i need a 500 words essay on cognitive learning theory in 4 hours

Journal of Management Studies 29:2 M a r c h 1992

'0022-2380 $3.50

TOP MANAGEMENT, STRATEGY AND ORGANIZATIONAL KNOWLEDGE STRUCTURES

M A R J O R I I ; A. LYLES

CH.ARLKS R . SCHVVENK

Indiana University

ABSTRACT

The cognitive aspects of strategic management and organizational learning have been receiving increasing interest from researchers (Fahey and Narayanan, 1986; Prahalad and Bettis, 1986). Researchers have called for more detailed descriptions of the ways that individual-level cognitions contri- bute to organizational-level strategies (Daft and Weick, 1984; Smircich and Stubbart, 1985). In this article, a theory of organizational knowledge struc- tures which include core and peripheral features is presented. From this theory propositions are suggested to guide future research on organizational learning.

INTRODUCTION

A review of the strategic management literature reveals an increased emph- asis on the top management team and its impact on the strategic behaviour and performance of the firm. One widely held perspective is that the top management team exerts influence over the members of the organization through its shared perspective of environmental events and organizational capabilities. Prahalad and Bettis (1986) suggest that this shared perspective can be described as a general knowledge structure that can store a shared dominant general management logic. Our purpose in this article is to expand the notion of organizational knowledge structures.

In the past decade, there has been extensive research on cognitive struc- tures and processes and their impact on organizational behaviour. This research addresses such topics as decision-makers' frame of reference (Ham- brick, 1981; Schwenk, 1988; Shrivastava and Schneider, 1984), assumptions (Mason and MitrofF, 1981), cognitive maps (Ford and Hegarty, 1984); schemas (Huff, 1983; Weick, 1979); belief systems (Dunbar'^^/ al, 1982); scripts (Gioia and Manz, 1985; Gioia and Poole, 1984); organizational learning (Argyris and Schon, 1978; Fiol and Lyles, 1985; Hedberg, 1981; Jelinek, 1979; Levitt and March, 1988; Lyles, 1988; Shrivastava, 1983) and interpretive systems (Bartunek, 1984; Daft and Weick, 1984; Meyer, 1982).

Address for reprinls: Marjorie A. Lyies. Indiana University, School of Business, 801 W. Michigan, Indianapolis, IN 46202-5I5I, USA.

156 M.A. LYLES AND C.R. SCHWENK

Most of tbese studies focus on a single concept and its definition and impact. Less effort is devoted to the study of tbe more important issue of tbe linkages between individual-level cognitions, tbe development of shared understandings, and organizational action. This has led to a ver\' fragmented view of these linkages.

Strategic management researcb has typically neglected the link between individual beliefs and cognitive structures, organization-wide beliefs and assumptions, and political processes. Hence, there are gaps in tbe research relating tbe interaction between individual-level understanditig and organiza- tional action. Some researcb assumes tbat organizations bave knowledge structures as do individuals. Otber research di.scus.ses tbe cognitions of key individuals but does not discuss tbe ways tbese are formed or modified by organizational events, or tbe extent to whicb understandings are shared by otbers in the organization.

Tbis article refiects the growing need for a more developed tbeoretical foundation that may serve as the basis for future empirical studies aimed at testing tbe existence and linkages of organizational knowledge and strategic management. We hope to theorize beyond tbe current research base .since a larger vision of the tbeory is needed, whereas experimentation bas focused only on elements of tbe tbeory. We bope to present a tbeoretical construct and to raise questions that lend themselves to later experimentation. We do not intend to prove tbe existence of an organizational 'mind' or 'cognitive structures' (Sandelands and Stablein, 1987). However, we do intend to address two futidamental problems: Wbat is an organizational knowledge structure? How migbt it develop and change?

In order to do tbis, we have bad to set some limits on wbat we can consider and use as evidence. Our focus will be on the development of organizational knowledge and how knowledge structures might change. We rely on tbe literature of tbe cognitive psycbologists and of the authors developing schema theory. Altbough we utilize this literature, we do not mean to imply tbat organizations have minds in the same sense that humans do. A basic assumption, bowever, is tbat tbere are shared understandings witbin organi- zations tbat influence organizational behaviour, though it is not necessary fbr knowledge to be sbared in order to influence behaviour. These are stored over time and may be retrieved (Brunsson, 1985; Daft and Weick, 1984; Hedberg, 1981).

T o help explain our tbeory we will use statements taken from interviews conducted with two senior executives of an international consulting firm tbat has just gone tbrougb a major strategic change. Tbese interviews provide a means of illustrating tbe major issues. Our intent however is illustration - these sentences do not represent scientific researcb or deeper knowledge structures. Our objectives are to identify sbared understandings through these statements.

This article examines tbree elements that influence the development of organizational knowledge. First, we will discuss the concept of schema and bow key decision-makers' schemata infiuence the development of widely held cause-and-effect beliefs. Tbe concept of 'schema' refers to individual-level knowledge while the term 'knowledge structure' refers to shared beliefs at tbe organizational level. Furtber, tbese beliefs have a structure.

MANAGEMENT, STRATEGY AND KNOWLEDGE 157

Second, we argue that there are certain core features of the organization knowledge structure which remain invariant over long periods of time, while there are peripheral features which change. We will then discuss the process of organizational knowledge development. Since we are dealing with a social theory of knowledge development, bargaining, consensus and dissensus become important processes for communicating and integrating alternative schemas. Finally we explore the implications of our theoretical discussion and give propositions to guide future research.

ORGANIZATIONAL KNOWLEDGE STRUCTURES

Argyris and Schon (1978) suggest that an organization is, at its root, a cognitive enterprise and it learns and develops knowledge. The knowledge structure serves to define expected relationships, behaviours, and actions for organizational members.

The organizational knowledge structure is different from organizational culture and climate in at least two important ways. The concept of knowledge structures deals with goals, cause-atid-effect beliefs, and other cognitive ele- ments. Thus, it is narrower than culture and climate, both of which refer more to affective or emotional elements. Further, the knowledge structure is more clearly linked to an organization's strategy for survival and more subject to change than an organization's climate or culture, neither of which changes readily or provides specific strategies for action for an organization.

The schemata of top managers have an important influence on the develop- ment of the organization's knowledge structure. Alba and Hasher (1983) suggest that the information contained within a schema may be encoded, stored and retrieved. A subset of schemas, called scripts, contains information about particular, frecjuently experienced events (Schank and Abelson, 1977). In-depth discussion of schema properties can be found in such works as Alba and Hasher (1983); Anderson (1983); Bartlett (1932); and Schank and Abelson (1977).

By their very nature, organizational knowledge structures differ from personal schemata because they are socially constructed and rely on consen- sus or agreement (Daft and Weick, 1984; Hedberg, 1981; Weick, 1979). Weick and Bougon (1986, p. 109) suggest that in developing knowledge structures, there are three stages that people go through to reach agreement: (1) agreement on which concepts capture and abstract iheir joint experience; (2) consensus on relations among these concepts; and (3) similarity of view on how these related concepts alTect each party. Some of the schema properties that are particularly relevant to organizational knowledge structures are encoding, storage, elaboration, forgetting, retrieval, modelling, modification of structures, adding new structures, and complexity of structures (Daft and Weick, 1984; Gioia'^and Manz, 1985; Lord and Foti, 1986).

Organizational researchers have shown that organizations develop shared frames of reference, recall of past events, the creation of stories atid myths, vicarious learning, unlearning and memories (Dunbar et al., 1982; Hedberg et al., 1976; HufT,'l983; Jelinek, 1979; Lyles, 1988; Manz and Sims, 1981; Martin, 1982). It is important to draw together the concepts of earlier

158 M.A. LYLES AND C.R. SCHWENK

researchers to specify the elements of organizational knowledge structures and tbe way they interact. Figure 1 describes the ways knowledge structures may change over time.

The process of developing the organizational knowledge structure is ongoing and continuous. As the figure shows, changes in a firm's environment may be perceived by many members and eacb may interpret it in different ways. However, when the change begins to affect company performance and when it cannot be adequately explained or predicted through the existing knowledge structure, it creates a challenge to that existing knowledge struc- ture and creates an interactive effect among the organizational participants, the environment and behaviour (Bandura, 1977; Davis and Luthans, 1980).

Political processes are evoked when different coalitions (such as operators, managers, and analysts and support staflT) within a firm who hold alternative schemas advocate their own positions through such mechanisms as task forces and special reports (Lord and Foti, 1986; Prahalad and Bettis, 1986; Taylor and Crocker, 1981). These different coalitions attempt to influence others and to gain the agreement of others about the coalition's interpretation of events (Lyles and Mitroff, 1985).

Changes in the organizational knowledge structure occur as a result of the impact of the interpretation of environmental events, results of past organiza- tional actions, the influence of the key decision-makers, and the advocacy position of coalitions within the firm. The political processes have to do with the negotiating and bargaining that takes place to gain support for alternative schemas. Socio-political themes, such as credibility and power, have been shown to influence the acceptance of particular views (Bower, 1970; Lyles and MitrofT, 1985; Mintzberg, 1983; PfefTer, 1981). The degree of consensus about goals and about the means of achieving them influences the efifectiveness of the firm (Bourgeois, 1980; Dess, 1987; Dess and Origer, 1987).

Once the key decision-makers sense and interpret changing environmental events, they frame them in the problem formulation process as problems, opportunities, or crises (Dutton and Jackson, 1987; Lyles and Mitroff, 1980; Mintzberg et al., 1976). These decision-makers have a strong influence on the development of the organizational knowledge structures since it is primarily they who interpret the importance of environmental events and who com- municate their view of the knowledge structure through speeches and state- ments. It seems likely that the key decision-makers' schemata closely reflect the 'core area' of the knowledge structure, and they influence the attitudes and beliefs of others in the organization by communicating and networking (Kotter, 1982; Tushman et al., 1989; Zaleznik, 1977). Once the key decision- makers have determined that the change challenges the core elements of the knowledge structure, they make changes in tbe know l̂edge structure and communicate these changes to others in the organization.

A critical point in our model of changes in organizational knowledge structures is tbe distinction between core features of the knowledge structure and peripheral features. Environmental changes which challenge core fea- tures of the knowledge structure evoke more profound changes in the know- ledge structure. Therefore, it is important that we discuss the differences between core and peripheral features.

MANAGEMENT, STRATEGY AND KNOWLEDGE 159

160 M.A. LYLES AND CR. SCHWENK

Core Set of Knowledge Structures Throughout the organization theory and philosophy of science literature are references to a core set of knowledge structures that explain beliefs and goals on which there is widespread agreement. These enable organizational mem- bers to understand the most basic purpose and mission of the firm. They provide a guide to expectations about organizational behaviour in the broadest sense of competitive situations. We suggest that this core set is simitar to what Abelson (1981) calls a metascript: it is abstract and lacks detail but still provides a guide to the firm's business situations. It has widespread agreement and acceptance among the managers.

For example, Fayol (1949) describes an 'espirit de corps' and a sense that 'union is strength'. Barnard (1938) suggests that organizations are co- operative systems that have a unifying purpose. It is the executive function to promote this widespread belief in the real existence of common purpose. It is belief that leads to co-operation, the development of a system of social action that overcomes the limitations of working alone. He suggests that it is the belief that is important, not necessarily the detailed understanding of the common purpose. Thompson (1967) refers to it as 'domain consensus'.

Even in other literatures, there is reference to core knowledge upon which there is widespread agreement. Kuhn (1962) suggests that in the development of science, there is a certain body of knowledge which provides a core upon which there is widespread agreement. Berger and Luckmann (1967) suggest that in 'secondary socialization', when individuals become committed to an institution, there must be an immersion in and commitment to the new reality. It is the acquisition of knowledge thai creates mutual identification. It is the internalization of the roles and attitudes of 'significant others' that creates knowledge about roles and language.

Propo.ntion 7: Firms are characterized by core elements of a generalized knowledge framework about which there is widespread agreement. The core facilitates understanding about the firm's general purpose, mission, and competitors.

For each organization there exists a core knowledge set about its mission, justification for its existence, and the basic business aim. This core would help to defme the most basic of the repertoire of organizationally-relevant know- ledge structures that describe the events and behaviours appropriate to the particular firm and the people within it. In order for the organization to exist, individuals and stakeholders must believe in the usefulness of the organiza- tion to serve some purpose and to carry out a mission. Barnard (1938), Cyert and March (1963) and Thompson (1967) develop this notion of a core set of values or beliefs that serve as motivating reasons for individuals to join together in an organization. Weick (1979) suggests that co-ordination within firms will be problematic until some agreement can be reached. It is necessary for sense-making. We propose that the core set provides the most basic elements of the organizational knowledge structure. It consists of the beliefs on which there is the most consensus among organizational members.

To illustrate these points, we use the consulting firm we studied. There was

MANAGEMENT, STRATEGY AND KNOWLEDGE 161

Table I. Elements of the firm's knowledge structure

Firm-in-industry:

Firm-client script:

Client script:

Competitor script:

Firm-in-industry

Firm—client script:

Client script:

Competitor script:

Firm-in-i nd us t ry:

Firm-client script:

Client script:

Competitor script:

1960s

Provide consulting services "Become al! things lo all people' Develop close relationship with client Provide primarily "manufacturing' expertise Client unsophisticated Loyal Heavy-industry, smokestack, manufacturing Provide broad base of services Big consulting firms Domestic

1970s

Provide consulting services Internationa! Co-ordinated activities across regions Purely business relationships Provide legs manufacturing expertise Becoming sophisticated about buying services Disloyal Making decisions based on shopping around Becoming specialists Growing faster than "us' International

1980s

Moving away from 'manufacturing' consulting Equal domestic vs. international Independent regions Provide client with best 'experts' 'Self them on us Sophisticated Many MBAs and ex-consultants in-house Few manufacturing Many small Niche strategies

a core knowledge structure. The first element of tbis structure was tbe belief tbat the company sbould serve a wide range of client needs. Tbis point was explained by one of our interviewees:

consulting . . . proliferated dramatically following World War II. The big- gest firms were those tbat developed a working relationsbip with the senior management ofa client company and advised on just about everytbing tbe client thougbt tbey needed advice on. Tbe way in wbich you really sold the service was to maintain a close personal relationship with the people running divisions and then whenever tbey needed some outside assistance, they would call you to come and give it. So, firms tended to become all things to all clients.

There was also agreement on who were the competitors.

162 M.A. LYLES AND CR. SCHWENK

Firms like McKinsey and Booz Allen, Arthur D. Little and some other smaller firms generally offered a variety of services. The CPA firms were somewhat more specialized, but they were big competitors.

Lord and Foti (1986) suggest that scripts are basic elements that help to order events and causal relationships. Table I summarizes the basic elements of the knowledge structure as we interpret them from the two interviews. Our purpose is to illustrate how two members of an organization begin to define the perception of ambiguous information and how this interpretation may begin to shape expectations about actions and behaviours. The examples in table I illustrate the basic elements of knowledge structures that are widely held in this particular firm.

To repeat briefly, this core set provides the initial knowledge structure through which the organizational members compare themselves to other organizations: it provides a basis for that comparison by providing knowledge about acceptable behaviour or actions for their firm. It can be used to show differences with what is acceptable behaviour in other organizations. This core is communicated and sponsored by top management in various ways including public and private statements, and symbolic actions. These actions are the means by which core knowledge structures are maintained.

Peripheral Structures

Proposition 2: Firms are characterized by elements of the knowledge struc- ture that support the core set, but about which there may not be wide- spread consensus.

Peripheral elements include knowledge about how to achieve the expectations established in the core set and how to interpret environmental and competitor signals - cause-and-elfect beliefs that interrelate the core set to actions and environmental factors. The peripheral structures are open to much more debate and disagreement than is the core set. Astley et al. (1982) suggest that because of the division of labour within firms, a variety of perspectives develop and this would lead to the development of different knowledge structures at the peripheral level.

Thus, we propose that the peripheral knowledge structure may ditfer from the core knowledge structure in two ways: the content and the degree of consensus. The core set contains knowledge about the most basic of the firm's purposes and goals, while the peripheral set contains knowledge about subgoals and about the behaviour or steps necessary to achieve those goals. The core set and the peripheral set are related and linked together. Indi- viduals apply their insights, heuristics, and schemas to attempt to achieve the subgoals that result in the accomplishment of the purpose of the firm.

In the case of our example firm, tlie peripheral set included beliefs about the clients and subgoals which would be served. One person stated:

Most partners thought that we needed to crystallize, clarify, and more precisely define our basic business strategy. I think most would agree with that, and they would probably need to make some changes in things that

MANAGEMENT, STRATEGY AND KNOWLEDGE 163

we see being done, but obviously not all agreeing on what those changes should be.

The clients were defined as 'the traditional smokestacks/heavy manufacturing businesses'. Expectations about appropriate organizational designs were also part of the peripheral knowledge structure.

We had been organized on a decentralized basis geographically. In our biggest office, we broke our consulting groups down into functional areas of activity. Within the regional organizations, they were not broken down that way.

The peripheral knowledge structure seems to contain knowledge about subgoals {e.g. business-level goals or ends) and about behaviours appropriate for achieving them (means). A number of studies have appeared recently that address the linkages between environmental assessment, means-ends congru- ence, and consensus. Ends are defined as business-level goals (such as rate of sales growth, net profit over five years, etc.) and means are competitive methods within that particular business (such as new product development, customer service, etc.). Representative of this line of research are Bourgeois (1980), Dess (1987) and Dess and Origer (1987). While these studies do not consider knowledge structures directly, they make a contribution to this thinking by addressing consensus about means-ends and the impact on organizational performance. Although these studies do not address corporate level strategy {e.g., what business do we want to compete in) and the results vary, they do seem to conclude that widespread consensus about means and ends does not lead to better performance.

Clearly the more agreement about the peripheral set and the core set, the more agreement there will be about specific courses of action necessary to ensure survival and the more single-minded will be the strategic thrust of the firm. This is the goal of congruity described by Barnard (1938) and Ouchi (1980). In tbis circumstance, the high level of consensus may lead to the development of strong scripts about behaviour in certain situations leading to poor adjustment or something similar to 'anchoring' (Kahneman etal., 1982).

COMPLEXITY AND RELATEDNESS

We propose that there are two characteristics of organizational knowledge structures that become important to understanding how these structures work. Over time and as an organization learns, experience about certain events and instances is built up. The storage of these experiences becomes more complex and more abstract over time (Martin, 1982). Thus, the first characteristic of the knowledge structures is that of complexity. Complexity refers to the amount of information or the number of elements within a knowledge structure. The second characteristic, relatedness, refers to the linkages between elements in the knowledge structure, particularly the links between core and peripheral elements.

164 M.A. LYLES AND CR. SCHWENK

Proposition 3: The organizational knowledge structure is likely to become more complex over time.

Complexity By complexity we mean the degree to which cognitive units are interrelated, creating a complex internal structure. Anderson (1983, p. 76) suggests that in knowledge representations, Marge knowledge structures must be encoded hierarchically, with smaller cognitive units embedded in larger ones'. In fact Anderson suggests that subhierarchies may become embedded within other hierarchies creating 'tangled hierarchies. It is necessary-' to discuss the concept of hierarchy in connection with knowledge structures, since some elements of the structures (core) are more important than others (peripheral).

The complexity of the knowledge structure influences the ability of organiza- tional members to retrieve elements that have become embedded in the hierarchical structure. This will influence their ability to respond to environ- mental changes and new situations. Levitt and March (1988) suggest that the memory of organizations is orderly but inconsistent. They suggest that knowledge becomes 'nested within' other knowledge, but not all events are recorded. Organizations do act as repositories which include the personnel, the rules of operation, stories, and computer bases.

Not all elements that are stored will be easy to retrieve. Which elements are retrieved depends upon the frequency of use, how recent they are, the usefulness, organizational proximity, preferences, self-interest and attention (Douglas, 1986; Levitt and March, 1988; March and Olsen, 1975; March et al., 1991).

The more complex the structure, the more the simple units may become hidden or forgotten. The original element or experience may become lost because of employee changes, because of aversion to the topic, or because the direct relationships have become unclear (March and Olsen, 1975). More complex structures allow more diverse information to be recognized and processed. On the other hand, a simple structure may cause decision-makers to ignore many environmental signals and to reject them because they are not recognized (Walsh and Fahey, 1986).

The term 'experience' refers to an organization's contact with the environ- ment which forms the basis for knowledge structures. The notion of complex- ity is based on the assumption that as a firm gains more experience and learns irom it, it will become more expert at what it is doing. As it becomes more expert, the knowledge structure builds on ihe base established by the past experiences. As the knowledge structure becomes more complex, it is able to encompass a greater number of new situations and problems.

The literature on experts supports this notion of the development of complexity in the knowledge structure. McKeithen et al. (1981) suggest that schemas of experts contain more links among the elements and possibly more efficient hierarchies. Lurigio and Carroll (1985) suggest that expert schemas contain more information and become more complex. It would seem that this concept applies also to organizations that learn from their experiences and that this influences how they view their strategic options for the future (Hedberg, 1981; Shrivastava, 1983; Tushman ^̂ a/., 1989).

MANAGEMENT, STRATEGY AND KNOWLEDGE 165

A n alternative assumption about the notion of complexity is based on more sociological theories. The.se suggest that organizations may not learn/rom past experiences per se. Instead the firms may utilize the current experience, interpret it, and develop new views that influence future behaviour ( M a r c h et al., 1991), Hence, the current experiences may not lead to improved perform- ance but to some changed behaviour. In fact M a r c h and Olsen suggest this may be 'superstitious' learning Afhere 'organizational behaviour is modified as a result of an interpretation of the consequences, but the behaviour does not affect the consequences significantly' (1975, p . 139).

Some environmental events will have an impact, be retained a n d be included in the knowledge structure, thus causing a new association to be developed (Meyer, 1982). Simple structures may develop over time into complex ones by the inclusion of new experiences that tie into the knowledge of goals (Schank and Abelson, 1977). D u r i n g this process of development, organizations with simple structures will a t t e m p t to find segments of the environment in which they can operate without being traumatized by too m a n y jolts (Meyer, 1982); or they will ignore the jolts, remaining unaffected by them and maintaining the viability of their structure.

O n the other hand a complex structure allows the recognition and interpre- tation of m a n y signals. Organizations that are diversified face highly uncer- tain environments, or multinationals have more varied experiences than others. They may develop more complex structures t h a t allow them to respond to a variety of environmental signals and that m a y even welcome periodic jolts ( T u s h m a n ?/a/., 1989).

Relatedness

Proposition 4: The organizational knowledge system is likely to have linkages between the core structure and the peripheral structure. The strength of this linkage is the degree of relatedness.

Relatedness represents the interrelationships between the core knowledge structure and the peripheral set. In cognitive psychology, the concept that lends some support to this notion is that of'spreading activation'. Anderson suggests that activation measures the likelihood that a piece of knowledge will be useful at a particular time and also that 'spreading activation amounts to the process of bringing information into working memory' (1983, p. 88). The degree of relatedness will influence the rate at which knowledge is retrieved and also which knowledge is retrieved. As the knowledge structures become tightly linked or coupled, it reduces the need to process every element. This seems to correspond to the 'strong' and 'weak' ideology that Brunsson (1985) identifies. •

If there is strong consensus among the organizational members regarding the core set and the peripheral set, we propose that they are tightly coupled. In this case there would be consensus not only about the basic mission of the firm, but also how to achieve it, and how to interpret environmental signals. On the other hand if there is consensus about the mission but disagreement about how to carry it out the core and peripheral elements would be loosely

166 MA LYLES AND C.R. SCHWENK

coupled. This point is illustrated by tbe following quote from one of tbe members of tbe consulting firm we studied:

We bad organized in a very decentralized way and so some people tbought tbat some of the suggestions that were being made would obviously lead to some changes in the organization structure. We had a lot of people who disagreed over our organization structure?.

We would characterize our sample firm as being loosely coupled. The degree of relatedness between tbe core and peripberal sets influences the firm's susceptibility to environmental cbange.

Proposition 5: When tbere is strong agreement about tbe core set and the peripberal set (tigbt coupling), tbe firm will need to maintain stability.

We propose tbat firms tbat bave a structure tbat is tigbtly coupled need to maintain stability and to avoid jolts. Consequently these firms seek to maintain tbe status quo and to avoid cbange. Tbey will be the forceful organizations Daft and Weick (1984) describe which attempt to cbange environmental factors. Top management in tbese firms avoids adapting tbeir schemas to tbe environment. Nisbett atid Ross (1980) suggest tbat when knowledge is embodied with strongly beld values, people will attempt to discredit uncongenial evidence. In schema theory', this is known as the perseverence effect tbat suggests resistance to change even in the face of contradictory evidence (Fiske and Taylor, 1984). Instead the top manage- ment may attempt to cbange the environment.

Otber firms wbicb are tightly coupled attempt to enact their environment such that they seek or perceive tbe environmental conditions that fit tbeir organizational structure. For example, a firm witb a tightly coupled structure might search for the same market conditions in diffet-ent countries so that it can use the same business methods over and over again. In otber words it searches for tbe conditions that match its knowledge and past experiences, tbus avoiding cbange. Lyles (1988) gives an example of a joint venture sopbisticated firm that utilizes the same tecbniques for joint venturing in every country regardless of tbe local conditions.

Tigbtly coupled firms have tbe greatest rigidity in tbe sense tbat tbey do not bave flexibility in responding to environmental changes. At tbe same time, because tbey attempt to maintain stability, they are most affected by rapidly cbanging environments. Their key decision-makers are interested in tbe reduction of debate or disagreement tbat migbt lead to views that contradict the tightly coupled relationships in the knowledge structure, and tbey prom- ote stability in the structure. Tbe key decision-makers in tightly coupled firms are interested in reducing disagreement because it could cause tbe 'uncou- pling' of the knowledge structure and create trauma for tbe firm.

Proposition 6: Wben there is disagreement (loose coupling), it enhances the firm's ability to adjust to change and to be flexible.

MANAGEMENT, STRATEGY AND KNOWLEDGE 167

Firms with a loosely coupled structure incorporate more disagreement and alternative interpretations of how best to carry out the firm's mission. Changes can be made more easily than in tightly coupled firms and there will be more flexibility in action taking and strategies. Environmental jolts do not affect such firms as greatly as tightly coupled firms (Weick, 1979). These firms can adjust their strategies to deal with changing environmental condi- tions and they can incorporate new information into their structures more easily.

Disagreement about the peripheral set and about the core-peripheral linkage may enhance the firm's ability to adjust to changes and to be flexible. Multiple views may exist and create the acceptance of diversity of action steps. These provide the basis for debate on alternative knowledge schemas about expectations of events and actions. Interestingly Prahalad and Bettis (1986) suggest that the recognition of the importance of strategic variety may be on the decline.

Relationship between Complexity and Relatedness The relationship between complexity and reiatedness is developed in table II. Each quadrant denotes a variation in strategic management determined by the organizational knowledge structure. Each also shows the areas of vulnera- bility.

Proposition 7: Loosely coupled firms with simple structures usually are young, emerging firms.

Quadrant I shows the conditions for a young, emerging firm in which there is widespread agreement about the basic mission, but little agreement about how to achieve it. It is open to different interpretations of the environment

I'ahlc II. Mainicnaiicc t)rknowledge strticlurc

Simple

"omplexity

slriiclure

Complex

Degree of relatedness

Loosely coupled

Adaptation: incremental changes Environmental interaction: some inclusion, re.silience

Strategic orientation: innovative Example: typical new ventures

Adaptation: continuous development Environmental interaction: Inclusion,

self-interested interpretation Strategic orientation: seeks

opportunity frequently in diverse industries

Example: The new General Motors

Tightly coupled

Adaptation: maintenance Environmental interaction: .stimuli

rejection enactment Strategic orientation; seeks single

industry Example: Polaroid

Adaptation: maintenance Environmental interaction: stimuli

rejection enactment Strategic orientation: seeks industry

that fits requirements Example: US Gypsum (USG Corp.)

168 M.A. LYLES AND C.R. SCHWENK

and because of its simple-loosely coupled relationship, it is resilient. It can be risk-taking, innovative, and experimental in its approach to strategic manage- ment as long as its actions are consistent with the simple core structure.

Proposition 8: Firms with a simple structure which is tightly coupled can be effective at strategic management if the environment is not rapidly chang- ing or if they enact their environment.

In Quadrant II are firms that have a simple structure that is tightly coupled to subgoals and beliefs about how to achieve them. Standard operating procedures and/or a strong commitment to the right way of doing things develop - as does a strong emotional commitment. There is frequently a sense of'knowing the right way of managing' that permeates the entire firm. Typical firms in this quadrant would have a single business orientation or would be in environmental niches. Up until recently, Polaroid fell into this quadrant with its strong, but simple, knowledge about its mission coupled with a strong belief about how to conduct its business.

Proposition 9: A complex structure that is loosely coupled leads the firm to more easily adapt to environmental changes.

A complex structure that is loosely coupled as in Quadrant III can perhaps be best illustrated by large firms in diverse product areas with diverse internal management systems and procedures. Choices about strategic opportunities are debated, and there is no consensus about the right methods ol achieving them. Consequently, we propose that environmental events are not filtered out but are likely to be recognized somewhere in the organization by fitting with the diverse structures. These are the firms that Prahalad and Bettis (1986) address as diversifiers.

Proposition 10: In a firm with a complex structure that is tightly coupled, strategic management should focus on maintenance and control of the environment.

The essential point in the last quadrant is that external constraints and rapid changes in the structure can serve to threaten the firm's survival. The best illustration of this type might be IBM who must reject and filter environmental events while attempting to compete in the rapidly changing information services industry. It has a strong knowledge of how best to carry out its strategy and well-developed management systems that need to be protected from change. It enjoys being in the powerful position of being able to enact its environment.

In summary, the present argument proposes that strategic actions are partially determined by the knowledge structure. The degree of complexity may influence a firm's repertoire of the definitions and understandings of how to handle different situations and events. Reiatedness will influence what is retrieved and how quickly: in tightly coupled firms tbere is no need for debate over things that everyone believes to be true. Because organizations are social

MANAGEMENT, STRATEGY AND KNOWLEDGE 169

systems, the degree of consensus is an i m p o r t a n t d e t e r m i n a n t of w h a t is learned and stored in the organizational learning structure.

STRATEGIC CHANGE AND KNOWLEDGE STRUCTURES

The core elements remain relatively stable over time. The patterns of deci- sions and actions represent the interpretation by management of the relation- ship between* the organizational knowledge structure and external reality. Interpersonal interactions partially determine the degree of consensus about the peripheral set.

Hence, the patterns of decisions, the amount of consensus and the organiza- tional structure are management's translation of the knowledge structure (Ranson et al., 1980). It provides them with an initial reference for assessing the environment and determining the meaning of environment cues. This view is supported by Weick whose selection process describes the superim- position of causal maps that provided meaning in the past to current puzzling situations such that the maps would 'provide a reasonable interpretation of what has occurred or they may confuse things even more. These maps are like templates that reveal configurations that may make sense or may not' (1979, p. 131). The relatively stable core set and the peripheral set determine the lens used by management for understanding environmental changes, interac- tions, and chains of events.

Although the core set of knowledge elements and cause-effect beliefs rarely change, it seems that the relationships represented in the knowledge structure do become more complex over time as more elements are incorporated (Narayanan and Fahey, 1982). When simple beliefs about cause-effect are not questioned and are strongly held, they remain unchallenged and widely accepted. They serve as a base for the development of more complex cause- effect elements. The simple relationships that form the foundation become buried and are rarely openly debated.

The more recent elements are the ones that serve as a focus for organizatio- nal debate. These usually do not initially address the relationships in the core. Thus the core set are usually unquestioned, embedded knowledge elements that remain unchallenged unless a major strategic event occurs. The periphe- ral sel, on the other hand, is the area in which organizational debate and alternative viewpoints emerge. The more tightly coupled the core set and the peripheral set, the more difficult it will be for the firm to make changes in its knowledge structure.

IMPLICATIONS AND CONCLUSIONS

As we mentioned at the beginning of the article, strategic management researchers are showing increased interest in the cognitions of the top management team. The theoretical propositions in this article are meant to begin a discussion about the existence and maintenance of organizational knowledge structures. One of the crucial points about these and the difference

170 M.A. LYLES AND C.R. SCHWENK

between an individual's schema is that the organizational knowledge struc- ture is built out of a social process. What we are trying to analyse is how the social process of creating organizational knowledge produces differences among firms. We think that Prahalad and Bettis (1986) began the discussion by addressing dominant logic in a very broad way. We have attempted to address the idea of collective knowledge more specifically in the hopes of identifying ideas to be tested.

Ours is not the only approach for understanding how organizations process information and take actions. More sociologically-oriented researchers have developed approaches which focus more on organizational behaviours and social processes. For example, Sandelands and Stablein (1987) in arguing for the 'concept of organization mind' have drawn analogies between the behaviour of individuals in an organization and neurons in an individual brain. They have also suggested that organizations are capable of parallel processing as is the human nervous system. Mary Douglas, in her book How Institutions Think (1986) has described the ways societies and the structure of institutions affect the cognitions of the individuals within them. She has provided examples of the ways organizational characteristics affect what their members remember and forget. Through an analysis of the textile and wine industries, she showed that changing technologies can affect the ways organi- zational participants classify information.

Though the sociologically-oriented approach certainly provides useful insights into the ways external factors may affect knowledge structures, we think our more psychologically-oriented approach may be more useful in understanding the detailed features of knowledge structures and how these features are related to each other. The approaches are complementary and we hope that future theoretical work will provide an integration.

There is a volume of research being developed that tests the relationships of cognitive elements and organizational behaviours. This research appears to support further work in the area of organizational knowledge structures. Future work may yield new ways of understanding such topics as strategic momentum (Miller and Friesen, 1980; Tushman et al., 1989), organizational learning (Lyles, 1988), and dominant general management logic (Prahalad and Bettis, 1986).

One obvious shortcoming of the previous body of work is that it has dealt only with organizational behaviours, with no attempt at assessing how knowledge is retained or used. Our theoretical proposals about organizational knowledge structures assume that there is a representation of complex events and their interpretations are stored in the collective knowledge of the firm. These knowledge structures may not be complete or accurate. What is stored and what is retrieved may depend on the existing knowledge structure, the events triggering retrieval and/or the interaction of the people. The degree of complexity and relatedness will influence these. Strategic responses to new situations may be the result of generalizing from the existing knowledge structure (Anderson, 1983).

This article is only a first step towards understanding how shared collective knowledge structures develop and shape the strategic actions of the firm. We

MANAGEMENT, STRATEGY AND KNOWLEDGE 171

bope tbat it may stimulate further research which will one day make it possible to understand tbe ways that tbe cognitions of executives affect tbe relationship between environmental change and strategic response.

REFERENCES

ABELSON, R . P . (1981). 'Psychological status of the script concept'. American Psycholog- ist, 36. 715-29.

Ai.BA, J . W. and HASHER, L . (1983). 'Is memory schematic?'. Psychological Bulletin, 9 3 , 2, 203-31.

ANDERSON,J. (1983). The Architecture of Cognition. Cambridge, Mass.: Harvard Univer- sity Press.

ARGYRIS, C . and SCHON, D . (1978). Organizational Learning: A Theory of Action Perspec- tive. Reading, Mass.: Addison-Wesley.

AsTLEV, W. G., AxELSSON, R., BUTLER, R. J., HICKSON, D . J . and WILSON, D . C . (1982). 'Complexity and cleavage: dual explanations of strategic decision-making*.

Journal of Management Studies, 19, 4, 357-75. B.'\Nr)L!R.\, A. (1977). Social learning Theory. Englewood CliiTs, N. J.: Prentice-Hall. BARNARD, C . I. (1938). Functions of the Executive. Cambridge, Mass.: Harvard Univer-

sity Press. BARTLEIT, F . C . (1932). Remembering: A Study in Experimental and Social Psychology.

Cambridge: Cambridge University Press. B.ARiUNEK., J . M. (1984). 'Chatiging interpretative schemes and organizational re-

structuring: the example of a religious order'. Administrative Science Quarterly, 29, 355-72.

BEROER, P . L . and LUCKMANN, T . (1967). The Social Construction of Reality. New York: Ooubleday Anchor.

BotiRGEOis, L. J . III. (1980). 'Performance and consensus'. Strategic Management Journal, 1. 227-48.

BOWER, J . L. (1970). Managing the Resource Allocation Process. Boston, Mass,: Graduate School of Business Administration, Harvard University.

BRUN.SSON, N. (1985). The Irrational Organization: Irrationality as a Basis for Organizational Action and Change. New York: Wiley.

CYERT, R. M . and MARCH, J. G. (1963). A Behavioral Theory of the Firm. Englewood ClifTs, N.J.: Prentice-Hall.

D.\rr, R. L. and WEICK, K . E . (1984). 'Toward a model of organizations as interpretation systems". Academy of Management Review, 9, 2, 284-95.

DAVIS, T . R. and LUTHANS, F . (1980). 'A social learning approach to organizational behavior'. Academy of Management Review, 5, 2, 281-90.

DESS, G . G . (1987). 'Consensus on strategv' formulation and organizational perform- ance: competitors in a fraE;mcnted industry'. Strategic Manai;^ement Journal, 8, 3, 259- 77.

DESS, G . G . and ORIGER, N . K . (1987). 'Environment, structure, and consensus in strategy formulation: a conceptual integration'. Academy of Management Review., 12, 2, 313-30.

DOUGLAS, M . (1986). How Institutions Think. Syracuse, N.Y.: Syracuse University Press.

DuNBAR, R. L. M., DUTTON, J . M. and TORBERT, W . R . (1982). 'Crossing mother: ideological constraints on organizational improvements'. Journal of Management Studies. 19. I, 91-108.

1 72 M.A. LYLES AND C.R. SCHWENK

DtrrroN, J . E. and J.ACKSON, S . E . (1987). 'Categorizing strategic issues: links to organizational action'. Academy of Management Revieiv, 12, I, 76-90.

FAHEV, L . and NARAY.\NAN, V. K. (1986). 'Organizational heliefs and strategic a d a p t a t i o n ' . Proceedings of the National Academy of Management, C h i c a g o , I I I . , 7 — 1 1 .

FAVOL, H . (1949). General and Industrial Management, ir. C. Stotirs. Londoti: Pitman. Fioi., C. M. and LVLES. M . A. (1985). 'Organizatiotial learning'. Academy of Manage-

ment Review, \0, A,^.^-^. FisKii, VV. T. and T.AVLOR, S. E . (1984). Social Cognition. Reading, Mass.: Addison-

Wesley. FORD, J . and HEGARIV, H . (1984). 'Decision makers' beliefs about the causes and

efiects of structure: an explorator\- study'. Academy of Management Journal, 27, 2 7 1 - 91.

GlolA, D. A. and MANZ, C . C . (1985). 'Linking cognition and behavior: a script processing interpretation of vicarious learning'. Academy of Management Review, 10. 3, 327-39.

GiOiA, D. A. and POOI.K, P. P. (1984). 'Scripts in organizational beiiavior'. Academy of Management Review, 9, 3, 449-59.

H.\MBRICK, D. C. (1981). 'Environmental strategy and power within top management teams'. Administrative Science Quarterly, 26, 253-75.

HEDBERC,. B . L . T. (1981). 'How organizations learn and unlearn'. In Nystrom, Paul C. and Starbuck, William H. (Eds.), Handbook of Organizational Design. Oxford; Oxford University Press.

HtDBi'RO, B. L. T., NvsiROM, C. and STARBUCK, W . H . (1976). 'Camping on seesaws? prescriptions for a self-designing organization'. Administrative Science Quarterly, 2 1 , 41-65.

HUFK, A. S. (1983). 'Industry influence on strategy reformulation". Strategic Manage- ment Journal, 4, 3, 119-31.

JEI.INKK, M . (1979). Institutionalizing Innovation: A Study of Organizational Learning Systems. New York: Praeger.

KAHNEMAN, D . , SLOVIC, P. and TVI;RSKY, A. (1982). Judgement Under Uncertainty. Heuristics and Biases. Cambridge: Cambridge University Press.

KorrER, J . (1982). General Managers. New 'Vork: Free Press. KL'HN, T . S . (1962). The Structure of Scientific Revolutions. Chicago: University of

Chicago Press. LEVITT, B . and MARCH. J. G. (1988). 'Organizational learning'. Annual Review of

Sociology, 14, 319-40. LORD, R . G . and FOTI, R . J . (1986). 'Schema theories, information processing, and

organizational behavior'. In Sims, Henry P. Jr., Gioia, Dennis A., and Associates, The Thinking Organization. San Francisco: Jossey-Bass, 20-48.

LURIGIO, A. J . and CARROLL, J . S. (1985). 'Probation officers' schemata of offenders: content, development and impact on treatment decisions'. /oMrna/ of Personality and Social Psychology, 48, 1112-26.

Lvr.ES. M. A. (1988). 'Learning among joint venture sophisticated firms'. Management International Review, 28, 85-98.

LVLES, M . A. and MnROFF, L L (1980). 'Organizational problem formulation: an empirical study'. Administrative Science Quarterly. 25, 102-19.

LYI.ES, M . A. and MiTROFF, L L (1985). 'The impact of sociopolitical influences on strategic problem formulation'. Advances in Strategic Management, 3, 6 9 - 8 1 .

MCKEITHEN, K . B., REITMAN, J . S., RUETER, H^ H . and HIRTLE, S . C . (1981). 'Knowledge organization and skill differences in computer programmers'. Cognitive Psychology, 13, 307-25.

MANZ, C . C . and SIMS, H . R . Jr. (1981). 'Vicarious learning: the influence of

MANAGEMENT, STRATEGY AND KNOWLEDGE 173

modeling on organizational behavior'. Academy of Management Review, 6, 1, January, 105-14.

MARCH, J . G. and O U S E N , J . P. (1975). 'Organizational learning and the ambiguity of the past'. European Journal of Political Research, 3, 147-71.

MARCH, J . G., SPOUIX, L . S. and TAMU?., M . (1991). 'Learning from .samples of one or fewer'. Organization Science, 2, 1-13.

MARTIN, J . (1982). 'Stories and scripts in organizational settings'. In Hastorf, A. and Isen, A. (Eds.), Cognitive Social Psychology. New York: Klsevier-North-Holland, 2 2 5 - 305.

MASON, R . O . and MITROFF, I. I. (1981). Challenging Strategic Planning Assumptions. New York: Wiley.

MF.YFR, A . (1982). 'Adapting to environmental jolts". Administrative Science Quarterly, 2 7 , 5 1 5 - 3 7 .

MILLER, D . and FRIESEN, P. (1980). 'Momentum and revolution in organizational adaptation'. Academy of Management Journal, 23, 591-614.

MINTZBERG, H . (1982). Power in and Around Organizations. Englewood ClifTs, N J . ; Prcntice-Hall.

MiNTZBKRC, H., RAJSINGHANI, D . and THEORET, A. (1976). 'The structure of "unstructured" decision processes'. Administrative Science Quarterly^ 2 1 , 246-75.

NAR.AY.ANAN, V. and FAHEY, L . (1982). 'The micro-politics of strategy formulation'. Academy of Management Review, 7, 25-34.

NisBEiT, R. E. and Ross, L. (1980). Human Inference: Strategies and Shortcomings of Social Judgment. Englcwood-CHfTs, N.J.: Prentice-Hall.

OUCHI, W. G. (1980). 'Markets, bureaucracies, and clans'. Administrative Science Quarterly, 25, 129-41.

Pn-KFER, j . (1981). Power in Organizations. Marshfeld, Mass.: Pitman. pR.\HALAD, C. K. and BEms, R. A. (1986). 'The dominant logic: a new linkage

between diversity and performance'. Strategic Management Journal, 7, 6, 485-501. RANSON, S., HiNiNGS, B. and GREENWOOD, R. (1980). 'The structuring of organizatio-

nal structures'. Administrative Science Quarterly, 25, I, March, 1-17. SANDEi-ANns, I., and STABLEIN, K . (1987). 'The concept ofan organization mind'. In

Ditomaso, N. and Bacharach, S. (Eds.), Research in the Sociology of Organizations. San Francisco: JAI Press.

SCHANK, R. C . and ABELSON, R. P. (1977). Scripts, Plans, Goals, and Understanding: An Inquiry into Human Knowledge Systems. Hillsdale, N.J.: Erlbaum.

SCHWENK, C . (1988). 'The cognitive perspective in strategic decision making'.youma/ of Management Studies, 25, 41-56.

SHR!V.\ST.\VA, P . (1983). 'A typology of organizational learning systems'. Journal of Management Studies, 20, 7-29.

SiiRiVA.sTAVA, P. and Sc:HNEiDr.R, S. (1984). 'Organizational frames of reference'. Human Relations, 37, 10, 795-^809.

SMIRCICH, L . and STUBBART, C . (1985). 'Strategic management in an enacted world'. Academy of Management Review, 10, 724-36.

T.AYLOR, S. E. and CROCKER, J . (1981). 'Schematic bases of social information processing". In Higgins, E. T., Herman, C. P. and Zazza, M. P. (Eds.), Social Cognition. Hillsdale, N J . ; Erlbaum.

THOMPSON, J . D. (1967). Organizations in Action. New York: McGraw-Hill. TUSHMAN, M . L . , VIRANY, B . and ROMANEI.LI, E . (1989). 'Effects of CEO and

executive team succession on subsequent organization peribrmance'. Columbia University Working Paper, February.

WALSH, J . P. and FAHEY, L . (1986). 'The rule of negotiated belief structures in strategy making'. Journal of Management, 12, 325-38.

174 M.A. LYLES AND C.R. SCHWENK

WEICK, K . E . (1979). The Social Psychology of Organizing. Reading, Mass.: Addison- Wesley.

WEICK, K . E . and BOUCON, M . G . (1986). 'Organizations as cognitive maps'. In Sims, Henry P. Jr., Gioia, Dennis A, and Associates (Eds.), The Thinking Organiza- tion. San Francisco: Jossey-Bass, 102-35.

ZALEZNIK, A . (1977). 'Managers and leaders: are they different?' Harvard Business Review, 55, 3, 67-78.