DIFFERENTIAL LOCALIZATION IN BUDGERIGARS - TOWARDS AN EXPERIMENTAL-ANALYSIS OF NAMING

In Experiment 1, 3 budgerigars (Melopsittacus undulatus) were trained with food reinforcement to make low- or high-frequency calls in respon se to different color stimuli, C1 and C2 (a color-naming task), using a gradual response-differentiation procedure and an automatic call-rec ognition system. Thus, a call within a certain frequency band was rein forced in the presence of C1 (''C1 call''), and a call within a differ ent band was reinforced in the presence of C2 (''C2 call''). In Experi ment 2, all 3 budgerigars were trained in a form-to-color matching-to- sample task, alternating trial by trial with either the color-naming t ask (2 birds) or an identity color matching-to-sample task (1 bird). S ample stimuli for the new matching-to-sample task were forms (F1 or F2 ) and comparisons were the same two colors (C1 and C2). Given Sample F 1 or F2, birds had to make a call to produce Comparison Pair C1 and C2 . With F1 as the sample, a peck on C1 was reinforced; with F2 as the s ample, a peck on C2 was reinforced. Although no particular call was sp ecified in the presence of Fl and F2, 2 birds made the C1 call in the presence of Fl and the C2 call in the presence of F2. In Experiment 3, the bird that failed to match form and color calls in Experiment 2 an d another bird were first trained in a color-to-form matching-to-sampl e task: C1 to F3 and C2 to F4. In this task, to produce the comparison pair of forms, a high call (or low for the other bird) was required i n the presence of C1, and a low call (or high) was required in the pre sence of C2. Both birds were then trained with an identity matching-to -sample task in which sample and comparison stimuli were the same two forms, F3 and F4. Trials on the identity task alternated with the colo r-to-form trials. Although no particular call was required in the pres ence of Samples F3 and F4, both birds came to make the C1 call in the presence of F3 and the C2 call in the presence of F4. Our technique pr omises to be useful for the study of emergent vocal relations in budge rigars and other animals.