Reward processing and schizophrenia

Despite being central to much behaviour relatively little is known about the biological processes underpinning affective responses and motivational processes. This is despite the fact that a number of psychiatric disorders include symptoms of anhedonia (a decrease in the pleasure produced by rewards), or amotivation (a reduction in the ability of rewards to motivate behaviour). Animal models are vital in investigating the pathology of diseases and the absence of reliable animal models of anhedonia or amotivation is hampering basic pre-clinical research. Schizophrenia is a debilitating brain disorder characterised by disturbed behaviour and abnormal mental functions. It manifests itself as three symptom clusters: 'positive' (additional to normal experience), 'negative' (lacking relative to normal experience) and 'cognitive' (abnormal information processing). Although there are good models of the cognitive and positive symptom clusters there are no reliable animal models of the negative symptoms. It is thus especially interesting that NMDA antagonists can mimic psychotic symptoms in humans and can also disrupt responding for food rewards in experimental animals (e.g. Gilmour, et al, 2009). However, recent research from our laboratory (e.g. Lydall et al, 2010) suggests that these pharmacological manipulations do not actually produce a direct analogue of anhedonia as it relates to the unmediated experience of pleasurable stimuli. The disconnections between these two strands of research suggest that the deficit produced by NMDA antagonists might relate specifically to the way in which animals learn that their actions are reliably connected to the delivery of pleasurable food rewards. It is already known that when an arbitrary response (e.g. pressing a lever) produces a particular food reward then the tendency to press the lever is controlled by the expectation of that food. However, with extended training on such instrumental tasks animals move to being controlled by stimulus-response habits that are not affected by the value of the reward itself (e.g. devaluing the food has no influence the lever press behaviour). It is also known that some frontal lesions and pharmacological manipulations related to physiological changes in schizophrenia influence this transition to habit-based behaviour. Thus, the first strand of this project will examine the ways in which the expectancy of a particular reward develops over instrumental training and how such expectancies influence the hedonic response to the reward itself. Further research will examine the effects of NMDA-based manipulations on the development of expectancies in instrumental behaviour and whether this interacts with the hedonic response elicited by the rewards. Our previous research found no evidence for a deficit in either licking measures of palatability or operant measures of motivation indicating that neither anhedonia nor amotivation is present in NMDA-based models of schizophrenia (see Dr Dwyer's research experience for details of the techniques used). However, other models remain to be investigated. In particular, the methylazoxymethanol acetate (MAM) model relies upon disruption of neurodevelopment resulting in neuroanatomical effects on frontal and temporal cortices resembling those in schizophrenia. Pilot research suggested that this model may produce a better analogue of the negative symptoms. If the number of lever press responses an animal is required to make for a given reward is steadily increased over trials, the point at which they stop working for that food gives an index of the motivational importance of the reward (the progressive ratio test): MAM treated rats demonstrated a reduction in progressive ration responding suggesting a deficit in motivation. Thus, the second strand of this project will examine whether the MAM model produces deficits in hedonic and instrumental responses indicative of a general analogue of anhedonia.