“…It seems logical, therefore, to predict relatively strong effects for Dangerous-White and Safe-Black trials, but these effects were not clearly evident in the data. To explain this apparent anomaly, it is important to bear in mind that the REC model predicts IRAP effects based on immediate and brief relational responses, not logical reasoning, which typically involves relatively elaborate and extended relational responding (Barnes-Holmes, Barnes-Holmes, & Cullinan, 2001). According to the REC model, therefore, the data from Experiment 2 indicate that frames of coordination (i.e., the verbal relation of equivalence or similarity) between "Safe" and "White" and between "Dangerous" and "Black" were relatively strong, but frames of distinction (i.e., the verbal relation of difference) between "Dangerous" and "White" and between "Safe" and "Black" were not (the word strong is used here simply to denote a high probability in immediate relational responding).…”