occur. A substantial number of exemplars is probably a universal requirement. Other requirements will likely depend on the concept to be learned and the particular species learning the concept. In the case of monkeys learning a same/different concept with auditory stimuli, contact with the sound source and possibly the fading procedure may have been crucial in addition to the large number of exemplars.
The message here is that we need to be more docile to any specific cognitive requirements of our species. We need to try out different procedures, and even different species. The pigeon, for example, may have been a bad choice of species to explore cognitive and memory tasks. In our well-controlled experimental environments the pigeon readily develops adverse reactions to changes, and one of these changes is novel transfer stimuli. Pigeons seem to be predisposed to learning responses to the absolute properties of individual stimuli rather than to the relative properties among stimuli. Many, or even most, cognitive tasks are relational ones that require training and learning against the predispositions of the pigeon. As our animal cognitive/memory tasks become better tuned to take advantage of the species predispositions and functional incompatibilities from procedures are minimized ( Sherry & Schacter, 1987), notions of capacity limitation for certain types of cognitive learning such as concepts may all but disappear, which should put even more pressure on the question of species' cognitive/intelligence differences.
Preparation of this article was supported in part by grants MH 35202 and MH 42881 to the author.