Disentangling representations of shape and action components in the tool network

Area of Science:

• Cognitive Neuroscience
• Neuroimaging
• Human Brain Imaging

Background:

• Understanding how humans process tools involves recognizing their shape and how to use them.
• Previous research suggests distinct brain networks for shape and action, but their independent representation remains unclear due to their correlation in daily tools.

Purpose of the Study:

• To disentangle the neural representations of object shape and tool-use actions.
• To investigate whether shape and action are independently represented in the human brain.
• To identify specific brain regions involved in processing distinct components of tool knowledge.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to record brain activity while participants viewed images of tools with orthogonally manipulated shape and action.
• Behavioral similarity judgments were collected to assess perceptions of object shape and action components (motion, hand configuration, movement magnitude).
• Information analysis and representational similarity analysis were applied to neural activation patterns within the tool-preferring network.

Main Results:

• The bilateral lateral occipitotemporal cortex showed strong shape representation but limited ability to differentiate tool-use actions.
• Frontal and precentral regions were primarily involved in representing kinematic action components.
• The left parietal cortex demonstrated coding for both object shape and tool-use actions.

Conclusions:

• The study successfully disentangled the neural representations of object shape from finer tool-use action components within the brain's tool network.
• Findings reveal both dissociable and associative roles for different brain regions in tool processing.
• This research clarifies the distinct contributions of various tool-preferring areas to understanding and interacting with tools.