We present a new model of remembering in the context of conditional discrim
ination. For procedures such as delayed matching to sample, the effect of t
he sample stimuli at the time of remembering is represented by a pair of Th
urstonian (normal) distributions of effective stimulus values. The critical
assumption of the model is that, based on prior experience, each effective
stimulus Value is associated with a ratio of reinforcers obtained for prev
ious correct choices of the comparison stimuli. That ratio determines the c
hoice that is made on the basis of the matching law. The standard deviation
s of the distributions are assumed to increase with increasing retention-in
terval duration, and the distance between their means is assumed to be a fu
nction of other factors that influence overall difficulty of the discrimina
tion. It is a behavioral model in that choice is determined by its reinforc
ement history. The model predicts that the biasing effects of the reinforce
r differential increase with decreasing discriminability and with increasin
g retention-interval duration. Data from several conditions using a delayed
matching-to-sample procedure with pigeons support the predictions.