The effect of rehearsal time on the temporal course of forgetting in short-term retention

The study was designed to investigate the effect of rehearsal time and rehearsal instruction on short-term retention of verbally presented consonant triples. Twenty-four male college students listened to recorded consonant triples which they were asked to recall after retention intervals of 0.5, 5.5, 10.5, and 15.5 sec. In 10 experimental conditions during the retention interval subjects were required to count backwards by threes from a recorded digit pair. The digit pair signaling the start of counting occurred immediately (0.5 sec.) after the triple or at delays of 5.5 and 10.5 sec. Once begun, counting continued until recall or until the word 'stop' was heard, in which case the remaining time to recall was 5 or 10 sec. In this manner length of the retention interval, amount of counting, and rehearsal time was varied separately and jointly in the 10 experimental conditions. In four control conditions only length of the retention interval was varied; counting was not required. No significant difference was found between one group which received instruction to repeat the consonant triples silently during rehearsal time and another group which received no specific instruction regarding rehearsal. Retention was found to increase with increasing rehearsal time of 0, 5, and 10 sec. prior to start of counting. This effect was particularly marked when time spent in counting was increased from 5 to 10 sec., and appeared whether length of the retention interval was constant or varied with rehearsal time. Unfilled time after counting but prior to recall was found to have an unexplained depression effect on retention. Results were discussed in terms of response competition theory and consolidation theory. A reformulation of the consolidation hypothesis was suggested involving increased resistance to forgetting of the target trace due to autonomous decay of incidental traces.