Exploring the edges of Pavlovian contingency space: an assessment of contingency theory and its various metrics

Three experiments tested water-deprived rats using a conditioned lick suppression paradigm to evaluate the edges of the contingency space proposed by R.A. Rescorla. Rescorla defined the contingency space based on the probabilities of an unconditioned stimulus in the presence and absence of the conditioned stimulus. However, the change in probability was unable to account for the experimental results without additional post hoc assumptions. Other contingency metrics were examined, of which the phi correlation coefficient accounted for the most variance, 76.4%.