IntroductionCognitive impairment is recognised as an important non-motor symptom in Parkinson’s disease (PD) and there is a need for evidence-based non-pharmacological interventions that may prevent or slow cognitive decline in this patient group. One such intervention is computerised cognitive training (CCT), which has shown efficacious for cognition across older adult populations. This systematic review aims to investigate the efficacy of CCT across cognitive, psychosocial and functional domains for people with PD and examine study and intervention design factors that could moderate CCT effects on cognition.Methods and analysisRandomised controlled trials investigating the effects of CCT in patients with PD without dementia, on cognitive, psychosocial or functional outcomes, will be included. The primary outcome is overall cognitive function. Secondary outcomes are domain-specific cognitive function, psychosocial functioning and functional abilities. We systematically searched MEDLINE, Embase and PsycINFO through 14 May 2020 to identify relevant literature. Risk of bias will be assessed using the revised Cochrane Risk of Bias tool. Effect sizes will be calculated as standardised mean difference of baseline to postintervention change (Hedges’ g) with 95% CI for each eligible outcome measure. Pooling of outcomes across studies will be conducted using random-effects models, accounting for dependency structure of effect sizes within studies. Heterogeneity will be assessed using τ² and I² statistic. Potential moderators, based on key study and intervention design factors, will be investigated using mixed-effects meta-regression models.Ethics and disseminationNo ethical approval is required. The findings will be disseminated in a peer-reviewed scientific journal.PROSPERO registration numberCRD42020185386.