Alcoholism is a serious disorder that is extremely difficult to treat. Understanding the neurochemistry of alcoholism will ultimately result in more effective treatment strategies. Recent research has indicated that the neurotransmitter dopamine (DA) codes prediction errors (the difference between expectation and outcome) regarding reward delivery. Prediction error signaling may be involved in the development of alcoholism and also may determine how well treatment-seeking alcoholics can maintain abstinence. The quantitation of the relationship between dopaminergic prediction error signals and alcohol consumption may provide investigators with a useful biomarker for future alcohol research on the interfaces between neurochemistry, cognition, and alcohol addiction. The long-term goal of this work is to increase our understanding of how DA is involved in the development and maintenance of alcoholism. The primary objective of this proposal is to compare the striatal dopaminergic prediction error signal among populations that differ in level of alcohol consumption, and to determine whether nontreatment-seeking alcoholics have different DA prediction error signals across addictive stimuli (i.e., alcohol versus caffeine). We will test the following hypotheses: (1) Compared to social drinkers, nontreatment-seeking alcoholics will have a greater decrease in DA in response to alcohol-related negative prediction error (alcohol administration expected but not delivered), and (2) In nontreatment-seeking alcoholics, alcohol-related prediction error will provoke a greater decrease in DA levels than caffeine-related prediction error. Changes in striatal DA levels will be detected using PET scanning and [11C]raclopride, a DA D2 receptor ligand that is sensitive to changes in endogenous dopamine levels. Twenty social drinkers and twenty nontreatment- seeking alcoholics will receive an IV infusion of either caffeine or alcohol on Day 1. On Day 2, subjects will be scanned during a baseline state and again during a negative prediction error state corresponding to the infusion they received on Day 1 (either caffeine or alcohol is expected, but not delivered). Quantitation of changes in DA levels as a function of prediction error will be achieved by calculating the change in D2 receptor availability during the prediction error scan relative to the baseline scan. The knowledge to be gained from the extension of this work will significantly advance the field of alcohol research by providing novel and important information about how DA prediction error signals are related to alcohol consumption. The neurotransmitter dopamine provides signals that indicate when an outcome does not meet expectation (prediction error). Characterizing the role of dopamine prediction error signal in social drinkers and nontreatment-seeking alcoholics will provide new and important information about the neurochemical processes involved in alcoholism. The knowledge to be gained from this proposal will set the stage for future studies which will use dopamine prediction error signal as a biomarker to investigate (1) why some, but not all, risky drinkers progress to full blown alcoholism, (2) why some alcoholic individuals are incapable of maintaining abstinence, and (3) if this biomarker can be used to predict to which treatment an individual alcoholic is likely to respond.
