Humans are equipped with multiple sensory channels, jointly providing both redundant and complementary information. A primary challenge for the brain is therefore to make sense of these multiple sources of information and bind together those signals originating from the same source while segregating them from other inputs. Whether multiple signals have a common origin or not, however, must be inferred from the signals themselves (causal inference, cf. “the correspondence problem”). Previous studies have demonstrated that spatial coincidence, temporal simultaneity, and prior knowledge are exploited to solve the correspondence problem. Here we demonstrate that cross-correlation, a measure of similarity between signals, constitutes an additional cue to solve the correspondence problem. Capitalizing on the well-known fact that sensitivity to crossmodal conflicts is inversely proportional to the strength of coupling between the signals, we measured sensitivity to crossmodal spatial conflicts as a function of the cross-correlation between audiovisual signals. Cross-correlation (time-lag 0ms) modulated observers’ performance, with lower sensitivity to crossmodal conflicts being measured for correlated than for uncorrelated audiovisual signals. The current results demonstrate that cross-correlation promotes multisensory integration. A Bayesian framework is proposed to interpret the present results whereby stimulus correlation is represented on the prior distribution of expected crossmodal co-occurrence.