This investigation showed that stimulant-dependent subjects (SDI) show impairments in time perception and in sensorimotor timing. Moreover, these impairments are task dependent and possibly specific for the duration of the intervals processed. The mediator models put forth here tested the hypothesis that (neuro-) psychological dysfunctions might carry the effects as mediator variables between stimulant dependence status and temporal processing abilities. We found partial support for this hypothesis, specifically for the estimation of the 53-second interval which is mediated by one form of self-reported impulsivity (non-planning impulsiveness). In contrast, impaired timing in SDI for the processing of 1 and 2-second intervals over a variety of tasks could not be explained by the included mediator variables. Taken together, these findings support the hypothesis that SDI have basic time processing dysfunctions within a specific time range (around 1 to 2 seconds) that are not due to altered levels of impulsivity or other cognitive dysfunctions.
Methamphetamine and cocaine abusers show altered function in prefrontal and striatal dopaminergic circuits (Baxter et al., 1988
; Chang et al., 2005
; Sekine et al., 2003
; Volkow et al., 2001a
). Dopaminergic systems in the striatum and prefrontal cortex are also thought to be involved in attention, working memory and impulse control (Arnsten and Li, 2005
; Cardinal et al., 2004
; Evenden, 1999
; Paulus et al., 2002
). As these brain systems have also been identified as guiding the estimation of time intervals and the timing of movement (Cuoll et al., 2004
; Matell and Meck, 2004
; Rubia and Smith, 2004
), we expected to find temporal processing impairments in SDI. Although our cross-sectional design limits our ability to draw strong conclusions regarding the causal effect of stimulant dependence on time processing, the current evidence points to the same specific brain structures and neural systems known to be involved in time perception and to be affected in SDI. Therefore, we propose that temporal processing abnormalities in SDI result from dysfunctions in fronto-striatal areas of the brain, whether that dysfunction contributes to or is a result of stimulant dependence.
Just as stimulant dependence might lead to brain changes underlying altered time processing, the reverse it also possible. For instance, the estimation of the 53-second interval was influenced by subjects' impulsiveness. However, higher impulsiveness and an altered sense of time in these SDI may have been an antecedent to initial drug use, i.e. a temperamental pre-disposition encouraging our tested SDI to fall into the habit of drug taking. Only longitudinal studies will reveal whether an altered experience of time in combination with other factors predisposes an individual to drug abuse. As is often the case in other complex behavioral domains, stimulant use and altered time perception may interact, leading to a mutually reinforcing effect on brain function.
Based on our results, one can conclude that long-term exposure to methamphetamine or cocaine is associated with impairments of processing temporal intervals. The fact that the statistical analyses found significant direct group effects especially for the processing of intervals of 1000 and 2000 ms should be taken with caution as it may imply only that in our specific tasks these intervals were more sensitive to group differences. For example, at longer intervals subjects might have used a strategy that subdivides the interval by subtle rhythmic movements, an automatic behavior that is hard to control for. This structuring of the intervals could have guided the temporal responses and made them more accurate. At longer intervals these strategies become more effective, therefore revealing timing impairments only at shorter interval lengths. In addition, the fact that in the temporal reproduction task we found significant group differences only for the 2-second interval (and not at 1 second) urges caution with our interpretation. It is, however, worth considering the results within the conceptual framework of the psychology of time. Researchers have established the categorical distinction between perception of duration
(for intervals up to 2 - 3 seconds) and estimation of duration
(for longer intervals). Events lasting only a few seconds are processed in the present as a perceptual whole, whereas longer intervals must be estimated from memory (Fraisse, 1984
; Pöppel, 1997
). It is possible that the impairments found in the SDI are specific to temporal processing in this shorter time range.
According to Rammsayer (1999)
a dissociation of effects as seen in our results on duration discrimination (impairment only for the processing of 1000 ms but not for 100 ms intervals) would favor the interpretation that timing processes per se
are not disturbed in SDI. Intervals with a length of up to some hundreds of milliseconds are supposed to be processed based on brain mechanisms outside of motor and cognitive control. The higher duration difference thresholds of SDI at intervals around 1000 ms, in contrast, would be interpreted as based on disturbances of additional cognitive processes that come into play only at longer intervals. Based on prior studies using tapping tasks, there is evidence that a specific temporal processing function occurs in a time range between approximately 300 ms and one or two seconds (Madison, 2001
; Mates et al., 1994
; Wittmann et al., 2001
). In fact, inferring from a recent synchronization tapping study employing a secondary attention task, Miyake et al. (2004)
concluded that in a time range between 450 to 1500 ms automatic processing that is not strongly affected by attention, whereas attention and working memory affect intervals in the range between 1800 and 3600 ms. This empirical finding is in accordance with similar theoretical proposals suggesting that an automatic timing system for shorter intervals can measure time without attentional modulation whereas a cognitively controlled timing system for supra-seconds intervals draws upon cognitive circuits of the brain (Lewis and Miall, 2003
). We show here that in timing performance with 1000 and 2000 ms intervals a variety of tasks are not mediated by attention or working-memory processes. First, timing disturbances occurred in SDI in the absence of impairments in most of the tasks of a battery of attention and working-memory tests. Second, an impairment of SDI in the backward digit span task did not mediate performance in time perception. Thus, it is possible that our group effects reveal a specific disturbance of a temporal processing mechanism that is active for durations of 1 to 2 seconds and which is not mediated by other cognitive processes.
In the 53-seconds time-estimation task SDI estimated the time interval to have lasted longer than did the control subjects. The fact that this group effect was mediated by greater impulsivity of SDI can be explained by models of prospective time perception. Subjects estimate the duration of a given interval as longer when the focus of attention is on the passage of time as opposed to a condition where the same interval is filled with activities that distract an observer from attending to time (Wittmann and Lehnhoff, 2005
; Zakay and Block, 1996
). An overestimation of time intervals is a sign of boredom or emotional distress that draws attention away from meaningful thoughts and actions and directs it to the passage of time (Danckert and Allman, 2005
; Twenge et al., 2003
). The higher impulsivity of SDI could lead to the subjective experience of being trapped in time during the time-estimation task. Then, SDI would focus their attention more on the passage of time and overestimate its duration. Although results are not unequivocal, a relationship between impulsivity and time perception also has been postulated for children with attention deficit hyperactivity (Barkley et al., 2001
) and patients with orbitofrontal cortex lesions (Berlin et al., 2004
). Our finding could have clinically relevant implications for the treatment of drug addicts with impulse control problems. It has been suggested that an altered sense of time could be one reason for impulsive individuals to discount the value of temporally delayed reinforcers more strongly and to persist in goal directed behavior that results in immediate or short-term gains at the expense of future or long-term interests (Barratt, 1983
, Takahashi, 2006
). For a person addicted to a drug, the benefit of resisting the temptation to use a drug, that is, to delay gratification, might lie subjectively too far in the future. Treatment programs could develop intervention strategies that manipulate the temporal delay of rewards or to cognitively restructure the perception of inter-temporal choices in order to shape more adaptive and health-promoting behavior (Monterosso and Ainslie, 2006
In summary, we found SDI to exhibit unmediated deficits in sensorimotor timing and an overestimation of a 53-second interval that was mediated by an increase in impulsivity. Methamphetamine and cocaine abusers show abnormal metabolic activity of the dopaminergic system and have structural brain changes in fronto-striatal regions, both factors that have been identified as important for contributing to changes in the processing of time. We therefore conclude that SDI have impairments in sensorimotor timing and that longer time intervals are overestimated due to more impulsivity. Diagnostic and therapeutic tools in psychiatry and neurology are being developed that assess/treat altered time-perception in various patient populations (Monterosso and Ainslie, 2006
; von Steinbüchel and Pöppel, 1993
). Patients addicted to methamphetamine and cocaine could become beneficiaries of such an approach.