The team, led by Mingyu Lu and colleagues at Shenyang Normal University’s College of Sports Science, searched eight databases, including PubMed, Embase, Web of Science, the Cochrane Library, CNKI, Wanfang Data and VIP, from their inception to January 2026. From 3,144 records initially identified, screening and deduplication left just 12 randomized controlled trials, all conducted in China and published between 2017 and 2025. Nine were two-arm studies and three were three-arm trials, together involving 676 individuals diagnosed with substance use disorders. Nine of the trials focused on methamphetamine dependence, one on heroin, and two did not specify the primary substance. Interventions lasted from four weeks to four months, with sessions of 25 to 60 minutes delivered two to five times per week.

The scope of the review was deliberately narrow. Participants had to meet DSM-5 diagnostic criteria for substance use disorder involving illicit drugs, excluding tobacco and alcohol, and studies had to compare chronic exercise against treatment as usual, which typically included drug prevention education, legal education, daily observation and hygiene training, with structured physical activity explicitly avoided. The outcome measures were the three classic pillars of executive function: inhibitory control, assessed with Stroop, Go/No-Go, Flanker or stop-signal tasks; working memory, assessed with N-back tasks; and cognitive flexibility, assessed with task-switching paradigms. Reaction time served as the primary metric throughout, since shorter latencies reflect better performance and all included studies reported it, whereas some did not report accuracy.

Why does executive function matter so much in addiction? Long-term drug use is associated with reduced activity in the prefrontal cortex and dysfunction of dopamine-mediated fronto-striatal circuits, which weakens inhibitory control, erodes working memory, increases impulsive decision-making and makes it harder to switch flexibly between behavioral strategies. The result is a vicious cycle in which automatic craving becomes difficult to suppress and compulsive drug-seeking intensifies, raising relapse risk. Mainstream treatments remain pharmacotherapy and psychosocial rehabilitation, but substitution drugs carry their own dependence risks, and psychological therapies face therapist shortages, poor compliance, high costs and elevated dropout rates. Exercise, endorsed by the World Health Organization as a low-cost, non-pharmacological adjunct, is thought to help by activating dopamine reward pathways, promoting brain-derived neurotrophic factor, enhancing neuroplasticity and improving cerebral blood flow.

The traditional pairwise meta-analysis delivered the review’s most robust finding. Compared with treatment as usual, aerobic exercise significantly improved inhibitory control, with a standardized mean difference of -0.70 and a 95 percent confidence interval of -0.94 to -0.47, and statistical heterogeneity was low to moderate at 33 percent. Egger’s test and funnel plot inspection found no publication bias. Subgroup analyses, however, revealed a striking dose-related pattern: sessions lasting 35 to 40 minutes and 60 minutes produced significant improvements, whereas a single 30-minute session did not reach significance, with a confidence interval spanning zero. The authors caution that some subgroups contained only one study, making these exploratory observations rather than evidence-based prescriptions. Sensitivity analyses using a leave-one-out approach confirmed the stability of the overall effect, although one study conducted on an indoor treadmill in an enclosed environment emerged as the main source of heterogeneity.

The network meta-analysis, which combines direct and indirect evidence to rank multiple interventions simultaneously, then produced the review’s headline comparisons. For inhibitory control, drawing on all 12 studies, resistance exercise ranked highest with a SUCRA value of 96.7 percent, followed by mind-body exercise at 68.0 percent, aerobic exercise at 35.4 percent and treatment as usual at 0.4 percent. Resistance exercise was significantly superior to aerobic exercise in pairwise comparison, with a standardized mean difference of -0.89. For working memory, based on six trials, mind-body exercise dominated with a SUCRA of 99.4 percent and was significantly more effective than both aerobic exercise and resistance exercise. For cognitive flexibility, from five trials, aerobic exercise led with a SUCRA of 68.3 percent and significantly outperformed treatment as usual, while resistance and mind-body exercise did not reach statistical significance against the control condition.

The authors offer tentative mechanistic explanations for these differential patterns, while flagging that most come from theory rather than direct evidence in this population. Resistance exercise’s possible advantage for inhibitory control may relate to its immediate strength feedback and visible load progression, which can enhance self-efficacy and psychological resilience, as well as its anaerobic character, which might stimulate release of brain-derived neurotrophic factor and insulin-like growth factor-1. Mind-body practices such as Tai Chi and Qigong resemble a form of dynamic meditation, integrating movement with mental concentration and controlled respiration; functional near-infrared spectroscopy research has found that Tai Chi enhanced activation of the right dorsolateral prefrontal cortex and its connectivity with the ventrolateral prefrontal cortex, regions central to working memory updating. Aerobic exercise’s edge for cognitive flexibility may rest on improved cerebral perfusion, since regular aerobic activity increases the diameter of major cerebral arteries and supports oxygen delivery to the prefrontal cortex.

None of these rankings should be taken at face value, and the review is unusually explicit about why. Five of the 12 trials were rated high risk of bias using the Cochrane ROB 2.0 tool, with the greatest concerns arising from deviation from intended interventions, because blinding participants and instructors in exercise trials is inherently difficult. The CINeMA framework rated the certainty of evidence as mostly low or very low, with only a handful of comparisons reaching moderate confidence. The network structure was sparse: some comparisons, such as resistance versus mind-body exercise, had no direct evidence at all and relied entirely on indirect pathways through treatment as usual. The authors also judged the transitivity assumption, which underpins all indirect comparisons, to be clinically fragile, given that the trials varied in substance type, participant demographics and intervention dosing. Non-significant inconsistency tests, they note, reflect limited statistical power rather than confirmed consistency.

Further limitations constrain generalizability. All included studies were conducted in China, and most participants had methamphetamine or heroin use disorders, so the findings may not extend to other substances, countries or treatment systems. Only immediate post-intervention effects were reported, with no long-term follow-up data, leaving the durability of any benefits unknown. The analysis treated the three executive function domains as independent outcomes despite their theoretical and neural overlap, which may inflate the risk of false-positive findings. Publication bias was not detected for inhibitory control or working memory, but the funnel plot for cognitive flexibility was asymmetrical, and small-study effects were apparent across all three outcomes.

What, then, can readers legitimately take away? The review’s own conclusion is measured: different exercise modalities may exert differential effects on specific executive function subdomains in Chinese individuals with substance use disorders, and any form of chronic exercise appears preferable to no exercise as an adjunctive measure in rehabilitation. But the authors insist the findings cannot establish definitive intervention rankings, a single optimal strategy, or precise exercise prescription parameters. Instead, they frame the work as directional evidence for designing future large-scale, multicenter randomized trials with head-to-head comparisons, clearly defined substance subgroups and thorough dose-response evaluation. Until such confirmatory studies arrive, the apparent hierarchy of resistance, mind-body and aerobic exercise for inhibitory control, working memory and cognitive flexibility respectively remains an intriguing hypothesis rather than a treatment guideline, and the global treatment gap, in which only one in twelve people with drug use disorders received care in 2023, still awaits better answers.

Subject of Research: Comparative effects of exercise modalities on executive function in individuals with substance use disorders

Article Title: Chronic effects of different exercise modalities on executive function in Chinese individuals with substance use disorders: a systematic review and network meta-analysis

Article References: Lu, M., Tian, Y., Wang, J., & Liang, Y. (2026). Chronic effects of different exercise modalities on executive function in Chinese individuals with substance use disorders: a systematic review and network meta-analysis. Frontiers in Psychology, 17, 1979254. https://doi.org/10.3389/fpsyg.2026.1979254

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1979254

Keywords: substance use disorders, executive function, network meta-analysis, aerobic exercise, resistance exercise, mind-body exercise, Tai Chi, inhibitory control, working memory, cognitive flexibility, methamphetamine, randomized controlled trials