Academic journal article Attention, Perception and Psychophysics

Working Memory Biasing of Visual Perception without Awareness

Academic journal article Attention, Perception and Psychophysics

Working Memory Biasing of Visual Perception without Awareness

Article excerpt

Published online: 19 October 2013

© Psychonomic Society, Inc. 2013

Abstract Previous research has demonstrated that the contents of visual working memory can bias visual processing in favor of matching stimuli in the scene. However, the extent to which such top-down, memory-driven biasing of visual perception is contingent on conscious awareness remains unknown. Here we showed that conscious awareness of critical visual cues is dispensable for working memory to bias perceptual selection mechanisms. Using the procedure of continuous flash suppression, we demonstrated that "unseen" visual stimuli during interocular suppression can gain preferential access to awareness if they match the contents of visual working memory. Strikingly, the very same effect occurred even when the visual cue to be held in memory was rendered nonconscious by masking. Control experiments ruled out the alternative accounts of repetition priming and different detection criteria. We conclude that working memory biases of visual perception can operate in the absence of conscious awareness.

Keywords Working memory . Interocular suppression . Conscious awareness . Nonconscious processing . Visual prior entry . Top-down modulation

According to the biased competition account of attention (Desimone & Duncan, 1995), an important function of visual working memory (WM) is to provide top-down modulation of sensory processing. A prime candidate for the source of topdown modulation via WM may be the prefrontal cortex (e.g., Desimone, 1996; Gazzaley & D'Esposito, 2007; Gazzaley & Nobre, 2012). Prefrontal feedback from the contents of WM can selectively modulate activity of visual neurons that code relevant stimulus features and therefore facilitate selection of matching over mismatching stimuli in the visual field. Indeed, the tenet that WM-feedback may automatically modulate perceptual selection processes has proven to be the case in a variety of behavioral paradigms. For example, top-down modulation from the contents of WM can guide attention to matching stimuli in the visual field (Bundesen, 1990; Duncan & Humphreys, 1989; Wolfe, Cave, & Franzel, 1989), even when observers have no explicit intention to voluntarily shiftattention to those stimuli (e.g., Carlisle &Woodman, 2011; Chen & Tsou, 2011; Downing, 2000; Huang & Pashler, 2007b; Olivers, Meijer, & Theeuwes, 2006; Pan & Soto, 2010; Soto, Heinke, Humphreys, & Blanco, 2005; though see Downing & Dodds, 2004; Woodman & Luck, 2007, for some opposite results).

A fundamental issue is whether conscious awareness is a necessary condition for WM guidance of perceptual selection to occur. Classically, WM is conceived of as a short-term store and control mechanism that enables the maintenance/ manipulation of critical information to meet behavioral goals, and WM has been assumed to operate on information that is consciously represented (Baddeley, 1986, 2003). Recent research, however, has provided intriguing evidence that WMdependent cognitive operations (including reading, doing arithmetical computations, complex perceptual sequence learning, and delayed visual discrimination) can be performed without awareness of the relevant cues (e.g., Hassin, Bargh, Engell, & McCulloch, 2009; Rosenthal, Kennard, & Soto, 2010; Sklar et al., 2012; Soto, Mäntylä, & Silvanto, 2011). For example, Soto et al. (2011) showed that subliminal visual cues can guide perceptual decision making in a delayed discrimination task, even when intervening visible distractors are presented during the retention interval. A study by Astle, Nobre, and Scerif (2010) also showed that subliminal cues could elicit biases of selection (and could modulate electrophysiological markers of spatial attention such as the N2pc) to items matching the subliminal cue. However, the study by Astle et al. is likely to have involved implicit repetition-priming effects, because observers were not required to actively process the subliminal cues and the role of top-downWMprocesses was not assessed. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.