Brain activity linked to the filtering of visual information under uncertain conditions

Brain activity linked to the filtering of visual information under uncertain conditions

By Di Catherwood, Graham Edgar and Dritan Nikolla University of Gloucestershire

Chris Alfor, University of the West of England

The aim of this investigation was to assess brain activity associated with patterns of “bias” in decision-making under conditions of uncertainty.

The project employed QASA, a model developed by the investigators, to track both knowledge and bias (i.e., the tendency to accept or reject information) during performance in two category formation experiments.

In one study, associated brain activity was monitored in 12 participants (Ps) using 128-channel EEG (EGI) recording, with Geo-source localisation software (Loreta model) then used to indicate task specific areas of EEG activation.

Visual displays were presented using E-prime software. Feedback was given to establish the target concept or category, and after this was acquired, the target category was suddenly changed. Shift in bias after target change was assessed.

One study (of 47 participants) involved an “abstract” category (a visual pattern defined by a particular colour and shape, etc., such as “red and square”) while the second study involved a simulated military scenario with the target being a flag on a building housing suspected terrorists (eg., “red flag with black cross”) and the task being to correctly identify the suspect flag before “bombing” the building.

In all cases, once the correct criterion was reached, the target was changed suddenly. The behavioural data indicate that with the change in category, Ps display either a “positive” shift in bias (i.e., more restrictive or discerning) or a “negative” shift (i.e., more lax or accepting).

EEG data for the abstract category task indicate that positive bias is linked to more prefrontal activity (frontal polar or orbitofrontal regions: Brodmann areas 10 and 11 respectively) than is the case for negative bias.

The results will be considered in terms of brain systems that filter stimulus-response contingencies and may account for bias patterns. The implications for real-world decision-making under uncertain conditions will be discussed.

©Typologos.com 2011