Differential Grouping of Features
Ronald A. Rensink, Cambridge Basic Research, Nissan Technical Center North America, Inc., Cambridge, Massachusetts, USA.

Perception, 29(suppl.): 99-100. [ECVP 2000; Groningen, The Netherlands.]


"Flicker" displays were created where an original and a modified image continually alternated, with brief blanks between them. Each image was an array of simple rectangles, half having one value of a feature (e.g., horizontal), the rest a different value (e.g., vertical). In half the trials, one of the items changed its value; in the remainder, no changes occurred. Observers were asked to detect whether a change was present in each trial.

As the display time in each alternation cycle is increased, search slopes become proportional to display time, with the proportionality constant a direct estimate of attentional capacity (Rensink, 2000, Visual Cognition 7, 345-376). Two display times were examined here: 80 ms and 800 ms. For orientation changes, slopes for the two conditions differed significantly, with the long-display slope corresponding to a capacity of 5.3 items. However, for contrast polarity, slopes did not differ, indicating a capacity of at least 10. It is suggested that visual attention still has a capacity of 4-5 units, but that items of similar polarity are grouped such that they effectively form a single unit. Such grouping was also found for size and color.


Back to main publications list.