Four studies revealed that a 2(1/2)-year-old chimpanzee (Pan), after 6 months of computer-controlled language training, proficiently reads projected word-characters that constitute the beginnings of sentences and, in accordance with their meanings and serial order, either finishes the sentences for reward or rejects them.

Lana, a chimpanzee sophisticated in the language Yerkish, was tested for free recall on lists consisting of from one to eight words randomly drawn from one of three taxonomic categories or on lists consisting of nine words with every third word from a different category. Serial position effects were observed for the four-to eight-item lists, with statistically significant first-item primacy effects on the seven-and eight-word lists and last-position recency effects on the six·, seven-, and eight-word lists. Also, above-chance clustering was observed in the recall of the nine-item categorized lists. These results were compared and contrasted with the serial position and clustering effects obtained in free recall with humans. Several alternative theoretical accounts of the results were considered.This study was designed to examine the free recall capabilities of Lana, a chimpanzee sophisticated in the language Yerkish (Rumbaugh, 1977). Lana was tested to determine if her episodic free recall performance would demonstrate the well established serial position (Murdock, 1962) and category clustering (Bousfield, Cohen, & Whitmarsh, 1958) effects that are obtained with humans. Serial position effects refer to the finding that when adult humans are given a list of, say, 20 words presented sequentially, their recall will be better both for words occupying the first few positions in the list (primacy effect) and for words occupying the last few positions in the list (recency effect) than for words occupying the middle positions of the list (Murdock, 1962). Category clustering refers to the finding that when adult humans are given a sequentially presented list of words that can be grouped together into a number of different taxonomic categories, they will tend to organize or cluster their recall by the categories even when the presentation of the list was not sequenced by category (Bousfield et al., 1958).Serial position effects have previously been investigated with infrahuman subjects in studies employing serial probe recognition tasks (SPR) (Sands & Wright, This research was supported by NIH Grant HD.{)6016, awarded to Duane M. Rumbaugh, and NIH Grant RR.{)0165, awarded to the Yerkes Regional Primate Research Center of Emory University, to whom the authors wish to express their gratitude for their support to the study. Requests for reprints should be sent to James P. Buchanan, Department of Psychology, University of Scranton, Scranton, Pennsylvania 18510. 1980a, 1980b; R. Thompson & Herman, 1977). R. Thompson and Herman (1977) presented tone lists of one to six items to a bottle-nosed dolphin. After hearing the tone list, the dolphin had to classify a subsequent tone as either old (in the list) or new. The probability of recognizing the old sound was highest for the most recent sound presented and dropped off sigmoidally for successively earlier sounds. The dolphin's performance, then, indicated a significant recency effect without any corresponding primacy effect. On the other hand, Wright (1980a, 1980b) report bot...

No abstract

No abstract