Modality-independent encoding of individual concepts in the left parietal cortex

Abstract

The organization of semantic information in the brain has been mainly explored through category-based models, on the assumption that categories broadly reflect the organization of conceptual knowledge. However, the analysis of concepts as individual entities, rather than as items belonging to distinct superordinate categories, may represent a significant advancement in the comprehension of how conceptual knowledge is encoded in the human brain. Here, we studied the individual representation of thirty concrete nouns from six different categories, across different sensory modalities (i.e., auditory and visual) and groups (i.e., sighted and congenitally blind individuals) in a core hub of the semantic network, the left angular gyrus, and in its neighboring regions within the lateral parietal cortex. Four models based on either perceptual or semantic features at different levels of complexity (i.e., low- or high-level) were used to predict fMRI brain activity using representational similarity encoding analysis. When controlling for the superordinate component, high-level models based on semantic and shape information led to significant encoding accuracies in the intraparietal sulcus only. This region is involved in feature binding and combination of concepts across multiple sensory modalities, suggesting its role in high-level representation of conceptual knowledge. Moreover, when the information regarding superordinate categories is retained, a large extent of parietal cortex is engaged. This result indicates the need to control for the coarse-level categorial organization when performing studies on higher-level processes related to the retrieval of semantic information.

The language network is recruited but not required for non-verbal event semantics

“…The non-causal nature of the language network activation during a non-verbal semantic task has important implications for the study of amodal/multimodal concept representations. A significant body of work has aimed to isolate "amodal" representations of concepts by investigating the overlap between regions active during verbal and nonverbal presentations of a stimulus (Bright et al, 2004;Devereux et al, 2013;Fairhall & Caramazza, 2013;Handjaras et al, 2017;Sevostianov et al, 2002;Thierry & Price, 2006;Vandenberghe et al, 1996;Visser et al, 2012;A. D. Wagner et al, 1997).…”

“…Despite this significant progress in dissociating linguistic and non-linguistic processing, the role of the language network in non-verbal semantics remains unclear. Neuroimaging studies that explicitly compared verbal and non-verbal semantic processing (Devereux et al, 2013;Fairhall & Caramazza, 2013;Handjaras et al, 2017;Vandenberghe et al, 1996;Visser et al, 2012, among others) often reported overlapping activation in left-lateralized frontal and temporal areas, which may reflect the engagement of the language network in non-verbal cognition. However, these studies have typically relied on group analysesan approach known to overestimate overlap in cases of nearby functionally distinct areas (Nieto-Castañón & Fedorenko, 2012) and/or do not report effect sizes, which are critical for interpreting the functional profiles of the regions in question (a region that responds similarly strongly to verbal and non-verbal semantic tasks plausibly supports computations that are different from a region that responds to both, but shows a 2-3 times stronger response to verbal semantics; see, e.g., Chen et al (2017) for discussion).…”

Common spatiotemporal processing of visual features shapes object representation

“…The descriptor (consisting of a vector with 512 elements) of each stimulus was then normalized and compared to each other stimulus using the pairwise correlation distance (1 - Pearson’s r). Second, a shape model was computed.Similarly to previous neuroimaging investigations on the same topic 48,49 , the medial-axis transform 18 was extracted from each manually segmented and binarized object silhouette. Then, shock-graphs skeletal representations were built, and their pairwise dissimilarity was computed using the ShapeMatcher algorithm (http://www.cs.toronto.edu/͂dmac/ShapeMatcher/; Van Eede, et al 50 ), which estimates the minimum deformation needed in order to match two different shapes 51 .…”

Emotionotopy in the Human Right Temporo-Parietal Cortex

“…Voxel-wise encoding 68,69 was performed using a multiple linear regression approach to measure the association between brain activity and emotion ratings, constituted by the six principal components.…”