抄録
Two experiments were performed to examine Hulse's cognitive rule-encoding theory and Capaldi's associative memory/discrimination learning theory of serial learning in rats as applied to both acquisition and extinction processes. In Experiment 1, three groups of rats were trained in a runway on a strongly monotonic decreasing series 14-7-3-1-0 food pellets (Group SM) or on a weakly monotonic 14-5-5-1-0 series (Group WM), or on a nonmonotonic 14-1-3-7-0 series (Group NM). Following acquisition, extinction trials were administered. In acquisition, Group SM showed more accurate anticipation of the 0-pellet trial than Group WM, and Group NM did not show any such anticipation. In extinction, Group SM and WM had a tendency to show less resistance to extinction than Group NM. In Experiment 2, two groups of rats were trained on 14-0-7-0-3-0-1-0 series (Group M) or 14-0-1-0-3-0-7-0 series (Group NM), and then they were given extinction trials. In aquisition, Group M showed slow running on the last three trials of the series but Group NM did not show such responding. In extinction, Group M showed less resistance to extinction than Group NM. The results of both experiments are shown to be consistent with the rule-encoding theory rather than the memory/discrimination learning theory.
