Commentary on <scp>L</scp>orains <i>et al</i>. (2014): A potentially important advance is understanding different types of gamblers

Lorains et al. 1 present the results of a study comparing strategic and non-strategic gamblers' performance on the Iowa Gambling Task (IGT) and a loss aversion task. In addition, they included a control group recruited from the general community who were matched to the problem gamblers on age, gender and estimated IQ. The control group did not have a gambling problem. The paper compares how two different types of gamblers perform on the IGT. The topic is important for two reasons. First, it clarifies exactly how the reasoning of problem gamblers differs from controls, and secondly, it takes into account the type of game that the participants play rather than assuming that they are all the same. The results indicate that problem gamblers who prefer strategy-based games such as sports bets and poker perform better than non-strategic gamblers on the IGT. This suggests that a model that may explain vulnerability to one type of gambling, such as electronic gambling machines (EGMs), may not necessarily apply to other types of games (e.g. strategic games). I also like the way they look into the performance on the IGT in more detail, examining the utility shape and loss aversion in the task, rather than just a simple comparison of the results. Another interesting finding is that they report that problem gamblers (regardless of subtype) and controls demonstrated similar recency and learning parameters, indicating that IGT performance may not be due to poor learning or memory. I have reported similar findings regarding implicit learning in Turner et al. 2. In that paper, we found no difference between problem and non-problem gamblers in terms of implicit learning, but an indication of a different strategy on the task. The cause of this difference in IGT scores cannot be determined in this study, because it is not known if the differences in this task are related to the cause of excessive gambling or are a result of excessive gambling. For example, it could be that differences in the participant's response to wins and losses led to their gambling problem, or it could be that this difference was the result of gambling over long periods of time. However, these findings point to a possible causal mechanism for problem gambling among EGM players that needs to be explored further. The paper has some weaknesses. The sample size is small, and the number of males and females is not equal for different types of games due to the fact that most strategic gamblers are male. In addition, I feel the study would have been stronger if they had also included non-strategic and strategic non-problem gamblers. Lorains et al. 1 matched their control on age, gender and estimated intelligence. They say nothing about the gambling habits of their control group. If they had contrasted their participants with non-problem gamblers who shared the same gambling preference, then they could have determined to what extent the results are related to problem gambling and to what extent they are related to gambling preference. For example, do people who do not have a gambling problem and enjoy playing on EGMs regularly perform significantly better on the IGT than those who have a gambling problem and play on EGMs? If so, it might indicate that the ability to learn from losses is a key factor in avoiding problematic gambling. Conversely, if non-problem gamblers who play EGMs perform as badly as problem gamblers who play EGMs, the results on the IGT might have more to do with their game preference than with gambling problems. In addition, they only discuss the results in terms of problem gamblers compared to their matched controls; however, the absolute values for these variables need some comment. In their Fig. 2, for example, the controls for the EGM group and the controls for the strategic groups appear to differ. In addition, the results suggest that strategic gamblers are much better at the IGT than the non-strategic non-problem gamblers. This is important, because it shows that strategic gamblers use strategies (skills); but this difference might be due in part to sex differences. Also in their Fig. 2, the results suggest that strategic gamblers may perform worse on the IGT compared to their matched control group. This difference is apparently not significant, but the confidence interval suggests that the effect might reach significance with only a few more participants. This would be a worthwhile direction for future research. Are strategic gamblers who have a problem simply less skilled than strategic gamblers who do not have a problem? In Turner & Fritz 3 we showed how chronically losing might be a result of being less skilled; however, we also discussed the role of emotional control and the possibility that it is emotional control that differentiates problem and non-problem strategic gamblers. The loss aversion findings are unclear. In the loss aversion task the controls for the non-strategic group and their controls scored higher in terms of loss aversion, but on the IGT the strategic group scored higher on loss aversion. This inconsistency is noted as a possible situational effect related to being in the action. Having a lower level of loss aversion might be related to the type of gambling. However, the IGT is not, in fact, a gambling task, and it is different from any actual commercial gambling game, so the generalizability of this result to gambling is unknown. An alternative explanation might be the pay tables of the gambling games that they typically play. In real casino games, strategic games typically require larger bets relative to the payoff than non-strategic gambles. Table games often have minimum bets of $10 or more, whereas EGMs have much lower minimum bets (e.g. $0.25, $0.05). Most people play more than the minimum, but none the less a single bet on a table game is typically much larger than on an EGM. In addition, many strategic games have an ‘even money’ payoff which means that the player has to risk $100 in order to win $100,* and the chances of a win are typically slightly less than 50%. In contrast, a bet on an EGM of 75 cents might give the player a one in 60 thousand chance of winning $2000. This difference in game design might be related to the lower loss aversion found for the strategic gamblers on the IGT; they have to risk more in order to win more. This is another area that needs to be explored further. The notion of what is strategic gambling also needs further exploration. Do these results apply only to sports bets and poker, or do they apply to other games? Some games, such as craps, are in fact games of pure chance, but are often played by people who think they are games of strategy. In addition, the payoff table is more similar to strategic games such as blackjack and sport betting than to non-strategic games such as an EGM. Do craps players perform on the IGT more like EGM players, or more like sports bettors? In summary, I like the direction that these authors have taken, both in terms of differentiating different types of problem gamblers and in terms of exploring in more detail the nature of the IGT task. I think this is a promising area of future study. None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.