Based on a functional approach to credibility judgments, the authors hypothesize that receivers’ judgments of senders’ credibility involve an evaluative dimension (i.e., good–bad) and are associated with approach and avoidance tendencies. In three experiments (total N = 645), participants (receivers) judged the credibility of suspects (senders) denying involvement in a mock theft. While watching or reading the message, receivers performed an approach-related (arm flexion) or an avoidance-related (arm extension) motor action. Although receivers’ affective evaluations of senders (good–bad) correlated strongly with credibility judgments in all three experiments, the results of the arm position manipulation were mixed. In Experiment 1, receivers in an arm flexion (vs. arm extension) state judged the sender as more credible, but only when informed beforehand about the upcoming credibility judgment. In Experiment 2 and 3, however, there was no evidence of an arm position effect on credibility judgments. A cross-experimental meta-analysis revealed that the effect of the manipulation was statistically indistinguishable from zero, Hedges’ g = 0.07, 95% CI [−0.09, 0.22], and provided strong support for the null hypothesis. Multiple interpretations of the results are discussed.