UBC Theses and Dissertations
Toward a theory of two-person interaction Reimer, William C.
In this dissertation, a conceptual framework for the study of interaction between two persons is presented. One specific aspect of that framework is selected and an experimental test which focuses on that aspect is conducted. This test is designed to begin the process of refining the original conceptualization. The conceptual framework utilized stresses the sequential and information-processing features of interaction. The responses of persons are considered to be the result of two processes: one in which an interpretation is made (the "interpretive process"), and one by which that interpretation forms the basis for a new response (the "decision process"). This two-step model of action is used in order to deal with some of the problems created when a simple one-step behavioral model is used to deal with cognitive and linguistic processes. Since sequential interaction is a central concern in this dissertation, the manner in which interpretations or decisions are changed over time is a crucial issue. It is proposed that the "interpretive process" is best accounted for by a threshold type of operation, whereas the "decision process" might best be dealt with by a more simple learning model. These suggestions are made in order to account for some of the resistance to change which the literature on expectations identifies, and at the same time, the flexibility of response which is found in situations of learning. Once this conceptual framework is specified, a more detailed elaboration of the "interpretive process" is begun. Two general types of threshold choice processes are described: one which predicts a change in choice after a run of events of the same type, and the other which predicts a change after the differences between two event types reaches a threshold. An experiment is developed which allows one to differentiate which threshold model best accounts for the choices made. Thirty-five subjects are used and the results support the difference threshold model as the one which accounts for most of the choices. However, the predictive power of the difference model at its maximum is only 84% of the choices made. There is, in addition, some evidence which suggests that the subjects might alter choice models under certain conditions. Finally, several weak points in the conceptual framework are identified, along with suggestions regarding strategies for future research. Refinements of the experimental design which include greater controls on motivating and memory factors are suggested. Such refinements would allow an even stronger test of the threshold models proposed. An alternative suggestion is that the reseach move to an elaboration of the relationship between events and interpretations or an elaboration of the "decision process" itself.
Item Citations and Data