Functional anatomy of a common semantic system for words and pictures

Abstract

The relationship between the semantic processing of words and of pictures is a matter of debate among cognitive scientists. We studied the functional anatomy of such processing by using positron-emission tomography (PET). We contrasted activity during two semantic tasks (probing knowledge of associations between concepts, and knowledge of the visual attributes of these concepts) and a baseline task (discrimination of physical stimulus size), performed either with words or with pictures. Modality-specific activations unrelated to semantic processing occurred in the left inferior parietal lobule for words, and the right middle occipital gyrus for pictures. A semantic network common to both words and pictures extended from the left superior occipital gyrus through the middle and inferior temporal cortex to the inferior frontal gyrus. A picture-specific activation related to semantic tasks occurred in the left posterior inferior temporal sulcus, and word-specific activations related to semantic tasks were localized to the left superior temporal sulcus, left anterior middle temporal gyrus, and left inferior frontal sulcus. Thus semantic tasks activate a distributed semantic processing system shared by both words and pictures, with a few specific areas differentially active for either words or pictures.

The left occipitotemporal system in reading: Disruption of focal fMRI connectivity to left inferior frontal and inferior parietal language areas in children with dyslexia

“…Previous studies investigating the left occipitotemporal cortex have indicated the existence of a hierarchy for visual word processing, progressing from simple letter percept in the occipital cortex to more complex features in the anterior inferior temporal regions (e.g., (Fernandez et al, 2001;Hagoort et al, 1999;Vandenberghe et al, 1996). More recently, a posterior to anterior VWF-System gradient of increasing print specificity was found in adults and adolescents (Brem et al, 2006;Vinckier et al, 2007) as well as in children (Brem et al, 2009;Van der Mark et al, 2009).…”

The neural substrates of inferential and referential semantic processing

“…Most of these studies, however, have focused on the distinction between different types of materials (i.e., verbal vs nonverbal), presented through the visual (Bright et al, 2004;Moore and Price, 1999;Vandenberghe et al, 1996;Vandenbulcke et al, 2006) and auditory (Dick et al, 2007;Thierry et al, 2003;Visser and Lambon Ralph, 2011) modalities. It is clear that the inferential versus referential distinction does not exactly map onto the verbal-nonverbal distinction.…”

“…Following this distinction, several studies have highlighted, in addition to a common left lateralized semantic network, material-specific activations, involving left hemispheric regions for verbal stimuli and right hemispheric regions for nonverbal stimuli (Thierry et al, 2003;Thierry and Price, 2006;Vandenberghe et al, 1996 as reanalyzed by Thierry and Price, 2006;Vandenbulcke et al, 2006). Specifically, some authors found that left middle and superior temporal regions were selectively more involved for verbal material, while the right midfusiform and right posterior middle temporal cortex were selectively more involved for nonverbal processing (Thierry and Price, 2006;Vandenberghe et al, 1996 as in Thierry and Price, 2006; and for converging evidence in patients with neurodegenerative pathologies see Butler et al, 2009).…”

“…Specifically, some authors found that left middle and superior temporal regions were selectively more involved for verbal material, while the right midfusiform and right posterior middle temporal cortex were selectively more involved for nonverbal processing (Thierry and Price, 2006;Vandenberghe et al, 1996 as in Thierry and Price, 2006; and for converging evidence in patients with neurodegenerative pathologies see Butler et al, 2009). …”

“…Our a priori prediction was that referential tasks, constituted by an association between words and pictures, would recruit the RFG, which has been reported to be specific for nonverbal tasks in both control (Thierry and Price, 2006;Vandenberghe et al, 1996 as reanalyzed by Thierry and Price, 2006;Vandenbulcke et al, 2006;Hocking and Price, 2009) and patient studies (Butler et al, 2009). Table 4 reports the peak coordinates localized in the RFG highlighted in the aforementioned studies.…”