Fish as a Natural Category for People and Pigeons1

Publisher Summary This chapter describes and compares the result of three experiments, one on human subjects and the other on pigeons, and draws conclusions about categorization in light of the results. The chapter establishes rankings of the photographs for human subjects, using reaction time to quantify the discriminability or acceptability of the fish or non-fish in a collection of 35-mm slides. Twelve human subjects responded to 160 underwater photographs to indicate whether they saw a fish or not. Four pigeons rapidly learned to sort underwater photographs being seen for the first time according to the presence or absence of fish in them. For a quasi-concept, a set of instances sorted according to the presence or absence of fish divided randomly, with as many fish in the negative class as in the positive. Pigeons learned to sort 80 stimuli into two categories. They learned quicker when the categories corresponded to fish versus non-fish than when the same stimuli divided into two arbitrary categories. Seven out of seven pigeons in two experiments learned to sort underwater photographs on the basis of the presence or absence of fish.