Comparisons of Sample Stimuli in Delayed Symbolic Matching-to-Sample: Some Results and Implications

Abstract

The relationship between properties of sample stimuli and matching accuracy was examined in four delayed symbolic matching-to-sample problems with pigeons. A peck on the center key of a three-key chamber presented one of two samples. Following one sample a response on the red side key was correct, and following the other sample a response on the green key was correct. In the first problem the temporal properties of a stimulus served as a sample. The correct side key response depended upon whether a yellow sample was “short” (.5 sec) or “long” (4 sec). The second problem compared food and blackout samples, and the third problem compared color and form samples. The fourth problem combined stimulus type and duration by examining.3- and 3-sec color and form samples. The main results were that (a) these sample stimuli produced retention curves that resembled those reported in previous research, and (b) no differences in matching accuracy were observed between.5 and 4-sec stimulus-duration samples, food and blackout, and color and form samples. These data can best be described by the rule statement “If sample A is presented, then peck one side key” and “If sample A is not presented, then peck the other side key.” This analysis suggests that some matching problems may be solved by considering the presence and absence of only one of the two sample types, and that the characteristics of the other sample become irrelevant.