Levinthal, D. A., & March, J. G., A Model of Adaptive Organizational Search, Journal of Economic Behavior and Organization, 2 (1981) 307-333.

This is a model of organizational change through adaptive search for new technologies. It permits the exploration of simultaneious organizational adaptation in search strategies, competences, and aspirations under conditions of enviornmental instability and ambiguity.

A technology is defined as a semi-stable specification of how a organiztional deals with its environment, functions, and prospers. It could be a production, normative, or consitutency function.

The basic premise of the model is that variation in organizational histories reflect the distributional characteristics of simple adaptation in ambiguous environments. It considers five effects seen in organizations:

1. Organizations distinquish between success and failure in allocating resources either to innovation and refinement.

2. Organizations modify search strategies with experience with them.

3. An increase in organizational experience affects the perceive efficiency of alternative search strategies.

4. Aspirations change with performance which affect definitions for success and failure.

5. Experience is confounded by random events and change.

R&D is seen as a search drawing from a distribution of search outcomes. The number of draws is a function of the size of the expenditure.

Organizational experience leads to three kinds of learning. One is adaptation of search strategies. One is improvement in search competencies. Finally they adapt their aspirations. Organizations also react to the environment by rules while modifying them at the same time. The organization has a goal by which it comapres achievement. If the target is higher than performance, the organization searches for solutions to the problem. If the opposite, organization slack accumulates and is channeled into things other than achieving goals (over-qualified personnel, relaxed management control, finding undiscovered improvement, etc.). Slack can help adaptation by maintaining a buffer of improvements against future adversity. In these situations there can also be "slack search" not justified by org. goals that can possible reap rewards to the organization elsewhere.

First order responses are rapid and tied to target/performance mismatch. Second order responses like changes in performance targets, tech opportunities, etc. are much slower. Aspirations adapt to actual performance. Success behavior is repeated. Values of alternatives are clarified through experience too.

Yet experience can be misleading. Organizations learn in confusing environments. When the environment changes must faster than the organization can adapt, superstitous learning can occur.