Wearable sensor-based haptic feedback systems for motor learning and gait reshaping: a review

Purpose This review aims to examine wearable sensor-based haptic feedback systems for motor learning and gait reshaping, with emphasis on the closed-loop chain from sensing to behavioral change. Design/methodology/approach This structured narrative review used a transparent search boundary. Literature was identified from PubMed/MEDLINE, Web of Science, Scopus, IEEE Xplore and Google Scholar for work published from January 2000 to May 2026 using combined terms for wearable sensors, haptic or vibrotactile feedback, biofeedback, gait, motor learning, Parkinson disease, stroke, older adults and foot-progression retraining. Studies were included when they reported wearable sensing, closed-loop or cue-based feedback, gait or motor-learning outcomes, validation metrics or deployment constraints; papers were excluded when they were unrelated to gait, lacked wearable sensing/feedback relevance, provided no usable quantitative or architectural information or duplicated a more complete report. Because outcomes and populations were heterogeneous, findings were synthesized as a structured narrative with quantitative comparison tables rather than a pooled meta-analysis. Findings Across the studies summarized in Table 2, the strongest support is for short-term modulation of stance symmetry, thigh extension, gait speed, arm swing and foot-placement targets; reported effects include p = 0.0493 for stroke stance-time symmetry, 14% stride-length and 18% gait-speed increases in older adults, a 229% arm-ROM increase and 2.4 degrees RMS foot-progression-angle tracking error with ICC = 0.94. One six-week knee-osteoarthritis gait-retraining study also reported a 20% reduction in first-peak knee adduction moment and retention of biomechanical and symptom changes at one month. These heterogeneous values do not support a single pooled effect size because sample sizes, comparators, cue schedules and endpoints differ. Sensor accuracy alone is insufficient; repeatability, calibration, latency, comfort, feedback dose, perceptual load, retention and transfer determine whether closed-loop systems remain useful outside the laboratory. Originality/value The review treats wearable haptic gait training as an integrated sensor-feedback-control problem rather than as a catalog of devices.