Objective <p>This study aims to systematically investigate activation in brain regions during early-stage motor sequence learning (MSL) and to identify convergent neural activation patterns, thereby advancing our understanding of the neural mechanisms underlying MSL.</p> Methods <p>This meta-analysis integrated functional magnetic resonance imaging (fMRI) data from 12 studies (219 participants) using the serial reaction time task (SRTT) paradigm, applying anisotropic effect-size seed-based <i>d</i> mapping (AES-SDM) to identify common brain activation regions underlying MSL. After the primary meta-analysis, subgroup analyses were conducted on the young adult group, implicit learning group, and bimanual/right-hand task group to investigate the specific effects of age and task type on brain region activation in MSL. Additionally, an independent fMRI dataset from 64 healthy participants performing SRTT was used to validate the meta-analysis results.</p> Results <p>We identified convergent activations during SRTT in cortico-basal ganglia-cerebellar circuit, encompassing the supplementary motor area (SMA), precentral gyrus (PreCG), rolandic operculum (ROL), and other cortical regions, alongside basal ganglia and cerebellum, collectively underscoring their critical roles in early-stage MSL. Subgroup analyses revealed age- and task-related modulation of neural activation patterns. Results were largely replicated in the independent validation dataset.</p> Conclusions <p>Our findings provide reliable evidence for the involvement of a cortico-basal ganglia-cerebellar circuit in MSL.</p> Significance <p>Our meta-analysis contributes to unraveling the brain activation patterns underlying early-stage MSL, offering insights for clinical interventions in motor dysfunction rehabilitation.</p>