Abstract: Impulsive drug-seeking has been one of typical characteristics of all drug-dependent individuals (Altman et al., 1996; Dawe et al., 2004; Dom, Hulstijn, et al., 2006), and questionnaire assessments based on self-reporting have been essentially taken as the conclusive metrical method of assessing impulsivity in the population. However, due to the varieties of meanings of “impulsivity” (Evenden, 1999; Dougherty et al., 2005), the correlation between the results of self-reporting and those of laboratory tasks has been marginal, or even rather weak (Dom, D’Haene, et al., 2006; Moeller et al., 2001; Reynolds et al., 2006; Dom et al., 2007). Moreover, researchers have pointed out that behavioral measures of impulsivity may be a better predictor of relapse risk than any measure utilizing self-reporting. In laboratory methods used to assess impulsivity, the Delay-Discounting Task (DDT) and Iowa Gambling Task (IGT) have been widely adopted to explore the high impulsivity characteristic of drug-dependent individuals. The DDT was propitious in assessing “myopic” behavior for future values in humans (Green et al., 1994; Rachlin & Siegel, 1994; Kirby & Herrnstein, 1995), but the function of IGT has remained unclear, as is its relationship with that of DDT (Bartzokis et al., 2000; Bechara, 2003; Monterosso et al., 2001; Dom et al., 2007). It is thus assumed in the present study that the high impulsivity characteristic of drug-dependent individuals is not only represented as a hyposensitivity to immediate rewards and an insensitivity to delayed value, but also as an immediate win-priority selection pattern and a tolerance to high risk. The present study explores the impulsive reaction patterns and the relationship in win-risk coincidence situations and delayed reinforcement situations in heroin abstainers in different abstinence phases.
In the present study, 61 male heroin abstainers (19~38 years old, M=29.25, SD=4.17) in a variety of abstinence phases (1~26 months, M=11.34, SD=7.09) and 32 normal male controls (18~39 years old, M=26.78, SD=7.18) were selected to undertake Chinese computerized versions of the IGT and DDT. In the IGT, subjects had to choose between four decks of cards. Unknown to the subjects, two piles offered high rewards but had potentially high losses, and hence were disadvantageous in the long run. The other two piles offered lower rewards but had much lower losses, and hence offered a net gain in the long run. In the DDT, subjects had to choose between hypothetical monetary rewards available immediately or following a delay. Delayed rewards were ¥1,000 and ¥10,000 in value, and the immediate-reward amount was adjusted until the choices reflected indifference. This process was repeated at each of eight delays (six hours to 25 years).
The results of the present study showed that (1) reaction patterns for pure win-cards were consistent between heroin abstainers and normal controls; however, the former still preferred to choose high win-cards in spite of potential high punishment, whereas the latter tended to transform the direction of card selection after getting corresponding feedback, to avoid potentially high risks; (2) whether the delayed rewards were ¥1,000 or ¥10,000, the delay-discounting rates (k) of heroin abstainers were much higher than those of normal controls; (3) the delay-discounting rates based on the DDT were not correlated with the decision-making performance (net score) based on the IGT, but the former was partially correlated with the transform ratio of selecting cards after a net loss based on the IGT; and (4) the net score and transform ratio of card selection after a net loss based on the IGT and delay-discounting rates based on the DDT in heroin abstainers did not improve in line with the abstinence time course.
From the present study, we conclude that (1) the impairment of the decision-making function in win-risk coincidence situations in heroin abstainers was demonstrated in their immediate win-priority selection pattern and tolerance to high risk; (2) the impairment of the decision-making function in delayed reinforcement situations in heroin abstainers was manifested in their hyposensitivity to immediate rewards and their insensitivity to delayed value, which led to a quick decay of long-time value; (3) the IGT was more sensitive to heroin abstainers’ tolerance to risk, while the DDT was more efficient in evaluating their insensitivity to long-term value; both corresponded with two key variables of decision-making—namely, risk and delay; and (4) impulsivity in heroin abstainers did not improve in line with the abstinence time course

Key words: heroin abstainers, behavioral impulsivity, the Delay Discounting Task, Iowa Gambling Task