Formulas for deriving the PCA Index Score.

(n) = any particular test session.x¯ = averaged Latency.p| = probability.Legend: The overall PCA Index Score, used for phenotype classification (see Fig. 6), was derived by averaging the individual PCA Scores for Days 4 and 5 of training. The PCA Score for each session was derived by averaging the three individual measures (Response Bias, Probability Difference, and Latency Score) for that particular session. Responses Bias is a proportion of lever presses/food cup entries in relation to the total number of responses. The Probability Difference was derived by subtracting the probability of food cup entries from the probability of lever presses. Latency score was the (averaged) difference between latencies to make food cup and lever responses (divided by the length of the CS duration; in this case 8 s).