Orientation decoding depends on maps, not columns.

Area of Science:

• Neuroscience
• Visual Perception
• Brain Imaging

Background:

• The representation of visual orientation in the primary visual cortex (V1) has been studied at a fine spatial scale, linked to columnar architecture.
• Current understanding suggests that decoding stimulus orientation from fMRI data relies on exploiting fine-scale neural organization.

Purpose of the Study:

• To investigate the spatial scale of orientation representation in human V1.
• To provide an alternative explanation for the success of multivariate pattern analysis (MVPA) in decoding orientation from fMRI data.
• To challenge the assumption that decoding relies on fine-scale columnar organization.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to measure brain activity in human V1.
• Analysis focused on identifying topographic maps of orientation preference at different spatial scales.
• Multivariate pattern analysis (MVPA) was employed to decode stimulus orientation from fMRI signals.

Main Results:

• Evidence for a coarse-scale topographic map of orientation preference in human V1 was found.
• This coarse-scale map is aligned with the angular-position component of the retinotopic map.
• The findings suggest that decoding success is explained by this coarse-scale map, not fine-scale columnar structure.

Conclusions:

• A coarse-scale orientation map in V1 offers a parsimonious explanation for successful fMRI decoding of stimulus orientation.
• The interpretation of decoding results should not assume reliance on or exploitation of fine-scale columnar organization.
• This coarse-scale organization provides a more comprehensive understanding of visual information processing in V1.