Opposite Effects of Recent History on Perception and Decision

Abstract

Recent studies claim that visual perception of stimulus features, such as orientation, numerosity, and faces, is systematically biased toward visual input from the immediate past [1-3]. However, the extent to which these positive biases truly reflect changes in perception rather than changes in post-perceptual processes is unclear [4, 5]. In the current study we sought to disentangle perceptual and decisional biases in visual perception. We found that post-perceptual decisions about orientation were indeed systematically biased toward previous stimuli and this positive bias did not strongly depend on the spatial location of previous stimuli (replicating previous work [1]). In contrast, observers' perception was repelled away from previous stimuli, particularly when previous stimuli were presented at the same spatial location. This repulsive effect resembles the well-known negative tilt-aftereffect in orientation perception [6]. Moreover, we found that the magnitude of the positive decisional bias increased when a longer interval was imposed between perception and decision, suggesting a shift of working memory representations toward the recent history as a source of the decisional bias. We conclude that positive aftereffects on perceptual choice are likely introduced at a post-perceptual stage. Conversely, perception is negatively biased away from recent visual input. We speculate that these opposite effects on perception and post-perceptual decision may derive from the distinct goals of perception and decision-making processes: whereas perception may be optimized for detecting changes in the environment, decision processes may integrate over longer time periods to form stable representations.

Serial dependence promotes the stability of perceived emotional expression depending on face similarity

“…Beyond orientation Fritsche et al, 2017;Liberman, Zhang, & Whitney, 2016;Manassi et al, 2017) and other basic features (motion: Alais, Leung, & Van der Burg, 2017;position: Manassi et al, 2018), serial dependence occurs at higher levels of perception as well. We have recently demonstrated that the continuity field is objectselective by showing that the perception of face identity is systematically biased towards identities seen up to several seconds prior, even across changes in viewpoint (Liberman et al, 2014; see also Taubert, Alais, & Burr, 2016;Taubert, Van der Burg, & Alais, 2016;Xia et al, 2016).…”

“…Previous studies have shown that the perception of orientation, numerosity, and other low-level stimulus features is serially dependent-systematically biased (i.e., pulled) towards similar visual input from the recent past (Cicchini, Anobile, & Burr, 2014;Corbett, Fischer, &Taubert, Alais, & Burr, 2016;Xia, Leib, & Whitney, 2016), as well as in feature space (object similarity; Fritsche, Mostert, & de Lange, 2017;Liberman, Fischer, & Whitney, 2014;Manassi, Liberman, Chaney, & Whitney, 2017). The spatio-temporal region over which current object features, such as orientation, are pulled by previously seen features is known as the Continuity Field (CF).…”

Reference Repulsion Is Not a Perceptual Illusion

“…Such a direct perceptual comparison minimizes the influence of working memory and post-perceptual decisions (Fritsche et al, 2017;Schneider & Komlos, 2008) and thus could shed light on the perceptual nature of the reference repulsion bias. To investigate whether discrimination judgments would alter the perceived orientation of a stimulus we computed biases between bins of trials with opposite and correct boundary judgments.…”

“…http://dx.doi.org/10.1101/366393 doi: bioRxiv preprint first posted online Jul. 10, 2018; 2018; Bliss et al, 2017;Fritsche et al, 2017;Papadimitriou et al 2015). Furthermore, the postperceptual nature of this repulsive bias away from an external reference delineates the effect from repulsive perceptual biases away from cardinal orientations, which may act as internal references (Rauber & Treue, 1998;Tomassini, Morgan & Solomon, 2010;Wei & Stocker, 2015).…”

“…Importantly, some of these contextual biases measured in participants' responses arise early during visual processing, for instance during the encoding of the physical stimulus or decoding of the sensory representation (Webster, 2015), whereas others arise at later post-perceptual decisional stages (Firestone & Scholl, 2016;Fritsche, Mostert, & de Lange, 2017), during working memory retention (Bliss, Sun & D'Esposito, 2017;Huang & Sekuler, 2010;Papadimitriou, Ferdoash & Snyder, 2015; Visscher, Kahana, & Sekuler, 2009) or when a motor response is formed (Pape & Siegel, 2016).…”