PhD Project Suggestion: Understanding the Primate Frontal Lobe

Howard Bowman


 

The prefrontal lobe is one of the most extensively studied regions in the human and non-human primate brain. Components of it have been variously associated with many different facets of cognition, such as, short-term memory; representing action and thought sequences; maintaining the current task goals; attentional control; inhibitory control; and enabling emotional responses to guide decision-making. However, there remains little absolute consensus on which of these theories is correct. One difficulty is that the neuroscience data obtained from prefrontal cortex can be opaque and difficult to interpret. The central tenet of the proposed PhD is that interpretation of this data can be advanced through neural network modelling.

            The proposed project would be a collaboration with John Duncan, Senior Scientist at the Medical Research Councilís Cognition and Brain Sciences Unit at Cambridge (http://www.mrc-cbu.cam.ac.uk/people/john.duncan/). John has undertaken behavioural, neuroimaging and single-cell recording studies of frontal lobe function over many years. This work has, for example, led to the adaptive coding theory of frontal lobe neurons (Duncan, 2001), which argues that frontal lobe units adjust their effective tuning according to the task being performed.

            A particular direction that the proposed PhD could take is to model Johnís single-cell data using the working memory system proposed in the simultaneous type/ serial token model (Bowman & Wyble, 2007). This theory proposes that working memory encoding is the process of associating perceived objects with episodic contexts through a binding pool. An initial research direction could be to model the dorsolateral and ventrolateral frontal areas that John has studied using the episodic binding mechanism encapsulated in the simultaneous type/ serial token model. A particular strength of this proposed research area is that predictions arising from the modelling could be directly tested in functional Magnetic Resonance Imaging or Electroencephalogram studies.

 

This project would be jointly supervised between Howard Bowman and John Duncan, Senior Scientist at the Medical Research Councilís Cognition and Brain Sciences Unit at Cambridge (http://www.mrc-cbu.cam.ac.uk/people/john.duncan/).

 

Bowman, H. and B. Wyble. (2007) The simultaneous type, serial token model of temporal attention and working memory. Psychological Review, 114(1):38-70

Duncan, J. (2001) An adaptive coding model of neural function in prefrontal cortex, Nature Reviews Neuroscience, 2(11), 820-829