Fate of irrelevant stimuli in Pavlovian conditioning

This thesis investigated the fate of irrelevant stimuli in Pavlovian conditioning. In Chapter 1, several theories of learning were evaluated with respect to findings related to blocking (Kamin, 1969) and the relative validity effect (Wagner, Logan, Haberlandt & Price 1968). The majority of these theories explain such effects by assuming that little is learned about irrelevant stimuli (e.g. Rescorla & Wagner, 1972 Mackintosh, 1975a Pearce & Hall, 1980). In contrast, the comparator hypothesis (e.g. Miller and Matzel, 1988) makes the assumption that learning about irrelevant stimuli occurs, but is not expressed. The three experiments reported in Chapter 2 tested this assumption using an extended version of the blocking procedure. In each case, an arrangement which, according to the comparator hypothesis, should cause a reversal of blocking failed to produce this result. The findings were, however, consistent with theories which assume that little is learned about irrelevant stimuli. The experiments reported in Chapters 3,4 and 5 were conducted to determine whether little is learned about these stimuli because animals do not attend to them (e.g. Mackintosh, 1975a Pearce & Hall, 1980). Discrimination training designed to measure associability changes was given, using an autoshaping procedure with pigeons, and a Pavlovian conditioning procedure with rats. The results ruled out several non-attentional interpretations, and suggested that an attentional process was involved, which, rather than directing the eyes towards relevant stimuli and away from irrelevant stimuli, operated once all stimuli had been perceived. The results were consistent with the theory of attention proposed by Mackintosh, although it is acknowledged that this theory has its limitations. In the final chapter, two hybrid theories that overcome these limitations were discussed.EThOS - Electronic Theses Online ServiceGBUnited Kingdo