Abstract
Recent studies in working memory have suggested that our explicit visual representation is limited to a few objects. However, there are two possible explanations for this limited capacity. One possibility is that the explicit visual representations decay and a few object representations remain after a delay period (memory limit hypothesis). Another explanation is that we explicitly perceive only a fraction of visual objects from the beginning (perceptual limit hypothesis). We tested these two hypotheses by manipulating the delay interval. A sample display, containing six stimuli regularly spaced on an imaginary circle, was presented for 3 s. The stimulus was a bar with orientation of 0°, 45°, 90°, or 135°. A small circle was attached at either side of the bar, creating eight possible orientations of sample stimuli. After 0 or 1 second (perceptual or memory condition, respectively), a cue was presented at one of the sample stimulus locations. Observers pressed a key to indicate the orientation of the sample stimulus at the cued location. Capacity was estimated by using a formula [number of the sample stimuli × (% correct − 12.5) / 87.5]. The measured capacity was 3.2 items in both 0-s and 1-s delay conditions. The measured capacity varied substantially across individuals, ranging from 0.5 to 5.8. Yet, the capacity limits in the 0-s and 1-s delay conditions were highly consistent within individuals (r = 0.96). These results support the perceptual limit hypothesis. The strong correlation between perceptual and working memory capacity suggests that they may be originated from the same limitation.