Preprocessing on the Basis of Frequency of Occurrence

The release from PI methodology was used to test Oldfield's (1966) suggestion that human memory is organized in part on the basis of frequency of occurrence. An observed release from PI identified frequency of occurrence as an encoding category and provided support for the underlying assumptions of Oldfield's model.

Influence of adult age on primary and secondary memory search

We compared young and elderly adults on the ability to search lists of words stored in primary memory (PM) and in secondary memory (SM). The results of Experiment 1 indicated that age differences in search performance were greatest under SM conditions. Older adults, unlike the young, appeared biased toward responding that probe items were not members of the memory sets stored in SM. As a result of this apparent bias, older adults committed a large number of errors on trials in which the probe was a member of the memorized list (i.e., positive probe trials) yet few errors on trials in which the probe was not a member of the list (i.e., negative probe trials). In addition, the responses of older, but not younger, adults to negative probe trials were found to be more rapid than were those to positive probe trials. In Experiment 2 we examined this pattern of responding and concluded that age differences were involved in the ability to encode memory sets and transfer them from PM to SM as well as in the ability to retrieve information from SM prior to conducting a memory search.

Influence of adult age on primary and secondary memory search

We compared young and elderly adults on the ability to search lists of words stored in primary memory (PM) and in secondary memory (SM). The results of Experiment 1 indicated that age differences in search performance were greatest under SM conditions. Older adults, unlike the young, appeared biased toward responding that probe items were not members of the memory sets stored in SM. As a result of this apparent bias, older adults committed a large number of errors on trials in which the probe was a member of the memorized list (i.e., positive probe trials) yet few errors on trials in which the probe was not a member of the list (i.e., negative probe trials). In addition, the responses of older, but not younger, adults to negative probe trials were found to be more rapid than were those to positive probe trials. In Experiment 2 we examined this pattern of responding and concluded that age differences were involved in the ability to encode memory sets and transfer them from PM to SM as well as in the ability to retrieve information from SM prior to conducting a memory search.

The retrieval of scaled English consonants

Two scales of the perceptual differences among English consonants had been constructed earlier, one from pairs of aural syllables and one from pairs of viewed one-syllable English words. In both instances the pairs of stimuli differed only in the initial consonants. In the present work 276 pairs of syllables were assembled in sets of six. A set was viewed sequentially for 1.5 sec/slide. Then 25 subjects who worked in groups of 4-6 attempted to write one randomly chosen omitted member of each pair on an answer form; another 25 subjects were asked to supply the opposite syllables of each pair. This task was termed "primary memory." Fifty additional subjects repeated the tasks but spent 10 sec counting backward by 3s before trying to recall and write the missing syllables, thus setting up a task of secondary memory. Response forms were scored for right, wrong, and omitted responses. All responses were then compared with each of the scales referred to above, and the mean value for each group of 50 subjects (25 who used the answer form in the ab order + 25 who used the ba order) was computed. The difference between the scores for the two tasks was statistically significant, as was the difference between the scores yielded by the two scales. The "visual" scale appeared to fit the data better than the "aural" scale. The procedures employed and the assumptions that were made yielded results that would not justify an opinion that an error in recall is typically "almost right." However, the data did reflect partial forgetting.

Lack of memory for nonattended items in dichotic listening

Three experiments were conducted to investigate whether semantic information presented to the unattended ear in a dichotic listening experiment has a memory effect after 30 sec. In all three experiments the Brown-Peterson paradigm was employed. Experiment 1 used the release from proactive inhibition technique and found no evidence that semantic material presented to the unattended ear produced proactive inhibitory effects on attended ear material. Experiment 2 suggests that this result was not due to a memory deficiency resulting from a lack of intent to learn. Experiment 3 indicates that repetition of the same information for three trials to the unattended ear did not enhance memory performance when the material was later presented to the attended ear. Assuming that some semantic processing does occur at the time of input, an interpretation based upon the levels-of-processing approach was suggested as a possibility for explaining the negative results.