Temporal discounting in heroin-dependent patients: No sign effect, weaker magnitude effect, and the relationship with inhibitory control.

Abstract

Impulsive behavior in heroin-dependent individuals and matched smokers was studied within the framework of temporal discounting. Two well-known effects were examined: the sign and magnitude effects (Kirby, 1997). The study also investigated the relationship between cognitive impulsivity and inhibitory control as measured by a Go/No-go task. Fifty-six heroin-dependent patients in the postmethadone treatment stage and an equal number of matched daily smokers participated in the study. The heroin-dependent patients showed higher discount rates in both gain and loss conditions. They also showed no sign effect by demonstrating similar discount rates in gains and losses. Heroin-dependent patients also exhibited a magnitude effect, which was weaker than that observed in smokers. Differential relationships between impulsivity and inhibitory control were found in the two groups. For smokers, decision-making and inhibitory control behaviors pertained to different psychological processes, whereas for heroin-dependent patients, the two behaviors were partially related. Finally, a significant correlation between length of heroin use and temporal discounting with small magnitude was found. The present study advances our understanding of the specific behavioral impulsive patterns in heroin-dependent individuals.

Hyperbolic Discounting: Value and Time Processes of Substance Abusers and Non-Clinical Individuals in Intertemporal Choice

“…First, Rachlin- and Green-models did not show a difference in 10 out of 14 studies, p -values>.1. In four conditions in [30], Rachlin-model fitted better, p -values <.05. Second, Rachlin-model was superior to the c -model in 8 out of 14 studies (Bickel B, Cheng heroin large, Estle 1 large, Estle 3 all three magnitude conditions, Cheng control small, and OU), p -values <.05, and marginally better in two studies (Cheng heroin small and Estel 1 small), p -values <.1.…”

“…The parameter s was significantly correlated with AUC in 9 studies (64.3%) in each of Rachlin and Green-models. In all, but the heroin and control participants in [30], the significant relationships between s and AUC were also negative.…”

“…In Rachlin- and Green-models k and s were significantly negatively related with AUC after controlling for the other parameter, which further demonstrated their more complex relationship in determining subjective value as exhibited in Figures 1 and 2. In addition, the net relationship between AUC and k, or s was also negative in [30]. Hence, in the c -model, k uniquely described the AUC.…”

“…In particular, the parameter s has been described as a person measure [24] and hence would not be expected to change with conditions. The k parameter of Green- and the base-model has often been used to describe individual differences as earlier described, but it has also been used to describe changes due to situations (e.g., to account for sign and magnitude effects [25], [30]); hence, the meaning of k may be broader to account for both person and situation based variability. The c parameter is directly indexing a valuation process and hence it is expected to vary with payoff level.…”

“…Table 1 details the characteristics of the studies [22], [28]–[30] included in the present manuscript. Each study is described in terms of number of participants, age, gender, education level and drug type (if substance abusers).…”