The polyvagal theory (Porges, 2007) proposes that physiological flexibility dependent on heart-brain interactions is associated with prosociality. So far, whether prosociality has a causal effect on physiological flexibility is unknown. Previous studies present mitigated results on this matter. In a randomized double-blind protocol, we used a generation of social closeness procedure against a standardized control condition in order to manipulate social affiliation as a prosocial interaction factor. High frequency heart rate variability (HF-HRV, indexing physiological flexibility), electromyographical activity of the corrugator supercilii (sensitive to the valence of the interaction) and self-reported measure of social closeness were monitored before, during, and after experimental manipulation. Cooperation was measured after the experimental manipulation as an index of behavioral prosociality. Data reveal no evidence toward and effect of the experimental manipulation on these measures. We discuss methodological aspects related to the experimental constraints observed in social psychophysiology. Implications for the experimental test of the polyvagal theory are approached within alternative theoretical frameworks.