Not all same-diVerent discriminations are created equal: Evidence contrary to a unidimensional account of same-diVerent learning

In Experiment 1, we trained four pigeons to concurrently discriminate displays of 16 same icons (16S) from displays of 16 diVerent icons (16D) as well as between displays of same icons (16S) from displays that contained 15 same icons and one diVerent icon (15S:1D). The birds rapidly learned to discriminate 16S vs. 16D displays, but they failed to learn to discriminate 16S vs. 15S:1D displays. In Experiment 2, the same pigeons acquired the 16S vs. 15S:1D task after being required to locate and peck at the odd-item in the 15S:1D displays. Acquisition of the 16S vs. 15S:1D task had little eVect on discriminative performance in the concurrent 16S:16D task, suggesting that a unidimensional entropy explanation for mastery of these two same-diVerent tasks is not viable. During testing, the birds transferred discriminative performance in both tasks to displays composed of diVerent visual stimuli. Such concurrent discrimination learning, performance, and transfer suggest that pigeons are Xexible in the way they process the displays seen in these two same-diVerent tasks. © 2005 Elsevier Inc. All rights reserved. * Corresponding author. Present address: The University of New Hampshire, Department of Psychology, 10 Conant Hall, Durham, NH 03824, USA. Fax: +1 603 862 4986. E-mail address: [email protected] (B.M. Gibson). 0023-9690/$ see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.lmot.2005.06.002 190 B.M. Gibson et al. / Learning and Motivation 37 (2006) 189–208