Answers are grounded in this paper and cite their sources.
Pairs of eight-letter sequences, zero-order or second-order approximations to English, were presented for 50, 125, or 200 msec. The materials were masked 0, 75, or 125 msec. after the offset of the initial presentation. By adding exposure duration to the delay of the masking stimulus, it was possible to compute the total time material was available for processing in iconic memory. Accuracy of report increased monotonically as a function of processing time, and the rate of increase (to an asymptote at about 200 msec) was greater for second-order sequences. In a second experiment, single zero-order and fourth-order approximations to English were presented for 40 msec. and were masked on either the left side (Letters 1-4) or the right side (Letters 5-8) after a delay of 0, 20, 40, 60, or 80 msec. In general, masking the left side reduced accuracy more than masking the right side, but as the delay of mask was increased, accuracy of identification increased and the difference between masking on the left and on the right decreased. Taken together, the results suggest that familiar letter sequences are input to memory more rapidly than are random letter strings and that the input mechanism proceeds sequentially from left to right. (Author)
No takes yet. Share an insight, caveat, or question.
Mewhort et al. (1969) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: