The short-term retention of nonhuman primates for a single sample or for two successively presented samples was assessed in four delayed matching-to-sample experiments with delays of .03, 4, 8, 16, and 32 sec. The single sample tasks included one (Experiment 1) or two (Experiment 4) distractor stimuli in the choice set (matching test). In the two successive samples tasks, the animals matched (reconstructed) the order of presentation of two samples with (Experiment 3) and without (Experiment 2) a distractor stimulus. Also, the possible combinations of eight stimuli (four colors and four shapes) were arranged to test the effects of sample set and choice set similarity. Taken together, analyses of the errors indicated that both sample and choice set similarity were significant sources of confusions in delayed matching. Order errors occurred independently of similarity but were a source of forgetting primarily at the longest delays (16 and 32 sec). Two exceptions to the similarity effect (second response errors in Experiment 3 and errors of an inexperienced group in Experiment 4) were observed. Possible reasons for the exceptions and several implications of these findings for theories of short-term memory are discussed.

No abstract

2 experiments were performed to clarify the reason that positive instanges-are superior to negatws'festances in conjunctive concept learning. In conjunction, all positive instances embody both relevant attributes (e.g., red and square). In nonconjunctive concepts, the positive-category includes other kinds of instances (e.g., inclusive disjunction: red or square or both). Using nonconjunctive rules, increasing the proportion of instances embodying both, relevant attributes improved performance with the proportion of positive instances held constant. Changing the proportion of positive instances had little effect when the composition of the positive category was held constant, confirming that it is -instances with both relevant attributes, rather than positive instances as such, that influence S's performance.

Two experiments were performed to determine the effect of sample duration (0.1, 2, and 4 sec), delay interval (.03, 4, 8, 16, and 32 sec), and type of stimulus (color and shape) on the matehing performance of rhesus monkeys. In Experiment I, the 15 possible delay-duration combinations were randomly presented in blocks of 15 trials. In Experiment 2, each duration was held constant and the five delays randomly presented. Then each delay interval was held constant with the three durations randomly varied. Matching performance increased as sample duration increased (ps< .01 and .005), while length of delay did not significantly affect performance. The type of stimuli paired in the matehing test significantly affected performance (ps< .05 and .10) with the shape/shape choices leading to the poorest performance. Stimulus discriminability and amount of training with brief sampie durations were implicated as significant determinants of matehing performance.In recent years, the delayed matching-to-sample (DMTS) task has been used extensively in nonhuman primate memory research. The wide use of the DMTS task to study animal memory is partially due to the noted similarities in task requirements between it and the short-term memory (STM) paradigm used with humans (e.g., Peterson & Peterson, 1959). Also, the effects of amount of interpolated activity, repetition of the stimulus, and stimulus difficulty in the DMTS tasks are generally similar to the effects of these variables on human STM (Jarrard & Moise, 1971). However, neither the animal DMTS nor the human STM literature has been consistent in reports of the effects of stimulus duration. A major aim of the present study was to investigate the relationship between stimulus presentation time and the degree of retention using a DMTS paradigm.The initial task of the subject in a STM experiment most likely involves the perceptual processes of identification or encoding of a stimulus item. Thus, temporal factors in the perceptual stage should strongly influence STM recall accuracy. That is, decreases in stimulus duration should restriet the time for perception and result in poorer recall. Although Aaronson (1967) concluded from her review of the human literature that the time interval between stimuli was more important in STM than stimulus duration per se, her conclusion has not

In two experiments, we explored the utility of using event-related brain potentials (ERPs) evoked during picture recognition to examine the cognitive and neural processes underlying primacy and recency effects. Each experiment consisted of 210 trials in which a recognition probe followed a 12-picture sequence (105 match and 105 nonmatch trials). The 105 match-probe trials consisted of 35 trials in which the probe matched a prime memory set item (Positions 1-3), 35 in which the probe matched a middle memory set item (Positions 6-8), and 35 in which the probe matched a recent memory set item (Positions 10-12). Behavioral results revealed recency but not primacy effects in both experiments. Recent probes, compared with prime and middle probes, evoked ERPs that were more positive from approximately 300 to 400 ms; this enhanced positivity occurred in a positive component peaking around 315 ms and a negative component peaking around 365 ms. These findings fit more closely with the notion of short-term memory as an activation of elements in long-term memory than as a distinct memory store (or stores) separate from long-term memory.