Industrial Marketing Management 35 (2006) 210 224

Learning process in new product development teams and effects on product success: A socio-cognitive perspective
Ali E. Akgu na, Gary S. Lynnb,*, Cengiz YNlmazc
b a Gebze Institute of Technology, Science and Technology Studies, Istanbul Cad. No. 101, 41400, Gebze, Kocaeli, Turkey Stevens Institute of Technology, Wesley J. Howe School of Technology Management, Castle Point on Hudson, Hoboken, NJ 07030, USA c Bogazici University, Department of Business Administration, 80815 Bebek, Istanbul, Turkey

Received 12 August 2003; received in revised form 25 November 2004; accepted 6 February 2005 Available online 4 June 2005

Abstract The study purports to develop and empirically test a model of team learning process and its effects on team performance in new product development teams. Using the socio-cognitive theory of learning in groups and organizations, several hypotheses were tested to show that the primer components of social cognition (that is, information acquisition, information dissemination, information implementation, unlearning, thinking, intelligence, improvisation, sense-making, and memory) form an interactive process model of the team learning phenomenon. By studying 165 new product development projects, it was shown: (i) that the eight primer socio-cognitive factors of information acquisition, information dissemination, information implementation, memory, thinking, improvisation, unlearning, and sense-making constitute interrelated sub-components of a higher-order team information-processing construct; (ii) that team intelligence is positively related to components of team information-processing; and (iii) that information-processing facilitates new product success primarily through the positive effects of superior information implementation. Theoretical and managerial implications of the study findings are discussed. D 2005 Elsevier Inc. All rights reserved.
Keywords: New product development; Organizational learning; Team learning; Social cognition

1. Introduction A majority of successful innovations is developed through the collective efforts of individuals in new product development teams (hereafter, NPD teams). NPD teams are organizational workgroups where individuals from diverse personal and organizational backgrounds come together for a limited (and usually, predetermined) time period and work in close collaboration towards creating, designing, developing, and marketing a new product (Pinto, 2002). The ultimate objective of all NPD teams is superior marketplace success of the new product. And, as widely noted in the relevant academic and popular literature (e.g., Grant, 1996; Leonard-Barton, 1992; Moorman & Miner, 1997; Nonaka &
* Corresponding author. Tel.: +1 201 216 8028; fax: +1 201 216 5385. E-mail addresses: aakgun@gyte.edu.tr (A.E. Akgu n), glynn@stevens.edu (G.S. Lynn), yilmazcengiz@gyte.edu.tr (C. YNlmaz). 0019-8501/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.indmarman.2005.02.005

Takeuchi, 1995), one of the major success factors in NPD teams is that the knowledge acquired by individuals within the team transcends beyond the individual mind and becomes a collective entity that facilitates the mission of the team (Madhavan & Grover, 1998; Moorman, 1995). Understanding and explaining the processes and procedures through which knowledge is generated, shared, disseminated, and utilized (that is, learned ) in NPD teams is therefore critical for understanding new product success . While considerable research has focused on the learning phenomenon in NPD teams (Meyers & Wilemon, 1989; Purser, Pasmore, & Tenkasi, 1992), particularly on the dynamics of team information-processing (Lynn, Reilly, & Akgu n, 2000; Moorman, 1995; Moorman & Miner, 1997, 1998), the current state of knowledge about the learning phenomenon in NPD teams needs to be expanded to include a greater understanding of the process of learning and its effects on project outcomes .

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

211

One way to expand the scholarship on NPD team learning is by cross-fertilization with different streams of research in this area (Akgu n, Lynn, & Byrne, 2003; Easterby-Smith, 1997). The organizational learning literature, for instance, offers a rich basis for identifying and understanding the key factors and mechanisms through which learning occurs in groups (Dixon, 1992; Huber, 1991). Similarly, studies on group psychology, small group researches, and social psychology emphasize the importance of social cognition for group and organizational learning process (Akgu n et al., 2003; Gruenfeld & Hollingshead, 1993; Higgins, 2000). Because the NPD process in most organizations is in essence a group activity that involves people throughout the organization and their reciprocal interactions, knowledge, behaviors, cognitions, and functional cultures (Madhavan & Grover, 1998), the integration of the social cognition literature into the theoretical articulations of learning in NPD teams might provide a better elucidation of the learning processes and practices as indicated by the future research suggestions of Akgu n et al. (2003) and Akgu n, Lynn, and Reilly (2002). Accordingly, the present study (i) incorporates the organizational learning theory and the socio-cognitive framework to develop a model of the NPD team learning process and (ii) operationalizes and empirically tests, in a nomological network, the effects of factors from both streams of research on new product success. In doing so, this article aims to provide managers with applicable guidance on how to facilitate the learning process and new product success in NPD teams.

cognitive representation of organizations and groups in general and organizational learning process in particular (see, Akgu n et al., 2003). Social cognition is an approach to the understanding of human social behavior, involving the investigation of the mental processes of people interacting with one another (Martin & Clark, 1990). Specifically, social cognition examines aspects of human information-processing; what it influences, and what it is influenced by, within the complex social interactions of groups (Gioia & Sims, 1986).1 As such, social cognition transcends beyond the cognitive approach (which emphasizes individual cognitive processes) and the structural approach (which, as in the organizational learning literature, emphasizes organizational and group routines; Gherardi, 1998; Schwarz, 1995), and amalgamates different views on the process of learning (such as behaviorism, cognition, and social construction; Akgu n et al., 2003). In this vein, as indicated by Akgu n et al. (2003), the sociocognitive perspective of organizational learning removes the learning-level dichotomy and constitutes a more grounded approach for understanding the learning process via reciprocal relations of both cognitive processes and social constructs. According to Akgu n et al. (2003), the sociocognitive view of organizational learning can be conceptualized as a higher-order construct (process), construed of cognitive processes and social constructs which are (1) distributed through the organization, (2) unfold overtime, (3) involve people in diverse functions and mind-sets, and (4) embedded in routines and institutional structures by means of organizational culture. Exemplars for such distinct but correlated features (sub-processes) include: & information/knowledge acquisitiongathering of data from various sources, including customers, competitors, economic assessments, financial statements, social reports, consultants, new members, acquisitions and mergers, and so on; & information/knowledge implementationuncovering and correcting product-related problems by collective utilization of relevant information; & information/knowledge disseminationdistributing and sharing information/knowledge in groups and organizations through a variety of means, including formal communication means (e.g. memos, reports, bulletin boards, and face-to-face meetings) as well as informal communication means (e.g., coffee-breaks, water-cooler discussions, hallway meetings, and so on); & memorystorage of skills and experiences of team members; & unlearningeliminating or changing how memory is manifested in organizations/teamssuch as changing team beliefs, norms, and values; & thinkinga process of decision-making, judgment, and
1 Different definitions of social-cognition can be seen in Ostrom (1984), and Akgu n et al. (2003).

2. Background 2.1. Social cognition and learning in teams Most of the literature on group and organizational learning has addressed concepts associated with cognition (Cook & Yanow, 1993). As a generic term, cognition is defined as the mechanism of information-processing (Reed, 1982) or data processing and storing in the human nervous system (Anderson & Ausubel, 1965). At the individual level, many sub-processes are required to explain the cognitive perspective of human informationprocessing, including the acquiring, forming, storing, manipulating, discarding, and implementing of information (Manis, 1966). Nonetheless, while a similar view of cognition has been discussed in higher-level social groups such as teams and organizations (Hedberg, 1981; Popper & Lipshitz, 1998), the insights and applications of individually oriented cognition are generally deemed less useful in understanding group and organizational learning due to differences in underlying assumptions and ontology (Cook & Yanow, 1993). Accordingly, scholars (e.g., Allard-Poesi, 1998; Gioia & Sims, 1986) have suggested that the sociocognitive perspective be used to explain the collective

212

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

creativity that includes: (i) problem identification (such as market analyses, customer needs and wants, technical assessment of the new product, and team member selection); (ii) decision-making (such as determining milestones and Finchstones_ during the development process, and budgeting and funding the project); and (iii) judgment (such as assessing test results, advantages and disadvantages of decisions, and selecting a process of product development); & intelligencethe cognitive ability and capability of the team to process, interpret, manupulate and use information; & improvisationsimultaneous planning and implementing of an action; and & sense-makingconstructing, filtering, organizing and framing of information in a meaningful way. Prior research has generally conceptualized the above constructs and their reciprocal interactions at the more general organizational level. In addition, aside from the research reported in Akgu n et al. (2002), which examines the dimensionality of the above-listed sociocognitive constructs in NPD projects (without focusing on the interactive nature of the processes in the team learning process and their impacts on new product success), empirical works regarding the joint roles of these constructs in the learning processes are rare. Given that the NPD context provides a more controllable, identifiable, and accurate unit definability for learning
Cognitive capability

processes (Akgu n et al., 2003; Meyers & Wilemon, 1989), which further indicates that the learning processes are more likely to be within the control of project management, the basic postulate of research is that each one of these constructs is critical for the team learning process in a nomological web and, thereby, for new product success.

3. Joint effects of learning processes on new product success The purpose of this section is to explore the joint and relative effects of the dimensions of social cognition in NPD teams on new product success. The theoretical model, depicted in Fig. 1, is based on the view that the nine factors (described above) can form an interactive, process model of social cognition that enables researchers and managers to better understand the team learning process and its role in developing successful new products. Following Akgu n et al. (2003), due to the complex and reciprocal nature of the relationships across the constructs of interest, the discussion of the hypothesized model is divided into three sub-sections for more coherent presentation. As shown in Fig. 1, the model suggests that new product success in NPD teams is a function of team information-processing, which involves a number of critical socio-cognitive constructs including information acquisition, information dissemination, inforProject success

Information-processing

Information acquisition

Information implementation

Information dissemination

Unlearning Team intelligence Thinking NPS

Improvisation

Sense-making

Memory

Covariances
Fig. 1. Proposed model.

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

213

mation implementation, memory, sense-making, improvisation, thinking, and unlearning. Each socio-cognitive dimension of information-processing is in turn theorized to be (i) interrelated reciprocally with others and (ii) influenced by the teams cognitive capability, that is, team intelligence. 3.1. Information-processing Information-processing refers to how organizations and groups perceive, interpret, store, retrieve, transmit and use information (Akgu n et al., 2003; Huber, 1991; Kiesler & Sproull, 1982). A similar conceptualization of information-processing has been adopted in the NPD literature (Adams, Day, & Dougherty, 1998; Lynn et al., 2000). However, studies on information-processing in the product development literature have narrowly focused on team information acquisition, implementation, interpretation, and memory (Adams et al., 1998; Moorman, 1995), as well as on how they affect the product development process and its outcomes (Lynn et al., 2000; Moorman, 1995). Further, information-processing constructs have generally been considered as constituting the conscious mode of processing. By contrast, works on the sociocognitive perspective of organizational learning have incorporated both conscious processing and motor or automatic processing (Akgu n et al., 2003). This latter view suggests that, in addition to the conscious mode of information-processing involving information acquisition, information dissemination, information implementation, thinking (deliberation and decision-making), sense-making, unlearning2, and information and knowledge storage (memory) the unconscious modes of information-processing, such as improvisation (Akgu n et al., 2003; Crossan, Lane, White, & Leo, 1996; Weick, 1993), are further needed to capture what constitutes the team learning process. In addition, the socio-cognitive view of learning process points out that the aforementioned information-processing constructs are reciprocally interrelated such that each acts both as a causal antecedent and a consequence of the others (Akgu n et al., 2003; Fiske & Taylor, 1991; Larson & Christensen, 1993). For instance, in an NPD project, team information acquisition impacts team memory because teams increase their skills, experience, absorbing capability, and/or memory by the accumulation of information; and team memory in turn effects team information acquisition because how teams receive information depends on previous knowledge and

skills. Team memory may also influence team sensemaking by determining how team members organize and filter information. This sense-making process, in turn, defines a framework for team information acquisition. These reciprocal relationships demonstrate the multidimensional nature of information-processing in teams (Akgu n et al., 2003). Consequently, and consistent with Akgu n et al. (2003)3, it is posited that substantively and statistically significant covariances exist between each pair of the information-processing factors: H1. Information-processing in new product development teams is best conceived as a multidimensional structure with eight correlated but distinct constructs: information acquisition, information dissemination, information implementation, unlearning, thinking, improvisation, sense-making, and memory. 3.2. Cognitive capability Using the concept of team intelligence as a surrogate for the teams cognitive capability, it is further posited that team intelligence acts as a driving factor for each informationprocessing construct. Note that this approach draws upon the literatures on epistemology (Goldman, 1986), cognitive and social psychology (Fiske & Taylor, 1991; Hunter, 1986), as well as management science (Glynn, 1996). For instance, Glynn (1996) and McMaster (1996) assert that intelligence helps organizations to obtain input from the environment, generate information, interpret information, discard information, and translate the resulting interpretations into action. Team intelligence is the functional intelligence of a group of people working as a unit, and relates to the teams capability and ability to process, interpret, manipulate and use information (Akgu n et al., 2002). More specifically, team intelligence is the ability and capability of the team to understand the changing market and technical dynamics, to articulate a clear vision for the project, to figure out an effective development process for the product/service, to gather information and knowledge regarding the market, technology, environment, industry and process, and to process and disseminate that information within and between teamsshortly, information-processing. In this

2 Although social cognition and cognitive psychology have extensively investigated knowledge elimination, and beliefs and attitude changesunlearning, as factors in information-processing (Crocker, Fiske, & Taylor, 1984), product-development studies have a lack of empirical works on unlearning in product-development teams.

3 At the organizational level, organizational culture links the sociocognitive constructs and, thereby, mediates the organizational learning processes. Especially, organizational culture reflects the covariance among the socio-cognitive factors-linking cognitive processes and social constructs; and reduces the confusion at the micro and macro levels of cognition (that is, individual, group, and organizational). However, at the project level, project visioning/goals, teamwork and trust reflects the covariance among the information processing factors and reduces the confusion at the micro and macro levels of cognition. Nevertheless, since visioning, trust and teamwork are embedded in NPD projects by their natures, team learning process model implicitly presented here covers them.

214

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

perspective, it is expected that the effect of team intelligence on new product success is mediated, fully, by each and every one of the information-processing factors. Therefore, it is hypothesized that: H2. Team intelligence is positively related to each component of team information-processing. 3.3. Project success After demonstrating the interwoven relations of sociocognitive constructs in NPD teams, the question of how these interactions affect performance (e.g. new product success, NPS) should be investigated from a pragmatic perspective. Note that there exist conflicting views on the effect of learning on performance in the organizational learning literature, probably due to varying levels of analyses and conceptualizations of the learning process (construct; Akgu n et al., 2003). A majority of these works has studied the learning process at an excessively broad unit of analysis (e.g., large organizations), or used a very narrow operational definition of the learning phenomenon (i.e., explained the learning process by a few elements of cognition, such as knowledge accumulation or utilization, and ignored the aggregated effects of other cognitive elements; Akgu n et al., 2003; Crossan, Lane, White, & Djurfeldt, 1995). On the other hand, whereas theoretical and empirical studies in the NPD literature indicate unequivocally that learning has a positive effect on project outcomes, the effects of sociocognitive processes (specifically information-processing constructs) on new product performance have been investigated in an isolated and linear manner in the NPD literature. For instance, Moorman (1995), in studying 92 product development projects, found that sense-making is positively associated with new product performance. Likewise, Moorman and Miner (1997) demonstrated that superior organizational memory positively affects creativity and short-term financial performance of new products. These authors also found in a subsequent study that, under highly turbulent environmental conditions, team improvisation positively influences product technical performance effectiveness (Moorman & Miner, 1998). By studying 281 new product development projects, Lynn et al. (2000) showed that information acquisition and information implementation positively affect NPS, and Cooper (1993) argued that information dissemination is the cornerstone of new product success. Based on these findings, it is posited that how teams handle and use information effectively is imperative for NPS and, therefore, each dimension of information-processing, within a nomological web , will facilitate the teams ability to develop new products with superior marketplace performance. H3. Each component of information-processing in new product development teams is positively related to new product success.

4. Research method 4.1. Measure development To test the above hypotheses, a questionnaire developed by Akgu n et al. (2002) was used. All constructs were measured using 11-point Likert scales ranging from Fstrongly disagree_ (= 0) to Fstrongly agree_ (= 10). For information dissemination, both informal and formal communication modes were used. The Finformal mode_ of information dissemination was measured using three items, whereas Fformal_ aspects of the communication processes were measured using two items. An exploratory factor analysis using varimax rotation was conducted to assess the dimensionality of this scale. As expected, a two-factor solution in which formal and informal communication items loaded on separate factors was obtained. Items loading on each dimension were then averaged to form two different composite indices representing formal and informal information dissemination for use in further analyses. For new product success (NPS), nine items developed by Cooper and Kleinschmidt (1987) were used. Each item captured a different indicator of new product success, such as whether teams met or exceeded sales, profit, managerial, and technical expectations. This operationalization was similar to Hackmans (1987) performance measure which also focused on meeting or exceeding the performance standards of people who receive and/or review the teams output (p. 323). Consistent with how such formative scales are analyzed, the nine items were combined to form a composite index of new product success (NPS). Following Lastovicka and Thamodaran (1991), in order to form a composite NPS score, factor scores resulting from an exploratory factor analysis of the scale (rather than using an unweighted linear composite of item scores) were used. A complete list of measurement items, which also includes those eliminated during the measurement-purification process, is provided in the Appendix. Descriptions of the operational logic used for the measurement of each construct are presented in Akgu n et al. (2002). 4.2. Sampling After designing and refining the questionnaire, contact persons from a variety of technology-based companies in the north-east of the USA were selected. The criterion for the selection of these contact persons was that the products launched by the firms of and associated with these people had to be commercialized at least 6 months prior to the data collection. Next, each contact person was asked to identify for each recently developed new product in his or her company a project manager, or a senior team member, or a department manager, or a director, who was directly involved in and highly knowledgeable about the NPD project. Because product or project managers are likely to have a Fbigger picture view_ of NPD projects than other team members, they

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224 Table 1 Sample statistics Industry Telecommunication Military/defense Software Computer/electronics Aerospace Manufacturing/machinery Communication Chemical Food Pharmaceutical Others (IT, oil, utility, and biotechnology) Total Number of NPD teams 37 33 14 13 12 12 12 8 8 7 9 165

215

were likely to provide more reliable and objective data. After the respondents had been selected, each was informed that his or her responses would remain anonymous and would not be linked to the company or its products. Note that the sampling procedure used in this study is similar to those used in many other studies on innovation (e.g., Ettlie & Rubenstein, 1987; Larson & Gobeli, 1988; Rochford & Rudelius, 1992; Thamhain, 1990). Of the 335 people asked to participate, 212 agreed and completed the questionnaire (a 63% response rate). However, of these 212 completed questionnaires, only 165 of them were used, after discarding some with missing data in one or more variables. Thus, the analyzable sample consisted of 165 NPD projects (each project representing a cross-functional new product development team) from 56 different companies. Table 1 depicts the primary industries that the firms in the sample operate in. 4.3. Measure purification and validation The procedures included assessments of item and scale reliability, unidimensionality, and convergent and discrim-

inant validity were used to validate measures. First, a series of exploratory factor analyses was conducted. A single factor was extracted for each multiple-item reflective scale in these analyses, using an eigenvalue of 1 as the cut-off point (which indicates that the measurement scales used in this study were unidimensional). Next, overall measurement quality was assessed using confirmatory factor analyses (Anderson & Gerbing, 1988). Maximum-likelihood LISREL 8.3 was used in the estimation and respecification of the confirmatory measurement model, and the sample covariance matrix was used as input. The initial measurement model included: (i) 24 observable variables (reflecting eight latent factors); (ii) the two composite scores of formal and informal communication processes (used as indicators of the information-dissemination factor); and (iii) the single-indicant new product success factor. However, this model did not display an adequate fit to the observed data. Specifically, three items from the unlearning scale and two items from the sensemaking scale tended to cross-load on other factors. An examination of these items revealed that deleting them would not deteriorate the content validity of the unlearning and sense-making scales (see the Appendix). Therefore, all of the five problematic items were eliminated in a step-bystep procedure. After the elimination of these five items, the resulting measurement model was found to fit the data reasonably well: v 2 (280) = 439.9, comparative fit index (CFI) = 0.90, incremental fit index (IFI) = 0.91, non-normed fit index (NNFI) = 0.88, and root-mean-square error of approximation (RMSEA) = 0.05. In addition, all items loaded significantly on their respective constructs (with the lowest t value being 2.50), providing support for convergent validity. As a check for discriminant validity, the variance extracted for each construct was greater than the squared latent factor correlations between pairs of constructs (see Table 2; Fornell & Larcker, 1981). Discriminant validity was also obtained using chi-square difference tests for all pairs of constructs

Table 2 Descriptive statistics for the scales, construct correlations, and reliability estimates Mean 6.24 5.78 6.94 5.15 6.91 5.55 5.27 6.21 5.36 6.01 S.D. 2.13 2.46 1.68 2.60 1.94 2.58 2.25 2.27 2.45 2.28 1 2 3 4 5 6 7 8 9 10 Information acquisition Information implementation Information dissemination Unlearning Thinking Improvisation Sense-making Memory NPS Team intelligence Variance extracted Composite reliability Coefficient alpha 1 0.78*** 0.50** 0.08 0.64*** 0.30** 0.66*** 0.45** 0.44** 0.63*** 0.55 0.79 0.70 0.45** 0.08 0.57*** 0.36*** 0.54** 0.44** 0.54*** 0.57*** 0.74 0.89 0.82 0.03 0.67*** 0.03 0.62*** 0.48** 0.24* 0.31** N.A. N.A. N.A. 2 3 4 5 6 7 8 9 10

0.04 0.35** 0.05 0.19* 0.08 0.12* 0.50 0.74 0.74 0.06 0.70*** 0.38** 0.35** 0.53*** 0.68 0.82 0.78 0.06 0.10* 0.16* 0.19* 0.75 0.89 0.85 0.55*** 0.44** 0.61*** 0.52 0.75 0.70 0.35** 0.39** 0.68 0.86 0.78

0.31** N.A. N.A N.A.

0.64 0.83 0.76

N.A.not available because the construct is measured with a formative scale or has less than three indicants. * p < 0.1. ** p < 0.05. *** p < 0.01.

216

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

(that is, correlations between all pairs of constructs were shown to be significantly below unity). The reliabilities of the multiple-item, reflective measures are reported in Table 2, along with construct correlations and descriptive statistics for the scales. All reliability estimates including coefficient alphas, average variance extracted for each construct, and LISREL-based composite reliabilities are well-beyond the threshold levels suggested by Nunnally (1978) and Fornell and Larcker (1981). It is therefore concluded that the measures are unidimensional and have adequate reliability, discriminant validity, and convergent validity. These find-

ings are similar to the findings of Akgu n et al. (2002), albeit using a different project dataset.

5. Analyses and results The model depicted in Fig. 1 was used to test the hypothesized relationships. The basic postulates of this model were that team learning: (i) is a multidimensional construct construed of several dimensions of team information-processing; and (ii) that it mediates the effect of team

Table 3 Parameter estimates Hypotheses H2 Path Team intelligenceYinformation acquisition Team intelligenceYinformation implementation Team intelligenceYinformation dissemination Team intelligence Yunlearning Team intelligenceYthinking Team intelligenceYimprovisation Team intelligenceYsense-making Team intelligenceYmemory Information acquisition YNPS Information implementationYNPS Information dissemination YNPS Unlearning YNPS Thinking YNPS Improvisation YNPS Sense-making YNPS Memory YNPS Information acquisition6information implementation Information acquisition6information dissemination Information acquisition 6unlearning Information acquisition6thinking Information acquisition6improvisation Information acquisition6sense-making Information acquisition6memory Information implementation6information dissemination Information implementation 6unlearning Information implementation6thinking Information implementation6improvisation Information implementation6sense-making Information implementation6memory Information dissemination 6unlearning Information dissemination6thinking Information dissemination 6improvisation Information dissemination6sense-making Information dissemination6memory Unlearning 6thinking Unlearning6improvisation Unlearning 6sense-making Unlearning6memory Thinking 6improvisation Thinking6sense-making Thinking6memory Improvisation 6sense-making Improvisation6memory Sense-making6memory Estimate 0.63 0.56 0.41 0.11 0.53 0.19 0.61 0.39 0.09 0.50 0.15 0.06 0.01 0.02 0.25 0.11 0.43 0.25 0.15 0.30 0.19 0.28 0.21 0.23 0.02 0.27 0.26 0.20 0.22 0.02 0.46 0.10 0.37 0.33 0.02 0.33 0.02 0.23 0.04 0.38 0.18 0.05 0.18 0.32 t -value 5.87 5.56 2.76 1.23 5.39 2.02 5.16 3.92 0.39 2.51 0.81 0.64 0.05 0.25 1.32 0.92 4.37 1.99 1.56 3.28 2.08 2.89 2.27 1.97 0.24 3.13 2.94 2.27 2.53 0.20 3.03 0.91 2.61 2.41 0.25 2.20 0.28 1.96 0.49 3.79 2.01 0.63 5.06 3.19

H3

H1

Nonsignificants at p < 0.05 level are in italics. v 2 = 440.91, df = 280, NNFI = 0.88, CFI = 0.90, IFI = 0.91, RMSEA= 0.05.

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

217

intelligence on new product success. Given that the eight information-processing constructs (that is, information acquisition, information implementation, information dissemination, unlearning, thinking, improvisation, sensemaking, and memory) have been hypothesized to constitute interrelated facets of a higher-order information-processing construct, the parameters representing the covariances across these constructs were allowed to be free during the estimation of the model. The above procedure was used, rather than incorporating a second-order information-processing into the hypothesized model, in order to (i) reveal the distinct relationships between team intelligence and each facet of information-processing, and (ii) explore the relative effect of each facet on new product success.4 The resulting fit indices indicated that the hypothesized model fits well to the observed data: v 2 (281) = 440.91; CFI = 0.90; IFI = 0.91; NNFI = 0.88; RMSEA= 0.05. Standardized parameter estimates were provided in Table 3. As shown in Table 3, a large number of the intercorrelations across dimensions of information-processing are positive and significant. Among the few nonsignificant parameter estimates are those representing the pairwise relationships between unlearning and: (i) thinking; (ii) sense-making; (iii) information acquisition; (iv) information implementation; and (v) information dissemination. In a similar vein, improvisation is found to be unrelated to: (i) thinking; (ii) sense-making; and (iii) information dissemination. Thus, H1 was partially supported. Concerning the directional effects of team intelligence on dimensions of team learning, the findings are largely in accordance with H2. Team intelligence is found to exert positive and significant effects on each and every facet of information-processing, except for unlearning. The strongest effect of team intelligence seems to be on information acquisition, followed in order by sense-making, thinking, information implementation, information dissemination, memory, and improvisation. Hence, H2 was partially supported. Finally, concerning the effects of information-processing constructs on new product success, only information implementation is found to exert a significant direct effect on new product success, thus largely failing to support H3. The parameter estimates for the direct paths linking the remaining information-processing factors with NPS are all nonsignificant. These estimates, however, represent the partial effects of each construct on NPSthat is, when all others are controlled for. A proper interpretation of these

results therefore suggests that information implementation acts as a key factor, mediating the effects of the other information-processing facets on NPS.

6. Discussion and implications The results of the study reveal several insights regarding the team learning process. First, the multidimensional structure of team learning process has long been a controversial issue in academia. The analyses of the interrelationships across facets of team learning process make a significant contribution to the current state of knowledge in this area. Note that, with regard to H1, the findings (showing significant bi-directional relationships across the information-processing factors) are consistent with most of the previous studies in the marketing and NPD literatures (see, Lynn et al., 2000; Moorman, 1995; Moorman & Miner, 1997, 1998; Sinkula, 1994 for more detail). Theoretical and managerial of implications of these observed bi-directional relationships across the information-processing constructs are discussed in further detail in the following paragraphs. Second, the relationships between team intelligence and the dimensions of team informationprocessing, as well as the relative and distinct effects of each dimension on NPS in new product development teams, are critical issues, particularly from a managerial perspective. Therefore, a major portion of the discussion focuses on the specific mechanisms through which team informationprocessing mediates the effect of team intelligence on NPS. For instance, findings indicate that the information acquisition construct (sub-process) is significantly related to team memory, thinking, sense-making, improvisation, as well as information dissemination and implementation. These results show that when an NPD team finds out technical, manufacturing, and/or marketing shortcomings of the product, i.e., a need for information acquisition, it is capable of building the necessary knowledge structure and skills (i.e., team memory). Specifically, team memory development is self-reinforced when new information is acquired and product-related problems are discovered. At the same time, team knowledge and skills may affect the teams ability to acquire information and/or to discover product-related problems. Further, information acquisition and team thinking reciprocally facilitate each other. A NPD team can generate alternatives and solutions to technical/ market/product problems by collecting data from different sources. In line with these findings, Garvin (1993) indicates that data collection is one of the critical steps towards solving new product problems in many team projects. Specifically, when a team discovers product-related problems, it can consider the advantages and disadvantages of many alternative solutions for an appropriate course of action, demonstrate curiosity about customer needs and wants, and elaborate on technical problems (for example, product performance shortcomings, cost overruns, and so

4 A second-order confirmatory factor analysis of a model depicting the eight interrelated facets as first-order factors loading onto a single information-processing factor was also examined. This model yielded acceptable fit indices as well ( m2 (222) = 418; CFI = 0.90; IFI = 0.89, NNFI = 0.86; RMSEA= 0.056). In addition, all first-order and second-order factor loadings were significant, thereby providing evidence for the plausibility of the thesis that information-processing is a multifaceted construct construed from the eight dimensions introduced in this research.

218

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

on). Simultaneously, when an NPD team considers the alternatives and shows curiosity about understanding product-related issues, its ability to discover technical, manufacturing and marketing shortcomings of the product increases. Note that the findings also demonstrate a bidirectional association between information acquisition and team sense-making. This finding suggests that an NPD team should organize information discovered from productrelated problems into meaningful patterns that can be understood easily by all team members. In this way, team members can interpret the marketing, technical, and manufacturing problems with minimal trouble. Concurrently, organizing market and technical information into meaningful ways helps teams to find out product-related problems more effectively. Finally, the results further suggest that information acquisition relates positively to information dissemination and information implementation. The discovery of product-related problems occurs simultaneously when team members develop a common vocabulary, conduct frequent informal water cooler/coffee maker communications with fellow project team members, and carry out formal communications through team meetings and memos. Similarly, information gathering from customers, competitors, markets, and technologies appears to play a critical role for the implementation of market and technical knowledge in organizations. New information can help anticipate emerging market needs, market movements, position and products of competitors and technological changes. Thus, in order to successfully implement the market plans, as well as to solve the customer problems and product development deficiencies, NPD teams need to develop the necessary skills to acquire information and to identify product-related problems. Next, team memory is found to be (1) positively associated with thinking, sense-making, information implementation, and information dissemination; and (2) negatively associated with unlearning and improvisation. Thus, team memory can help an NPD team to organize and understand the market and technical-related information sense-making. Specifically, team memory serves as a lens through which information is interpreted and sense-making process is handled. This is because extant team knowledge and skills reflect a repository of past experiences and provide an understanding or framework of how things work during the course of the project. At the same time, organizing, coding and sorting information during the project may enhance the knowledge and skill repository of teams, that is, team memory. The results also indicate that team memory can help team members to think , reason, and solve problems easily. Teams identify problems, generate alternatives, and select the best courses of action by the use of their knowledge and skill repositories. Team members cannot think deeply about any major issue unless they have a fair amount of relevant factual knowledge. At the same time, team memory is developed when team members consider advantages and disadvantages of alter-

native solutions and demonstrate curiosity about understanding customers and technical processes. The results further indicate that team memory and team unlearning are negatively related. That is, because team memory creates rules that may become inflexible during the project, an NPD team with a strong memory structure might find it very difficult to deviate from its previous knowledge store, thus inhibiting the unlearning. This finding provides a good example as to how core competencies may become core rigidities during the course of a project. Finally, the results further reveal team memory is negatively associated with team improvisation. This finding is consistent with Moorman and Miners (1998) findings, and suggests that a strong memory structure inhibits deviations from planned activities. Existing knowledge may lead to perception rigidity and thereby hinder the enactment of action as it unfolds. Next, team thinking is found to have positive bidirectional relationships with information dissemination, sense-making, and implementation/utilization. Thus, whereas thinking may facilitate the dissemination of information, communication of information across team members may as well help teams to consider alternative solutions for the problems encountered, thereby fostering further thinking. Indeed, it is common knowledge that solutions derived through effective thinking processes should be disseminated and communicated through the team for an effective NPD process. In addition, sensemaking, organizing the information into meaningful ways, appears to help team members understand each other and thereby to lead to effective collective considerations of alternative problem solutions. For instance, Dougherty, Borrelli, Munir, and OSullivan (2000) show that team sense-making in product development helps team members to solve problems and make decisions effectively by creating a shared web of meanings. When people have identical interpretations for the same information, they can collectively manipulate (i.e., think) that information for generating new ideas, detecting errors, giving decisions, etc. Finally, team thinking is also related to the information implementation/utilization. As Clark and Wheelwright (1993) indicate, when NPD teams generate alternative solutions to market and technological problems, analyze opportunities and threats, and carefully and repeatedly consider the product development process as well as their own structure, they can use information more effectively, act creatively, and initiate new insights essential for product success more easily. In regard to team improvisation, it was found that there exists a positive bi-directional relationship between team improvisation, information acquisition, information implementation, and unlearning. These findings demonstrate that a team can discover the marketplace and technology related problems of new products and correct such problems also by acting rather than by merely following a rigid and welldefined plan. Note, however, that any information obtained

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

219

through improvisation should be utilized (implemented) by the entire team for the team to benefit fully from the improvisation. If the knowledge of improvised actions is not shared by the individual team members who have performed that action, the knowledge generated will remain tacit and will not be beneficial to the project team as a whole. Note that the results further show that team unlearning, that is, belief changes, positively relates to team improvisation, and vice versa. This finding is interesting because the relationship between team unlearning and team improvisation has not been investigated in the past. The positive path coefficient between these two facets of team information-processing suggests that when a team acts without strict adherence to a plan, it may have a stronger temptation to modify its beliefs. In contrast, when a team follows plans programmed beforehand, it might not perceive a need to revise and evaluate its current belief structures.5 Regarding H2, it was shown that team intelligence is positively associated with team information acquisition, information dissemination, sense-making, information implementation, memory, and thinking. These findings are consistent with the organizational intelligence literature, and highlight the importance of Fteam competency_that is, a teams technical, market, and information-processing capabilities . If a team has the ability to gather information from different sources (both within the organization and from customers, vendors, suppliers, libraries, consultants, and so on), as well as to generate different market and technology scenarios, it can more successfully: (i) code and sort out information; (ii) discover the technical and market-related product problems; (iii) consider the different alternative solutions to problems; (iv) leverage the knowledge and skills of team members; (v) disseminate and communicate the information throughout the team; (vi) correct productrelated problems; and (vii) launch the new product. Interestingly, however, this research has failed to demonstrate a significant relationship between team intelligence and team unlearning. Team capabilities are found to be unrelated to the extent to which beliefs are changed during the course of the project. It appears that unlearning relates more closely to other information-processing components, particularly to improvisation. Consequently, it can be concluded that belief changes in NPD teams are generally initiated and influenced by action rather than by cognitive capabilities (cf. Sproull, 1981). Finally, with regard to H3, this research reveals that, when all information-processing factors are considered jointly, only information/knowledge implementation exerts
5 It is also important to note that no significant covariance has been observed between team unlearning and other cognitive elements. This does not mean that unlearning is isolated from other information-processing factors, however. Rather, an overall assessment of our findings suggests that team unlearning relates indirectly to other information-processing factors via team memory and team improvisation.

a significant direct effect on NPS. This finding is consistent with the studies of Moorman (1995) and Lynn et al. (2000), which suggest that utilizing information and knowledge to solve market-related and technical-related problems during the NPD process is a key determinant of the new products marketplace success (Moorman, 1995). Marketplace success is determined largely by the degree to which a project team corrects for the product problem areas, particularly regarding sources of customers dissatisfaction, and incorporates lessons learnt during the project into the new product development process. Consequently, it appears that all other information-processing factors influence NPS indirectly through the mediating role of information implementation. For instance, after customer information is acquired and disseminated during a project, applying this knowledge reduces ambiguity and uncertainty associated with market and product definitions, and team members can thereby make and implement better and faster decisions towards a more successful new product (Bacon, Beckman, & Wilson, 1994). Overall, this research reveals that integrating cognitive and social processes under the umbrella of social cognition provides benefits to NPD team project managers and scholars in viewing the dynamic nature of the NPD learning process. Note that, in cross-functional product development teams, technical and market knowledge is socially distributed because teams are not isolated in an organization and team members take action in the presence of others, such as suppliers, other project team members, managers, customers, and so on. Accordingly, there is a dynamic and multi-path knowledge network that reaches to all those involved in the NPD process, both inside and outside the team (Wind & Mahajan, 1997). Therefore, managers should consider the full set of socio-cognitive factors, and should attempt to facilitate conditions for an effective socio-cognitive process, in order to achieve successful NPD. Ignoring one or some of the elements of social cognition may reduce or halt the real impact of the learning process during the project. For instance, management should enhance changes in team beliefs when rapid changes in technological sophistication and marketplace dynamics exist. Specifically, management should break established team mental models and project infrastructures by supporting action changes and encouraging new behaviors. Effective unlearning can be attained by promoting cross-team communication and team experimentation, changing the people during the project, and creating an artificial stress accompanied with managerial support. Similarly, project managers are advised to plan project activities in a flexible manner that allows changes as the project evolves for an effective improvisation. Furthermore, managers should promote the dissemination and then usage or utilization of new knowledge. It is important to make sure that new knowledge does not become local and is understood and internalized by the entire team. Knowledge generated during improvisation and not used throughout the

220

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

project should be embedded into team memory. Therefore, managers should foster a documentation and filing system to guard against loss of critical information. Also, managers should form a small group to code and cluster information into topics, sources, types or importance levels, in order to help team members to understand and internalize information and knowledge. Then, these information clusters might be entered into a central database which everybody could easily access for an effective team sense-making. Managers should also enhance the teams thinking capability. Managers can encourage continuous experimentation and dialogue, and facilitate the usage of simulations/scenarios and what-if analyses. Finally, managers should increase the teams cognitive capability, that is, team intelligence, by considering the teams ability to acknowledge, recognize, monitor, and discriminate information and knowledge. Team intelligence can also be increased by spreading information throughout the team and improving members access to a variety of information sources and different points of views. Note that the socio-cognitive approach in this study draws attention to a dynamic team learning process that uses all cognitive processes. Teams can arrange the magnitude and composition of each learning construct, i.e., effectively switch the cognitive gears (Louis & Sutton, 1991), contingent upon the environmental conditions and activities deemed necessary for effective learning. In this regard, managers may emphasize some learning constructs more than others for an effective NPD project. For instance, when an NPD team is performing in a turbulent environment, the management could echo the importance of team improvisation and unlearning (Eisenhardt & Tabrizi, 1995). In a similar vein, when there exists uncertainty in the environment, managers can encourage team intelligence, thinking, and information acquisition more often than other learning constructs. It is important, however, to understand that all learning constructs are to some extent needed for an effective NPD process. This study also shows that team learning not only consists of information and knowledge but also encapsulates feelings, beliefs, actions, and emotions. In this sake, managers should take into account the feelings, emotions, beliefs, and actions of team members if they wish to leverage team learning effectively. Most specifically, managers should manipulate the fear, hope, excitement, despair, and anxiety of team members to create a safe psychological environment that will ultimately lead to superior learning performance. The NPD literature has identified team processes as being team member communications, team structure, cross-functional integration, team size, stage-gate processes, rewards, and so on. However, these types of constructs denote the nature of the interaction processes rather than states of the interaction processes (Marks, Mathieu, & Zaccaro, 2001). As Marks et al. (2001) have noted, team members convert inputs to outcomes through

a series of structural, cognitive, verbal, and behavioral parameters. The socio-cognitive framework introduced in this research enlightens the states of the team interaction process by using cognitive and contextual mediators. In this vein, understanding the socio-cognitive processes that team members use in NPD projects will enable managers to control and manage the project, to select, train, develop, and reward personnel for teamwork, more effectively.

7. Limitations and future research There are some methodological and theoretical limitations in this study which nevertheless highlight fruitful avenues for future research. Foremost among these is the fact that the study used single sourcing and self- and retrospective reporting. Because teams with a relatively small number of members were used and respondents were selected with utmost care among those who are most knowledgeable about the teams procedures, it is not expected that key informant bias represents a major limitation in this study. Nonetheless, scholars (e.g., Aviolo, Yammarino, & Bass, 1991; Gupta & Beehr, 1982) have argued that studies employing single-source methodology can be biased by artificially high inter-correlations produced by overall response tendency. In order to assess the degree to which such a common method variance has inflated (or, perhaps, deflated) the parameter estimates, additional analyses were conducted. The measurement model was re-estimated after adding the original model a new common method (single source) factor onto which all measurement items are allowed to load. Each measurement item in this model loaded onto both its respective latent factor in the original measurement model and the common method factor. The latent factor correlations obtained from this model were then used in a path-analytic investigation of the parameters in Fig. 1. Note that these new parameter estimates are essentially adjusted for possible common method variance. The statistical significance of these adjusted estimates were the same as the unadjusted estimates; there were only slight differences in the magnitudes of parameter estimates; and the correlation between the adjusted and unadjusted estimates is 0.93 ( p < 0.001). Based on these analyses, it was concluded that single sourcing may not be a major source of bias in this research. However, it should be noted that the responding managers may have their own ego-involved biases, and these might have influenced their responses. Finally, since retrospective reports were used, respondents answers might have been influenced by memory distortion and halo effect. This study has theoretical limitations as well. Identifying these limitations might provide fruitful avenues for future researchers to extend and expand the inquiry in this area. First, the presented model in this study was general

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

221

and did not capture the possible moderating effects of environmental turbulence and uncertainty. Prior research has shown that the effect of cognitive factors on individual, group, and organizational performance can vary substantially with environmental conditions. For instance, under turbulent conditions, improvisation and unlearning might produce more desirable results for team performance (Weick, 1979), whereas under uncertain conditions, team deliberation and sense-making might be preferable (Langer, 1978). Second, the proposed model can be expanded and a more complete model of learning can be investigated by adding new constructs (e.g., emotion). Third, each learning construct can be expanded both theoretically and empirically. For instance, the operationalization of team unlearning can be improved by including changes in team routines and beliefs. Likewise, team intelligence or team competency can be investigated in greater detail. Fourth, other resultant products of social cognition such as group social identity, relations between group veterans and newcomers, and group transactive memory system could be investigated in future studies of NPD teams.

Information implementation Overall, the market perceived this product had fewer problems than what was considered normal in the industry. Most of the lessons learned pre-launch were incorporated into the product for full-scale launch. *Overall, the team did an outstanding job uncovering product problem areas with which customers were dissatisfied. Overall, the team did an outstanding job correcting product problem areas with which customers were dissatisfied. Thinking *Team members generated an abundance of alternative solutions to overcome problems encountered during the project. Team members considered advantages and disadvantages of many alternative solutions before selecting an appropriate course of action. During the project, team members were extremely curious about understanding customers needs and wants. During the project, team members were extremely curious about identifying technical problems (for example, product performance shortcoming, cost overruns, and so on). Improvisation The team figured out the new product development process as it went along as opposed to following a rigid well-defined plan. The team improvised in developing this product as opposed to following the plan strictly. The team improvised in commercializing this product as opposed to following the plan strictly. Memory *There was a well-defined procedure for developing this product. *There was a well-defined procedure for commercializing this product. There was an abundance of knowledge readily available on how to develop this product. There was an abundance of knowledge readily available on how to commercialize this product. *There were adequate skills on the team to develop this product. There were adequate skills on the team to commercialize this product. Intelligence During the project, the team had the ability to transfer customer needs to product design specifications. During the project, the team had the ability to generate different market and technology scenarios *During the project, the team had the ability to gather information from different functions within our organization.

8. Conclusion Learning is a broad concept involving different definitions, constructs, epistemologies, and anthologies. Although a broad research base on the subject is in existence, it remains a complex and confusing concept. One goal of this research was to leverage the NPD team learning literature by cross-fertilization with other literatures. In pursuit of this, social cognition was examined. Another goal of this study was to operationalize social cognition and to test it empirically in new product development teams and new product performance. It was shown empirically that team learning process cannot be explained by reducing the scope to one or another element. Rather, learning process is an aggregation of all socio-cognitive components. Furthermore, it was found that team intelligence is the driving force of team information-processing, and that team information-processing has a positive influence on the success of new product projects. At the same time, different informationprocessing components were examined simultaneously, thus providing evidence as to how teams learn consciously and unconsciously. The socio-cognitive perspective introduced in this research thus constitutes a useful addition to the extant research on the performance of NPD teams.

Appendix A. Measures We used a Likert scale of 0 (= Fstrongly disagree_) to 10 (= Fstrongly agree_).

222

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224

During the project, the team had the capability to gather information from outside the organization (for example, customers, vendors, suppliers, libraries, consultants, and so on). Unlearning The teams beliefs changed pre-prototyping to postlaunch regarding: & & & & & the features customer demanded; the features that were technically possible; the rate of technological improvements; the new product development process needed; *the steps needed to be followed to develop this product; & *the steps needed to be followed to bring this product to the market successfully; & *the rate of market acceptance. Information acquisition The team did an outstanding job in discovering technical shortcomings of this product. The team did an outstanding job in discovering manufacturing shortcomings. The team did an outstanding job in discovering marketing shortcomings of this product. Information dissemination Informal Team members developed a common vocabulary when discussing the project. Team members conducted frequent informal communications at water cooler/coffee maker with fellow project team members. Team members conducted frequent informal communications at lunch or after work with fellow project team members. Formal Team members conducted frequent formal communications through team meetings with fellow project team members. Team members conducted frequent formal communications through memos with fellow project team members. Sense-making Information collected by the team (for example, test results) was coded and sorted to be understood easily by other team members. Market information was organized in meaningful ways. Technical information was organized in meaningful ways. *Technical information was summarized to reduce its complexity. *Market information was summarized to reduce its complexity.

New product success This product: & met or exceeded volume expectations; & met or exceeded sales dollar expectations; & met or exceeded the first year number expected to be produced and commercialized; & met or exceeded overall sales expectations; & met or exceeded profit expectations; & met or exceeded return on investment expectations; & met or exceeded senior management expectations; & met or exceeded market share expectations; & met or exceeded customer expectations. Notes: *these items were eliminated during the measurepurification process.

References
Adams, M., Day, G., & Dougherty, D. (1998). Enhancing new product development performance: An organizational learning perspective. Journal of Product Innovation Management, 15, 403 422. Akgu n, A. E., Lynn, G. S., & Byrne, J. (2003). Organizational learning: A socio-cognitive framework. Human Relations, 56(7), 839 868. Akgu n, A. E, Lynn, G. S., & Reilly, R. R. (2002). Multi-dimensionality of learning in new product development teams. European Journal of Innovation Management, 5, 57 72. Allard-Poesi, F. (1998). Representations and influence processes in groups: Toward a socio-cognitive perspective on cognition in organization. Scandinavian Journal of Management, 14, 395 420. Anderson, R., & Ausubel, D. (1965). Readings in the psychology of cognition. Holt, New York Rinehart and Winston, Inc. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411 423. Aviolo, B. J., Yammarino, F. J., & Bass, B. M. (1991). Identifying common methods variance with data collected from a single source: An unresolved sticky issue. Journal of Management, 17, 571 587. Bacon, G., Beckman, S., & Wilson, E. (1994). Managing product definition in high-technology industries: A pilot study. California Management Review, 36(3), 32 56. Clark, K., & Wheelwright, S. (1993). Managing new product and process development. MA Harvard Business School Press. Cook, S., & Yanow, D. (1993). Culture and organizational learning. Journal of Management Inquiry, 2, 373 390. Cooper, R. G. (1993). Winning at new products: Accelerating the process from idea to launch. Reading, MA Addison-Wesley Publishing Comp. Cooper, R. G., & Kleinschmidt, E. (1987). Success factors in product innovation. Industrial Marketing Management, 16, 215 223. Crocker, J., Fiske, S., & Taylor, S. (1984). Schematic bases of belief change. In J. R. Eiser (Ed.), Attitudinal judgment (pp. 197 225). New York Springer-Verlag. Crossan, M., Lane, H., White, R., & Djurfeldt, L. (1995). Organizational learning: Dimensions for a theory. International Journal of Organizational Analysis, 3, 337 360. Crossan, M., Lane, H., White, R., & Leo, K. (1996). The improvising organization: Where planning meets opportunity. Organizational Dynamics, 20, 20 35. Dixon, N. (1992). Organizational learning: A review of literature with implications for HRD professionals. Human Resource Development Quarterly, 3, 29 49.

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224 Dougherty, D., Borrelli, L., Munir, K., & OSullivan, A. (2000). Systems of organizational sensemaking for sustained product innovation. Journal of Engineering and Technology Management, 17, 321 355. Easterby-Smith, M. (1997). Disciplines of organizational learning: Contributions and critiques. Human Relations, 50, 1085 1112. Eisenhardt, K., & Tabrizi, B. (1995). Accelerating adaptive processes: Product innovation in the global computer industry. Administrative Science Quarterly, 40, 84 110. Ettlie, J., & Rubenstein, A. (1987). Firm size and product innovation. Journal of Product Innovation Management, 4, 89 108. Fiske, S., & Taylor, S. (1991). Social cognition. New York McGraw Hill. Fornell, C., & Larcker, D. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18, 39 51. Garvin, D. (1993). Building a learning organization. Harvard Business Review, 71, 78 91. Gherardi, S. (1998). Competencethe symbolic passe-partout to change in a learning organization. Scandinavian Journal of Management, 14, 373 393. Gioia, D., & Sims, H. (1986). Introduction: Social cognition in organizations. In H. P. Sims, & D. A. Gioia (Eds.), The thinking organization: The dynamics of organizational social cognition (pp. 1 11). San Francisco, CA Jossey-Bass. Glynn, M. A. (1996). Innovative genius: A framework for relating individual and organizational intelligence to innovation. Academy of Management Review, 21, 1081 1111. Goldman, A. (1986). Epistemology and cognition. Massachusetts Harvard University Press. Grant, R. (1996). Prospering in dynamically-competitive environments: Organizational capability as knowledge integration. Organization Science, 7, 375 387. Gruenfeld, D., & Hollingshead, A. (1993). Socio-cognition in work groups. Small Group Research, 24, 383 405. Gupta, N., & Beehr, T. (1982). A test of the correspondence between selfreports and alternative data sources about work organizations. Journal of Vocational Behavior, 20, 1 13. Hackman, R. (1987). The design of work teams. In J. W. Lorsch (Ed.), Handbook of organizational behavior (pp. 315 342). Englewood Cliffs, NJ Prentice Hall. Hedberg, B. (1981). How organizations learn and unlearn. In P. C. Nystrom, & W. H. Starbuck (Eds.), Handbook of organizational design, vol. I. Oxford, UK Oxford University Press. Higgins, T. (2000). Social cognition: Learning about what matters in the social world. European Journal of Social Psychology, 30, 3 39. Huber, G. (1991). Organizational learning: The contributing processes and the literatures. Organization Science, 2, 88 115. Hunter, J. (1986). Cognitive ability, cognitive aptitudes, job knowledge, and job performance. Journal of Vocational Behavior, 29, 340 362. Kiesler, S., & Sproull, L. (1982). Managerial response to changing environments: Perspectives on problem sensing from social cognition. Administrative Science Quarterly, 27, 548 570. Langer, E. (1978). Rethinking the role of thought in social interactions. In J. H. Harvey, W. Ickes, & R. F. Kidd (Eds.), New directions in attribution research. Hillsdale, NJ Erlbaum. Larson, J., & Christensen, C. (1993). Groups as problem-solving units: Toward a new meaning of social cognition. British Journal of Social Psychology, 32, 5 30. Larson, E., & Gobeli, D. (1988). Organizing for product development projects. Journal of Product Innovation Management, 5, 180 190. Lastovicka, J., & Thamodaran, K. K. (1991). Common factor score estimates in multiple regression problems. Journal of Marketing Research, 28, 105 112. Leonard-Barton, D. (1992). Core capabilities and core rigidities: A paradox in management in managing new product development. Strategic Management Journal, 13, 111 125.

223

Louis, M. R., & Sutton, R. (1991). Switching cognitive gears: From habits of mind to active thinking. Human Relations, 44, 55 76. Lynn, G., Reilly, R. R., & Akgu n, A. E. (2000). Knowledge management in new product teams: Practices and outcomes. IEEE Transactions on Engineering Management, 47, 221 231. Madhavan, R., & Grover, R. (1998). From embedded knowledge to embodied knowledge: New product development as knowledge management. Journal of Marketing, 62, 1 12. Manis, M. (1966). Cognitive processes. Belmont, CA Brooks/Cole Publishing Company. Marks, M., Mathieu, J., & Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. The Academy of Management Review, 26, 356 376. Martin, L., & Clark, L. (1990). Social cognition: Exploring the mental processes involved in human social interaction. In M. W. Eysenck (Ed.), Cognitive psychology: An international review (pp. 195 216). Hillsdale, NJ Willey. McMaster, M. (1996). The intelligence advantage: Organizing for complexity. Newton, MA Butterworth-Heinemann. Meyers, P., & Wilemon, D. (1989). Learning in new technology development teams. Journal of Product Innovation Management, 6, 79 88. Moorman, C. (1995). Organizational market information processes: Culture antecedents and new product outcomes. Journal of Marketing Research, 32, 318 335. Moorman, C., & Miner, A. (1997). The impact of organizational memory on new product performance and creativity. Journal of Marketing Research, 34, 91 106. Moorman, C., & Miner, A. (1998). The convergence of planning and execution: Improvisation in new product development. Journal of Marketing, 62, 1 20. Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company: How Japanese companies create the dynamics of innovation. New York Oxford University Press. Nunnally, J. (1978). Psychometric theory. New York McGraw-Hill. Ostrom, T. (1984). The sovereignty of social cognition. In R. Wyer, & T. Srull (Eds.), Handbook of social cognition (pp. 1 38). Hillsdale, NJ Lawrence Erlbaum. Pinto, J. (2002). Project management. Research Technology Management, 45, 22 37. Popper, M., & Lipshitz, R. (1998). Organizational learning mechanisms: A structural and cultural approach to organizational learning. The Journal of Applied Behavioral Science, 34, 161 179. Purser, R., Pasmore, W., & Tenkasi, R. (1992). The influence of deliberations on learning in new product development teams. Journal of Engineering and Technology Management, 9, 1 28. Reed, S. (1982). Cognition: Theory and applications. Belmont, CA Wadsworth, Inc. Rochford, L., & Rudelius, W. (1992). How involving more functional areas within a firm affects the new product process. Journal of Product Innovation Management, 9, 289 299. Schwarz, N. (1995). Social cognition: Information accessibility and use in social judgment. In E. Smith, & D. Osherson (Eds.), Thinking (pp. 345 376). Cambridge, MA MIT Press. Sinkula, J. (1994). Marketing information processing and organizational learning. Journal of Marketing, 58, 35 45. Sproull, L. (1981). Beliefs in organizations. In P. C. Nystrom, & W. H. Starbuck (Eds.), Handbook of organizational design, vol. 1. (pp. 223 224). New York Oxford University Press. Thamhain, H. (1990). Managing technologically innovative team efforts toward new product success. Journal of Product Innovation Management, 7, 5 18. Weick, K. (1979). The social psychology of organizing (2nd edR). New York Random House. Weick, K. (1993). Organizational redesign as improvisation. In G. P. Huber, & W. H. Glick (Eds.), Organizational change and redesign (pp. 346 374). New York Oxford University Press.

224

A.E. Akgu n et al. / Industrial Marketing Management 35 (2006) 210 224 Gary S. Lynn (PhD, Rensselaer Polytechnic Institute) is a tenured Associate Professor at Stevens Institute of Technology, USA. Cengiz Yilmaz (PhD, Texas Tech University) is an Associate Professor of Marketing in the School of Business Administration at Bogazici University, Turkey.

Wind, Y., & Mahajan, V. (1997). Issues and opportunities in new product development: An introduction to the special issue. Journal of Marketing Research, 34, 1 12. Ali E. Akgu n (PhD, Stevens Institute of Technology) is an Associate Professor of Science and Technology Studies at Gebze Institute of Technology, Turkey.