تاثیر قاب بندی موقعیتی و پیچیدگی رهبران کاریزماتیک ایدئولوژیک و عمل گرا : بررسی با استفاده از یک شبیه سازی کامپیوتری

Revisiting the work of Weber [Weber, M. (1921). The theory of social and economic organizations. New York: Free Press], Mumford and colleagues (e.g., [Strange, J. M., & Mumford, M. D. (2002). The origins of vision: Charismatic versus ideological leadership. Leadership Quarterly, 13, 343−377; Mumford, M.D. (2006). Pathways to outstanding leadership: A comparative analysis of charismatic, ideological and pragmatic leaders. Mahwah, NJ: Erlbaum.]) explored the thesis that in addition to charismatic leadership, there exist at least two additional pathways to outstanding leadership: ideological and pragmatic. Despite the compelling results of initial studies, however, questions remain as to when and under what situational conditions these three leaders operate most effectively. As such, an experiment was conducted to investigate two noteworthy contextual influences: 1) situational congruence with a leader's mental model and 2) environmental complexity. The experiment made use of a computerized leadership simulation where participants took on the role of a university chancellor. Results indicate that leader type, complexity, and situational framing were critical factors in determining leader performance on multiple game performance criteria as well as creative process criteria. Implications and avenues for future research are discussed.

There is little denying the impact leaders have on our lives (Bass, 2008 and Yukl, 2006). Leaders possess both the capacity for outstanding achievement and at the same time, the faculty for deleterious harm (e.g., Bennett, 1976 and Hyde, 1971) – this is particularly true for outstanding leadership, or leaders that “have a disproportionate impact on the institutions in which we work and the broader world in which we live” (Mumford, Strange, & Bedell-Avers, 2006, p. 3). Not surprisingly then, there are few areas of organizational behavior as investigated as leadership – an area of research that continues to grow each year (Hunt, 1999 and Lowe and Gardner, 2000). Given the substantial influence leaders possess, it is not surprising that outstanding leadership has been examined by a number of researchers, using a variety of approaches. It is apparent, however, that the most frequently applied models fit into the charismatic or transformational leadership framework (Lowe & Gardner, 2000). Although variants of these approaches exist (e.g., Conger and Kanungo, 1998, Bass, 1985 and House, 1977), there is general consensus that these approaches fall under the rubric of vision-based, or broadly-framed charismatic leadership (Hunt, 1999, Schyns et al., 2007, Yukl, 1999a and Yukl, 1999b). In fact, most charismatic and transformational approaches can trace their roots to the early work of Weber (1921) who used the term “charismatic-authority,” suggesting that this style of management was based on “devotion to the specific and exceptional sanctity, heroism, or exemplary character of an individual or person.” (Miller, 1963, p. 10). Although the charismatic model has produced compelling results (e.g., de Hoogh, den Hartog, and Koopman, 2004 and Lowe et al., 1996) some researchers have noted that not all leaders operate in a traditionally charismatic fashion, yet are still able to achieve exceptional outcomes. In a case study of Benjamin Franklin, for example, Mumford & Van Doorn (2001) observed that Franklin did not employ a vision-based approach to leadership, but rather operated in a more pragmatic, problem-solving manner and in doing so instituted such influential policies as the introduction of paper currency and paved roads. Similarly, in a review of 11 outstanding companies, Collins (2001) found the selected business leaders to be characteristically pragmatic, rather than charismatic, in their leadership style. Along these lines, a few researchers have posited that the charismatic approach may actually be limiting in some circumstances, suggesting that pragmatic approaches may be more effective in situations characterized by uncertainty (Khurana, 2002 and Pasternack and O'Toole, 2002). In a discussion of 40 Fortune 500 firms, for example, Gary (2002) noted: “when the road ahead is unclear, vision can only take you so far” (p. 3). In light of the observation that not all outstanding leaders fit the charismatic mold, and that charisma or visionary leadership can be insufficient at times, Mumford and colleagues (e.g., Strange and Mumford, 2002 and Mumford, 2006) returned to Weber's (1921) original work to develop a more comprehensive approach to understanding outstanding leadership. Weber suggested that in addition to charismatic leadership, managers could also employ a “rational” approach that emphasized a pragmatic leadership style as well as a “traditional” or ideological approach, where the leader's influence rested in the belief in “the sanctity of immemorial traditions” (Miller, 1963, p. 10). Stated more directly, the extended leadership framework developed by Mumford and colleagues is comprised of three leadership styles: charismatic, ideological and pragmatic leaders, or the CIP model of leadership. Empirically, this extended leadership model has been investigated and validated in a wide range of studies examining behaviors such as creative problem-solving (Bedell, Hunter, & Mumford, 2008; Mumford, Bedell, Hunter, Espejo, & Boatman, 2006), communication strategies (Mumford et al., 2006), political tactics (Mumford et al., 2006), Leader–Member Exchanges (LMX) (Mumford et al., 2006), mental-model formation (Strange & Mumford, 2002), Machiavellianism (Bedell-Avers, Hunter, Angie, & Vert, 2006), and early-life experiences (Ligon, Hunter, & Mumford, 2008). The results of these studies have been compelling and indicate that although the leader types do indeed differ on a number of key variables, each type is uniquely capable of outstanding achievement (Mumford, Strange, & Hunter, 2006). It should also be noted that although the model was initially examined via the investigations of historically notable, outstanding leaders, recent evidence suggests that the model may also be applicable to more typical forms of leadership. For example, Bedell et al. (2008) examined the CIP model using a sample of undergraduate students engaged in a leadership problem-solving task. The results of the study were consistent with theoretical predictions derived from the outstanding leadership model, thereby lending some degree of support for extending the model to a broader range of leaders. Moreover, early studies that served as an impetus for pragmatic leader investigations examined more typical or day-to-day leadership (e.g., Connelly et al., 2000). This indicates that the CIP model has, at least in part, its roots in more typical forms of leadership (Mumford & Van Doorn, 2001). Barring a few notable experimental studies (e.g., Bedell et al., 2008) the primary method used to investigate the CIP model has been historiometric – where historical data was quantified and subsequently analyzed using traditional statistical techniques (Simonton, 1990). Although this method has a number of unique advantages with regard to the study of outstanding leadership (e.g., enables access to numerous high-level leaders), it also has a number of limitations. For example, the historiometric approach is limited in that many environmental and situational conditions may only be controlled for rather than investigated directly. Similarly, little is known regarding how environmental influences, such as complexity (Uhl-Bien, Marion, & McKelvey, 2008) may impact leader behaviors as they encounter crises or problems inherently characterized by high levels of ambiguity and change (Hunt, Boal, & Dodge, 1999). In sum, an extended model of outstanding leadership was developed in response to the observation that not all outstanding leaders operate in a charismatic fashion. Initial studies examining the CIP model have produced compelling results, strongly suggesting that there exist distinct yet equally viable leadership paths to outstanding achievement. Despite answering many questions, however, these studies and the methods applied present a number of new questions regarding charismatic, ideological and pragmatic leaders and the contexts they operate in. Thus, the aim of the present effort is to (1) experimentally investigate two important influences on leader behavior: situation framing and environmental complexity and more centrally, and (2) explore how these factors might interact to differentially influence the performance of charismatic, ideological, and pragmatic leaders. 1.1. An extended approach to outstanding leadership – The CIP model Although the typology put forth by Weber (1921) served well as a framework, it was necessary to consider from a process-perspective how the three leadership types actually influence their followers. Accordingly, the sensemaking portion of the model was developed and emerged from the observation that outstanding leaders often emerge during times of crisis – a point noted by a number of leadership scholars (e.g., Hunt et al., 1999, Mumford, 2006 and Rivera, 1994), as well as leaders themselves. John F. Kennedy, for example, noted that as a leader he was grateful for the crises he faced because they allowed him the opportunity for outstanding achievement (Perret, 2001). What is unique with regard to the CIP model of leadership, however, is the consideration of how these leaders respond to crises. Certainly, all outstanding leaders are compelled to offer some form of sensemaking to their followers; to provide an interpretation of the situation and offer direction and comfort during times of stress and ambiguity. In fact, sensemaking stands as an important influence mechanism for several reasons. Sensemaking activities allow for leaders to articulate their goals and paths to attaining those goals at a time when subordinates are most ready to receive direction (House, 1971). Moreover, sensemaking provides followers the perception of control, allowing them to face the crises in a functional manner as well as simultaneously providing motivation via increased perceptions of self-efficacy (Tierney & Farmer, 2002). Given the critical role that sensemaking plays in leadership (e.g., vision development, motivation) and the fact that all leaders do not influence followers in the same way, the extended model explores the underlying cognitive differences that would explain how leaders interpret and make sense of the situation. In other words, the CIP model is founded in the notion that charismatic, ideological, and pragmatic leaders think about crises differently and as such provide alternative, yet equally effective, forms of sensemaking to followers. Based, at least in part, on their early-life experiences, Strange & Mumford (2002) suggested that the three leader types form differing prescriptive mental models to help guide sensemaking efforts during ill-defined, ambiguous, crisis events. Generally speaking, mental models are cognitive representational systems that help individuals understand how a system operates as well as providing guidance as to how to function within that system (Johnson-Laird, 1983). The prescriptive mental models used by leaders, then, represent an idealized image of the current social system – they represent a situation as it could be rather than as it currently is (i.e., a descriptive mental model). These prescriptive mental models are typically derived from a reflection on and analysis of the current situation – an analysis that is often prompted by perceived deficiencies in the current system (Zhou & George, 2001). Specifically, Strange & Mumford (2002) suggested that prescriptive mental models are formed from consideration of: (a) goals that may be pursued, (b) causes of goal attainment and (c) other alternative models that may provide insight into the current system. These mental frameworks, moreover, are also significantly shaped by prior life experiences (Ligon et al., 2008). Once formed, these prescriptive mental models provide leaders with a foundation and guiding framework for helping subordinates cope with the stress, anxiety, and confusion associated with crisis events. Mumford (2006) suggested that although the mental models employed by the three leader types are used to for the same purpose (i.e., to guide sensemaking activities), they are unique and distinct from one another. Specifically, he suggested that these mental model differences may be summarized along seven key mental-model features: (a) timeframe, (b) type of experiences available, (c) nature of outcomes sought, (d) type of outcomes sought, (e) focus in model construction, (f) locus of causation and (g) controllability of causation. A summary of these differences, taken from Mumford (2006), may be seen in Table 1. Centrally tied to the present effort is a careful consideration of how these differences in mental-model formation might interact with the situation to influence leader performance. More explicitly stated, the aforementioned differences in prescriptive mental models have implications for when the three leader types are likely to succeed, or conversely, demonstrate performance decrements. With respect to the CIP, for example, charismatic leaders are theorized to employ a future-oriented timeframe within their respective mental models – emphasizing future goals and future goal attainment. Further, these future-oriented visions are typically formed in such a way as to indicate to the subordinate that following that vision will lead to a more positive, brighter future. Witness the “I have a dream” speech given so effectively by Martin Luther King, Jr. (Bennett, 1976), whereby the promise of a better future was used to provide comfort to followers in need of guidance and hope. Under conditions of high complexity, however, the future may be less clear thereby reducing the influence and impact of these leaders in such conditions (Kukalis, 1991 and Plumlee, 2003). Similarly, because charismatic leaders often pursue multiple goals, increases in complexity may make it difficult to determine or prioritize which goals to pursue. Stated another way, there is the potential for leaders with a multiple goals emphasis to become overwhelmed by excessive amounts of information when attempting to pursue multiple outcomes. Moreover, charismatic leaders are also vision-based and present their view of the future in a positive, almost utopian, light. Under many conditions, this vision may be highly appealing to followers and result in high-level achievement, a point well documented by previous research (e.g., de Groot et al., 2000, Judge and Piccolo, 2004 and Lowe et al., 1996). At times however, being bound to a personal vision may be limiting, resulting in an overly narrowed focus and ultimately reduced performance (e.g., Payne, 1973). Pragmatic leaders, on the other hand, are present-focused and problem-solving oriented. For example, in their review of Benjamin Franklin, a pragmatic leader, Mumford & Van Doorn (2001) observed that Franklin's outstanding achievements were obtained via an emphasis on problem-solving rather than vision formation. Moreover, the researchers observed that although Franklin engaged in high levels of intellectual stimulation personally, he did not attempt to motivate subordinates via the traditional transformational leadership method and instead relied on rational persuasion with an emphasis on solving problems. As such, it seems reasonable to suggest that pragmatic leaders are likely to prefer working on the specific task provided to them. Put another way, to the extent they are allowed to focus directly on problem-solving, they are likely to excel. Their general lack of affect and emotion in sensemaking suggests that they are likely to demonstrate a consistency in performance across a number of conditions, even those of high complexity (Bedell et al., 2008). This also indicates, however, that pragmatic leaders are not likely to develop the emotionally provocative visions evidenced by charismatic and ideological leaders – when, indeed, such visions may be desired by followers (Mumford, 2006). Finally, the mental models of ideological leaders are typically founded in “righting” the “wrongs” of previous events, resulting in an emphasis on the past rather than the future (Mumford, 2006 and Strange and Mumford, 2002). This emphasis on the past certainly has its advantages. For example, ideological leaders are often able to perform well under many complex conditions where it is difficult to predict what the future might hold – they are able to use previous events as an anchor to provide sensemaking and guidance to their followers. It is important to keep in mind, however, that under some conditions a singular, narrow focus on past events can have its drawbacks as well and may produce overly rigid thinking due to a lack of openness to emergent and new ideas (Gerring, 1997 and Mumford, 2006). Pasternack, Williams & Anderson (2001), for example, reference the case of Eastman Kodak where an over-emphasis on the past resulted in adverse and “deadly slow decision-making” (p. 7). Rigid thinking, moreover, may also result vis-à-vis an over-emphasis on ideological leaders' personal beliefs and values when engaging in sensemaking activities. For example, in their study of Machiavellianism, Bedell-Avers et al. (2006) found that ideological leaders were the most unwilling to compromise their personal values to attain organizational goals. This strict emphasis on ones' beliefs also results in an inward focus, which, at times, may compound their problematic “tunnel-vision” approach to solving a given problem (Mumford, Scott, & Hunter, 2006). In some instances, however, an emphasis on beliefs and values may help this leader type in situations where other leaders may become deterred. Witness Jane Adams' fight to establish the Hull House in an area of Chicago most would not willingly enter (Linn, 1935). Finally, this leader type also tends to have transcendent goals and may prefer broad tasks (e.g., a global mission) to those that are more narrowly focused (Mumford et al., 2006). 1.2. Situation framing and leader mental model In the selection of historical biographies for their recent investigation of the CIP leadership model, Mumford et al. (2006) sampled a wide range of leaders from varying domains including military, business, religious, and political arenas. The end result was an initial collection of more than 250 historical leader biographies. Examining the patters of leaders across these arenas, the researchers observed that certain leader types emerged more commonly from specific domains. For example, the majority of charismatic leaders appeared to emerge from the political arena – an observation not too surprising given that this arena often rewards the promise of a better future. Pragmatic leaders, on the other hand, were most often found in business settings, a setting where discrete problem-solving is often sought out. Finally, it was common for the researchers to witness ideological leaders emerging from social justice domains – domains emphasizing past wrongs and a strong commitment to ones' personal beliefs. Although multiple examples running contrary to this trend certainly exist (e.g., Ronald Reagan, an ideological politician), the general self-selection into, and emergence from, common domains resulted in a knowledge gap regarding how the leader types may perform under varying conditions (Mumford, 2006). More central to the present effort, we know little about how the three leader types may perform in situations best-suited, or framed, for another leader type. For example, the question remains as to how a charismatic leader may perform in a characteristically “ideological” situation (e.g., a situation calling for a commitment to ones' beliefs and values), or how a pragmatic leader might perform in a situation requiring the generation of a new, future-oriented vision (i.e., a charismatic-framed situation). Thus, a primary aim of the present effort was to examine meaningful conditions that may provide clues as to how these leaders might perform in varying circumstances. The role of the situation in understanding how leaders behave is hardly a new area of investigation. For example, early LPC models examined leader–member relations, task structure and positional power as key situational influences (Fiedler, 1978). Other models such as Path-Goal Theory (House, 1971) also focused on the situation, including aspects of the task as well as the subordinate. Some models have even gone so far as to suggest that the situation may be such that a leader is not even necessary (e.g., Kerr & Jermier, 1978) – albeit with questionable empirical support (e.g., Dionne, Yammarino, Atwater, & James, 2002). Finally, recent considerations of situational influences on leadership demonstrate that research continues to focus on the context in which leadership operates (e.g., Sternberg and Vroom, 2002 and Vroom and Jago, 2007). Thus, it is clear from the aggregate of these studies that the situation makes a difference in how leaders behave (Vroom & Jago, 2007). If it is granted that the situation is important in understanding leadership, the emergent question for the present effort becomes: What aspects of the situation, specifically, are critical to investigating charismatic, ideological and pragmatic leaders? The answer, it seems, is inherently tied to the leaders' respective mental models (Mumford, Marcy, Eubanks, & Murphy, in press). Based on earlier discussion, it is clear that mental models play a critical role in how the differing leader types behave. Accordingly it is also reasonable to suggest that how a situation is framed might impact leaders differentially. Specifically, given the pragmatic leaders' flexibility and malleability via their focus on solving a problem rather than causes, we expect relative consistency in task performance across conditions (Collins, 2001 and Pasternack and O'Toole, 2002). This malleability was witnessed in a historiometric study by Bedell-Avers et al. (2006), who found that pragmatic leaders were the least rigid and most capable of adjusting their behaviors to suit the situation. Ideological leaders, on the other hand, should perform well in conditions viewed as neutral or consistent with their core set of beliefs and values. Under more ideologically-focused conditions, however, ideological leaders may suffer as they try to reconcile their personal beliefs with those called for by the situation resulting in a decrease in performance. Gerring (1997) discusses the dogmatism characterizing ideological leaders – particularly when situations are ideologically charged. In their study of Machiavellianism, for example, Bedell-Avers et al., (2006) found that ideological leaders were the most rigid of the three leader types and were typically unwilling to adjust their behaviors to accommodate a given circumstance. Along these lines, Mumford et al. (2006) found that ideological leaders spent substantial time focusing on how to reconcile potential problem solutions with their core beliefs and values when solving ill-defined problems. Finally, given the charismatic leader's ability to form viable, compelling visions, we expect charismatic leaders to perform most strongly in the charismatic situations – or situations framed in such a way as to allow for the development of those visions. As a whole, the above suggests that we would expect to find an interaction between leader type and situation framing, leading to our first study hypothesis. A summary of study hypotheses and predicted mean trends, may also be seen in Table 2: Hypothesis 1. Situational framing will differentially impact the three leader types resulting in a two-way performance interaction between situation framing and leader type. 1.3. Complexity and leadership Examination of leadership research, broadly, would seem to indicate that the majority of studies implicitly assume leadership is a relatively static phenomenon occurring in a generally stationary environment. This implicit assumption has been criticized by a number of scholars (e.g., Hunt and Ropo, 2004 and Uhl-Bien et al., 2008) – criticism warranted with even a cursory consideration of the situations and contexts leaders face on a daily basis. Consider, for example, the constituencies leaders may interact with: lower-level subordinates, key lieutenants, customers, suppliers, other leaders, their superiors, competitors, or even other organizational leaders, all of whom may also interact with one another in some fashion (Mintzberg, 1973 and Yukl et al., 2002). If we also consider the complex cognitive tasks leaders must engage in, such as planning, forecasting and problem-solving, the contextual complexity increases nearly exponentially (e.g., Marta et al., 2005 and Mumford, Bedell, and Hunter, 2008). Furthermore, all of these interactions occur in an environment that can change very rapidly (Uhl-Bien et al., 2008). These environmental changes, moreover, will then impact any or all of the above issues leaders face in potentially reciprocal fashion. Thus, it may better be stated that leadership is a form of managed chaos, with leaders clearly operating in ambiguous, dynamic, rapidly-changing contexts (Hunt and Ropo, 2004, Marion and Uhl-Bien, 2001, Mumford et al., in press and Uhl-Bien et al., 2008). Not surprisingly, there has been a recent call for greater emphasis on the impact and role of complexity in understanding leadership (cf. Uhl-Bien et al., 2008). From the above discussion, it seems clear that contextual complexity is an important aspect in understanding leadership. What is less clear at this point, however, is how complexity explicitly relates to the CIP model of leadership. The answer, it seems, involves the consideration of the mental models leaders use to guide sensemaking efforts during crises. As complexity is increased, for example, elements such as timeframe and number of goals emphasized by the leader are likely to impact how a leader performs. In some instances, complexity may make performance more difficult and in others, impact performance negligibly, if at all. To illustrate, consider a charismatic leader in conditions of high complexity. Under highly complex conditions, an emphasis on vision formation and future goal attainment may prove problematic (Gary, 2002 and Khurana, 2002), resulting in a performance decrement. This point was explicitly noted by Pasternack & O'Toole (2002) in their discussion of Hewlett Packard and the decisions made by the future-oriented, charismatic CEO. Specifically, Pasternack and O'Toole noted that in this highly complex condition, the over-emphasis on charismatic behaviors, explicitly devoted to “reinventing the business in fundamental ways” (p. 6) resulted in significant loss for the organization. Pragmatic leaders, on the other hand, are less likely to be impacted by complexity given their emphasis on problem-solving versus future-oriented vision formation and would likely perform in a consistent fashion across differing conditions of complexity – a point observed in his review of 11 outstanding companies (Collins, 2001). Moreover, Bedell et al. (2008) found that pragmatic leaders performed the most consistently in their experimental study of problem-solving. Finally, ideological leaders are likely to perform well under conditions of complexity via their emphasis on a core set of beliefs – beliefs that help guide and filter information given to them (Strange & Mumford, 2002). As a whole, the above would seem to suggest that complexity will impact each type differently, leading to our second hypothesis: Hypothesis 2. Complexity will impact the three leader types differently, resulting in a two-way performance interaction between complexity and leader type. Along similar lines, it seems reasonable to assume that complexity will also interact with situation framing to produce differences in performance as well. Consider, for example, a situation in which a leader is instructed to focus on the future and asked to focus only on those elements external to the problem – conditions congruent with a charismatic-framed situation. Intuitively, it seems reasonable to assume that under conditions of low complexity, framing a situation in this manner may result in an increased focus and provide general direction about what elements of the problem are most important. Now consider what may happen as task or environmental complexity is increased. Research suggests that this discrete focus, induced via situation framing, may actually reduce the leaders' capacity to work with multiple causes and, in turn, hamper performance ( Bercovitz et al., 1997, Dosi and Lovallo, 1997 and Kahneman and Lovallo, 1993). In contrast, a pragmatic-framed situation given its emphasis on problem-solving as well as malleability with respect to method of problem-solution (Mumford et al., 2006) may help guide information processing in high-complexity conditions resulting in increased performance. Finally, the focus induced via the ideological situation is likely to produce consistency across complexity conditions. As a whole the above suggests that how a situation is framed may operate differentially with complexity to produce differing performance outcomes. Stated more directly, the above would seem to predict a complexity by situation framing interaction, leading to our third hypothesis. Hypothesis 3. The performance observed across the situation-framing conditions will vary as complexity is increased, resulting in a two-way interaction between situation framing and complexity. Finally, and most central to the present effort, we must consider the interactive effects of leader type, situation framing, and complexity. As noted earlier, it may be difficult for a charismatic leader to emphasize future-oriented goals in highly complex conditions (Gary, 2002, Kukalis, 1991, Pasternack and O'Toole, 2002 and Plumlee, 2003). Thus, we would expect charismatic leaders to demonstrate performance decrements in high complexity, charismatic-framed situations. In these highly complex conditions, however, charismatic leaders may be adequate problem-solvers if, indeed, asked to be (i.e., placed into a pragmatic-framed situation). Ideological leaders, on the other hand, may suffer from over-rigidity if placed into ideological-framed situations due to a resulting over-emphasis on past events, as well as unwillingness to compromise their core beliefs and values (Bedell-Avers et al., 2006). This performance decrement relationship, moreover, may be compounded as complexity is increased and they default intuitively to the fundamental characteristics of their respective mental model. Finally, pragmatic leaders, given their flexibility and malleability, are likely to perform well across most conditions with a slight preference for complexity given their partiality for engaging in complex tasks (Collins, 2001 and Mumford and Van Doorn, 2001). On the whole, the above suggests that the three leader types are likely to demonstrate performance differences when placed in varying complexity and situation-framing conditions. Hypothesis 4. The three leader types will perform differently when placed in varying complexity and situation-framing conditions resulting in a three-way interaction among leader type, complexity, and situation framing. 1.4. Creative problem-solving when facing crises Although it is useful, and indeed necessary, to consider the impact of the situation on relevant outcome criteria, it is just as critical to consider the processes leaders use to deal with problems (Hunter et al., 2007 and Mumford et al., in press). As noted earlier, outstanding leaders typically emerge during times of crisis (e.g., Rivera, 1994). As leaders face these crisis events, events characterized as ill-defined and complex, they must develop new and novel approaches to problem-solving (e.g., Reiter-Palmon & Illies, 2004). As such, thinking creatively appears to be a critical process of outstanding leadership (Mumford and Licuanan, 2004, Mumford et al., 2003 and Shalley and Gilson, 2004). This notion of creativity being critical to leader performance has been investigated in a number of studies. Some researchers have directly illustrated the impact of leadership on creative performance. For example, West et al. (2003) found that leadership, via inducing task clarity, was significantly related to innovation. Other studies have examined potential moderators and boundary conditions of leadership and creative performance. For example, Baer, Oldham, and Cummings (2003) explored the role of the task, finding that task-challenge served as a moderator of leadership and creative performance. Studies on creative climate have shown that leaders may be more or less effective under certain conditions (Shalley & Gilson, 2004). Finally, in an experimental study examining the creative thinking of leaders, Mumford et al. (2003) found that the creativity of the followers, as well as the actions of leaders, was critical to creative performance. Although additional examples exist, these studies appear to indicate that (a) creative problem-solving is critical to leadership and (b) that the situation plays an important role in understanding leader creativity. What remains open to question, then, is what aspects of the situation are most relevant to impacting the creative performance of leaders? Although there are a number of potential factors that may impact creative performance (e.g., rewards, creative climate), goal clarity appears to be among the strongest. For example, in a series of studies, Shalley, 1991 and Shalley, 1995 found that having clear, focused goals was positively related to creativity. Similar results were also found by Carson & Carson (1993) and West et al. (2003) who examined the impact of clarity on creative outcomes. Accordingly, we would expect that a situation framed in a manner that provides the greatest clarity would result in the highest levels of creativity. Upon examination of the theorized mental-model frameworks, it was determined that the ideologically framed situation provided the most clarity and focus. Specifically, due to the emphasis on fewer goals as well as a narrower, internal (versus external) focus we felt the ideological-framed situation would result in the greatest goal clarity. Thus, we predicted that this situation would result in the highest creativity. This prediction led to the fifth study hypothesis: Hypothesis 5. Situation framing will impact creative performance, with creative performance being highest in the ideologically framed situation. In addition to studies investigating leadership and creativity broadly, there exist two studies that may provide insight into the present effort, directly. The first, conducted by Mumford et al. (2006), employed the historiometric approach to analyze the creative problem-solving of 120 historical leaders. The researchers examined the differences among the eight creative processes known to be used in creative thinking (e.g., Mumford, Baughman, & Threlfall, 1996). Examining multiple problem-solving events at multiple points in the leaders' careers, the researchers found an interesting pattern of differences among charismatic, ideological and pragmatic leaders, with the exception of one notable “non-finding.” Specifically, the researchers failed to observe a significant difference among the leader types on overall creative achievement – an observation consistent with the fundamental tenets of the CIP leadership model (Mumford et al., 2006). Where the leaders did differ, however, were in the processes emphasized during creative problem-solving. Pragmatic leaders, for example, emphasized early stage processes, charismatic leaders emphasized idea generation and ideological leaders emphasized idea-evaluation with respect to their personal beliefs and values. The most critical aspect of predicting creative performance, however, was found in how well the leaders integrated ideas and potential solutions in relation to the complex demands placed on them by the external environment. In a second study experimentally examining differences in creative performance across charismatic, ideological and pragmatic leaders, Bedell et al. (2008), manipulated the type of problems given to the three leader types. Specifically, problem types varied as a function of being in a structured (school) or unstructured (social) domain. In addition, the researchers manipulated whether these leaders were, or were not, designated as leaders within the problem-scenarios provided. With respect to solution creativity, researchers found that pragmatic leaders would evidence typical responses across most conditions – a finding consistent with other researchers investigating pragmatic leadership (Collins, 2001, Gary, 2002 and Pasternack and O'Toole, 2002). Ideological and charismatic leaders, on the other hand, evidenced notable differences in creative performance across conditions. For example, ideological leaders generated solutions that were rated more creatively in the more structured, less complex condition, yet evidenced decreased performance when structure was decreased. Along these lines, we would expect to find a similar pattern of creative performance for ideological leaders under conditions of high and low complexity, respectively. Finally, charismatic leaders appeared to prefer less structured situations, producing solutions that were rated as more creative under such conditions, although only in socially-oriented domains. This observation is consistent with Yadav, Prabhu, and Chandy (2007) who found that leaders with a future-orientation were associated with increased innovation. As such, we would expect charismatic leaders to produce more creative responses in high-complexity situations. As a whole, these observations and those noted above indicate that situational complexity may interact with leader type, resulting in creative performance differences across conditions. These observations led to our sixth and final hypothesis. Hypothesis 6. Complexity will impact the creative performance of the three leader types resulting in a two-way interaction between complexity and leader type.

4.1. Limitations Before turning to the broader conclusions of the present study, it is important that a number of limitations first be addressed. To begin, the present study made use of a classic experimental approach and as such, is limited with regard to generalizing the findings to “real-world” situations. Although this approach was necessary to examine the study manipulations in controlled conditions, there are certainly a number of potential boundary conditions that may impact the generalizability of the findings that are unaccounted for. Such boundary conditions stand as important and viable areas for future research. Second, the nature of the study task allowed for only specific cognitive indicators of leader performance to be examined. As such, a number of important leadership behaviors could not be examined, including: subordinate reactions to sensemaking, leader–member exchange, inspirational motivation, intellectual stimulation, or other relationship-based influence and motivational tactics (e.g., Bass, 1985, Bass and Avolio, 1990 and Graen and Uhl-Bien, 1995). Although the study would have been richer with the inclusion of such relationship data, the focus on leader decision making and problem-solving was intentional, given that the vast majority of leadership studies fail to examine critical cognitive aspects of leader behavior (Hunter et al., 2007 and Mumford et al., in press). It must be noted, however, that the behaviors examined in this effort are limited by design and caution is warranted in drawing conclusions beyond these primarily cognitive indicators. Third, the situational-framing manipulation used in the study must be viewed with a degree of caution. Despite having 100% agreement amongst judges assessing the representativeness of the mental model framing, the use of a written scenario is an abstract representation of reality and is therefore a noteworthy limitation. Fourth, the leader measure employed in the study was ipsative in nature and is therefore only appropriate for categorizing leaders into preferred types (Baron, 1996 and Chan, 2003). More directly, total scores on the various leader types (e.g., a participant's degree of pragmatic leadership) could not be obtained. Given this, we concede that a non-ipsative, normative measure may provide a greater indication of these preferred leader styles and enhance results even further by providing potential interaction information. Moreover, the measure has not been examined in relation to any other measures of leadership (e.g., MLQ). That said, caution is warranted in viewing charismatic leaders in this study as empirically analogous to charismatic leaders in other studies using more common leadership measures. Similarly, a manipulation check was not employed for the pragmatic and ideological leaders; thus some degree of caution is warranted in considering the categorization results of the measure. Although the measure has demonstrated adequate validity, broadly, by means of the substantive emphasis on item selection procedures as well as interrater agreement during measure development, it is hoped that future studies will examine the links between these types of measures directly, adding to the overall validity of the CIP leadership measure. Fifth, caution must be used when drawing comparisons between previous conceptualizations of charismatic leadership (e.g., Conger & Kanungo, 1998) and charismatic leadership discussed with respect to the CIP model. Although both approaches trace their roots back to initial conceptualizations by Weber (1921) and therefore share most of their fundamental characteristics (e.g., both are vision-based, emotionally evocative, and future-goal oriented), the foundation of the CIP model in sensemaking and emphasis on mental-model development is likely to result in at least a few model differences. Moreover, until the empirical relationship between the two conceptualizations is established we must be cautious when making overly generalized comparisons between the approaches. Sixth, the use of undergraduate students in the present effort brings into question the liberal use of the term “leader” when describing their behaviors on a relatively controlled experimental task. Accordingly, it is open to question as to how applicable a model of “outstanding” leadership is to more mundane, typical leadership. Results in several studies, however, (e.g., Bedell, Hunter, and Mumford, 2008 and Ligon, Hunter, and Mumford, 2008) suggest that there is some stability in the leader pathways and that undergraduate students produce results consistent with hypothesized predictions as well as the original theoretical framework. With respect to the stability of pathways, Ligon et al. (2008) found that life experiences in early and late adolescence were predictive of later-life leader categorization. Thus, we feel that an experimentally-based undergraduate sample is a reasonable first step in exploring the leadership boundary conditions considered in the present study. That being stated, it is important to be cautious when generalizing from undergraduate samples and as such, this certainly stands as a study limitation. Finally, the statistical approaches applied (i.e., controlling for covariates) precluded specific cell-mean testing and as such, caution is warranted in interpreting differences among individual conditions. In this regard, cell-mean differences may only be viewed as exploratory. Although this limitation is regrettable, we believe there is still value in interpreting basic trends across conditions as they correspond to fundamental theory and extant findings. 4.2. General findings Even bearing these limitations in mind, we believe that the present effort makes a number of noteworthy contributions to the study of leadership broadly as well as to the CIP model of leadership directly. The first is that the use of a complex computer simulation appears to be a viable approach to examining leadership behaviors. The simulation used in the present effort allowed for a dynamic approach to leadership investigation, as well as the generation of objective performance criteria. Moreover, the experimental task enabled investigation of the cognitive processes leaders may engage in as they solve complex problems. The pairing of both objective and subjective performance criteria as well as the use of a relatively high fidelity, dynamic simulation allowed for relatively strong conclusions to be drawn with regard to the CIP model of leadership. The second, and arguably most important finding, is that the type of leader, the situation and the task complexity all interact to produce differing performance and process outcomes. This interaction was observed on two, relatively unrelated (r = .08), indices of game performance. Moreover, two-way interactions were also observed for strategic creativity on the task. On the whole, these results suggest, rather convincingly, that both complexity and situation framing must be considered when examining charismatic, ideological and pragmatic leaders. In fact, the lack of significant main effects observed for any of the study criteria indicates that the context may play a primary role in understanding performance differences among the leader types. The results of the study also provide clues to how the various leaders performed under differing situations and conditions of complexity. Although caution is warranted when interpreting these trends, they do provide general clues as to the circumstances in which leaders excel or demonstrate performance decrements. Specifically, cell means across conditions and criteria seem to indicate that charismatic leaders performed well in charismatic situations – but only in low complexity conditions. As complexity increased, charismatic leaders seemed to prefer the pragmatic situation to either the ideological or charismatic situation. This finding is consistent with Gary (2002), Kukalis (1991), Pasternack & O'Toole (2002) and Plumlee (2003) who suggested that vision formation may prove difficult in ambiguous conditions. At first glance, this pattern is somewhat contradictory with the findings of Bedell et al. (2008) that seemed to suggest charismatic leaders preferred less-structured conditions. It is important to bear in mind, however, that this was only the case in socially-oriented conditions – conditions best-suited for charismatic leaders (Mumford, Antes, Caughron, & Friedrich, 2008) Thus, it seems that the future-oriented focus of the charismatic leaders' mental model may lead to performance decrements on certain cognitive tasks, but may serve them well on more socially-oriented problems. Pragmatic leaders also produced a rather interesting pattern of results. Under conditions of low complexity, pragmatic leaders seemed to prefer the ideological situation. This finding may appear somewhat surprising, given the minimal interest in providing a vision to followers and general lack of emotion displayed by pragmatic leaders. The clues to interpreting this finding, however, may be in the focus induced by the ideological situation. In this situation, participants were asked to place an emphasis on the past, think more internally, and limit their attention to specific research-oriented tasks. It seems reasonable to believe that all of these aspects of the situation may contribute to a greater task focus. If we consider this in light of the leader's general preference for problem-solving, we see that pragmatic leaders may simply enjoy the liberation of being allowed to focus on a clear, finite objective. Along related lines was the finding that pragmatic leaders performed similarly across most conditions in high-complexity conditions. This finding is consistent with those in Bedell et al. (2008), where pragmatic leaders performed similarly across most study conditions. It seems that, as complexity increases, pragmatic leaders may pay less attention to the context and simply go about solving the problem. Of the three leader types, ideological leaders may have produced the most dynamic pattern of results, in some cases producing the lowest means observed, in others demonstrating the highest means observed. In terms of high performance, ideological leaders showed a preference for the pragmatic situation – but only in highly complex conditions. Thus, it appears that their relatively rigid beliefs serve them well in dealing with complexity, but only in situations where they are asked to be problem-solvers. With regard to lower performance, ideological leaders seemed to perform poorly in the ideological situation, overall. This finding is somewhat counter-intuitive, as one may predict that congruence between leader type and situation type (i.e., ideological leader in an ideological situation) would result in higher performance. When we consider the ideological leader's relatively narrowly framed mental model, however, it seems that the pairing may result in a form of “tunnel-vision” (Mumford, 2006). The results observed on the creativity criteria also indicate that this may be the case as ideological leaders were less creative as complexity increased, suggesting that they may turn to their mental model in times of stress and ambiguity, resulting in more narrowed thinking. It is important to bear in mind, however, the observed differences across game criteria, where ideological leaders seemed to perform more strongly in broader, rather than narrower tasks. Thus the decrements may only hold for certain tasks these leaders engage in. Hopefully, future research will explore this important finding more directly. Finally, the study results do offer insight into the importance of the situation, as well as the role of complexity in understanding leader behavior. With respect to the situation, it was interesting to note that the ideological situation produced more creative strategies overall. This finding is not wholly surprising given the sizable literature on goal-setting and creativity (e.g., Amabile and Gryskiewcz, 1987, Pinto and Prescott, 1988 and Shalley, 1995). It is interesting to note the focusing effects of the ideological mental model and how they may impact the various leader types. Pragmatic leaders, for example, seemed to prefer the ideological situation, whereas ideological leaders suffered from potentially compounding rigidity effects. With respect to complexity, it was interesting to note that a main effect was not observed in either the game performance criteria or the creativity criteria. The results of this study, particularly the observed interactions would seem to indicate that complexity stands as a powerful moderator, at times enhancing patterns that exists and, potentially more critical, resulting in reverse performance relationships. In either case, it appears important to consider the moderating role of complexity in studies of leadership, as the presence or lack thereof, may notably alter relevant leadership outcomes. 4.3. Implications The results of this study have a number of noteworthy theoretical and practical implications for understanding the CIP model of leadership. From a theoretical standpoint, results suggest that a greater emphasis should be made in considering the context these leaders operate in, lending credence to the recent call for an increased focus on complexity when investigating leadership (Hunt and Ropo, 2004, Marion and Uhl-Bien, 2001 and Uhl-Bien et al., 2008). Investigation of complexity as well as other boundary conditions, moreover, should be considered in light of the differing leader's mental models. It is clear that how leaders frame a given problem and make sense of an ambiguous, crisis situation impacts how they approach its solution. A greater emphasis, therefore, should be placed on understanding these relationships, and more importantly, investigating these processes from a substantive standpoint. Although the present study provides clues to these relationships, it is important that future research explore each aspect of the leader's mental model explicitly and directly. From a practical perspective, the results of the present effort would appear to speak to the importance of considering the situation when selecting leaders for, or placing leaders into given leadership roles. Consider, for example, the hiring of an individual for the position of Chief Executive Officer (CEO). If the organization operates in a highly turbulent, changing environment – environments often faced by entrepreneurs (Cogliser & Brigham, 2004) – selecting a charismatic leader may prove a poor choice if stakeholders are requesting a new, future-oriented vision. If this organization, however, were to select a pragmatic leader, we might expect to see consistent performance, independent of the environment. Finally, ideological leaders may prove highly effective if allowed to focus on problem-solving among a broad range of performance outcomes. Of course, such applications still stand as speculative at this point and future field-based studies are required before any substantial conclusions may be drawn. Still, the results seem to underscore the practical importance of context and environment in understanding effective leadership. In addition to offering insight into leadership hiring and placement, the present effort also speaks to the types of errors the differing leader types may make once in positions of influence. For example, it is possible that under complex crisis conditions, ideological leaders may inherently place too great of an emphasis on ideological concerns and not on problem-solving, particularly if the problems they face are laden with ideological issues and those around them are rallying around such issues. Pragmatic leaders, on the other hand, may operate too distantly from their followers in situations where emotional appeals would be welcomed, and indeed, necessary. Charismatic leaders, finally, show particular weakness in high-complexity situations where they are asked to focus on a new vision rather than more discrete problem-solving. The results of this study, then, would seem to highlight the importance of appropriately framing the situation relative to leaders' respective mental models. Previous studies indicate that key lieutenants, in particular, may play a critical role in this respect (Mumford et al., 2006). Although speculative, it seems reasonable that the leaders' close and trusted cadre may help them deal with the errors they are most likely to make by helping frame the situation in light of the respective leader's mental model. Thus, investigation of both errors and the role key lieutenants play in guiding and limiting those errors stand as critical and exciting areas of future research. In sum, the results of the present effort have demonstrated unique differences among the three leader types. Although these leaders did not perform differently on the various outcome criteria in general, they did demonstrate interesting differences when the situational context was investigated. It is our hope that future efforts will continue to explore the impact of these contexts – particularly with respect to additional performance and process criteria.