Abstract
Ambiguous figures such as the Necker cube elicit regular perceptual alternations between two appropriate interpretations. Previous work has shown that higher working memory capacity (WMC) may promote perceptual stability of a currently-perceived interpretation. Specifically, higher-WMC individuals experience fewer perceptual alternations of the Necker cube than lower-WMC individuals (Allen et al., VSS2010). A possible explanation is that interpreting an ambiguous figure requires a form of problem-solving or hypothesis-testing (Gregory, 1972), something that higher-WMC individuals excel at during high-level cognitive tasks (Dougherty & Hunter, 2003). Presenting a Necker cube intermittently rather than continuously dramatically reduces perceptual alternations. This phenomenon, a type of perceptual memory (Leopold et al., 2002), is studied here to test whether higher-WMC individuals experience fewer perceptual alternations than lower-WMC individuals under intermittent-presentation conditions. In other words, does higher WMC predict better perceptual memory? Participants indicated perceptual alternations of a Necker cube during 2-min trials from four conditions. In the Continuous condition, the cube was presented continuously, as in Allen et al. (VSS2010). In three Intermittent conditions, the cube was presented for 1 sec and then disappeared for (i) 1 sec (Short-Intermittent condition), (ii) 5 sec (Medium-Intermittent condition), or (iii) 10 sec (Long-Intermittent condition). Replicating previous results, WMC was negatively correlated with perceptual alternations in the Continuous condition (r=-0.31, p<0.02). Additionally, WMC was negatively correlated with perceptual alternations in the Short-Intermittent (r=-0.29, p<0.03) and Long-Intermittent conditions (r=-0.32, p<0.02). Results from the Medium-Intermittent condition approached significance (r=-0.21, p<0.10). Overall, higher WMC did indeed predict better perceptual memory. The results support a novel role for working memory in a lower-level form of problem-solving than is typically associated with WMC. Moreover, they suggest that working memory aids in disambiguating visual information by maintaining an interpretation of a stimulus even when the stimulus is not seen, and without any instruction to do so.
Meeting abstract presented at VSS 2013