Word frequency and accumulative proactive inhibition.

Abstract

An interference theory of forgetting which emphasizes PI from linguistic associations predicts that the PI in the recall of successive lists of words should build up more rapidly if the lists are made up of high-frequency words than if made up of low-frequency words. Ss learned 4 lists of 16 pairs to a criterion of 12 correct, each successive list being recalled after 24 hr. The words in the lists learned by 1 group were AA 2-syllable words; for the other group, the frequency was 1 per million. PI increased directly as a function of number of prior lists but did not differ as a function of word frequency; thus, theoretical expectations were not confirmed.

Conservation of meaning as a factor in forgetting new associations.

“…Finally, the present data are consistent with other recent studies (e.g., Underwood, 1967) in failing to find a relationship between word frequency and retention. On the other hand, none of these studies have controlled for m while varying frequency.…”

Retention of verbal discrimination lists as a function of number of prior lists, word frequency, and type of list.

“…Such associations might influence the amount of cumulative interference. In a previous study using paired-associate lists (Underwood & Ekstrand, 1967) the amount of interference was not influenced by word frequency. Still, it is possible that for the verbal discrimination task, interlist associations could influence the amount of cumulative interference, if it occurs.…”

Unit-sequence interference in short-term memory: Facilitation versus interference factors.

“…Correspondingly, since proactive inhibition (PI) for terminal items in lists appears to increase with list length (See Postman et al, 1964), PI might also cumulate because of increasing set length with each successive word recalled. Further, such PI might vary with word frequency (see Underwood & Ekstrand, 1967). Turnage found a Frequency X Time interaction (averaged over all four tests) and a Frequency X Number of Prior Tests interaction whose forms seemed in accord with interference predictions.…”

“…Due to problems involving differences in degree of original learning, the present data are ambiguous on this point. However, Underwood and Ekstrand (1967) reported trends indicating more PI due to repeated testing for LF-L than for HF-H lists in a paired-associate study.…”