A Detection and Decision Process Model of Matching to Sample
Anthony A. Wright University of Texas Health Science Center, Graduate School of Biomedical Sciences
The topic of this chapter is matching to sample by pigeons and the decision processes that pigeons make when they are choosing between the comparison stimuli. Since the topic of this volume is signal-detection theory (SDT), I should mention at the outset that the model presented in this chapter shares with signal-detection theory a similar detection framework, but the decision aspects and their consequences are quite different.
Matching to sample is a favorite paradigm with which to study animal learning and memory. The diagram in Fig. 9.1 is a schematic of the matching-to-sample procedure. In the first panel (9.1a) the pigeon pecks a sample stimulus (e.g., red) which is the shape of a vertical bar, recessed behind a clear pecking key. The sample goes off, and two comparison stimuli appear behind similar pecking keys on either side of the center sample (panel 9.1b). In this example the matching comparison is on the left and the nonmatching one on the right. In the second panel, the pigeon pecks the matching comparison stimulus and is rewarded with mixed grain.
The focus of this chapter is on the pigeon's decision processes that precede and include the comparison choice. Let us consider several possibilities of how subjects could, in general, go about making this choice. One possibility is that after observing the sample stimulus (making their observing response to it), pigeons could stand back and observe both comparison stimuli simultaneously so that the match, or closest match, is obvious, and then make a decision. But it will be shown that pigeons do not do this. Another possibility is that pigeons could first look at each comparison stimulus individually, and then choose one of them, the one that most closely matches the sample stimulus. But it will be shown that