Abstract # 121:

Scheduled for Saturday, June 19, 2010 08:30 AM-08:40 AM: Session 20 (Medallion Ballroom B) Oral Presentation


AGE-RELATED DEFICITS IN REVERSAL LEARNING IN VERVET MONKEYS (CHLOROCEBUS AETHIOPS SABAEUS)

A. P. Grand1,2, K. R. Finnie2,3, J. L. Sousa2,3 and J. D. Jentsch1,2,4
1University of California, Los Angeles, Department of Psychology, Los Angeles, CA 90095, USA, 2University of California, Los Angeles/Wake Forest University Vervet Research Colony, 3Wake Forest University, Department of Comparative Medicine, 4University of California, Los Angeles, Department of Psychiatry and Biobehavioral Sciences
line
     

As part of an extensive pedigree-wide phenotypic characterization, subjects [N=102; 2-24 years of age] were presented with behavioral tests designed to assess the ability to learn a rewarded response (discrimination/acquisition), remember that learned rule (retention), and to subsequently update their behavior when the rule was changed (reversal). Using a custom-made Wisconsin general test apparatus, subjects were presented with 3 boxes fitted with distinct visual stimuli and were trained to select one particular stimulus since it was always associated with a food reward. Subjects were trained on 6 successive visual discriminations and were then tested for their ability to learn, retain and reverse 6 additional discrimination problems. Dependent variables included average trials to criterion for the acquisition, retention, and reversal phases and average number of perseverative errors (selecting the previously rewarded stimulus) during the reversal phase. ANOVAs were conducted to determine differences between age categories (juvenile, young adult, middle-age adult, old-age adult); age effects for trials to criterion for reversal [F(3,92)=4.35, P=0.007] and number of perseverative errors during reversal [F(3,92)=4.10, P=0.009] were found. Post-hoc analyses demonstrated that compared to middle-age and old-age animals, the youngest animals took fewer trials to meet criterion [P=0.002; P=0.016 respectively] and made fewer perseverative errors [P=0.002, P=0.05 respectively]. These results indicate that the ability to update behavioral responses during adaptive learning is much more sensitive to normal aging than is simple feedback-based learning.