To investigate the effects of sensory amplitude electrical stimulation (SES) delivered by glove electrode during task-specific exercise on arm movement, function, and sensation in chronic stroke. The design was an intervention pilot study, pre-test, post-test, follow-up design. The settings used were a university research laboratory and home-based intervention. Participants comprised of 11 individuals with chronic stroke (7.2 ± 4.1 years post onset) and moderate arm paresis, 10.82/20 ± 2.27 on the Stroke Rehabilitation Assessment of Movement (STREAM) - Arm Subscale. Participants were seven males and four females (mean age: 59 years). Participants were recruited from university-based database. Intervention- Participants engaged in task-specific training at home for 30 min, twice daily, for 5 weeks, while receiving SES via glove electrode. Participants received supervised task practice at least twice during intervention period for 1 hour. Main outcome measures- Jebsen-Taylor Hand Function Test (JTHFT), STREAM - Arm Subscale, Motor Activity Log-14 (MAL-14) - Amount and Quality Subscales, and Nottingham Stereognosis Assessment (NSA). Significant changes were found in group mean pre- and post-test comparisons on the NSA (P = 0.042), MAL amount subscale (P = 0.047), and JTHFT (with writing item 29 excluded) (P = 0.003) and in pre-test to follow-up comparisons on NSA (P = 0.027) and JTHFT (writing item excluded) (P = 0.009). There was no significant change on the STREAM (P = 1.0). Individuals with a greater baseline motor capacity determined by STREAM scores (P = 0.048) and more recent stroke (P = 0.014) had significantly greater improvements. Combining task-specific training with glove-based SES in chronic stroke resulted in changes in arm sensation and function that were maintained at 3-month follow-up.