Cognitive and Neural Computations of Semantics

In our everyday lives, we rely on existing relations among elements in our environment (i.e., semantic information) to interact efficiently with the world.

This information can either be used to facilitate understanding by exploiting redundant (congruent) evidence or to signal out salient stimuli by highlighting unexpected (incongruent) elements.

This duality raises fundamental questions about how and when our brains utilize stored semantic knowledge as its influence seems to vary depending on the specific cognitive domain.

This seemingly paradoxical state represents a cognitive puzzle that questions whether the presence of (in)congruent contextual information in a given situation has a positive or negative impact on how we perceive, process, and remember information.CONNECTS seeks to address the paradoxical effects of semantic congruity across various cognitive domains by providing a unified framework.

The proposed framework builds on the Transfer Appropriate Processing principle and brings it to a neural representational level.

By examining the transformations of neural representations, it is possible to quantify the degree of overlap in cognitive computations as a measure of appropriate transfer.

Thus, CONNECTS dissolves the paradox by proposing that performance would be optimal when the required cognitive computation is oriented towards the same stimulus properties emphasized by the semantic (in)congruity of the stimulus.

This proposal not only reconciles conflicting evidence on specific domains but also provides a domain-agnostic framing of the conundrum that ensures its integrative aim.CONNECTS combines a solid theoretical foundation with cutting-edge neuroscientific techniques.

The project's multi-method approach, including behavioural, neural (fMRI and EEG), and computational data from artificial neural networks.

This approach offers a comprehensive exploration of the phenomenon which is a core requirement for a unifying framework.