaving a psychological imprint onmodern Americans (Tocqueville, 2000, p. 1835; Turner, 1921). Inrisking their personal safety to pursue advancement and opportu-nity, the pioneers embodied the assertive power that frequentlyemerges in modern Western theories of charismatic and transfor-mational leadership (Anderson & Galinsky, 2006; Antonakis &House, 2002; Galinsky et al., 2003; Kark & Van Dijk, 2007;Sternberg, Kaufman, & Pretz, 2003). We propose that the visual rep-resentation of ‘‘pushing the frontier” persists among modern Amer-icans, who typically position leaders at the forefront of their groups,where they can chart the group’s direction, independently forgenew ground, and inspire followers forward to develop new terrainsand redefine their environments through innovation (Fig. 1a).

Power as group-focused action.Other research suggests contrasting consequences of power.

Rather than being poor perspective-takers, powerful people canexhibit superior skills of individuating followers (Overbeck & Park,2001). And rather than asserting their will on the environmentwithout focusing on their groups, powerful people often act to ben-efit their groups by contributing to public goods (Galinsky et al.,2003). To resolve these conflicting findings, Chen, Lee-Chai, andBargh (2001) argue that communally-oriented people are morelikely than exchange-oriented people to respond to power in altru-istic ways. Given the Asian focus on interpersonal relationships(Markus & Kitayama, 1991), Zhong et al. (2006) offer a cross-cul-tural argument, suggesting that Asians primed with power refrainfrom assertive action, which could disrupt the group and social or-der. The authors find that while subliminal power primes facilitateproduction of assertion words (e.g., change, challenge, oppose, move,start) among European Americans, the primes facilitate productionof restraint words among Asians (e.g., hold, restrain, withhold, stay,cease).

Taken together, this research suggests that within interdepen-dent Asian contexts, which foster interpersonal and group respon-sibilities over individual assertion (Markus & Kitayama, 1991),


power might come to be associated with different action orienta-tions. Rather than neglecting others’ perspectives, Asian leadersmight be especially likely to direct their attention and action towatching, protecting, and relating to their groups. Indeed, theGLOBE studies found that self-protective leadership, which priori-tizes individual and group protection, was an aspect of leadershipthat differed cross-culturally. Nordic- and Anglo-Europeans viewedthis dimension as inhibiting good leadership, but people from Con-fucian cultures and South Asians rated it more favorably (Houseet al., 2004).

In identifying the historical antecedents of group-focused leaderaction, Hsu (1983) notes that Asian leaders typically did not promotethe inherently risky strategy of ‘‘pushing” the frontier by pursuingopportunities and innovation, recognizing that such change couldthreaten group stability. Instead, they often adhered to proper roles,maintained social order and stability, and protected the group frominternal and external threats (Fu & Chiu, 2007; Northouse, 2001;Zemba, Young, & Morris, 2006). Given their protective stance and fo-cus on group responsibilities (Kark & Van Dijk, 2007), researchershave evoked paternal imagery to represent Asian leadership (Chenget al., 2004; Chow, 2008; Erez & Earley, 1993; Fu, Wu, Yang, & Ye,2008; Low, 2006). Chinese leaders are responsible for cultivatingren (benevolent good feelings) within the group (Dai & Zheng,2002; Graham & Lam, 2003), and Japanese employees address super-visors as ‘‘old men” and ‘‘fathers,” even expecting them to presideover weddings and family ceremonies (Dore, 1973).

At this point, note that we are not arguing that Asians are lesscreative or innovative than Americans, only that these culturesmight have found different balances between assertive andgroup-focused leader action. Indeed, some researchers argue thathistorically, ancient Chinese civilizations were more technologi-cally advanced than the Greeks (Nisbett, Peng, Choi, & Norenzayan,2001). In other words, the ancient Chinese and Greeks were able topursue similar goals through different cultural values. Althoughthe Chinese discouraged assertive change, seeing it as a threat togroup stability, they encouraged technological developmentthrough pragmatism. Using a physical metaphor, Hsu (1983) ob-served that Chinese leaders preferred to look backwards to the pastfor answers, rather than look forward.

Drawing from these arguments, we refine how followers under-stand leader action cross-culturally. Rather than simply focusingon asserting one’s will in the environment and approaching oppor-tunities without sensitivity to others, Asians might expect leaderaction to also focus on group relationships.

Representing power as group focus: trailing-behind imagery.We suggest that the group-focused roles associated with Asian

paternal leadership are embodied in trailing-behind imagery.Although this position constrains the leader’s ability to view theopportunities that lie ahead, it enables the leader to monitor thegroup, protect it from harm, and ensure stability. Like the frontierimage, this defensive posture also evokes linear imagery (e.g.,

Table 1Hypotheses.

Blazing the trail Trailing

Spatial metaphor Leader in the front LeaderPrevalence of spatial

metaphorMore prevalent in Western culturalcontexts

More p

Leader’s focus Focused on assertion FocuseLeadership evaluation – More favorably evaluated in US

– Contingent on assertion success– Mo– Con

weSituational facilitator At the presence of opportunities At the

‘‘holding the line”), as it sustains, supports, and stabilizes groupmotion rather than pushing it forward (Fig. 1b).

Testing for cultural differences in leadership imagery

We summarize our key constructs, hypotheses and predictionsin Table 1. First, we define the key constructs as follows:

Trailblazing leadership. Trailblazing leadership is the visual rep-resentation of assertive power (Galinsky et al., 2003; Keltner et al.,2003). The trailblazing leader’s actions are approach oriented, fo-cused on the pursuit of opportunities and rewards. To gain re-wards, leaders assert their will on the environment, engaging inbehaviors that can be innovative and risky.

Trailing-behind leadership. The visual representation of group-fo-cused leadership (Cheng et al., 2004) is the trailing-behind leader.Trailing-behind leaders focus their attention on the group, andtheir actions are protection-oriented, defending group stability.

Predictions

We propose that cultural differences in leader imagery arise be-cause Americans and Asians differentially emphasize the leader’ssocial role as promoting assertive action or watchfully protectingthe group. We suggest that both Americans and Asians recognizeassertive action and group focus as complementary roles of leader-ship. However, the American focus on assertive power (Burns &Brady, 1992; Hoppe & Bhagat, 2008; Kim, 2005; Lee et al., 2000;Therivel, 1995) causes preferences for trailblazing leaders whoforge innovative paths before them, unless group protection is sali-ent. While Asians recognize assertive power, they also focus ongroup and interpersonal relationships (Chen et al., 2005; Leeet al., 2000; Markus & Kitayama, 1991) and are more likely thanAmericans to represent leaders who trail behind followers towatch over them.

These arguments manifest themselves in four specific predic-tions. First, we study both cultures’ assumptions about the leader’sposition (Study 1):

Hypothesis 1. Americans are more (less) likely than Asians toposition leaders in the front (at the back).

Next, we show cultural differences in how people evaluate frontand back leaders’ effectiveness, and test whether perceptions thatthe leader is assertive versus group-focused mediate these culturaldifferences (Study 2).

Hypothesis 2a. Compared to Asians, Americans are relativelymore likely to evaluate front leaders as effective.

Hypothesis 2b. Compared to Americans, Asians are relativelymore likely to evaluate back leaders as effective.

the group Hypotheses (Study)

at the backrevalent in Asian cultural contexts Hypothesis 1 (Studies 1a and

1b)d on groupre favorably evaluated in Asiatingent on success in addressing group

lfare

Hypotheses 2–3 (Study 2)

presence of threats Hypothesis 4 (Study 3)


Hypothesis 3a and b. The effect described in Hypothesis 2 is med-iated by the perceptions of the leaders’ relative focus on (a) asser-tion and (b) group welfare.

Finally, situational priming can simulate these cultural differ-ences, even within primarily American samples (Study 3). GivenKeltner et al.’s (2003) arguments that reward-rich environmentstrigger opportunity seeking whereas threat triggers a focus onprotection:

Hypothesis 4. Opportunity salience versus threat salience shouldincrease (decrease) the likelihood to select leaders in the front (atthe back).

Fig. 2. Fish display used in Study 1: (a) front fish higher than back fish and (b) frontfish lower than back fish.

Overview of the studies

In three studies, we explore how people spatially representleaders and followers. Studies 1a and b are cross-cultural compar-isons demonstrating that Asians are more likely to select back lead-ers than are Westerners. Study 1a uses a stimulus in which collegestudents from the United States and Singapore identify the leaderin a school of cartoon fish (Morris & Peng, 1994). In addition to thisetic, experimenter-designed stimulus, Study 1b employs an emicmethod (Pike, 1967) in which participants draw their own imagesof groups, and independent coders rate how they orient leadersand followers.

In Study 2, we assign Singaporeans and Americans to evaluateeither front or back leaders. We also explore whether Singaporeansand Americans differ in rating these leaders and test whether per-ceiving the leader as assertive or group-focused mediates the ef-fects of culture and position on leader evaluations.

Finally, Study 3 simulates the cross-cultural differences in a pri-marily Western managerial sample. Given Keltner et al.’s (2003)arguments that assertive power emerges in resource-rich environ-ments and a protection orientation arises in threatening environ-ments, Study 3 situationally primes opportunity versus threat toexplore the conditions that facilitate these leader representations.We hypothesize that when threats are salient, even chronicallyopportunity-focused Americans will be more likely to locate lead-ers behind groups, where they can fulfill a protective role. Whilethe back position facilitates the leader’s ability to watch over thegroup and protect it from threats, the front position enables lead-ers to see what lies ahead and aggressively pursue opportunities.

Study 1: Americans and Asians identify the leader’s position

To test Hypothesis 1—the basic contention that Americans(Asians) are more likely than Asians (Americans) to choose leadersin front (back)—we performed two investigations. Study 1a dem-onstrates the effect when participants choose a leader in an exper-imenter-provided stimulus and Study 1b replicates this patternbased on participants’ self-generated drawings of groups.

Study 1a: selecting the leader in an experimenter-created stimulus

Study 1a adapted Morris and Peng’s (1994) cartoon fish stimu-lus, which is influential in cross-cultural research because peopleanthropomorphize the fish, importing their implicit cultural theo-ries about social interactions instead of using knowledge aboutspecific groups and organizations. Prior research with this stimulusindicates that Americans make dispositional attributions about thefirst fish (they view it as a leader), while Chinese view that fishmore situationally (they assume that the group is chasing it; Mor-ris & Peng, 1994).

Rather than examine attributions about the first fish, we askedAmericans and Asians to simply identify the leader fish. Further, gi-

ven that past research suggests that height might influence percep-tions of power (Giessner & Schubert, 2007; Schubert, 2005), wecounterbalanced the heights of the leaders and followers. Wehypothesized that, regardless of height, people in both cultureswould infer that the front fish is the leader, but Asians would bemore likely than Americans to select back leaders.

MethodParticipants. We collected data in two waves. In the first wave, par-ticipants were 36 American (54% male) and 78 Singaporean (58%male) undergraduates who received $3 for their participation. Inthe second wave, participants were 66 American (46% male) and48 Singaporean (58% male) undergraduates who received $1 fortheir participation.

Procedure. This study employed a 2 (Culture: American versus Sin-gaporean) � 2 (Leader Height: high versus low) between-subjectsdesign. We asked American and Singaporean participants to viewthe cartoon and, ‘‘Circle the fish that you think is most likely tobe guiding the other fish.” In the first wave of data collection, par-ticipants viewed a stimulus in which one fish swam ahead and an-other swam behind (see Fig. 2a), and in the second wave, theyviewed an identical stimulus except that we reversed the heightof the front fish so that it was lower than the group (see Fig. 2b).

ResultsBecause few participants chose a middle fish as the leader (in

the first wave, one Singaporean and no Americans; in the secondwave, two Americans and two Singaporeans), we eliminated thiscategory. We then performed a Culture (American, Singapore-an) � Leader Height (high versus low) logistic regression, with Lea-der Selection (front versus back) as the dependent variable.

As predicted, we found a main effect of Culture, b = �1.20,SE = 0.54, p = .027, while the main effect of Height, b = �0.59,SE = .46, p = .20, and the Culture � Height interaction, b = �15.67,SE = 1087.10, p = .99 were nonsignificant. Thus, 94% of Americansselected the front fish as leader (six selected the back fish). By con-trast, only 80% of Singaporeans did so, and the rest (24 partici-pants) selected the back fish as leader.

2 A one-way ANOVA (American versus Singaporean) revealed that the two culturalgroups did not differ in the number of followers that they drew, F(1, 189) = 1.90,p = .17, g2

p ¼ :01 (Americans: M = 12.66, SD = 11.15; Singaporeans: M = 10.37,SD = 11.78).


Study 1b: examining leader imagery through emic methods

Study 1a employed an etic approach: that is, we imposed anexperimenter-created stimulus upon participants in both Ameri-can and Singaporean cultures. Critics of etic approaches withincross-cultural psychology have noted that experimenter-createdstimuli might restrict and constrain participants’ responses (Pike,1967). To supplement this test with an emic approach, that is,one that relies upon participants’ self-generated perceptions, wedesigned a study in which participants drew their own images ofleaders and followers.

MethodParticipants. Participants were 98 American (46% men, 41% wo-men, 13% missing gender data) and 95 Singaporean (41.1% men,54.7% women, 4.3% missing gender data) undergraduates whowere compensated $1 for their participation.

Procedure. Participants received the following instructions via acomputer that ran Microsoft Word:

Show us how you would design a group. Use the blue figure torepresent the leader. Use the orange figures to represent the fol-lowers. To arrange the group, copy, paste, and drag the figuresyou would like to use into the box.

In the center of the computer screen, participants saw a box,and at the bottom of the screen, there was a blue figure that repre-sented the leader and an orange figure that represented followers.We restricted participants to one leader, but they could include asmany followers as they chose. They could copy, paste, and dragthese figures, position them anywhere on the screen, and orientthem in any direction. When participants completed their draw-ings, we saved and printed their work.

Measures. Two coders (one of Asian descent, the other of Euro-pean–American descent) independently coded participants’ draw-ings, and resolved any coding discrepancies through discussion.Given that followers generated their own drawings, they drewgroups in various shapes beyond Study 1a’s linear representation.We therefore created a basic coding scheme to capture the leader’sfocus of attention:Number of followers facing the leader’s back. The inter-rater reliabil-ity was .99. Consistent with the trailblazing representation, we ex-pected Americans to depict relatively more followers watching theleader’s back.Number of followers that the leader watches. The inter-rater reli-ability was .98. Consistent with the trailing behind representation,we expected Singaporeans to depict leaders watching relativelymore followers, as compared to looking outwards at the environ-ment or frontier.

To create the two key dependent variables, we also instructedcoders to count the total number of followers (inter-rater reliabil-ity = .99). We formed a proportion of the group that the leaderwatched or that watched the leader to control for the size of thegroup that the participants drew.

ResultsWe first computed the proportion of followers facing the lea-

der’s back. The distribution of the proportions was highly skewed(skewness = 1.49, t = 8.42, p < .001), and a square-root transforma-tion reduced skewness to 0.75 (t = 2.75, p < .01). Analysis of vari-ance performed on the transformed proportions revealed thatAmericans represented more followers as facing the leader’s back,F(1, 187) = 4.40, p < .05, g2

p ¼ :023 (Americans: M = 24.11%,SD = 27.17%; Singaporeans: M = 19.80%, SD = 29.13%).

We next computed the proportion of followers that the leaderwatched. Once again, the distribution of the proportions was

highly skewed (skewness = 2.39, t = 13.56, p < .001), and asquare-root transformation reduced skewness to 0.75 (t = 0.33,ns). Analysis of variance performed on the transformed proportionsrevealed that Singaporeans (M = 41.19%, SD = 35.82%) representedleaders as watching a greater proportion of followers than didAmericans (M = 30.69%, SD = 47.03%), F(1, 189) = 7.98, p = .005,g2

p ¼ :041.2

Discussion of Studies 1a and 1bStudy 1 provides evidence of cross-cultural differences in the

physical positions people associate with leadership (Hypothesis1). Study 1a found that Singaporeans were more likely than Amer-icans to choose a leader in the back within a stylized, experi-menter-provided stimulus. Consistent with this effect, Study 1btapped participants’ freely generated representations and foundthat Americans represented more followers facing the leader’sback, and Singaporeans were more likely to draw leaders watchingthe group. While much research has noted cultural differences inpeople’s values (Hofstede, 1980) and attributions (Menon, Morris,Chiu, & Hong, 1999) about groups, this study begins to answerthe critical, but less-studied question of how culture also shapesthe mental pictures people carry about leaders and their groups(see Brewer & Chen, 2007).

Note that nearly every American and the majority of Singapore-ans selected the front fish as the leader in Study 1a, perhaps be-cause social and business organizations frequently assign leadersto front positions. However, Singaporeans exhibited more variedchoices, with a sizable minority selecting the back fish. BecauseSingapore is a multicultural, industrialized society, with heavyAsian and Western influences (Inglehart, 1997; Li, Ngin, & Teo,2008), Singaporeans exhibit more balanced leader representationsrecognizing both faces of leadership. To help understand whatthese cultural differences mean, we next explore the basic typesof action people associate with the leader’s position.

Study 2: Group focus mediates cultural differences in leaderevaluations

Study 2 tests Hypotheses 2 and 3 in a cross-cultural experiment.We consider two issues. First, we explore cultural differences inevaluating the effectiveness of front and back leaders, and testwhether, as compared to Singaporeans, Americans are relativelymore likely to evaluate front leaders as effective (Hypothesis 2a).Likewise, as compared to Americans, we test whether Singapore-ans are more likely to evaluate back leaders as effective (Hypothe-sis 2b). Second, we test whether perceiving front and back leadersas assertive (Hypothesis 3a) and group-focused (Hypothesis 3b),respectively mediate ratings of leader effectiveness. This study alsochecks our assumption that people associate front leaders with anassertion orientation and back leaders with group focus.

MethodParticipants. The participants were 107 American (55% male, 43%female, two missing gender data) and 97 (48% male, 50% female,two missing gender data) Chinese Singaporean undergraduates.Participants received $3 for their participation.

Procedure. The design included two independent variables, eachwith two levels: Culture (American versus Singaporean) and Lea-der Position (leader in front versus leader at the back). Participantsviewed a cartoon depicting the ‘‘human” work team, and we desig-


nated the leader by circling either the figure in the front of thegroup or the figure behind it (see Fig. 1). Participants then an-swered a questionnaire that contained the key measures.

Measures. Three orthogonal factors emerged in a factor analysis ofthe questionnaire items. The first factor tapped leader effectivenessratings:

Overall leader evaluation (two items, alpha = .86). ‘‘Has goodleadership qualities”; ‘‘Has poor leadership qualities” (reversescored).

The other two factors tapped the types of action people associ-ated with the designated leader. One factor captured perceptionsthat the leader would engage in assertive, bold action to transformthe environment in innovative ways:

Assertive action (three items, alpha = .86). ‘‘Is a risk taker”; ‘‘Is aninnovator”; ‘‘Is creative.”

The final factor captured the leader’s role in monitoring thegroup’s welfare:

Group-focused action (three items, alpha = .84). ‘‘Is receptive tothe ideas of others in the group”; ‘‘Attends to the needs of the restof the group”; ‘‘Cares for others in the group.”

Participants responded to all questions on a Likert scale(1 = highly unlikely; 7 = highly likely).

ResultsLeader evaluation. We first tested for cultural differences in perceiv-ing leader effectiveness (Hypothesis 2). We performed a Cul-

3

3.5

4

4.5

5

eropagniSSU

Front leaderBack leader

3

3.5

4

4.5

5

5.5

eropagniSUS


3

3.5

4

4.5

5

5.5

eropagniSSU


a

b

c

Fig. 3. Assumptions about leaders at the back and front: (a) leader evaluation, (b)assertion, and (c) group focus (Study 2).

ture � Position ANOVA on leader evaluation (Fig. 3a). We found asignificant Culture � Position interaction, F(1, 200) = 4.18, p = .042,g2

p ¼ :020, partially supporting Hypothesis 2a. The two groups didnot evaluate the front leader differently (Americans: M = 4.70,SD = 1.14; Singaporeans front: M = 4.58, SD = 1.32), t(202) = 0.69,ns, failing to support Hypothesis 2a. However, in support of Hypoth-esis 2b, Singaporeans (M = 4.04, SD = 1.36) saw the back leader asmore effective than did Americans (M = 3.41, SD = 1.42),t(202) = 3.24, p < .01. There was a significant main effect of positionwhereby both cultures saw front leaders as more effective than backleaders, F(1, 200) = 24.49, p < .001, g2

p ¼ :11. The main effect of cul-ture was not significant, F(1, 200) = 1.86, p = .17, g2

p ¼ :009.

Mediation Analysis. Next we tested Hypothesis 3. To test whetherassertion and group focus mediate the effect of culture and posi-tion on leader evaluation, we first test whether we obtained a sig-nificant Culture � Position interaction on assertion and groupfocus separately.

Assertion as mediator. When we performed a Culture � Position AN-OVA on assertion (Fig. 3b), results showed that the expected Cul-ture � Position interaction was not significant, F(1, 205) = 0.097,p = .76, g2

p ¼ :00. Thus, assertion fails to be a mediator for thecross-cultural differences in leader evaluation, failing to supportHypothesis 3a. We obtained a main effect of position on assertion,F(1, 200) = 67.14, p = .00, g2

p ¼ :25. This effect confirmed ourassumption that participants from both cultures similarly repre-sented front leaders as more assertive than back leaders (Americansfront: M = 4.45, SD = 1.15; Americans back: M = 3.22, SD = 1.09; Sin-gaporeans front: M = 4.94, SD = 1.01; Singaporeans back: M = 3.61,SD = 1.18). We also obtained a main effect of culture on assertion,F(1, 200) = 8.12, p = .005, g2

p ¼ :039, whereby Singaporeans evalu-ated both front and back leaders as more assertive than didAmericans.

However, assertion did mediate the main effect of position onleader evaluation. When we entered the main effects of cultureand position, their interaction, and the mediator, assertion, in theregression with leader evaluation as the dependent variable,although the main effect of position remained highly significant,its effect was significantly attenuated, b = �0.53, p = .006, Sobel’sz = 2.19, p = .028, while the effect of assertion remained significant,b = 0.45, p < .01. Thus, both cultures considered front leaders to bemore effective than back leaders, and this was partially mediatedby perceiving those leaders as more assertive.

Group focus as mediator. We next performed a Culture � PositionANOVA on group focus (Fig. 3c), and obtained the expected Cul-ture � Position interaction effect necessary for mediation,F(1, 200) = 8.53, p = .004, g2

p ¼ :041. Singaporeans viewed the backleader as more group-focused than did Americans, t(202) = 6.73,p < .001), but the two groups did not differ in their perception ofthe front leader being group-focused, t(202) = 0.70, ns (Americansfront: M = 3.77, SD = 1.24; Americans back: M = 4.19, SD = 1.28;Singaporeans front: M = 3.89, SD = 1.21; Singaporeans back:M = 5.29, SD = 1.05). We also obtained a main effect of position,F(1, 200) = 29.23, p < .01, g2

p ¼ :13, which confirmed our assump-tion that both cultures saw back leaders as more group-focusedthan front leaders. There was also a significant main effect of cul-ture, F(1, 200) = 13.29, p < .001, g2

p ¼ :062, with Singaporeans rat-ing both front and back leaders as more group-focused than didAmericans.

We tested for mediation by entering the main effects of cultureand position, their interaction, and the mediator, group focus, inthe regression, with leader evaluation again the dependent vari-able. The Culture � Position interaction was no longer significant,b = 0.27, p = 0.33, but the effect of group focus remained significant,

Table 2Leader selection with threat and opportunity priming (Study 3).

Front leader Middle leader Back leader

Opportunity priming 27 (43.5%) 32 (51.6%) 3 (4.8%)Threat priming 23 (41.1%) 22 (39.3%) 11 (19.6%)

Note. We report counts of participants who selected leaders in each position, withthe percentage from each priming condition in parentheses.


b = .40, p < .001 (Sobel’s z = 2.67, p = .008), confirming Hypothesis3b. We conclude that Asians’ relatively higher preferences for lead-ers in back derived from their inference that such a leader watchesover the group.

DiscussionThese findings suggest that physical position is a metaphor for

the social roles associated with leadership (e.g., assertively pursu-ing innovative opportunities versus protectively focusing on thegroup). In this study, we confirmed our assumptions with data thatboth Americans and Singaporeans viewed front leaders as assertivetrailblazers and back leaders as watching over the group. While thecultures did not differ in perceiving front leaders as assertive,Asians did rate the back leader as more group-focused than didAmericans. Further, although the two cultures did not differ intheir evaluations of front leaders (Hypothesis 2a), Asians saw theback leader as more effective than did Americans (Hypothesis2b). The cultural difference whereby Singaporeans rated back lead-ers as more effective was mediated, not by assertion (Hypothesis3a), but by Singaporeans’ perceptions that back leaders were moregroup-focused (Hypothesis 3b).

These findings help explicate the dual face of Asian leadershipperceptions. Whereas assertive views of power dominate Americanleadership perceptions, Asian perceptions are more nuanced.Whereas Study 1 showed that most Singaporeans chose front lead-ers, several represented leaders in back. And Study 2 showed thatalthough Singaporeans largely shared American views of frontleaders and assertive power, they were more likely to perceiveback leaders favorably because they assumed that back leaderswere focused on the group. Given the pervasive Western influencein Singapore, it is unsurprising that Singaporeans might be accus-tomed to front leaders and assertive power. However, althoughhighly counterintuitive to Westerners, the prevalence and valua-tion of the back leader representation among Singaporeans reflectsthe crucial Asian leadership function of watching over the group.We next simulate, in an American context, the underlying mindsetthat produces these cultural differences. We test whether situa-tional cues might provoke even Americans to shift away from theapparently obvious cultural assumption that leaders precedefollowers.

Study 3: Shifting managerial assumptions about leader position

To test Hypothesis 4, Study 3 replicates these cross-cultural dif-ferences in a predominantly Western managerial setting. Even if avalue is not chronically active within a culture, if we activate itthrough priming (Gardner, Gabriel, & Lee, 1999), people’s percep-tions and behaviors might mirror what is typical within culturesthat endorse such values. Thus, whereas Study 1 found that nearlyall Americans assumed that the leader was the figure in front of thegroup, Study 3 attempts to attenuate this intuitive cultural beliefwithin a predominantly American sample by activating the mind-set typically present in Asian cultural contexts.

Specifically, Study 2 established that beliefs about the leader’sgroup protective actions mediated cultural differences in people’sevaluations about front and back leaders. Keltner et al. (2003) ar-gue that these types of action orientations emerge within specificenvironmental conditions. Because assertive power arises in envi-ronments rich in reward and opportunity (Keltner et al., 2003), wepredict that when opportunities are salient, Americans reinforcetheir chronic preferences for an innovative trailblazer who standsin front of the group, viewing and capitalizing upon the novelopportunities that lie ahead. By contrast, because threats activatea more vigilant, protection-oriented state (Higgins, 1997; Keltneret al., 2003) that is chronic within Asian contexts (Lee et al.,2000), we predict that Americans thinking about threats will be

more likely to exhibit the patterns we have documented amongAsians. That is, they will situate the leader at the back to protec-tively watch over the group. Thus, rather than demonstratingmechanisms through mediation as in Study 2, Study 3 directlymanipulates the mindsets that facilitate these patterns cross-cul-turally. Additionally, Study 3 examines these leader representa-tions within a managerial population.

MethodParticipants. We surveyed 123 managers in a part-time MBA pro-gram at an American university. Because these managers partici-pated in a class exercise, we did not pay them, but debriefedthem in class. Their average age was 30.16 years (SD = 4.28),74.6% were male, 78% were American, and the rest were other eth-nicities (9.8% Indian and 3.3% Chinese). We excluded one partici-pant whose age was 5.35 standard deviations older than that ofthe other participants.

Procedure. To manipulate the mindsets that would facilitate theserepresentations, participants were randomly assigned to receiveone of two instructions, either ‘‘Please list three key threats thata company might face,” or ‘‘Please list three key opportunities thata company might face.” Following this, they viewed a cartoon thatdepicted a ‘‘human” team (i.e., Fig. 1 with no leader circled). Wetold them: ‘‘Circle the figure that you think is most likely to be incharge of this team.”

ResultsWe performed a multinomial regression with Priming Condi-

tion (opportunity versus threat) as the independent variable andLeader Selection as the categorical dependent variable (front ver-sus middle versus back). Condition was significant, v2 (1,N = 119) = 6.74, p = .034. As compared to threat-primed partici-pants, opportunity-primed participants selected more leaders inboth the front (b = 1.46, SE = 0.71, p = .040) and middle (b = 1.67,SE = 0.71, p = .018) than leaders in the back. We report the numberand percentage of participants who selected front, middle, andback leaders in the opportunity versus threat priming conditionsin Table 2.

DiscussionThis study reveals the conditions that underlie the cultural dif-

ferences we have documented in Studies 1 and 2. In support ofHypothesis 4, we found that temporarily salient threats and oppor-tunities can shift preferences for front and back leaders, even with-in a predominantly American sample.

Specifically, managers primed with opportunity relied more onleaders in the front, who are poised to scout for opportunities, andmanagers primed with threat were more likely to select back lead-ers who can vigilantly resolve threats to the group’s security.

General discussion

The American Psychologist’s special issue on leadership urgedresearchers to integrate culture into their models (Avolio, 2007).In response, this paper examines how culture shapes people’s lead-


ership imagery (Giessner & Schubert, 2007; Raghubir & Valenzuela,2006; Schubert, 2005). Because American culture emphasizesassertion (e.g., pursuing opportunities, asserting one’s will on theenvironment, innovation, and risk-taking), Americans mentally sit-uate innovative leaders at the forefront of groups. And becauseAsian cultures inculcate group focus (e.g., relational goals, protect-ing the group from threat, maintaining group stability), Singapore-ans are more likely to position leaders at the back, watching overthe group. Thus, although Westerners may regard it as intuitiveand obvious that leaders would and should precede their followers,we suggest that this familiar assumption is culturally bound.

These studies offer several opportunities for future research.Most notably, we have not assessed the prevalence of this leaderimagery in Asian contexts other than Singapore. Because Singa-pore’s most well-known leader, Lee Kwan Yew, derived his leader-ship philosophy from Confucianism (Dai & Zheng, 2002; Tu, 1998;Zakaria, 1994), we expect similar patterns in other Confucian cul-tures, which similarly inculcate group focus.

Second, abstract concepts such as power and leadership can beunderstood through many images. Although we compared backversus front leaders, people associate other physical positions withleadership. Whereas this research shows that goal orientationstrigger front-back imagery, popularity cues might heighten prefer-ences for central leaders (Raghubir & Valenzuela, 2006), and hier-archy cues might heighten preferences for leaders at the top(Giessner & Schubert, 2007; Schubert, 2005). Front-back imageryco-exists alongside other metaphors for leadership because eachmetaphor only partially captures the concept (Lakoff & Johnson,1980).

Finally, we deliberately employed cartoon images of fish (Study1a) and people (Studies 1b, 2, and 3) so that participants relied ontheir abstract implicit theories of leadership rather than on specificknowledge about particular types of groups (Morris & Peng, 1994).To extend the external validity of these findings, research mightexamine physical position in real organizations (e.g., office design,seating arrangements at meetings). Similarly, further researchcould consider how the type of group and relationships betweengroup members and the leader shape these inferences (e.g.,close-knit versus less coordinated groups, Ip, Chiu, & Wan, 2006;Lickel et al., 2000).

Theoretical implications

In spite of these limits, the research suggests two key implica-tions for leadership theory. First, while much research notes howleaders use symbols to influence followers (Pfeffer, 1981; Weick,1987), less research has considered how followers actually inter-pret those symbols. The present research shows that intuitive lead-ership theories might be grounded in less formalized ways(Barsalou, 1999). In addition to holding declarative beliefs aboutleaders, people also simulate leader behaviors and situate their ac-tion in a spatial context (Barsalou, 1999). Specifically, a front orback physical position might become ‘‘entrenched” in memory asa situated concept, activating a broad pattern of beliefs associatedwith it (e.g., cultural values around assertion or protection, Nieden-thal, Barsalou, Winkielman, Krauth-Gruber, & Rie, 2005). While re-search on symbolic leadership often assumes that leaders candeploy symbols (including position) with few constraints (Pfeffer,1981), our research questions this assumption by showing howperceptual symbol systems arise from the cultural context andhow they become entrenched in the mind, limiting the effective-ness of the leader’s symbolic repertoire.

A second implication for leadership theory emerges when com-paring the content of this imagery cross-culturally. Specifically,leaders are expected to both take the group to the Promised Land

and to protect the group from harm’s way. These functions areembodied in the spatial imagery we identify. However, becauseof the strong association of leadership with approach goals inAmerican contexts, research has often overlooked group-protec-tive-functions.

As a result, our research questions several fundamentalassumptions about power and its consequences (see Zhong et al.,2006 for related arguments). First, we have reviewed prominentrecent research that conceptualizes power as triggering an ap-proach orientation (Keltner et al., 2003) and assertive action(Galinsky et al., 2003). Classic leadership research similarly por-trays charismatic leaders transforming their environments andoverturning the status quo with groundbreaking innovations(Antonakis & House, 2002; Bass, 1981; Ehrhart & Klein, 2001;Galinsky et al., 2003; Kark & Van Dijk, 2007; Meindl et al., 1985;Sternberg et al., 2003; Weber, 1947; Zaleznik, 1989). Related tothis assertive conception of power, other research assumes powerprovokes ‘‘asocial” consequences, reducing perspective-taking(Galinsky et al., 2006) and creating psychological distance fromother people (Lee & Tiedens, 2001). By studying leadership func-tions and their associated imagery in a culture that also privilegesrelational goals, we reveal the dual functions of leadership, high-lighting the leaders’ role in both finding opportunities and watch-ing over the group (Kark & Van Dijk; 2007; Zhong et al., 2006).While we have observed how both of these functions emergeand shape leader evaluations, further research might considerhow to facilitate a different kind of power—one that triggersgroup-focused, protection-oriented, prosocial action (Chen et al.,2001; Galinsky et al., 2003; Overbeck & Park, 2001).

Implications for leadership in practice

We next consider the organizational implications of this con-trasting leadership imagery. The most obvious implication is that,if the prototypical positions that people associate with leadershipdiffer cross-culturally, people might misdiagnose social relationsin inter-cultural interactions. For instance, Americans might as-sume that the Asian businessman in the back of the room is anassistant, rather than the key decision maker. This inference, basedon a culturally-bound perceptual representation, could be sociallyand financially costly.

Likewise, if leaders who fit prototypes are especially likely to beevaluated favorably (Giessner & van Knippenberg, 2008; van Knip-penberg & Hogg, 2003), the cultural differences we have identifiedsuggest that inter-cultural interactions might suffer because cul-turally diverse leaders fail to fit prototypes. For instance, Americanfollowers might resent a leader in back who monitors them, lead-ing to hostility on cross-cultural teams. Similarly, although a front-leader fits American leader prototypes well, some Asian followersmight prefer a back leader who watches over the group.

Finally, although this paper has emphasized a causal processwhereby leadership imagery emerges from cultural assumptions,positions might also foster leadership styles by facilitating particu-lar leadership behaviors. If leaders are randomly assigned to take afront versus back position, it is possible that they might shift intheir styles, priorities, and values. One reason why position couldcreate leadership style is through the vantage point it offers. Fromthe back position, the leader has a holistic view (Nisbett et al.,2001) that facilitates genchi genbutsu (in Japanese, ‘‘deeply observ-ing the actual situation in detail,” Liker, 2004, p. 173). Thus,whereas American automobile CEOs often initiated radical actions(Zhong et al., 2006), storming in from the outside to abruptly shakeup company culture and direction, Japanese automobile CEOs wereinsiders who understood the group, built on the company’s priormomentum, and stepped in only to ensure stability (Liker, 2004).


Similarly, Nelson Mandela associated the back position with a pro-tective leadership style noting, ‘‘A leader. . . is like a shepherd. Hestays behind the flock, letting the most nimble go out ahead,whereupon the others follow, not realizing that all along they arebeing directed from behind” (Mandela, 1994, pp. 25–26). Thus,the relationship between symbol and culture is mutually reinforc-ing: cultural values shape salient symbols, these symbols leadmanagers to design organizations that locate leaders in front ofor behind groups, and these physical positions in turn facilitateleadership styles consistent with cultural expectations, reinforcingthe symbols.

Conclusion

In sum, leadership imagery matters because it both reflects andshapes how people think about leaders (Lakoff & Johnson, 1980).Drawing from historical narratives featuring George Washingtonor Abraham Lincoln as trailblazers challenging the frontier, or inthe story of Lee Kwan Yew, Singapore’s benevolent protector(Kitayama et al., 2006), contemporary national leaders legitimatetheir power through consistent imagery and storylines, wearingcowboy hats as they visit their ranches or offering fatherly wisdomto ensure they fit the prototype (Giessner & van Knippenberg,2008; van Knippenberg & Hogg, 2003). By defining the leader’s po-sition—both socially and spatially—culture colors the pictures wesee when we imagine following the leader.

Acknowledgment

We are grateful for the insightful comments from the reviewers,the editor, Daan van Knippenberg, and Adam Galinsky, Will Felps,Steffen Giessner, Priya Raghubir, Thomas Schubert, Ana Valenzue-la, Chen-bo Zhong, and participants at the Erasmus LeadershipConference. This research was funded by the University of ChicagoBooth School of Business, the Kauffman Foundation, and the Tem-pleton Foundation.

References

Anderson, C., & Galinsky, A. (2006). Power, optimism, and risk-taking. EuropeanJournal of Social Psychology, 36, 511–536.

Andrzejewski, S. (1954/2003). Military organization and society: The sociology of workand organization. London: Routledge.

Antonakis, J., & House, R. J. (2002). The full-range leadership theory: The wayforward. In B. J. Avolio & F. J. Yammarino (Eds.), Transformational and charismaticleadership: The road ahead (pp. 3–33). Oxford, England: Elsevier Science.

Avolio, B. J. (2007). Promoting more integrative strategies for leadership theory-building. American Psychologist, 62, 25–33.

Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22,577–609.

Bass, B. M. (1981). Stogdill’s handbook of leadership: A survey of theory and research(Rev. ed.). New York: Free Press.

Bass, B. M., & Klubeck, S. (1952). Effects of seating arrangement on leaderless groupdiscussions. Journal of Abnormal Psychology, 47, 724–727.

Boroditsky, L. (2000). Metaphoric structuring: Understanding time through spatialmetaphors. Cognition, 75, 1–28.

Brewer, M. B., & Chen, Y.-R. (2007). Where (who) are collectives in collectivism?Toward conceptual clarification of individualism and collectivism. PsychologicalReview, 114, 133–151.

Burns, D. J., & Brady, J. (1992). A cross-cultural comparison of the need foruniqueness in Malaysia and the United States. Journal of Social Psychology, 132,487–495.

Chemers, M. M. (2001). Leadership effectiveness: An integrative review. In M. A.Hogg & R. S. Tindale (Eds.), Blackwell handbook of social psychology: Groupprocesses (pp. 376–399). Oxford, UK: Blackwell.

Chen, H., Ng, S., & Rao, A. (2005). Cultural differences in consumer impatience.Journal of Marketing Research, 42, 291–301.

Chen, S., Lee-Chai, A. Y., & Bargh, J. A. (2001). Relationship orientation as amoderator of the effects of social power. Journal of Personality and SocialPsychology, 80, 173–187.

Cheng, B.-S., Chou, L.-F., Wu, T.-Y., Huang, M.-P., & Fahr, J.-L. (2004). Paternalisticleadership and subordinates responses: Establishing a leadership model inChinese organizations. Asian Journal of Social Psychology, 7, 89–117.

Chow, I. H.-S. (2008). Culture and leadership in Hong Kong. In J. S. Chhokar, F. C.Brodbeck, & R. J. House (Eds.), Culture and leadership across the world: The GLOBEbook of in-depth studies of 25 societies (pp. 909–946). Mahwah, NJ: LawrenceErlbaum Associates.

Dai, C., & Zheng, Z.-G. (2002). Managing talent in China: Confucian origins. In C. B.Derr, S. Roussillon, & F. Bournois (Eds.), Cross-cultural approaches to leadershipdevelopment (pp. 155–167). Westport, CT: Quorum Books.

Dore, R. P. (1973). British factory–Japanese factory: The origins of national diversity inindustrial relations. Berkeley: University of California Press.

Eaton, A. A., Visser, P. S., Krosnick, J. A., & Anand, S. (2009). Social power and attitudestrength over the life cycle. Personality and Social Psychology Bulletin, 12,1646–1660.

Ehrhart, M. G., & Klein, K. J. (2001). Predicting followers’ preferences for charismaticleadership: The influence of follower values and personality. LeadershipQuarterly, 12, 153–179.

Ellyson, S. L., & Dovidio, J. F. (Eds.). (1985). Power, dominance, and nonverbalbehavior. New York: Springer-Verlag.

Erez, M., & Earley, P. C. (1993). Culture, self-identity, and work. New York: OxfordUniversity Press.

Fiske, A. P. (2004). Four modes of constituting relationships: Consubstantialassimilation; space, magnitude, time, and force; concrete procedures; abstractsymbolism. In N. Haslam (Ed.), Relational models theory. A contemporaryoverview (pp. 61–146). Mahwah, NJ: Lawrence Erlbaum.

Fiske, S. T. (1993). Controlling other people: The impact of power on stereotyping.American Psychologist, 48, 621–628.

French, J. R. P., & Raven, B. H. (1959). The bases of social power. In D. Cartwright (Ed.),Studies in social power (pp. 150–167). Ann Arbor, MI: Institute for Social Research.

Fu, J. H.-Y., & Chiu, C. (2007). Local culture’s responses to globalization: Exemplarypersons and their attendant values. Journal of Cross-Cultural Psychology, 38(5),636–653.

Fu, P. P., Wu, R., Yang, Y., & Ye, J. (2008). Chinese culture and leadership. In J. S.Chhokar, F. C. Brodbeck, & R. J. House (Eds.), Culture and leadership across theworld: The GLOBE book of in-depth studies of 25 societies (pp. 877–907). Mahwah,NJ: Lawrence Erlbaum Associates.

Galinsky, A. D., Gruenfeld, D., & Magee, J. (2003). From power to action. Journal ofPersonality and Social Psychology, 85, 453–466.

Galinsky, A. D., Magee, J. C., Inesi, M. E., & Gruenfeld, D. H. (2006). Power andperspectives not taken. Psychological Science, 17, 1068–1074.

Gardner, W. L., Gabriel, S., & Lee, A. Y. (1999). ‘‘I” value freedom but ‘‘we” valuerelationships: Self-construal priming mirrors cultural differences in judgment.Psychological Science, 10, 321–326.

Giessner, S. R., & Schubert, T. W. (2007). High in the hierarchy: How vertical locationand judgments of leaders’ power are interrelated. Organizational Behavior &Human Decision Processes, 104, 30–44.

Giessner, S. R., & van Knippenberg, D. (2008). ‘‘License to fail”: Goal definition,leader group prototypicality, and perceptions of leadership effectiveness afterleader failure. Organizational Behavior and Human Decision Processes, 105, 14–35.

Graham, J. L., & Lam, N. M. (2003). The Chinese negotiation. Harvard Business Review,81, 82–91.

Grint, K. (2005). Leadership: Limits and possibilities. New York: Palgrave Macmillan.Grothe, M. (2004). Oxymoronica: Paradoxical wit and wisdom from history’s greatest

wordsmiths. New York: Harper Collins.Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52,

1280–1300.Hofstede, G. (1980). Culture’s consequences: International differences in work-related

values. Thousand Oaks, CA: Sage.Hoppe, M. H., & Bhagat, R. S. (2008). Leadership in the United States of America: The

leader as cultural hero. In J. S. Chhokar, F. C. Brodbeck, & R. J. House (Eds.),Culture and leadership across the world: The GLOBE book of in-depth studies of 25societies (pp. 475–543). Mahwah, NJ: Lawrence Erlbaum Associates.

House, R. J., Hanges, P., Javidan, M., Dorfman, P. W., & Gupta, V. (2004). Leadershipand organizations: A 62 nation GLOBE study. Thousand Oaks, CA: Sage.

Hsu, F. L. K. (1983). Rugged individualism reconsidered. Knoxville: The University ofTennessee Press.

Inglehart, R. (1997). Modernization and postmodernization: Cultural, economic, andpolitical change in 43 societies. Princeton, NJ: Princeton University Press.

Ip, G. W.-M., Chiu, C.-Y., & Wan, C. (2006). Birds of a feather and birds flockingtogether: Physical versus behavioral cues may lead to trait-versus goal-basedgroup perception. Journal of Personality and Social Psychology, 90, 368–381.

Kark, R., & Van Dijk, D. (2007). Motivation to lead, motivation to follow: The role ofthe self regulatory focus in leadership processes. Academy of ManagementReview, 32, 500–528.

Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition.Psychological Review, 110, 265–284.

Kim, K. H. (2005). Learning from each other: Creativity in East Asian and Americaneducation. Creativity Research Journal, 17, 337–347.

Kipnis, D. (1972). Does power corrupt? Journal of Personality and Social Psychology,24, 33–41.

Kitayama, S., Ishii, K., Imada, T., Takemura, K., & Ramaswamy, J. (2006). Voluntarysettlement and the spirit of independence. Evidence from Japan’s ‘‘NorthernFrontier”. Journal of Personality and Social Psychology, 3, 369–384.

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of ChicagoPress.

Lee, A. Y., Aaker, J. L., & Gardner, W. L. (2000). The pleasures and pains of distinctself-construals: The role of interdependence in regulatory focus. Journal ofPersonality and Social Psychology, 78, 1122–1134.


Lee, F., & Tiedens, L. Z. (2001). Is it lonely at the top? Independence andinterdependence of power-holders. In B. M. Staw & R. I. Sutton (Eds.). Researchin organizational behavior (Vol. 23, pp. 43–91). Greenwich, CT: JAI Press.

Li, J., Ngin, P. M., & Teo, A. C. Y. (2008). Culture and leadership in Singapore:Combination of the East and the West. In J. S. Chhokar, F. C. Brodbeck, & R. J.House (Eds.), Culture and leadership across the world: The GLOBE book of in-depthstudies of 25 societies (pp. 947–968). Mahwah, NJ: Lawrence Erlbaum.

Lickel, B., Hamilton, D. L., Wieczorkowska, G., Lewis, A., Sherman, S. J., & Uhles, A. N.(2000). Varieties of groups and the perception of group entitativity. Journal ofPersonality and Social Psychology, 78, 223–246.

Liker, J. K. (2004). The Toyota way: Fourteen management secrets from the world’sgreatest manufacturer. McGraw-Hill: New York.

Low, P. K. C. (2006). Father leadership: The Singapore case study. ManagementDecision, 44, 89–104.

Magee, J. C., Galinsky, A. D., & Gruenfeld, D. H. (2007). Power, propensity tonegotiate, and moving first in competitive interactions. Personality and SocialPsychology Bulletin, 33, 200–212.

Mandela, N. (1994). Long walk to freedom. The autobiography of Nelson Mandela.Boston & New York: Little Brown.

Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for cognition,emotion, and motivation. Psychological Review, 98, 224–253.

McArthur, L. Z., & Post, D. L. (1977). Figural emphasis and person perception. Journalof Experimental Social Psychology, 13, 520–535.

Meindl, J., Ehrlich, S., & Dukerich, J. (1985). The romance of leadership.Administrative Science Quarterly, 30, 78–102.

Menon, T., Morris, M. W., Chiu, C. Y., & Hong, Y. Y. (1999). Culture and the construalof agency: Attribution to individual versus group dispositions. Journal ofPersonality and Social Psychology, 76, 701–717.

Morris, M. W., & Peng, K. (1994). Culture and cause: American and Chineseattributions for social and physical events. Journal of Personality & SocialPsychology, 67, 949–971.

Naisbitt, J. (1982). Megatrends: Ten new directions transforming our lives. New York:Warner Books.

Niedenthal, P. M., Barsalou, L., Winkielman, P., Krauth-Gruber, S., & Rie, F. (2005).Embodiment in attitudes, social perception, and emotion. Personality and SocialPsychology Review, 9, 184–211.

Nisbett, R., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and system ofthoughts: Holistic versus analytic cognition. Psychological Review, 108,291–310.

Northouse, P. G. (2001). Leadership: Theory and practice. Thousand Oaks, CA: Sage.Overbeck, J. R., & Park, B. (2001). When power does not corrupt: Superior

individuation processes among powerful perceivers. Journal of Personality andSocial Psychology, 81, 549–565.

Pfeffer, J. (1981). Power in organizations. Marshfield, MA: Pitman Publishing.Phillips, J. S., & Lord, R. G. (1986). Notes on the practical and theoretical

consequences of implicit leadership theories for the future of leadershipmeasurement. Journal of Management, 12, 31–41.

Pike, K. L. (1967). Language in relation to a unified theory of structure of humanbehavior. The Hague: Mouton.

Raghubir, P., & Valenzuela, A. (2006). Center-of-inattention: Position biases indecision-making. Organizational Behavior & Human Decision Processes, 99,66–80.

Ritter, B. A., & Lord, R. G. (2007). The impact of previous leaders on the evaluation ofnew leaders: An alternative to prototype. Journal of Applied Psychology, 92,1683–1695.

Schubert, T. (2005). Your Highness: Vertical positions as perceptual symbols ofpower. Journal of Personality and Social Psychology, 89, 1–21.

Schyns, B., Meindl, J. R., & Croon, M. A. (2007). The romance of leadership scale:Cross-cultural testing and refinement. Leadership, 3, 29–46.

Simpson, J. A., & Weiner, E. S. C. (Eds.). (1989) (2nd ed.). The Oxford English dictionary(Vols. 1–20). Oxford: Clarendon Press.

Sternberg, R. J., Kaufman, J. C., & Pretz, J. E. (2003). A propulsion model of creativeleadership. The Leadership Quarterly, 14, 455–473.

Sun Tzu (1963). The art of war (S.B. Griffith, Trans). Oxford: Clarendon Press.Taylor, S. E., & Fiske, S. T. (1975). Point of view and perceptions of causality. Journal

of Personality and Social Psychology, 32, 439–445.Therivel, W. A. (1995). Long-term effect of power on creativity. Creativity Research

Journal, 8, 173–192.Tiedens, L. Z., & Fragale, A. R. (2003). Power moves: Complementarity in submissive

and dominant nonverbal behavior. Journal of Personality and Social Psychology,84, 558–568.

Tocqueville, A. D. (2000). Democracy in America (H. Mansfield, & D. Winthrop, Eds. &Trans.), Chicago: University of Chicago Press (Original work published 1835–1840).

Turner, F. J. (1921). The frontier in American history. New York: Henry Holt andCompany.

Tu, W.-M. (1998). Confucius and confucianism. In W. H. Slote & G. A. Devos (Eds.),Confucianism and the family (pp. 3–36). Albany, NY: State University of NewYork Press.

van Knippenberg, D., & Hogg, M. A. (2003). A social identity model of leadershipeffectiveness in organizations. Research in Organizational Behavior, 25, 243–295.

Weber, M. (1947). The theory of social and economic organization (T. Parsons, Trans.).New York: Free Press.

Weick, K. E. (1987). The management of eloquence. In D. R. Hampton, C. E. Sumner,& R. A. Webber (Eds.), Organizational behavior and the practice of management.Glenview, IL: Scott Foresman and Company (pp. 581–586).

Zakaria, F. (1994). Culture is destiny: A conversation with Lee Kuan Yew. ForeignAffairs, 73, 2.

Zaleznik, A. (1989). The managerial mystique: Restoring leadership in business. NewYork: Harper & Row.

Zemba, Y., Young, M. J., & Morris, M. W. (2006). Blaming leaders for organizationalaccidents: Proxy logic in collective-versus individual-agency cultures.Organizational Behavior and Human Decision Processes, 101, 36–51.

Zhong, C., McGee, J. C., Maddux, W. W., & Galinsky, A. D. (2006). Power, culture, andaction: Considerations in the expression and enactment of power in East Asianand Western societies. In E. A. Mannix, M. A. Neale, & Y. Chen (Eds.). Research onmanaging in teams and groups (Vol. 9, pp. 53–73). Greenwich, CT: ElsevierScience.

Blazing the trail versus trailing the group: Culture and perceptions of the leader’s position

Documents