Dopaminergic signaling of uncertainty and the aetiology of gambling addiction

Although there is increasing clinical recognition of behavioral addictions, of which gambling disorder is the prototype example, there is a limited understanding of the psychological properties of (non-substance-related) behaviors that enable them to become 'addictive' in a way that is comparable to drugs of abuse. According to an influential application of reinforcement learning to substance addictions, the direct effects of drugs to release dopamine can create a perpetual escalation of incentive salience. This article focusses on reward uncertainty, which is proposed to be the core feature of gambling that creates the capacity for addiction. We describe the neuro-dynamics of the dopamine response to uncertainty that may allow a similar escalation of incentive salience, and its relevance to behavioral addictions. We review translational evidence from both preclinical animal models and human clinical research, including studies in people with gambling disorder. Further, we describe the evidence for 1) the effects of the omission of expected reward as a stressor and to promote sensitization, 2) the effect of the resolution of reward uncertainty as a source of value, 3) structural characteristics of modern Electronic Gaming Machines (EGMs) in leveraging these mechanisms, 4) analogies to the aberrant salience hypothesis of psychosis for creating and maintaining gambling-related cognitive distortions. This neurobiologically-inspired model has implications for harm profiling of other putative behavioral addictions, as well as offering avenues for enhancing neurological, pharmacological and psychological treatments for gambling disorder, and harm reduction strategies for EGM design.

Effects of bet size and multi-line play on immersion and respiratory sinus arrhythmia during electronic gaming machine use

Electronic Gaming Machines (EGMs) are regarded as a relatively harmful gambling product, and are associated with psychological immersion (the 'machine zone') and physiological arousal. Specifically, immersion is a phenomenon of attention manifesting as an intense focus on the game at the expense of peripheral stimuli and goals. Past research has indicated significant levels of immersion in response to modern multi-line EGMs when the overall bet is increased, which further scales with risk for problem gambling (Dixon et al., 2014). The present study sought to separate the effects of multi-line play and bet size, on measures of immersion and cardiac activity. Seventy-six male undergraduate students played an authentic EGM on each of 4 pre-defined play strategies while providing electrocardiogram data. The strategies varied the number of paylines and the bet multiplier. From the physiological data, we extracted Respiratory Sinus Arrhythmia (RSA), a marker of task attention derived from heart rate variability. We found that immersion ratings were significantly greater when both paylines and bet-size were high. Importantly, selectively increasing the paylines, but not bet multiplier, produced significant increases in immersion. RSA change indicated parasympathetic withdrawal, consistent with increases in attention during EGM use, but did not differentiate game settings. These results suggest that multi-line EGMs capture attention across a range of play-styles, and that immersion may be effectively amplified by multi-line play.

Can event-related potentials serve as neural markers for wins, losses, and near-wins in a gambling task? A principal components analysis

Originally, the feedback related negativity (FRN) event-related potential (ERP) component was considered to be a robust neural correlate of non-reward/punishment processing, with greater negative deflections observed following unfavourable outcomes. More recently, it has been suggested that this component is better conceptualised as a positive deflection following rewarding outcomes. The current study sought to elucidate the nature of the FRN, as well as another component associated with incentive-value processing, the P3b, through application of a spatiotemporal principal components analysis (PCA). Seventeen healthy controls played a computer electronic gaming machine (EGM) task and received feedback on credits won or lost on each trial, and ERPs were recorded. The distribution of reward/non-reward outcomes closely matched that of a real EGM, with frequent losses, and infrequent wins and near-wins. The PCA revealed that feedback elicited both a frontally maximal negative deflection to losses, and a positive deflection to wins (which was also sensitive to reward magnitude), implying that the neural generator/s of the FRN are differentially activated following these outcomes. As expected, greater P3b amplitudes were found for wins compared to losses. Interestingly, near-wins elicited significantly smaller FRN amplitudes than losses (with no differences in P3b amplitude), and may contribute to the maintenance of gambling behaviours on EGMs. The results of the current study are integrated into a response profile of healthy controls to outcomes of varying incentive value. This may provide a foundation for the future examination of individuals who exhibit abnormalities in reward/punishment processing, such as problem gamblers.