Objects seen as scenes: neural circuitry for attending whole or parts

Abstract

words)Depending on our goals, we pay attention to the global shape of an object or to the local shape of its parts, since it's difficult to do both at once. This typically effortless process can be impaired in disease. However, it is not clear which cortical regions carry the information needed to constrain shape processing to a chosen global/local level. Here, novel stimuli were used to dissociate functional MRI responses to global and local shapes. This allowed identification of cortical regions containing information about level (independent from shape). Crucially, these regions overlapped part of the cortical network implicated in scene processing. As expected, shape information (independent of level) was mainly located in category-selective areas Highlights  Modified Navon figures allow dissociation in time of fMRI responses for the global/local levels. Shape-invariant hierarchical level information was found in scenes selective areas, whereas level-invariant shape information was found in object-and faces-selective areas. Level and shape regions were divided by the mid-fusiform sulcus (MFS) in VTC cortex, and each type of region connected into its own pathway. Having separate level/shape pathways could facilitate selective-attention, but foster illusory conjunctions.

Intra-V1 functional networks predict observed stimuli

“…However, coding of shape that is tolerant to changes in size is present in higher-order-visual areas such as LOC (Grill-Spector, 2003). In our previous study, information about shape invariant with respect to level was found in these same object-selective cortices (lateral ventral occipitotemporal and lateral occipital regions, including LOC) (Valdés-Sosa et al, 2020). So why did the invariant classification of letter identity fail?…”

“…(Flevaris et al, 2011, 2014). In a prior study by our group using activation-based MVPA (Valdés-Sosa et al, 2020), information about level (independent from shape) was found in scene-selective cortex (medial ventral occipitotemporal and middle occipital areas). This area could be a source of feedback contributing to the intra-V1 network effect of level.…”

“…Second, intra-subject cross-classification was also accurate for the abstract level, transferring learning across letters identities. Both cross-classifications depend on information not strictly tied to the retinotopic pattern of stimulation (Kaplan et al, 2015; Valdés-Sosa et al, 2020); they are somewhat more abstract than for the specific discriminations. In the case of invariance for level, this implies that the details of line configurations distinguishing ‘E’ and ‘U’ were abstracted away (Hübner & Volberg, 2005).…”

“…The data used here have been previously described in other publications. A summary description is provided below; see Valdés-Sosa et al, 2020 and Valdés-Sosa et al, 2022, for a full explanation.…”

Co-fluctuations of neural activity define intra-V1 networks related to perceptual organization

“…This experiment is described in more detail in (Valdés-Sosa et al, 2020b). The stimuli were projected on a screen at the subject’s feet, viewed through an angled mirror fixed to the MRI head-coil, and were generated using the Cogent Matlab toolbox (http://www.vislab.ucl.ac.uk/cogent.php).…”