Premium
Theory content and explanatory power of simulation models
Author(s) -
Tuggle Francis D.
Publication year - 1978
Publication title -
behavioral science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 45
eISSN - 1099-1743
pISSN - 0005-7940
DOI - 10.1002/bs.3830230315
Subject(s) - explanatory power , explanatory model , context (archaeology) , computer science , set (abstract data type) , process (computing) , econometrics , function (biology) , metric (unit) , psychology , mathematics , statistics , epistemology , engineering , operations management , paleontology , philosophy , evolutionary biology , biology , programming language , operating system
A concise methodology for the efficient modeling of behavioral phenomena in living systems at the levels of organisms, groups, organizations, societies, and supranational systems is proposed. The methodology helps resolve questions such as: When a model is to be made more complex? What variables are useful ones to add? When should the model development process cease? The methodology is founded upon (1) the precise identification and delineation of the set of behavioral phenomena to be explained, and (2) the development of a meaningful metric of the content or “size” of alternative explanatory models. From these data, measures of explanatory power (relative amount of the phenomena explained) and explanatory yield (the average amount of explanation per unit of theory content) may be derived. Explanatory power is shown to be an increasing function of theory content, and explanatory yield is shown to be a decreasing function of theory content. The methodology is illustrated in detail in the context of six successive simulation models of the cognitive behaviors of a subject solving a job shop scheduling task. The success of the final model in providing a complete description of one large class of the subject's behavior corroborates the usefulness of the methodology.