Does what we remember influence what we see? Effects of visual working memory on perceptual discrimination
Poster Presentation: Saturday, May 17, 2025, 8:30 am – 12:30 pm, Banyan Breezeway
Session: Visual Memory: Objects and features
Schedule of Events | Search Abstracts | Symposia | Talk Sessions | Poster Sessions
Zexuan Niu1 (), Andrew Hollingworth1; 1University of Iowa
Visual Working Memory (VWM) has been proposed to be maintained by sustained activity in early visual cortex. This sensory recruitment hypothesis is supported by evidence of bidirectional influences between VWM and perception. VWM report can be biased systematically by new perceptual information. In addition, VWM content can influence psychophysical thresholds in a perceptual discrimination task (e.g., Teng & Kravitz, 2019), suggesting that VWM content biases perceptual experience. The latter evidence is provocative, as VWM is classically considered a form of visual memory that does not support conscious perception. Here, we examined the possibility that effects attributed to an interaction between VWM and perception are generated instead by interactions within VWM itself. We adapted the Teng and Kravitz (2019) method, in which participants completed a color discrimination task while holding a color in VWM. In the original study, the discrimination stimuli were spatially separated, and VWM may have been required to maintain color information across serial shifts of focal attention between stimulus locations, allowing for biases within VWM to influence discrimination performance. Here, we systematically varied the spatial separation between discrimination stimuli. When the spatial separation was relatively large, we successfully replicated the original effect indicating that the VWM color influenced discrimination thresholds. As spatial separation was reduced, reducing the demand to shift attention between locations, the effects of VWM on discrimination thresholds were likewise reduced. Crucially, when the discrimination stimuli abutted—so that discrimination could be achieved by attending to the color-contrast border, without the need to switch attention between locations—the effects of VWM on discrimination thresholds were eliminated. Together, our results indicate that effects of VWM on perceptual discrimination thresholds arise only when VWM is needed to support spatially separated comparisons and that, at least for the present task, VWM content does not directly alter perceptual experience.