Which theory best explains why sensation seekers are motivated to engage in risky recreational behavior such as mountain climbing?

Within the last decade there has been an unprecedented increase in the availability of legalized forms of gambling in the US, accompanied by widespread social legitimization of gambling as an acceptable leisure activity. Although gambling is a harmless recreational activity for most persons, it can become problematic for others. However, little basic laboratory research has been attempted to examine problem gambling. Abt (1987) discussed the paucity of empirical research on problem gambling as well as the lack of data on the processes of `normal' gambling. It has been argued that the population of gamblers constitutes a continuum of involvement, not discrete groups such as `compulsive' and `social' gamblers (Dickerson, 1993). Pathological phenomena such as loss of control, persistence when losing, and related cognitive distortions should be examined as risk factors in regular gamblers rather than retrospectively in those seeking treatment.

`Chasing' behavior has been described as the most significant step in the development of pathological gambling (Lesieur, 1979). Chasing is defined as the continuation of gambling, often with increased wagers, after a sequence of losing bets (Dickerson, 1993). Chasing is described as a downward spiral of options and involvement (Lesieur, 1979). As losses accrue, the gambler's selection of options (sources of money) contracts while his breadth of involvement expands. Chasing is defined by gambling and losing, followed by more gambling to `get even'. The more money that is lost, the more intense becomes `the chase' (Lesieur, 1984). Chasing is practically ubiquitous among problem gamblers. In fact, it is hard to imagine how one could develop the complex of personal, economic and social problems characteristic of problem gambling without the persistent pursuit of gambling losses.

In the extant literature, chasing behavior has almost exclusively been discussed in terms of returning to gamble on another day with the express intention of recouping lost money. This conceptualization might usefully be termed, `between-session' chasing. However, a `within-session' conceptualization is a useful point of departure for understanding the individual determinants of chasing. A tendency to gamble too long within a particular session must be a contributing factor to the development of gambling problems. Problem gamblers have difficulty quitting, regardless of whether they are losing or winning. When they are losing, they tend to chase their losses by investing additional money that they had not planned on using. Cash machines, conveniently located within the gambling venue are one typical source of new funds. At best, gamblers may get even; at worst they suffer a crippling loss. Even problem gamblers win on occasion. However, a good run of cards or a sizable jackpot never seems to be enough. The winnings are frequently reinvested immediately. At best, only some of the winnings are lost; at worst all of the winnings are lost and then the original stake is lost too. This type of `within-session' chasing behavior is important because it contributes to crippling and demoralizing losses. These losses are then the object of `between-session' chasing of the type described by Lesieur. That is, as soon as a new stake can be secured (by means legal or illegal), the problem gambler will try again.

Examining the determinants of within-session chasing is a valid goal for laboratory-based research. This study was intended to address the following question: How do individual differences affect the probability of gambling longer or more heavily than one should? If personality traits or cognitive styles can be shown to contribute to within-session chasing, didactic or cognitive-behavioral interventions may be suggested. Such strategies could be applicable not only to problem gamblers, but to the broad continuum of gamblers as part of a harm-minimization approach.

Sensation-seeking is a basic personality factor with several biological correlates and a high heritability (Zuckerman, 1979, Zuckerman, 1994a). Sensation-seekers are defined as those who seek novel, varied or complex sensations or experiences and are willing to take risks for the sake of such experience. Individuals differ in their optimal level of catecholamine systems arousal (Zuckerman, 1984). High sensation seekers have higher optimal levels of arousal (i.e. they feel better when there is a high level of activity in brain norepinephrine and dopamine systems). An arousal theory of gambling suggests that the uncertainty and monetary risk of gambling provides a higher level of stimulation and arousal which high sensation-seekers desire (Zuckerman, 1994a). Thus the arousal or excitement produced by gambling activity is rewarding to the high sensation seeker, even exclusive of the prospect of winning money. Therefore, high sensation-seekers' priority is to maintain arousal by playing as long as possible, despite the punishing effects of monetary losses.

However, various studies have failed to find an association between sensation seeking and gambling (e.g. Blaszczynski, Wilson, & McConaghy, 1986, Blaszczynski, McConaghy, & Frankova, 1990). However, these results could be due to methodological problems in the studies such as the failure to control for age, the type of gambling involved or differences between gamblers seeking or in treatment and those who are still gambling (Zuckerman, 1994a). Most of the studies finding no differences between gamblers and controls use gamblers in treatment or seeking treatment. It is also crucial to take into consideration the heterogeneity of gambling activities. Coventry and Brown (1993) suggested that high sensation seekers may only engage in certain types of gambling such as casino games and race track, and that lows prefer less stimulating outlets such as off-track betting parlors. It was also suggested that gamblers become specialists, concentrating all their efforts in just one arousal seeking activity: gambling. They would therefore not score highly on a sensation seeking scale involving a variety of sensation-seeking activities.

Impulsivity contributes to gambling problems. McCormick (1993) noted a significant positive relationship between the severity of gambling and impulsivity in a large sample of substance abusers. Patients who were negative or moderate for gambling problems were lower in impulsivity than a group who were probable pathological gamblers, who were in turn lower than a group of severe pathological gamblers.

There has been considerable variation in the definition of impulsivity across studies. Impulsivity has been conceptualized as: the failure to evaluate the risk or danger inherent in a situation; acting without thinking; the inability to plan ahead; lack of response moderation; and deficits in passive avoidance learning (Helmers, Young & Pihl, 1995).

Gray, Owen, Davis and Tsaltsas (1983) proposed that impulsive persons are relatively more sensitive to stimuli associated with reward than with stimuli associated with punishment. The basic assumption of this theory, namely that differences in learning are a function of the interaction between personality type and reinforcement contingencies, has received considerable empirical support (Farmer & Nelson-Gray, 1995). According to Wallace, Newman and Bachorowski (1991), impulsives focus selectively on reward stimuli and disregard stimuli indicating that response has the potential for harmful consequences. Since all gambling forms are intrinsically biased against the gambler, gambling presents a good method for testing Gray's hypothesis. Impulsivity should hinder response inhibition when punishment (losing) outweighs reward (winning). Within-session chasing is one such situation.

Helmers et al. (1995) pointed out an important distinction between measures of sensation seeking and measures of impulsivity. For example, measures of sensation seeking may be associated with risk-taking behaviors that are often carefully planned to minimize physical danger, such as mountain climbing or other risky sports. Impulsivity measures may capture more of a lack of planning or the underestimation of risk. However, impulsivity and sensation seeking are moderately related (Zuckerman, 1979, Zuckerman, 1994a, Zuckerman, 1994b) and together they comprise one of five broad factors of personality (ImpSS; Zuckerman, Kuhlman, & Camac, 1988, Zuckerman, Kuhlman, Thornquist, & Kiers, 1991).

Sociocognitive factors have been hypothesized to affect gambling behavior. According to Abt, McGurrin and Smith (1985), any model of gambling behavior should take into account the player's interpretations and evaluations of the meaning of gaming activities within a social and cultural context and it should incorporate the gambler's cognitive activity as a central mechanism. A number of irrational beliefs and cognitive errors in gamblers may contribute to persistence, level of involvement and specific behaviors in gambling. Serious gamblers entertain irrational beliefs about gambling, such as the self-perception that one has special skills and insight needed to beat the odds (Walker, 1992). The gambler believes that persistence must eventually be rewarded. Lesieur (1979) describes the decision to chase losses as infinitely logical to the gambler: as long as he is still betting, he has a chance to get even. Walker (1992) describes the illusion of control and the belief in luck as two very important components of irrational thinking in gamblers.

Gamblers may view their own gambling activity as an important opportunity to convey a favorable image of themselves (Holtgraves, 1988). Gambling can be seen as a means to present oneself as daring, glamorous, romantic and exciting. Breen and Zuckerman have devised a questionnaire to assess the strength of attitudes and beliefs specific to gambling (see Section 2).

Laboratory research in gambling has been criticized for lack of external validity (e.g. Anderson & Brown, 1984, Dickerson & Adcock, 1987). Student volunteers with little or no gambling experience are often given complex theoretical gambles in artificial surroundings, without real money at stake. Such experiments have questionable generalizability to gambling in the real environment. Breen (1996) conducted a pilot study designed to improve on the generalizability of laboratory gambling paradigms. Subjects were male college students who were either probable problem gamblers or nonproblem gamblers. They gambled under realistic conditions in the laboratory using real money. The outcomes of the gambles were prearranged so that subjects won predominantly at first; however the rate of losing increased steadily in proportion to the number of bets placed. This experimental design served as an analogue of a situation which frequently occurs in real gambling. That is, it is not uncommon for the player to be ahead of the `house' for a time. Obviously, it is optimal to quit when one is winning. In this experiment, subjects determined their own fate by their objective gambling behavior. `Chasing' under these conditions was operationally defined as reinvesting one's winning repeatedly and continuously until all winnings, plus the original stake are lost. However, no differences were found between `chasers' and `nonchasers' on personality factors such as impulsivity or sensation-seeking, or on a cognitive measure of beliefs about gambling. It was thought that several weak points in the method could have obscured the results. Thus, the present research implemented several methodological changes designed to improve on the ecological validity of the pilot study.

The previous discussion suggested that impulsivity, sensation-seeking and beliefs and attitudes to gambling may contribute to both the willingness to begin gambling, and the amount of participation (i.e. the persistence of responding) once gambling. We hypothesized that there would be significant positive associations between measures of impulsivity, sensation-seeking, and gambling attitudes and beliefs, and `chasing' within the gambling session. Chasing was defined as opting to gamble until all available money was lost.

Subjects

All the subjects in the experiment reported that they had gambled for money at some time in their lives. Only male undergraduates were selected because it is difficult to obtain an adequate sample of females who are experienced enough in gambling. For three consecutive semesters, every student in introductory psychology courses was screened by asking them “how often do you gamble?” Subjects (N=248) were selected on the basis of self-reported frequency of involvement in gambling, ranging from

Results

The expected positive correlations were observed between GABS and SOGS scores (r=0.38, p<0.001). GABS was minimally but significantly correlated with the IMP subscale of ImpSS (r=0.13, p<0.05). GABS was significantly correlated with the SS subscale of ImpSS (r=0.22, p<0.001).

There were three general outcomes possible. Subjects could: (1) decline to gamble; (2) gamble and quit while they still had some money left; or (3) gamble until all money was lost. Those who decided not to gamble and took

Discussion

Laboratory research in gambling has rightly been criticized in the literature for lacking in ecological validity. In this experiment, an attempt was made to create a valid analogue of gambling. Validity questions were directly addressed in the design of the experiment. For example, subjects were given US$10 in cash for their participation, regardless of whether or not they gambled. If they decided to gamble, most of them felt that they were risking their own money. Each subject who chose to

Recommended articles (6)

• Research article

  Stress and gambling

  Current Opinion in Behavioral Sciences, Volume 31, 2020, pp. 8-12