Conceptual numerousness judgments by squirrel monkeys.

Two monkeys were reinforced for responding to the card which displayed fewer number of entities (three randomly selected sizes of filled circles) than the other card in any given pair. Area and brightness cues were controlled (at least for the successive numerousness discriminations), as were specific-pattern learning cues. Training proceeded in the order 2 versus 7 (2:7), 2:6, 2:5...2:3, 3:7, 3:6...3:4, etc., until it was judged that the monkeys were unlikely to attain the stringent criteria for discrimination. Both monkeys met criteria on the 7:8 discrimination, and one monkey met criteria on the 8:9 discrimination. It was concluded that the monkeys' numerousness judgments were made on a conceptual basis and that, among nonhuman animals, the evidence for such judgments appears to be limited to apes and monkeys.