fMRI activity patterns in human LOC carry information about object exemplars within category

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Neuroimaging

Background:

• The lateral occipital complex (LOC) in the occipito-temporal cortex processes object information.
• Traditional fMRI analysis struggles to differentiate object-specific signals within the LOC.
• Understanding fine-grained object representation is crucial for cognitive neuroscience.

Purpose of the Study:

• To investigate object-related information in fine-grained fMRI activity patterns within the LOC.
• To explore the capabilities of multivariate pattern recognition for analyzing brain activity.
• To compare object discrimination in the LOC with early visual cortex.

Main Methods:

• Utilized functional magnetic resonance imaging (fMRI) with multivariate pattern recognition (support vector classification).
• Analyzed distributed fMRI signals from the LOC and early visual cortex.
• Examined generalization of object information across changes in stimulus size and viewpoint.

Main Results:

• Above-chance discrimination of object categories and within-category exemplars was achieved in the LOC.
• Exemplar-specific information in the LOC generalized across changes in size and viewpoint.
• Early visual cortex also predicted object identity across size and viewpoint changes, but showed a dissociation regarding size discrimination compared to the LOC.

Conclusions:

• Directly evoked fMRI activity patterns in the LOC differ for individual object exemplars.
• Multivariate pattern recognition offers a novel method to study neuronal representations using fMRI.
• The findings highlight distinct roles of the LOC and early visual cortex in object representation and size invariance.