Commentary on Brunborg et al. (2012): Future possibilities for investigating gambling and diminished aversive conditioning


Research into gambling is at an early stage and is dogged by low levels of interest, by erratic funding and by the unfortunate influence of the gambling industry [1, 2]. This exploratory study by Brunborg and colleagues [3] highlights how classical conditioning plays a critical role in the associative learning with gambling. This form of conditioning is already familiar in the way stimuli such as particular symbols, images or other sensory events manage to form positive associations with gambling, as when the sound of horses’ hooves at a racetrack or the ‘chunka-chunka’ of a win on a slot-machine become associated with the excitement of a potential win. Their study suggests that for pathological gamblers, while their receptivity might increase to positive associations, their receptivity to negative associations appears to decrease.

The results present several intriguing possibilities. First, if this diminished aversive conditioning connects somehow with a previously established propensity for pathological gambling, then this could provide a means by which to detect those at risk and to assist them in adjusting to this vulnerability. In other words, it could provide a basis for early intervention. However, addictive behaviour emerges typically within a context involving a range of influences such as social skills, cognitive deficits, genetics, trauma and family of origin issues [4, 5]. The chance of tracking this back reliably to one of these influences seems a distant hope.

A second possibility might be that such conditioning emerges more as part of what is going on during the process of forming an addictive relationship, where it is commonplace to encounter a series of aversive life events involving harms to both the gambler and those around. For example, as a relationship to slot machines intensifies, the gambler often faces severe financial losses which are likely to result in family strife, employment problems and court involvements. For most people, any one of these events would be enough for them to think twice about pursuing gambling further but, for a pathological gambler, such events are unlikely to result in diminished involvement. Perhaps a key functional ability involves switching off concerns about these troubles in order for the gambling relationship to continue uninterrupted.

A third and most likely possibility involves a mixture of both prior tendencies and acquired learning. The two might even feed off each other. For example, a person's prior inclination for impulsivity might interact with learning how to ignore negative consequences and this might, in turn, promote greater impulsivity and so forth. Unfortunately, these interactional effects make it difficult to discern the extent to which prior tendencies or acquired learning contributes to the behaviour; in most situations it is likely to be a mixture of both. As with many other complex behaviours, the mixing of influences from both the person and the context prevents clear tracking of causes.

In whichever direction further research is pursued, the paper adds weight to claims that learning associated with addictive relationships operates in a different way than normal consuming behaviour. Of course, their findings will require further replication, ideally with a larger sample of gamblers with varying levels of addictive involvement, but they are consistent with the increasing body of research literature highlighting how conditioning with addictive behaviour involves different dynamics than other forms of learning [6-8].

The authors point out how their results challenge the idea of aversive therapy as an effective treatment intervention, but they also have implications at other levels. From a public health perspective, their findings throw into question assumptions that view money expended on gambling as an extension of normal consumer behaviour. While pathological gamblers in the whole population are small in number, they contribute disproportionately to the money expended and, accordingly, to the profits accrued. Profit beneficiaries—including industry, government and community recipients—rationalize their gains by depicting gamblers as neo-liberal actors capable of weighing up both the benefits and the risks in a rational fashion [9]. However, with a large proportion of these profits coming from pathological gamblers, this study helps to draw a very different picture: one in which gambling profits are derived by exploiting distortions in learning, in this case a diminished appreciation of negative consequences.

Declarations of interest



  • School of Population Health, University of Auckland, Auckland, New Zealand. E-mail: