March, J. G., & Olsen, Johan, P., The Uncertainty of the Past: Organizational Learning Under Ambiguity", European Journal of Political Research, 3 (1975) 147-171.
Previous models have ignored cognitive and evaluative limits of learning in organizations, relying on more perfectly rational closed loop cycles. Individuals in organizations modify their understanding in a way that is intendedly adaptive even though faced with ambiguity about what happened, why it happened, and whether it is good.
They also specify a theory based on "liking, seeing, trusting, contact, and integration in an organization.
Organizational intelligence comes from either rational calculation about future consequences is used to choose between alternatives, or learning by experience where feedback from previous experience is used to choose among alternatives. Learning happens through experimentation, evaluation, and assessment.
Yet we know that rationality is limited in awareness of alternatives, precision of information, and clarity and consistency of goals. There is not organizational preference function.
Current learning understanding sees is as rational adaptation. While people may not be able to predict consequences, they should be able to understand what happens and why, and know success from failure. Failure to learn is from rigidities and lack of motivation in the organization.
The authors believe that there are also limitations to learning by experience. The goals that define success and failure are often ambiguous or in conflict, what happened is unclear, and causality is complex. People come to believe a particular interpretation of history that impacts future actions (and may in fact be wrong).
The Complete Cycle of Choice
1. The cognitions and preferences held by individuals affect their behavior.
2. The behavior (including participation) of individuals affects organizaitonal choices.
3. Organizational choices affect environmental acts (responses).
4. Environmental acts affect individual cognitions and preferences.
Yet this pervasive model has some significant limitations in certain situations.
Individual Beliefs and Actions
Most organizational theory believes that action follows attitude, and attitude is stable over the course of a choice. Yet people move in and out of choce situations. People devote varying amounts of time and attention to making specific decisions. The capacity for beliefs is much more than the capacity for action.
In situations where time is scarce, people attend to the choice situations most attractive with the highest expected return. Yet any complex social structure can weaken the connection between individual behavior and individual beliefs and preferences. Roles, duties, and obligations are behaviorally important to involvement. People often make decisions because they believe they are expected to. They also have multiple preferences that can conflict with pure self-interest theories. Attention focus rather than utility can explain more of behavior.
Beliefs and preference result from behavior as well (note the cycle above). Motives and intentions are discovered after the fact.
Individual Action and Organizational Choice
In groups, a decision process transforms the behavior of individuals into something that could be called organizational action. Much of modern organizational theory centers around organizational action as the sum of individual actions.
The authors find a loose connection between individual and organizational action. In reality, the formal decision-making process is connected to the maintenance of the organizational as a social unit as well as collective action for substantial results. There is also a more complex, dynamic interaction between particpants, problems, and solutions in an organization. The process is highly influenced by the environment. The dynamic situational forces can produce organizational choice and action not desired by anyone in the organization.
Organizational Change and Environmental Response
The environment responds to organizational choice, though it is also a loose coupling at times. Much of enviromental action has nothing to do with the organization.
Environmental Response and Individual Beliefs
Individuals will interpret environmental response differently from others and even from themselves (in a a different context). Objective reality is seldom achieved. Environmental actions are often ambiguous, hard to observe or properly understand. The environment is too complex for comphension
Implicaitons for Theories of Choice Situations
People are often in a world of which they have only modest control, though they try their best to understand and react to a situation or outcome. The four-point cycle is often broken and loosely coupled. It can be overwhelmed by outside forces.
Learning Under Ambiguity
Choice situations provide occasions for argumentation and interpretation as well as decision-making. The changes in attitudes and beliefs as a result of the decision making is also an important outcome of the process. Organizations can rewrite history and faciliate retrospective learning.
But how do organizations learn? March's model assumes a modification of adaptive rationality - as an organization gains experience, it learns more and more about coping with its environment and formalizes this learning into rules.
Another model by Cyert and March elaborates the complete cycle by:
1. There are a number of states in the organization. At any time, the organization prefers certain states over others.
2. Uncontrollable external disturbances impact the organization.
3. There are also decision variables in the organization manipulated by decision rules.
4. Each combination of external forces and decisions changes the state of the system.
5. A decision rule that leads to a preferred state is likely to be reused in the future. than in the past.
Incomplete Learning Cycles
Role-constrained learning
In this situation individual learning has not impact on behavior. Often this is because role-definition or standard operating procedures inhibit action.
Supersittious Experiential Learning
In this case the cycle proceeds and learning happens, but the organizational actions have little impact on the environmental response. Much of learning in medicine and administrative techniques have this aspect.
Audience Experiential Learning
Here individual action has little impact on organizational action. Learning occurs, but adaptation does not necessarily. Much of learning in politics or research happens this way.
Learning under ambiguity.
Here it's hard to tell what happened or why. Causal connections must be inferred.
Information, Incentives, Cognitive Structures, and Micro-Development
Tradition theory assumes that learning is adaptive -- experience produces wisdom and improved behavior. Yet modern organizations develop myths, legends, and illusions. They develop conflict over myths.
People do modify their behavior in a way that is intendedly adaptive, but they are operating under ambiguous conditions.
Different parts of the organization see different "worlds". Records of actions and history are not completely accurate or comprehensive or retriveable. The variation in organizational memory affects organizational beliefs.
Incentives can also favor one interpretation of a situation than another, or affect the definition of success or failure.
Previous attitudes and beliefs also form the struture for future cognition and can dramatically affect learning and attitude and behavior modification. Learning is a form of attitude formation.
Beliefs are also sensitive to the timing, order and context of information. There is a capacity limit to processing new information and relating it to past information.
An Example -- Seeing, Liking, and Trusting
Organizational participants generally try to make sense of things, imposing order, attributing meanings,a dn providing explanations.
What a person "sees" includes how he defines actions and outcomes, his theories about the world, and his interpretation of those theories. What a person "likes" is his values and tastes and affective sentiments. Futhermore,
He sees what is to be seen.
He likes what is to be liked.
He sees what he expects to see.
He likes what he expects to like.
He sees what he is expected to see.
He likes what he is expected to like.
A organizational participant perceives reality and relates it to past experience (there is a chance at objective reality). Also there are truly things in a person's self-interest that they will like.
Yet people also have expectations of reality that affect what they ultimately see. What they like is largely based on their set of values, attitudes and opinions that are difficult to change.
Social norms also affect what a person sees and likes. Objective reality, attitude structures, social reality, and social norms thus impact on behavior. Yet in situations of ambiguity these factors don't completely agree.
The degree of organizational memory and retrieval can add ambiguity to interpreting past events.
Organizations consist of individuals with:
1. Varying patterns of interaction with each other
2. Varying degrees of trust in each other
3. Varying degrees of integration into the organization (accept responsibility for the organization) The opposite is alienation.
4. Varying orientations to events in the phenomenal world, to the extent that the event is seen, liked, degree of relevance, and extent to which the event was controlled by different individuals.
We assume individuals interpret events with some cognitive consistency. Cognition and attitudes are interelated. Yet we know from social psychology (Bem) that ambiguity, short attention spans, and tolerance for inconsistency can affect this process. From these facts we can propose that:
1. An integrated person will see what he likes, an alienated person will see what he dislikes.
2. An integrated person will like what he sees, an alienated person will dislike what he sees.
3. An person will like what others he trusts like, and dislike what other he distrusts like.
The source of information often determines its credibility. Sentiments diffuse through the contact network characterized by variations in trust. They spread positively across trust relationships, and negative across distrust relationships. A person probably likes what the people he is most in contact with like.
4. A person will see what others he trusts sees (and vice versa). People see what conforms to group perceptions.
5. A person will like those who produce "like" events and prevent "dislike" events.
6. A person will believe the people who he trusts produce "like" events and those he distrusts produce "dislike" events.
7. A person will see events as relevant if he agrees about them with people he trusts and disagrees with people he distrusts.
8. A person will be active to the extent that seeing, liking, and trusting are unambigous.
9. A person will seek contact with people he trusts and avoid contact with people he distrusts.
10. A person will feel integrated into a organization to the extent he like the relevant events that he sees.
Therefore these propositions emphasize the impact of interpersonal connections in organizations and the affective connection between the organization and participant on the development of beliefs.
Conclusion
People act, observe the conseqences of their action, make inference about those consequences, and draw implications for future action. The process is adaptively rational. If the information is accurate, the goals clear and unchanging, and the environment stable, the process will result in improvement over time.