The current thesis aimed to explore the associations between hazardous drinking, automatic processing of alcohol-related cues and behavioural impulsivity. In addition to assessing the impact of these processes on hazardous drinking in isolation, the specific prediction of dual process models of addiction, that the association between automatic cognitive processes and hazardous drinking would be moderated by behavioural impulsivity, was also tested. These general research questions were investigated cross sectionally in young adult student populations (chapter 3), / adolescents (chapter 4), and older adults (chapter 5). The aim of the study described in chapter 6 was to investigate the effect of a priming dose of alcohol on these processes and their association with alcohol-seeking behaviour. Finally, chapter 7 was an investigation into the mediators and moderators of the effects of ego depletion on drinking behaviour from the perspective of dual process models of addiction. Automatic processing of alcohol-related cues predicted drinking behaviour in all cross sectional studies, although the specific aspects of automatic processes that predicted drinking behaviour differed. The effects of the alcohol prime and anticipated effects of alcohol on different measures of cognitive bias were inconsistent, and automatic cognitive processes only predicted ad-lib drinking in non-intoxicated individuals. Impulsivity only had a direct association with drinking within the sample of older adults, and neither the alcohol prime or ego depletion manipulation increased either measure of behavioural impulsivity. There was evidence that the alcohol priming effect was the result of general impairments in executive cognitive functioning. In the cross sectional studies support for dual process models of addiction was only found in the adolescent sample, in which impulsive decision making moderated the association between attentional bias and problem drinking in adolescents. There was evidence that increases in automatic approach responses towards alcohol-related cues following an alcohol prime were the result of impairments in executive cognitive function. Although ego depletion resulted in increased alcohol consumption this was not due to increased behavioural control by automatic cognitive processes. The results from the current thesis offer considerable support for incentive- motivational models of addiction, arid-highlight the importance of the roles of specific aspects of automatic cognitive processing in different samples. As behavioural impulsivity was only associated with hazardous drinking in older adults this suggests that cumulative experience with alcohol is necessary before these processes impact behaviour. Although support for dual process models of addiction was only found in adolescents, previous research would suggest that the predictions of these models are most likely to be found in such samples.