Transfer in perceptual learning following stimulus predifferentiation.

Abstract

experiments on stimulus predifferentiation were conducted to test the acquired distinctiveness of cues and differentiation hypotheses of perceptual learning. In the first, 240 Ss were given a recognition test following stimulus predifferentiation training. Observation training yielded superior recognition of 6-point shapes and distinctiveness pretraining yielded superior recognition of 24-point shapes. The former result was consistent with differentiation theory and the latter with acquired distinctiveness. In the second, 30 Ss received stimulus predifferentiation training followed by a discriminative transfer task. Distinctiveness pretraining yielded greater positive transfer than observation, a result consistent with acquired distinctiveness of cues.

Serial learning of nonsense shapes as a function of instructions controlling labeling

“…If groups given neutral instructions contain fewer labeling Ss and if labels facilitate recognition of complex but not of simple shapes, Group L should learn the high-complexity list with fewer errors than Group N, but Group N should commit fewer errors on the low-complexity list. These predictions follow from earlier results (e.g., Ellis & Muller, 1964). However, if we are correct in claiming that labels are superfluous in this task, then the interaction should not be significant.…”

“…Previous investigators (e.g., Ellis & Muller, 1964) have found an interaction between complexity and type of pretraining (observation and labeling), such that recognition of high-complexity shapes was superior for label pretraining, but recognition of low-complexity shapes was superior for observation pretraining. This two-way interaction was not significant in the present task, although there is a slight trend in that direction.…”

“…In a typical predifferentiation study (cf. Ellis & Muller, 1964), Ss given labeling pre training received a fixed number of paired-associate trials (with feedback) in which to learn a label response (E-supplied) to a shape stimulus. Observation pretraining consists of presenting the shapes for the same number of trials under instructions to inspect and discriminate among them; no feedback is given.…”

Linguistic influences on visual memory

“…The Whorfian hypothesis is better supported by the literature which documents the influence of labeling on memory of visual forms (Ellis, 1968;Ellis & Muller, 1964;Santa & Ranken, 1968. These studies show that providing subjects with unique label training for a set of nonsense shapes can improve recognition of the shapes at least when the stimulus is complicated (Ellis & Muller, 1964), or when there is a large number of stimuli to be maintained (Santa & Ranken, 1972).…”

“…These studies show that providing subjects with unique label training for a set of nonsense shapes can improve recognition of the shapes at least when the stimulus is complicated (Ellis & Muller, 1964), or when there is a large number of stimuli to be maintained (Santa & Ranken, 1972). The effects on recognition are strongest when the labels are in some way representative of the visual stimuli (Ellis, 1968).…”