Open field, panel length discrimination by homing pigeons (Columba livia)

Abstract

Pigeons were trained to find a food reward in a square environmental space where two opposing walls had longer panels attached to them, and the other two opposing walls had shorter panels attached to them. Magnitude discrimination theory would lead to the prediction that there would be asymmetrical discriminative performance between experimental groups; rewarded associations with longer panels were expected to yield better correct choice performance than to shorter panels. Interestingly, pigeons trained with food rewards abutting the long panels did not outperform pigeons trained with food rewards abutting the short panels, although discrimination performance in both groups at the end of training was modest. Reversal training following the initial discrimination training provided the strongest evidence of successful initial learning, and more importantly, that there was no difference in learning performance between the two groups. The difference in the performance of pigeons compared to previous studies in rats is most easily explained by the different experimental procedures employed, most notably the use of appetitive (pigeons-food reward) compared to aversive (rats-escape in a water maze) motivation. Alternatively, the data support the hypothesis that in similar training environments pigeons, compared to rats, are more likely to combine panels of different lengths into a more holistic geometric representation of space.