Spatial Learning in South American Opossums and Armadillos

This experiment compares spatial learning in the South American opossum (Didelphis albiventris) and armadillo (Chaetophractus villosus) in a Y-maze. Ss learned to turn to one arm of the maze for food reinforcement. To earn further reinforcements, they had to return to the start-box (after consuming the food) where they were restrained for a fixed intertrial interval. The number of entries to the unbaited arm en route to the goal-box (Ri errors), and in the way back to the start-box (Rf errors) were the dependent measures. Opossums and armadillos did not differ in Ri errors. However, only armadillos mastered the task of returning directly to the start-box after each reinforcement. Moreover, a significant within-session improvement in Rf errors was observed in armadillos, but failed to appear in opossums. Results are discussed in terms of species differences in sensitivity to temporal delays.