James S. McDonnell Foundation
About UsFunding OpportunitiesGrant AdministrationMeetingsFunded Grants
  Home > Grants > Archived Grants > 1998 McDonnell - Pew Program in Cognitive Neuroscience  

 

 
 
  Massachusetts Institute of Technology
Principal Investigator: Jonathan D. Wallis
Cortical Representation of Affective Working Memory


Damage to ventromedial prefrontal cortex (vmPFC) in both humans and monkeys produces changes in emotional and social behaviour. It is not clear whether these changes are due to a difficulty in processing the emotional, or 'affective', content of stimuli, for example whether it is pleasant or unpleasant, or whether they reflect a more basic difficulty in recognising what the stimulus is. It is also not clear how the vmPFC processes information. Certain areas of prefrontal cortex (PFQ are thought to be responsible for holding a mental representation of information in a short-term 'working' memory store, that is a store for information which you are currently processing, or 'working on'. The 'domain-specific processing' hypothesis argues that different areas of PFC are thought to be responsible for holding representations of different sensory 'domains'. For example, there is evidence that dorsolateral PFC is responsible for representing 'where' an object is, whereas ventrolateral PFC is thought to represent 'what' the object is. Since damage to vmPFC impairs emotional behaviour one would predict that vmPFC would represent affective information in working memory. Previous studies, however, emphasized that the vmPFC acts as a long-term memory store responsible for associating stimuli with their reward value. It is possible that neuronal activity related to an affective representation has not been recorded because the appropriate behavioural paradigm has not been used, namely one which requires the monkey to constantly update a mental representation of affective information

The proposed study has two objectives. First, to examine the type of information neurons in the vmPFC encode, i.e. whether they encode information about objects, emotional information, or a combination of the two and whether this information is represented in a working memory store. Second, to examine whether the ventrolateral PFC maintains an object representation and the vmPFC maintains an emotional representation, as the 'domain-specific processing' hypothesis would predict. These objectives will be achieved by recording from neurons in the vmPFC and the ventrolateral PFC of an awake monkey while that monkey is remembering a specific object, and whether or not that object was 'pleasant' or 'unpleasant' (rewarded/unrewarded). Neurons might 'remember' a specific object, the reward, or a combination of the two, i.e. whether a specific object was pleasant or unpleasant. If activity is maintained over a delay period in which the object and reward are no longer present, this will indicate that the monkey is maintaining a representation of the information. Since the external stimulus does not need to be present for its reward value to be internally represented, this provides a mechanism whereby an individual may think about rewards and goals some distance into the future, and plan the appropriate behaviour for obtaining them. This would provide a novel mechanism to explain the profound emotional disturbances which accompany PFC injury. For example, such patients are disinhibited and impulsive, perhaps because they are unable to plan complex behaviour since they cannot maintain a representation of the goal, which might bias their behaviour towards simpler responses directed towards more immediate rewards.

The anatomical localisation of affective representation would have implications for a number of psychiatric disorders thought to be dependent on vmPFC pathology such as depression, obsessive-compulsive disorder, sociopathy and attentional deficit disorder. For example the patient with obsessive-compulsive disorder might be forced to endlessly repeat an unrewarding behaviour because they are unable to hold in mind information regarding future goals, which would allow them to plan an alternative, more satisfying behaviour. Future studies could look at the influence of drugs on the putative neural representation of emotion, which might lead to the development of novel treatment strategies for these disorders.

 
 
   
SearchCopyrightSite Map