engaged in by gamblers seeking treatment. Fisher and Griffiths (1995) argue that England's legal "fruit machines" (slots) are especially risky for adolescents. They claim that game machines, better than other technologies, can be designed and programmed to encourage frequent gambling. Gupta and Derevensky (1996) asked heavy and light video-game-playing children (ages 9-14) in Canada to complete a questionnaire and to play a computer blackjack game. The high-frequency video game players were more likely to report being regular gamblers. Heavy-playing boys also bet more on the blackjack tasks. The authors speculate that experience with video games, in which practice can improve performance, leads teenagers to have the illusion that gambling machine games are somehow solvable. Griffiths (1990) found that troubled teenagers (problem gamblers, those who had been charged with crimes) were likely to hang out in video arcades and to play fruit machines frequently. However, this study and others on the correlates of children's machine gambling are only suggestive of a causal link between playing game machines and pathological gambling, and reasonable alternative explanations exist. For example, background and personal factors leading British adolescents to get into trouble could also lead them to hang out in arcades, play slots, and also to have illusions of skill in their gambling and other areas of their lives.
If new game machines such as video poker machines can be tailored to their users, they might be able to deliver more effective reward contingencies. Such an effect could increase the probability of problem gambling. Kilby (1987) discussed an older rating system for casino players whereby records were kept of frequent patrons' conversions of currency to chips. Those who cashed in more money might be given more "comps" such as free food, drinks, or games. Today, plastic club cards used with game machines are a far more sophisticated version of the old system; they can record exactly how much a gambler is wagering on which types of games. In theory, these cards can track gambler preferences, wagers, and outcomes; future rewards and games can be "personalized" to those patterns (Popkin and Hetter, 1994). The cards also can be used to tally frequent gambler credits, encouraging loyalty to the casino or other venue.
Telecommunications technology also could be used in tailor-