Cross-modality effect in implicit learning of temporal sequence

Although implicit learning of temporal structure in the uni-modal are widely studied, the effect of cross-modal switching on implicit learning of temporal sequence is currently unknown. The present study adopted the modified serial reaction time (SRT) task based on the temporal sequence. After learning, the attribution test was used to access the consciousness state of acquired knowledge. A total of 116 healthy participants were randomized to four groups based on the sensory channels in the front and back halves of the task, AA (auditory-only), VV (visual-only), AV (auditory-visual), and VA (visual-auditory). The results showed that temporal sequences were acquired implicitly for all groups. In particular, the results also showed that temporal sequences can be successfully switched across modalities during implicit learning, regardless of modality switching directions. Most importantly, a significant gain was found only in the VA group during the learning transition compared to the VV group. The attribution test showed similar results. These results supported the argument that temporal perception was based on an internal clock model, as well as temporal sequence information is dependent on auditory representations.