My research has focused on whether the behavioural features of mobile gambling affect user behaviours in a way that is more likely to increase the risk of harm, or whether it is more likely that mobile gambling is attractive to people with a problem with gambling in the same way other technologies appear to be. It is predicted that the combination of the highly perseverative schedules of reinforcement found in gambling, when taken in conjunction with the manner in which mobile play affects latencies between gambles, makes mobile gambling particularly prone to risky outcomes. Mobile gambling is also associated with forms of play that are more attractive to people prone to problem gambling, such as in-play betting.

My research programme has consisted of conducting psychometric analyses of problem gambling assessment data, and then taking these findings into laboratory and field environments by getting participants to play on simulated gambling games. The first strand initially looked at nationally representative survey data to test whether problem gambling assessments better capture a categorical or dimensional model (James, O’Malley & Tunney, 2014), suggesting that a categorical model was a better fit of the data. This suggests closer similarities with an addiction model. Subsequently we studied whether assessments consistently measure similar subtypes of gambler and whether these similarities were captured across multiple self-report assessments that ostensibly measure different models of gambling disorder (James, O’Malley & Tunney, 2016a). This analysis consistently identified three groups of gambler, one showing minimal harms, a second systematically endorsing two behaviours (loss-chasing and preoccupation), and a third endorsing multiple problems including items indicating a loss of control over gambling. This research is currently being extended to further model these groups, pooling data from multiple surveys to look at the common sociodemographic predictors of membership of these subtypes.

The findings from this have been taken to be tested in laboratory settings using a simulated gambling machine where participants were exposed to different behavioural features that affected the decision to stop gambling in the face of extended, unavoidable losses (James, O’Malley & Tunney, 2016b). In this study participants played on a simulated slot machine for real money, in which timing and win rate were manipulated, and the effect these had on the decision to stop gambling in the face of extended, unavoidable losses was measured. This suggested that the characterising feature of mobile gambling – intermittent, relatively low reinforcing play, is more difficult to extinguish and these gamblers appeared to chase losses more. We translated this to the field using two apps in which participants played on a simulated scratchcard game over a period of several weeks for real money. Both collected behavioural data, and the second study collected further information about the context in which the app was used. Across both studies there was evidence of significant perseverance in spite of consistent losses, influenced by behaviour and risk-taking traits. This data is also being used to model behavioural and cognitive features that may also explain perseverative gambling play such as near-misses

References

James, R. J. E., O'Malley, C., Tunney, R. J. On the latent structure of problem gambling: a taxometric analysis. Addiction 109 (2014) 1707.

James, R. J. E. O'Malley, C., Tunney, R. J. Loss of control as a discriminating factor between different latent classes of disordered gambling severity. Journal of Gambling Studies (2016a).

James, R. J. E. O'Malley, C., Tunney, R. J. Why are some games more addictive than others: the effects of timing and payoff on perseverance in a slot machine game. Frontiers in Psychology - Decision Neuroscience (2016b).