Do ventral visual category-selective regions engage during passive comprehension of verbal scene descriptions?

Determine whether functionally specific regions along the human ventral visual pathway (e.g., the fusiform face area and the parahippocampal place area) are engaged during passive comprehension of verbal descriptions of visual scenes, in the absence of explicit mental imagery instructions.

Background

The paper proposes that deep language understanding requires exporting information from the core language system to extra-linguistic brain regions. For perception, prior work shows that category-selective visual areas (e.g., face- and place-selective regions) are activated when participants are explicitly instructed to imagine percepts.

However, it remains unresolved whether these functionally specific ventral visual regions are also recruited during natural, passive language comprehension that describes visual scenes without instructing imagery. Establishing this would provide direct evidence for exportation to perceptual systems during everyday language understanding.