Social Experience and Motivated Cognition Shape the Representation of Concepts: A Behavioural and Functional Neuroimaging Study
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
The ventrolateral anterior temporal lobe (vlATL), including the anterior inferior temporal and fusiform gyri, is the centre of a supramodal and category-general hub for semantic representation. Dorsolateral ATL regions, such as the temporal pole (TP) and anterior middle temporal gyrus (aMTG), have been associated with more specialised semantic function, including social concept processing, although the precise location and laterality of these effects vary across studies, prohibiting firm conclusions. The ‘socialness’ of concepts has been defined in numerous different ways which, along with uncontrolled effects of other semantic dimensions, could explain inconsistencies in previous behavioural and neuroimaging findings. Therefore, we used a new set of validated socialness norms to compare social and non-social concept processing, while controlling for other important semantic variables, including concreteness and valence. Moreover, we report the first investigation of a potential link between socialness and motivation. This was aimed at unpacking the informational content or processes that make social concepts unique, and particularly a putative role of reward processing. We report two pre-registered experiments, one behavioural (N = 89) and one using fMRI (N = 30), examining the orthogonal effects of socialness and motivation on semantic decisions. Consistent with other recent findings, social concepts were processed more accurately than non-social concepts. Left TP activation was driven by social concepts but not non-social concepts, and some evidence suggests there is also right TP selectivity for social concepts. All concept types engaged the left vlATL and bilateral aMTG suggestive of a category-general semantic function. Multivariate pattern analyses found distinguishable local activation patterns associated with social vs non-social concepts in the left TP, aMTG, and vlATL, and for high vs low motivation concepts in the left TP and aMTG. The activation patterns encoding socialness in the aMTG generalized across levels of motivation, suggesting that socialness is encoded independently of motivation- related information. These findings align with multiple representation theories that claim social and reward-related experiences contribute to semantic representation. They also particularly pinpoint the left TP as having a preference for social concepts.