Two experiments studied the peripheral discriminability of a target differing in its line slope (a tilted T) and in its line arrangement (an L) when presented in briefly flashed displays of upright Ts. The results showed that: (a) an Land a tilted T were equal in discriminability when attention was focused or concentrated on one display position, (b) the discriminability of an L decreased while the discriminability of a tilted T was not statistically significantly affected as the number of display positions that attention needed to be paid to increased, and (c) the reaction time to find a disparate tilted T was less than that to find a disparate L The results are interpreted as supporting the hypothesis that, under distributed attention in peripheral vision, the visual system is more sensitive to differences in line slope than to differences in line arrangement. The results are discussed in connection with hypotheses of how selective attention affects the discrirninab ility of a target.Recently, Beck and Ambler (1972) have reported that a masking flash that controlled the time available to attend selectively to individual letters in a display increased the discr iminab ility of a target differing in its line slopes relative to that of a target differing in its line arrangement. They hypothesized that in peripheral vision differences in line slope are more discriminable than are differences in line arrangement when attention is distributed and not concentrated or focused. This paper presents two experiments designed to compare further the peripheral discriminability of targets differing in their line slopes and in their line arrangements when presented in briefly flashed displays. Experiment I studied how discriminability was affected by prior information regarding the position of a target in a display. Experiment Il investigated the reaction time differences to targets differing in their line slopes and in their line arrangements. Hypotheses of how selective attention affects the discriminability of a target will be considered in the discussion section. EXPERIMENT IThe displays consisted of a circular arrangement of eight letters flashed for 50 msec. An S was given the task of discriminating whether a display contained all upright Ts or a disparate letter _ The disparate letter could be either an L or a tilted T. The lines of an L have the same slopes as the lines of an upright T but differ in their arrangement. The lines of a tilted T have the same arrangement as the lines of an upright T but differ in their slopes. The letters in the display were covered by "This research was supported by National Science Foundation Grant GB-24884. The co mpu ter facilities were made available through ARPA Contract r44620-67C-0099. The paper was written while the first au thor was a visiting member of the Computer Science Center at the University of Maryland. The preparation of the paper was supported in part by National Science Foundation Grant GJ-32258X. 225 masks 50 msec after the offset of the display. A within-Ss design i...

No abstract

No abstract

A familiarity effect in these experiments is defined as a subject's ability to respond more rapidly to a familiar stimulus than to an unfamiliar stimulus. In the first experiment, responding faster to familiar letters (upright) than to unfamiliar letters (inverted) occurred only when the two stimulus types were presented in a random order. These results were interpreted in terms of the effects of processing strategy changes. The second experiment compared the responding of Japanese and American subjects to Japanese and English letters. American subjects responded faster to English letters and Japanese subjects responded faster to Japanese letters. This familiarity effect was obtained even when stimulus presentation was organized by letter type and subjects knew which letter type to expect. The final experiment compared English and Japanese letters in a memory search task. The rate of search for Japanese letters was slower than for English letters. However, no zero-intercept difference was obtained. The evidence indicates that familiarity does not affect an initial encoding process, but it can affect a comparison process.

Two experiments were run in which the effects of rehearsal strategy on memory for spacing and frequency were examined. In both studies subjects were given either free rehearsal instructions or instructions to rehearse only the currently presented item (restricted rehearsal instructions). In accord with predictions concerning the effects of changing the distribution of rehearsals of each item in the list, accuracy of judged spacing was greater and judged frequency was less under restricted rehearsal instructions than under free rehearsal instructions.