A NOTE ON MEASURING RECURRENCE

Abstract

Previously reinforced, and later extinguished, responding recurs when environmental conditions change. When studying recurrence, its magnitude (i.e., how much responding recurs) can be assessed by means of absolute and relative measures. These measures, however, highlight different aspects of recurrence that, if not taken into account, may lead to ambiguous interpretations of results. In the present article, first, absolute and relative measures of recurrence are described. Next, interpretations of hypothetical results are presented to illustrate that these measures provide different, but complementary, information about recurrence. That is, absolute response rates in the Test phase and response rates in the Test phase expressed as a proportion of response rates in the immediately preceding phase index how much responding recurs after undergoing extinction. On the other hand, response rates in the Test phase expressed as a proportion of response rates in the Training phase index how closely responding approaches its rate when previously reinforced. Considering the different, but complementary, information provided by each measure might lead to more precise descriptions and interpretations of empirical findings in the experimental analysis of recurrence.