The explanation of loss of control of gambling behaviour (so-called pathological gambling) presents a considerable challenge for general theories of addiction for two main reasons. First, unlike many other addictions, gambling does not involve the ingestion of substances that alter psychopharmacological states. Second, like many other addictive activities, the majority of the population participates to some degree (Walker Reference Walker1992b), yet only a small percentage of gamblers develop problems with their gambling behaviour.
The role of decision-making in theories of gambling addiction has not featured as prominently as it should, given the central importance of decision making to gambling (e.g., not only deciding when to initiate a gambling episode and when to cease an episode, but deciding what to bet on and how much to wager). The identification within a single theoretical framework of different types of decision-making vulnerabilities, and their association (to varying degrees) with underlying neurophysiological mechanisms, affords the gambling research community an overarching framework in which to map out decision-making deficits. However, the argument that the explanation of problem gambling is attributed to Vulnerability 6 (the illusion of control) is, at best, premature. Moreover, the relationship between conscious (explicit) and unconscious (implicit) cognitive processes during gambling needs to be considered in relation to the types of vulnerabilities that Redish et al. outline.
The application of dual processes theories to addiction (see Wiers & Stacy Reference Wiers and Stacy2006a; Reference Wiers and Stacy2006b for a recent overview) and the role of two distinct types of decision-making processes in gambling (e.g., Bechara et al. Reference Bechara, Damasio, Tranel and Damasio1997; Evans & Coventry Reference Evans, Coventry, Wiers and Stacy2006) cuts across a number of the vulnerabilities outlined. It has been argued (Evans & Coventry Reference Evans, Coventry, Wiers and Stacy2006) that the explanation of gambling behaviour should be seen in the context of two different types of decision making – implicit and explicit systems – mirroring processes with associated neurocorrelates that have been identified in human reasoning research. This opens up three possibilities regarding how decision-making mechanisms might affect gambling behaviour in light of these two systems. First, explicit cognitive processes, in the form of “cognitive distortions,” may account for increased gambling behaviour. Second, implicit (“evolutionary older”) cognitive processes may account for increased gambling behaviour. Third, the interaction between these two different systems of decision-making may account for the behaviour. Fourth, decision-making involving both or either of these in tandem with additional (non-decision-based) mechanisms may explain gambling behaviour.
The notion that cognitive distortions underlie loss of control of gambling behaviour (Wagenaar Reference Wagenaar1988) is problematic. First, cognitive distortions when it comes to decision making under conditions of uncertainty are present in the general population as a whole (hence in the gambling population at large), and not just in the relatively small percentage of problem gamblers. So cognitive distortions do not explain the change from so-called normal gambling behavior to the point where someone loses control of their behaviour (see Coventry Reference Coventry, Marotta, Cornelius and Eadington2002; Sharpe Reference Sharpe2002 for discussion). In tandem with another variable, such as the requirement for a specific high arousal state associated with positive hedonic tone (Vulnerability 2), the role of cognitive distortions in gambling is perhaps more appealing. Coulombe et al. (Reference Coulombe, Ladouceur, Desharnais and Jobin1992) reported an increased incidence of cognitive distortions during gambling for regular as compared to occasional video poker players, together with a significant correlation between arousal increases observed during play and the frequency of (verbalised) erroneous beliefs about gambling. However, Coventry and Norman (Reference Coventry and Norman1998) found no association between level of gambling behaviour and number of cognitive distortions under more controlled circumstances using a more robust coding scheme for cognitive distortions. So the claim that cognitive distortions underlie loss of control of gambling is controversial at this time.
The importance of implicit (unconscious) processes in relation to human decision-making and reasoning has been demonstrated across a wide range of decision-making and reasoning tasks (Evans Reference Evans2003). In relation to gambling as a learnt behaviour, people can learn to predict complex patterns through this implicit system, without the necessary acquisition of explicit knowledge. For example, the implicit system is particularly good at pattern recognition and identifying sequential dependencies (so critical for tasks such as language learning). On its own, though, it seems unlikely that loss of control of gambling behaviour can be causally explained by implicit learning. However, the unconscious nature of this system provides a possible key to the role of this system in the explanation of the development of problem gambling behaviour. Evans and Coventry argue that the possible desire for dissociative states and experiences (escaping from the vulnerabilities associated with everyday life; see Kuley & Jacobs Reference Kuley and Jacobs1987) provides a plausible motive to maintain a decision-making system that infers patterns where they do not aid future prediction (e.g., in roulette). Usually the explicit system kicks in when the implicit system does not serve us well, but for gamblers who desire being in an unconscious state, the explicit system may rather act as means of providing post hoc rationalisations (Coventry & Norman Reference Coventry and Norman1998) or confabulations that serve to maintain the use of implicit processes. If this view is correct, verbalised cognitive distortions are not causes of continued gambling behaviour, but rather, are a means of maintaining behaviour dominated by a different system of (unconscious) processing.
Consistent with the view that unconscious implicit processes are important for continued gambling, Diskin and Hodgins (Reference Diskin and Hodgins1999; Reference Diskin and Hodgins2001) have shown that there is indeed attentional narrowing, an inability to keep track of time, and the experience of dissociative states, in regular gamblers when they gamble.
Decision-making within a dual process framework, in tandem with a cluster of variables associated with positive hedonic tone and escape, illustrate that the explanation of gambling behaviour needs to involve multiple constructs, and multiple vulnerabilities. Detailed process studies of on-line gamblers are desperately needed to identify how the explicit and implicit systems interact over time in the transition from regular to problem gambler.
