Abstract
The capacity of visual working memory is limited to representing three simple objects that no longer exists in the external environment. Interestingly, we previously found that this severe capacity limit also applies to representing objects that remained visible (Tsubomi, Fukuda, Watanabe, & Vogel, 2013). Here, we further investigated a critical factor that determines this capacity limit for representing visible and no-longer visible information. To do this, we asked participants to observe eight sample color squares for 1s and then tested their memory immediately or after a 1 second-long blank retention interval with a bicolor test probe providing two alternatives to choose from as the original sample color. Replicating our previous finding, three-item capacity limit was observed for both with and without retention interval. In the critical conditions, we cued the test location prior to the onset of the test probe using two types of cues with varying sample-to-cue ISIs. When a dot cue was presented, we found that the cuing effect was best observed when the cue was presented while the sample color squares were present, and this cueing effect gradually decreased over time after the sample squares disappeared. By contrast, when a color patch cue was presented, we did not observe reliable cuing benefit across all ISIs. These results indicate that the interference caused by the onset of the color patch cue or the bicolored test probe itself is a critical factor that limits visual capacity to three items independent of the retention interval. In other words, only a few object worth of visual information can be maintained robustly against the interference at a given moment regardless of whether the objects are present or not. Meanwhile, there is another type of large-capacity visual representations that are vulnerable to interference and decay over time after the objects disappear.
Meeting abstract presented at VSS 2017