Mapping visual symbols onto spoken language along the ventral visual stream

Area of Science:

• Neuroscience
• Cognitive Science
• Psycholinguistics

Background:

• Reading transforms visual symbols into sounds and meanings.
• The ventral occipitotemporal cortex (vOT) is crucial for this process.
• Understanding neural representations in vOT is key to decoding reading mechanisms.

Purpose of the Study:

• To investigate neural representations in the ventral occipitotemporal cortex (vOT) supporting the visual symbol to sound/meaning transformation in reading.
• To examine how artificial orthographies and learning impact these neural representations.
• To map the hierarchical organization of reading representations in the vOT.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to measure neural activity.
• Participants learned novel words in artificial orthographies.
• Representational similarity analysis (RSA) was applied to fMRI data to compare neural patterns.

Main Results:

• Right vOT and posterior left vOT showed sensitivity to visual similarity.
• Left vOT encoded letter identity, with representations becoming more position-invariant along a posterior-to-anterior gradient.
• Anterior left vOT exhibited similar patterns for words sharing sounds/meanings across different orthographies.

Conclusions:

• The vOT exhibits a hierarchical organization for reading.
• Representations shift from visual sensitivity to abstract, invariant codes conveying spoken language information.
• This gradient facilitates the transformation of written symbols into meaningful speech.