Skip to main content

From perception to conception: How the brain processes meaningful concepts

Final Report Summary - PERCEPCON (From perception to conception: How the brain processes meaningful concepts)

How do we understand what we see? The research carried out in the Percepcon grant addressed this central issue of how the brain processes visual objects as meaningful entities; how perception becomes conception. Going beyond vision alone, we developed a feature-based cognitive theory of semantic representations based on large-scale property norm data (the Conceptual Structure account). Cognitive models based on large normative datasets are well-suited to capture statistical regularities within and between concepts, providing both category structure and basic-level individuation. We combined the semantic model with a neurobiological account of visual object processing in the ventral temporal cortex.

In a large set of studies using both time-sensitive magnetoencephalography [MEG] and spatially-sensitive fMRI, we found that visual object recognition involves dynamic processes of transformation from low-level early visual analyses through superordinate category to basic-level conceptual representations. We found that different kinds of semantic representations are developed over time, from early categorical [animals, tools] representations in the fusiform to object-specific [dog, hammer] representations in the perirhinal cortex, suggesting a conceptual hierarchical of processing along the ventral stream similar to the visual hierarchy of processing. Our research shows that feature-based models of object meaning provide a unifying set of principles which account for the different types of semantic representations of objects that evolve over time along the ventral stream.

A review of this work has recently been published by Alex Clarke and Lorraine K Tyler in Trends in Cognitive Science, “Understanding what we see: How we derive meaning from vision”.2015, 19(11): 677-687.