Objective: The aim of this study was to evaluate the distribution of generated stress around implants and adjacent bone tissue using different implant-retained overdenture designs through photoelastic analysis. Methods: Over an edentulous human mandible, achieved from a human model, 2 or 4 microunit analog abutments were embedded (Master; Conexao Systems Prosthodontics, Sao Paulo, Brazil), settled in the interforaminal region. Three models of photoelastic resin (Araltec Chemicals Ltda, Hunstman, Guarulhos, Sao Paulo, Brazil), with 2 or 4 incorporated implants and microunit abutments, were obtained from molds using silicone for duplication. Inclusion, finishing, and polishing procedures were applied on the frameworks. This study was based on 3 different mechanisms of implant-retained mandibular overdentures: O'ring (GI), bar-clip (GII) (both with 2 implants), and their association (GIII) (with 4 implants). After the adaptation of each overdenture system on the photoelastic models, 100-N alternate occlusal loads were applied on back-side and frontside regions. The photoelastic analysis was made with the aid of a plain polariscope linked to a digital camera, Sony Cybershot > 100, which allowed visualization of the fringes and registration of images on digital photographs. Results: The results demonstrated higher tension concentrated over the GIII, with a flat distribution of stress to the posterior ridge and overload on the posterior implants. GI showed the smaller stress level, and GII, intermediate level; there was distribution of stress to the posterior ridge in these 2 groups. Conclusion: The use of bar attachment proved to be a better alternative, because it showed a moderate level of tension with a more uniform stress distribution and possessed higher retention than did the ball system.22623322336State of Sao Paulo, Brazil [2005/03595-9]State of Sao Paulo, Brazil [2005/03595-9]