Organizational Routines Are Stored as Procedural Memory: Evidence from a Laboratory Study

• Published in: Organization science (Providence, R.I.) R.I.), 1994-11, Vol.5 (4), p.554-568
• Main Authors: Cohen, Michael D, Bacdayan, Paul
• Format: Article
• Language: English
• Subjects: Behavior patterns; Card games; Declarative memory; Experimental psychology; Group Dynamics; Learning; Machine learning; Memory; Observational learning; Observational research; Organization development; Organizational Behavior; Organizational behaviour; organizational learning; organizational memory; Organizational research; procedural memory; Psychology; Research Applications; routines; skills; Social psychology; standard operating procedures; Studies
• ISSN: 1047-7039; 1526-5455
• DOI: 10.1287/orsc.5.4.554

Organizational routines—multi-actor, interlocking, reciprocally-triggered sequences of actions—are a major source of the reliability and speed of organizational performance. Without routines, organizations would lose efficiency as structures for collective action. But these frequently repeated action sequences can also occasionally give rise to serious suboptimality, hampering performance when they are automatically transferred onto inappropriate situations. While the knowledgeable design and redesign of routines presents a likely lever for those wishing to enhance organizational performance, the lever remains difficult to grasp because routines are hard to observe, analyze, and describe. This paper argues that new work in psychology on "procedural" memory may help explain how routines arise, stabilize and change. Procedural memory has close links to notions of individual skill and habit. It is memory for how things are done that is relatively automatic and inarticulate, and it encompasses both cognitive and motor activities. We report an experiment in which paired subjects developed interlocked task performance patterns that display the chief characteristics of organizational routines. We show evidence from their behavior supporting the claim that individuals store their components of organizational routines in procedural memory. If routines are stored as distributed procedural memories, this may be the source of distinctive properties reported by observers of organizational routines. The paper concludes with implications for both research and practice.