Abstract
The perceived direction of motion (e.g., up/down) in an ambiguous third-order motion stimulus can be changed by instructions to attend to a particular color (Blaser et al., 1999) or by prior practice in a color-search task (Tseng et al., 2000). In these experiments, subjects performed thousands of consecutive trials attending to only one color. Tseng found that sensitization to that color survived for a month. Is there attentional sensitization when observers shift attention between colors every N trials, N = [1, 200].
Procedure. In our third-order ambiguous-motion paradigm, even frames contain red/green isoluminant gratings, odd frames contain high/low contrast texture gratings. Apparent motion is determined by figure-ground, i.e., the movement of salient areas. Salience is determined by the difference from the gray background—areas of high contrast or of high color saturation have greater salience. Attention to a color produces a change in motion-direction perception that is equivalent to an increase in saturation, i.e., an increase in salience. Three different attend cues for each trial were used: letters, color patchs, and spoken color names. After observers attended to red stripes, attention was switched to green every N trials, and vs vs. The results were compared to the no-instruction condition.
Results. Most observers failed to perceive motion above 8 Hz. Below 4 Hz, all modes of instruction and all values of N produced shifts in observers' psychometric functions equivalent to increasing the saturation of the attended color by 10–20%. This effect of “fast,” voluntary attention, while highly significant, is half of what we measured with prolonged attention to the same color.