Orientation decoding across cortical layers with high-field 7 Tesla fMRI

Understanding how the brain processes uncertainty in visual information is crucial for unravelling the mechanisms of decision-making. When we perceive things, like the orientation of an object (e.g. is the painting hanging straight on the wall), our brains process this information in complex ways. We want to find out whether changes in how uncertain we feel about our visual decisions are due to the initial sensory input (feedforward), or if they come from our attention and expectations (feedback).

This research project uses high-resolution 7 Tesla fMRI brain imaging to investigate how the brain encodes and processes visual uncertainty across different cortical layers. The study focuses on how participants judge the orientation of visual stimuli and how their brains represent the uncertainty of these judgments. Initial findings from the primary visual area (V1) showed that participants' responses could be decoded from all cortical layers. However, the relationship between uncertainty and behaviour in V1 was not particularly informative.

To build on these preliminary results, the research team aims to extend the analysis to the secondary visual cortex (V2) and potentially higher visual areas. We will use advanced model-based decoding algorithms, specifically the TAFKAP model-based decoding algorithm, to estimate uncertainty and presented orientation from layer-specific fMRI data.

The project's main objectives include implementing and refining the TAFKAP algorithm for estimating uncertainty and presented orientation from fMRI data, analysing the relationship between decoded uncertainty in different cortical layers and behavioural variability, and developing data visualisation pipelines for publication.

MDAP's expertise will be crucial in implementing advanced data analysis techniques, optimising computational efficiency, and creating effective data visualisations. This collaboration will enable our team to extract maximum value from the existing dataset and overcome current analytical limitations.

This cross disciplinary study, combining neuroscience, psychology, and advanced imaging techniques, has significant implications for understanding how the brain processes visual information and makes decisions under uncertainty. It could provide empirical support for predictive coding models and offer insights into neurological or psychiatric disorders affecting sensory processing, decision-making and visual function.