Acute effects of cocaine in two models of inhibitory control: Implications of non-linear dose effects - Fillmore et al. - 2006 - Article

What is this article about?

Stimulant drugs’ effects on performance has been well known for many years. These stimulants can decrease fatigue, increase vigilance, and speed reaction time (RT), prolong effort and generally increase productivity or work output. Specifically, these stimulant drugs have been shown to enhance the ability to inhibit behavioral responses. Different stimulants such as methylphenidate (Ritalin) and d-amphetamine have been shown to improve inhibitory control in healthy adults and children with attention deficit hyperactivity disorder (ADHD).

Tasks that study this effect are based on the stop-signal model. These tasks measure an individual’s ability to inhibit behavioral responses. They require quick, accurate responses to go-signals and the inhibition of these responses during stop-signals. The go-signals often consist of letter pairs (O and X for example) which are presented visually with one go-signal at a time, on a computer screen. The participants can respond to these letters by pressing one of two computer keys. In the stop-signals, the participants need to inhibit their responses. These stop-signals are often tones, sometimes accompanying a go-signal. The stop-signals occur at variable stimulus onset asynchronies (SOAs) with respect to the letter (50 ms or 300 ms) and they occur only at for example 25% of the trials. This means that subjects must overcome their tendency to respond to a go-target when they hear a stop-signal. This indicates their level of inhibitory control. Inhibitory control is often modelled as the mean latency to inhibit responses, called the stop-signal reaction time (SSRT). This is the time that is needed to inhibit the pre-potent response when the stop-signal occurs. The time that participants need to inhibit is less than the time that is required to respond. SSRTs are related to the number of successfully inhibited responses, with longer SSRTs associated with less successful response inhibitions. Longer SSRTs thus suggest weak inhibitory control, which might be due to a slow inhibitory process.

The evidence with regards to the role of stimulant drugs in inhibitory control are not consistent. Some show that stimulants impair inhibitory control in some contexts. The authors suggest that one critical factor in determining facilitation of inhibitory control is dose. Doses that are effective in facilitating one type of behavior may actually be detrimental to other types of behavior. With regard to inhibitory control, some studies report U-shaped dose-response curves after Ritalin.

Research about non-linear dose response effects could help to understand how changes in cognitive functions might maintain or escalate stimulant abuse. Alterations in inhibitory control might be the most likely contributor to abuse potential. When there are impairments in inhibitory control, this could lead to a lower ability to stop drug-acquisition, which in turn leads to a lower ability to stop drug-taking.

Cocaine and amphetamine users may be motivated to self-medicate attentional deficits and hyperactive/impulsive tendencies. However, whether a stimulant drug facilitates or disrupts a cognitive function may depend on the dose.

In the present study, the authors examine the possibility that a U-shaped dose-effect function on SSRT might also be evident in response to an abused stimulant. They used the stop-signal model to examine the acute effects of four different doses of oral cocaine HCI (0,100,200, and 300 mg) on SSRT in a group of adults with a history of cocaine use. They also examined the generalization of the dose-response effects to a different model of inhibitory control. The dose effects on the SSRT measure of inhibitory control were compared to a measure of response inhibition obtained from another task which is also used commonly to study drug effects on inhibitory control, the go-no-go model.

What were the methods used?

Participants

There were 12 adult participants, with nine men and three women. They all had a history of cocaine. The mean age was 42 years. There were 8 African Americans and 4 Caucasian participants.

The volunteers had to have a minimum of grade 8 education, reading ability, correct vision, and no self-reported psychiatric disorders. They also had to: score at least 4 on the 14-item, self-report Drug Abuse Screening Test (DAST), a self-report of past week cocaine use, and test positive for the presence of cocaine or benzoylecgonine in their urine.

All of the participants smoked cocaine in the form of crack. No volunteer was in treatment for their substance use.

Materials

Stop-signal task

This task has been described before.

Cue-dependent go-no-go task

The cued go-no-go RT task is another measure of inhibitory control. Participants are presented with a cue, that provide information about the target stimulus that is likely to follow. The cues have a high probability of signaling the correct target. The go-cue conditions are the most interesting, since these cues generate the tendency to respond faster to targets. However, sometimes subjects must overcome this tendency and inhibit their response, if this cue is followed by a no-go target. Therefore there are often failures in inhibiting the response if a no-go target is displayed after a go-cue. This effect of the go cue condition is sensitive to the effects of psychoactive drugs, including stimulants and depressants.

Drug effect questionnaire (DEQ)

The DEQ consists of 15 items which are sensitive to cocaine effects. The items were: any effects, active/alert/energetic, bad effects, good effects, high, irregular heart beat/racing, like, anxious/nervous, pay for this drug, rush, shaky/jittery, take this drug again, talkative/friendly, nauseated/queasy, and sluggish/fatigued/lazy. The items were presented on the monitor, and participants rated each item using the computer mouse to select among five responses: not at all, a little bit, moderately, quite a bit, and very much.

What were the findings?

The results in this study show that cocaine improves the ability to inhibit responses, measured by both models (stop-signal and go-no-go task). In the stop-signal model, cocaine reduced the time to inhibit a response. In the cued go-no-go model, there were drug-induced decreases in the number of failures to inhibit responses. The dose-response functions differed depending on the measures.

In the stop-signal task, there was a quadratic dose-response function: 100 mg and 200 mg cocaine produce faster SSRTs compared with placebo. There were no significant speeding effects to the 300 mg dose: in this case, participants’ mean SSRT was nearly identical to placebo.

In the cued go-no-go task, there was a more orderly, linear improvement as a function of dose.

None of the tasks showed any cocaine effects on response activation, which was measured by RT to go targets. This means that the cocaine-induced improvements in response inhibition did not reflect speed-versus-accuracy trade-offs.

The U-shaped dose-response function in SSRT is in line with the studies about methylphenidate in children with ADHD. It seems to be true that the facilitating effects of stimulant drugs are limited to intermediate doses: above these doses there is no improvement and there could even be impairing effects. These findings implicate that intermediate doses of cocaine can help the user to restore cognitive functioning. This might lead the user to repeatedly take the drug. However, as they do this, their inhibitory control can become impaired and this can lead to impulsivity, perseverative responses and binge use of the drug. In line with this, cocaine abusers show patterns of impulsivity and perseverative behavior. They show hypoactivity (lower activity) in their cingulate and dorsolateral prefrontal cortal regions. These areas are associated with inhibitory control, and this could be due to their long-term cocaine use. Cocaine users also show enhanced sensitivity to stimulant drugs. This could also lead to impulsive behavior in response to higher drug doses.

It is hard to generalize the findings to other populations than cocaine users. Cocaine users are characterized by poor inhibitory control, but it is possible that the cocaine-induced facilitation of inhibitory control is specific to individuals with poor baseline levels of inhibitory control. Other studies have shown that d-amphetamine can improve response inhibition on a stop-signal task by speeding SSRT, but the facilitation was only for individuals who displayed slow response inhibition at baseline. Therefore it is not so clear whether facilitating effects of stimulants might be limited to individuals with poor inhibitory control.

What can be concluded?

There is a lot more research needed about the relation between drug effects on cognitive functions and their abuse potential. A delay-based assessment of control that has received little research interest is the delay discounting model. Some studies have shown that d-amphetamine decreased discounting of delayed monetary rewards in healthy adults. It is interesting to see whether there is a similar control-enhancing effect of a stimulant drug. Now, it is still unclear if inhibitory mechanisms might also influence discounting behaviour. Therefore, drug studies that compare stimulant drugs to discounting and other delay-based assessments of control could provide a better understanding of the role of impulsivity in stimulant drug use.