BACKGROUND AND PURPOSE: Action observation (AO) generates motor simulations in the brain, activating areas involved in actual movement and facilitating imitation. This randomized controlled trial aimed to analyze if observing actions performed by a personally selected referent improves balance in asymptomatic individuals compared to observing a standard referent. METHODS: A total of 165 healthy participants were randomized into three groups: a control group observing landscape videos, a standard referent group observing an unidentifiable model, and a chosen referent group selecting models based on personal similarity. Balance (static and dynamic) and strength were assessed at baseline, immediately post-intervention, at 15 days, and at 4 weeks. AO exercises related to balance were viewed three times per week for four weeks. RESULTS: Key findings revealed favorable changes in static balance for both the standard and chosen referent groups, particularly in challenging single-leg stances with eyes closed. Strength improvements were observed only in the standard referent group. Average adherence to the protocol was 91%. CONCLUSION: AO enhances balance and strength in healthy individuals, with referent characteristics significantly impacting outcomes. The findings suggest referent selection is a crucial factor in AO's efficacy, emphasizing AO's therapeutic potential, particularly for populations with mobility challenges. Further research should refine referent selection to optimize clinical results.