Individual differences could explain the failure in transitive inference formation in pigeons using probabilistic reinforcement

In propositional logic, it is stated that for if A is predicated for every B, and B for every C, A must necessarily be predicated of every C. Following a similar logical process, it can be said that If A > B and B > C, then A > C, this is called transitive inference (TI). Piaget developed a verbal task to evaluate TI in children. Subsequent studies adapted this task for animals using a conditioned discrimination between five-terms sequence of stimuli A + B-, B + C-, C + D-, and D + E-. If subjects prefer B over D during test, it is assumed that TI has occurred. In this experiment, we analyzed the effects of task complexity on TI by using a five-terms sequence of stimuli associated with probabilistic outcomes during training, in pigeons. Thus, both stimuli are reinforced in each pair but with different probability, 0.8 for + stimulus and 0.2 for the-stimulus. We found that performance during C + D- pair is impaired and preference in the test pair BD is affected. However, this impairment is dependent on individual differences in performance in C + D- pair. We compare our findings with previous research and conclude that Pavlovian mechanisms, as well as ordering of stimuli, can account for our findings.

Automated summary

This experiment analyzed the effects of task complexity on transitive inference in pigeons. It found that performance in the C + D- pair was impaired and preference in the test pair BD was affected. However, this impairment was dependent on individual differences in performance in the C + D- pair.