Representation of serial order: a comparative analysis of humans, monkeys, and pigeons

In a serial-order task, subjects are trained to respond to five stimuli in a specific order, (e.g., A-->B-->C-->D-->E) to obtain a reward. Once the serial-order task is learned, a pairwise test is administered in which all ten pairwise combinations of the five stimuli are presented to the subject. Humans and monkeys respond at high levels on all pairwise items, that is, they respond to the items in the pair in the order in which they appear in the original sequence. Given pair BD, for example, humans and monkeys first respond to item B and then item D. Examination of the latencies reveals that humans and monkeys show a first-item effect, a missing-item effect, and a symbolic-distance effect, all in line with the view that they form a representation of the order of the five items and use that representation to guide their behaviour during the pairwise test. In contrast, pigeons perform poorly on the pairwise test, and show no first-item, missing-item, or symbolic-distance effects, suggesting that although pigeons can learn to respond to five items in a specific order, they do not form a representation of the sequential order of those items. We discuss the extent to which the difference in performance between birds and primates on the serial-order task reflects a difference in cognitive abilities, or whether the difference can be attributed to noncognitive factors.