Co-Authors:
Benjamini, L., Agricultural Research Organization, The Volcani Center, Bet Dagan, Israel
Abstract:
Six naive brown-necked ravens (Corvus corax ruficollis) and four herring gulls (Larus argen- tatus) were tested in oddity learning experiments following the (a) concurrent and (b) con- secutive methods of problem presentation. The experiments included training stages us- ing problems constructed out of a limited number of stimuli, followed by test stages in which new oddity problems were presented. The correction method was used throughout the experiments. The ravens and gulls tested in the experiment using the concurrent method of presenta- tion demonstrated rapid learning of the oddity problems. Oddity learning was tested by the achievement levels attained in the first presentations of new oddity problems, which were between 90 and 92 % correct responses. Achievement levels obtained in the experiment using the consecutive method of presen- tation differed markedly between the species. Analysis of the results showed use of two forms of learning; learning to respond to oddity and learning of visual discrimination be- tween the specific stimuli presented. The extent to which each form of learning occurred differed with the species. The gulls’ achievements in the training stage were due mainly to discrimination learning of each problem individually. Improvement in oddity learning oc- curred only in the test stage, when new stimuli were introduced and frequently exchang- ed. In contrast to the gulls, the ravens’ achievements in the consecutive experiment resulted mainly from oddity learning. Discrimination learning occurred when further im- provement in oddity learning was retarded consequent to the introduction of new test stimuli. The effect of the introduction of unfamiliar stimuli on the improvement in oddity learning was thus inverse in the two species. Considering the results of the two experiments it was concluded that both the gulls and ravens are highly capable of oddity learning, ant that the gulls have a stronger tendency to learn the discriminations between the stimuli presented at the expense of learning to resp- ond to oddity. The occurrence of either oddity learning or discrimination learning by the gulls depended strongly on the experimental procedure. The reliability of the generalization of responding to oddity as a criterion of oddity learning is discussed. Both species showed more efficient generalizability of responding to oddity when new problems were constructed out of familiar stimuli. However, some facts indicated that this phenomenon was partly due to the restriction of stimulus variety in the experimental situation. © 1994, Brill. All rights reserved.
תפריט נגישות
ניגודיות עדינה
מונוכרום
הדגשת קישורים
חסימת אנימציה
פונט קריא
סגור
איפוס הגדרות נגישות
להורדת מודול נגישות חינם
