How pigeons discriminate the relative frequency of events

This study examined how pigeons discriminate the relative frequencies of ev ents when the events occur serially. In a discrete-trials procedure, 6 pige ons were shown one light nf times and then another nl times. Next, they rec eived food for choosing the light that had occurred the least number of tim es during the sample. At issue were (a) how the discrimination was related to two variables, the difference between the frequencies of the two lights, D = nf - nl, and the total number of lights in the sample, T = nf + nl; an d (b) whether a simple mathematical model of the discrimination process cou ld account for the data. In contrast with models that assume that pigeons c ount the stimulus lights, engage in mental arithmetic on numerons, or remem ber the number of stimuli, the present model assumed only that the influenc e of a sample stimulus on choice increases linearly when the stimulus is pr esented, but decays exponentially when the stimulus is absent. The results showed that, overall, the pigeons discriminated the relative frequencies we ll. Their accuracy always increased with the absolute value of the differen ce D and, for D > 0, it decreased with T: Performance also showed clear rec ency primary and contextual effects. The model accounted well for the major trends in the data.