Neurocognition in stimulant addiction: commentary on Kendrick et al (2021)

It would be easy to assume that that abusing stimulant drugs such as amphetamine, MDMA (‘Ecstasy’), and cocaine deleteriously impacts the brain and cognitive function. However, logically it could be the case that any neurobehavioural changes seen in drug abusers might have antedated drug taking, and so have been initially latent. An important issue is then to decide whether such changes might have been predisposing or causal factors for drug taking. They might even be considered as possible intermediate phenotypes or endophenotypes according to Gottesman's original concept (Gottesman and Goulds, 2003)—suggestive of possible constitutional or genetic factors leading to addiction. Such a hypothesis is commonly addressed by following the first degree relatives of the drug abuser to determine whether the neurobehavioural changes are also present in them. Of course, any such change might also be attributable to other familial factors, for example, the common environment, including possible stressful factors that may also contribute to neurobehavioural changes.

However, assuming that stimulant drug taking itself may also lead to brain and behavioural changes, based on evidence from animal studies where drug administration can be carefully controlled (George et al., 2008), and on correlations of duration or extent of drug talking with structural brain changes (Ersche et al. 2011a), several important questions remain: to what extent do any neurobehavioural changes produced by the drugs themselves contribute further to a drive to compulsive stimulant use (or addiction)? What is the more general nature of these changes, which might, for example, include such functions as memory and attention? To what extent would these cognitive deficits impede rehabilitation? Can they be remediated? Or do they recover over time? A further important issue is whether the type of drug abuse—for example, opioids, cannabis, and alcohol, as well as polydrug abuse including stimulants—affects the outcome.

Several studies have investigated these questions using functional neuroimaging methods generally in humans and neural interventions in animal studies, and there is support for most, if not all, of these cascading causal facets of stimulant addiction. Thus, there is evidence, for example, of impaired response inhibition processes in stimulant abusers and their siblings that correlates with a reduction of white matter in the vicinity of the right inferior frontal gyrus (Ersche et al., 2012). This supports a hypothesis that impulsivity may lead to addiction, presumably by promoting risky decision making. It fits with evidence in experimental animals that impulsive responding is associated with a down-regulation of dopamine D2 receptors in the nucleus accumbens, mirroring results in human cocaine abusers (Dalley et al., 2007). This tendency to impulsivity in rats has been shown to predict compulsive cocaine seeking, and one prominent hypothesis links this to a devolution of control from ventral to dorsal striatum, accompanied by impaired top-down control from the prefrontal cortex over habit development, known to be mediated by the dorsal striatum, specifically the putamen. Hence, compulsive drug seeking is hypothesized in part to represent a change in the balance between goal-directed behaviour and habits in favour of the latter (Everitt and Robbins, 2016). Whether this bias is also evident before drug taking or in first degree relatives is, however, more controversial. For example, some studies have observed evidence of a larger putamen in the first degree relatives of stimulant abusers (Ersche et al., 2013), which could be interpreted to represent a greater propensity to habit learning, if one assumes (as for the celebrated example of the hippocampus of taxi drivers, McGuire et al., 2000) that greater use implies greater structural change. Additionally, recent studies of resting state functional connectivity have suggested less connectivity of the ventromedial prefrontal cortex and caudate nucleus, both in stimulant abusers and in their relatives, suggestive of an impaired ‘goal-directed system’, in relation to the ‘habit system’ (Ersche et al., 2020).

An important limitation of many of the studies described here is that they are cross-sectional in nature; that is, they generally measure a snapshot of neurocognitive function at a particular time that is almost invariably after drug abuse has begun. Definitive evidence of causal factors in addiction should ideally include measures at several points in an individual's life history, i.e. before drug abuse, during the course of abuse, and ideally after long-term abstinence, to determine whether any neurobehavioural or cognitive effects are permanent. Largely on the basis of on cross-sectional studies, however, it is generally believed that for many individuals there are long-lasting neurocognitive deficits that diminish over time during abstinence (Ersche et al., 2006).

Hence, the current study by Kendrick et al. (2021) is of special interest as it examines brain structure and cognitive function in a prospective study of 17 very early stimulant users, repeated at 12 months. Two sub-groups of drug users, apparently matched according to demographic and premorbid indices, are identified according to subsequent LOW or HIGH stimulant usage. A comprehensive cognitive test battery reveals deficits in the HIGH group in verbal and working memory and attentional function that correlated strongly with reductions in grey matter in the putamen. This study is interesting in deflecting focus from the reductions in grey matter that have been reported to occur in the prefrontal and temporal cortex in stimulant abusers and that have been associated with cognitive impairments (Porrino et al., 2010). The reduction in putamen volume is not consonant with the simple notion that putamen size reflects habitual propensity to compulsive behaviour. However, it is possible, as the authors suggest, that this change may occur at a later point in stimulant addiction. The possible involvement of the putamen in cognitive deficits is intriguing in view of the considerable evidence that cognitive impairment in Parkinson's disease, including in working memory and attention, may be partly due to dopamine-dependent reductions in the dorsal striatum, including the putamen (Cheesman et al. 2005 ). While such cognitive deficit is usually linked to caudate nucleus involvement, a recent study has revealed correlations of cognitive and attentional impairments to reduced putamen volume (Hunerli et al., 2019). Moreover, planning impairments in patients with obsessive-compulsive disorder have been associated with reduced connectivity of the putamen with the dorsolateral prefrontal cortex (Vaghi et al., 2017). A recent publication in marmoset monkeys has shown impairments in serial reversal learning following inactivation of the putamen (Jackson et al., 2019), behaviour that is often impaired in stimulant abusers (Ersche et al., 2011b). Other recent studies have highlighted the putamen as an early focus of effects of stimulant abuse. Thus, ongoing stimulant use has been associated with reduced functional connectivity of the ventrolateral putamen, supramarginal gyrus, and insula, presumably because of the effects of stimulants on dopamine function in the striatum (Ersche et al., 2020). Moreover, stimulant abusers have recently been shown to exhibit reduced glutamate turnover in the putamen, correlating with their tendencies to ‘automaticity’ as measured by a habit questionnaire (Ersche et al., 2021). The present study therefore places a new focus on how the putamen may also contribute cognitive impairments in stimulant abusers. It will be interesting in future studies of a larger sample of these patients to track the evolution of these changes, especially following abstinence.