Top-Down Feedback Controls the Cortical Representation of Illusory Contours in Mouse Primary Visual Cortex

Abstract

Visual systems have evolved to recognize and extract features from complex scenes using limited sensory information. Contour perception is essential to this process and can occur despite breaks in the continuity of neighboring features. Such robustness of the animal visual system to degraded or occluded shapes may also give rise to an interesting phenomenon of optical illusions. These illusions provide a great opportunity to decipher neural computations underlying contour integration and object detection. Kanizsa illusory contours have been shown to evoke responses in the early visual cortex despite the lack of direct receptive field activation. Recurrent processing between visual areas has been proposed to be involved in this process. However, it is unclear whether higher visual areas directly contribute to the generation of illusory responses in the early visual cortex. Using behavior, in vivo electrophysiology, and optogenetics, we first show that the primary visual cortex (V1) of male mice responds to Kanizsa illusory contours. Responses to Kanizsa illusions emerge later than the responses to the contrast-defined real contours in V1. Second, we demonstrate that illusory responses are orientation-selective. Finally, we show that top-down feedback controls the neural correlates of illusory contour perception in V1. Our results suggest that higher-order visual areas may fill in the missing information in the early visual cortex necessary for illusory contour perception.

The promise of layer-specific neuroimaging for testing predictive coding theories of psychosis

“…1a&b). Laminar fMRI can be used to study visual illusions, such as the Kanizsa illusion, for which it has been shown that illusory contours are signalled in the deep layers of V1 (Kok et al, 2016) in line with subsequent animal studies (Pak et al, 2020). Individuals with schizophrenia tend to be more resistant to various illusions such as the Kanizsa illusion (Spencer et al, 2003(Spencer et al, , 2004; but also see Foxe et al, 2005).…”

Higher-order visual areas broaden stimulus responsiveness in mouse primary visual cortex

“…While areas such as LM, AL and PM are all considered HVAs, LM is thought to be part of the mouse ventral stream, while both AL and PM are attributed to the dorsal stream (Wang et al, 2012). Thus, the roles of area LM versus AL and PM in modulating activity in V1 might be different, also in view of the specific functions of LM in higher order visual processing, such as illusory contour detection and shape categorization (Khastkhodaei et al, 2016; Matteucci et al, 2019; Pak et al, 2020; Tafazoli et al, 2017). Previous studies indicated that inactivation of LM primarily affects the superficial layers of V1 (Pafundo et al, 2016); conversely, we observed that inactivation of AL or PM elicited a modulation that was similar for all the major subpopulations of V1.…”

“…Several studies have investigated what functions each of these areas might fulfill in visual processing. A wide range of functions has been found, complementing V1 in orientation discrimination and contrast detection (Jin and Glickfeld, 2020), spatial integration (Murgas et al, 2020), perception of higher-order visual features (Khastkhodaei et al, 2016) and illusory contours (Pak et al, 2020), object and shape representations (Matteucci et al, 2019; Tafazoli et al, 2017). Furthermore, HVAs partially overlap with the rodent posterior parietal cortex, and have been implicated in several functions beyond simple visual processing, for instance multisensory integration (Meijer et al, 2020; Olcese et al, 2013; Song et al, 2017), (multi)sensory evidence accumulation and decision making (Erlich et al, 2015; Hanks et al, 2015; Licata et al, 2017; Raposo et al, 2014) and navigation (Harvey et al, 2012; Krumin et al, 2018).…”

Postnatal Development of Visual Cortical Function in the Mammalian Brain

“…The visual system's ability to group discrete elements with similar orientation into a continuous contour that can be detected is known as contour integration (Field et al, 1993). Neural correlates of contour integration include horizontal connections and feedback connections (Taylor et al, 2014;Pak et al, 2020). Earlier studies based on computational network models proposed to understand illusory contour responses suggest that neurons in higher order visual areas such as V2 might pool orientation selective feedforward inputs from lower order areas like V1 (Mignard and Malpeli, 1991).…”