Associative and free recall of paired-associates under conditions of response articulation and % OSM partial reinforcement

Abstract

Free and associative (S-R) response recall was compared under % OS~1 partial reinforcement (100%, 50%, 25%) Kausler, 1963; Goulet & 8olso, 1966;Schulz, 1965; for % OSM). However, the results of these studies have been inconclusive in regard to isolating the specific mechanisms by which partial reinforcement has its negative effect. The present experiment attempted to delineate the source of the negative effectof% OSMpartial reinforcement.Two hypotheses have been put forth to explain the effect of % OSM partial reinforcement. Schulz (1965) Psychon. Sci., 1966, Vol. 5 (6) L. R. GOULETWEST VIRGINIA UNIVERSITY noted that the manipulation of % OSM should affect the associative phase of PA learning but not the response learning phase in that Ss' opportunity to learn the responses is not changed. Goulet & 8olso (1966), however, stated that the effect of % OSM, affected response learning indirectly through Ss' decreased attention to responses not paired with stimuli on particular trials. To support the latter hypothesis, Goulet & Solso had Ss learn a PA list where half of the pairs were learned under 25% OSM partial reinforcement, with the remaining half learned under 100% OSM. After four and eight trials Ss were given two unpaced recall tasks, a free recall task involving the response recall without the aid of the stimulus terms and an associative recall task which involved presenting a list of the stimuli to S with instructions to recall the responses and pair them with the appropriate stimulus terms. Replicating previous studies, a pronounced effect of%OSMon associative learning was found. Also, a parallel, negative effect was found for the free response recall measure, an effect not expected in that Ss' opportunity for response learning was not changed by the different % OSM treatments. These results suggested that % OSM, at least in part, affected PA learning indirectly through the response learning stage. The present experiment tested this hypothesis by testing for free recall and associative or S-R recall of responses under conditions where Ss were or were not forced to attend to the responses.Under the conditions where Ss I attention to responses was (W) required, a null effect of % OSM on free recall was expected, with a substantially reduced effect of % OSM on associative recall. Under conditions where Ss' attention to responses was not (WN) required, parallel negative effects of % OSM were expected for both free and associative recall. MethodExperimental Design. 1. Sixty general psychology students were randomly assigned to treatments of a 2 by 2 by 3 mixed factional design, with type of recall (free or associative) and type of attention (W, WN) asbetween-S factors and % OSM (100%, 50%, 25%) as a within-S or mixed-list factor. The recall measures, as mentioned previously, involved having Ss recall all responses in the list without stimulus cues (free response recall) or providing the list of stimulus terms to S with instructions to recall the responses and pair them with the appropriate stimuli. Attention ...

Degree of learning and transfer of training

“…List 2 was practiced for 10 trials or to a criterion of one errorless trial, whichever occurred last. To assure attention to the stimulus terms when the responses were deleted (Goulet, 1966), the Ss were asked to pronounce each stimulus and response when they appeared in the window of the drum.…”

Further data on degree of learning and transfer of training

“…List 2 was practiced for 10 trials or to a criterion of one perfect trial, whichever occurred last. To assure attention to the responses when the stimuli were deleted (Goulet, 1966) the Ss were asked to pronounce all words as soon as they appeared in the window of the drum.…”

“…The present experiment extended that of Goulet (1967) by varying % aSM rather than % aRM. The % aSM methodology involves deleting the stimulus terms on a fixed percentage of the trials, a procedure which retards only associative learning (Goulet, 1966;Schulz, 1965) as long as Ss are forced to attend to the responses on trials where the stimuli are deleted (Goulet, 1966;Goulet & Sol so , 1966). Degree of List 1 practice (learning?)…”