Mankind collects the information from multi-sensory modalities when detecting and perceiving objects and events. Such multimodal inputs, after being achieved by the brain, are rapidly integrated to form cognitive processes, achieve perception, and finally guide people’s behaviors. Audio-visual integration is extremely common in daily life and plays an important role in human cognitive activities. Therefore, audio-visual cross-modal interaction has become a hot topic in recent years. This paper gives an overview of audio-visual interaction and describes the achievement by focusing on the enhancement of cognition with congruent audio-visual stimuli, the adjustment of perception with incongruent visual-auditory information, and emphasizes the development of uni-sensory dominant mechanism, based on which  the outlook of  research directions is discussed.