On the interaction of the subject and the experiment in the matching model

Two types of matching designs, static and dynamic, are differentiated. While all matching designs are logically the same in terms of the probability model which determines chance level of performance, an attempt is made to demonstrate that there is an interaction between the tactics, strategies, and actual knowledge ofS and the type of design used which will, in turn, lead to different results even whenS's knowledge is held constant. It is suggested that analogous situations may exist within the framework of the traditional psychometric model.The work reported in this paper was supported by a National Science Foundation Undergraduate Research Participation Grant (NSF-G15797) to the senior author. The authors are indebted to Joel E. Greene for his advice and criticism.