AbstractWhile numerous studies have shown that visual perceptual learning (VPL) occurs as a result of exposure to a visual feature in a task-irrelevant manner, the underlying neural mechanism is poorly understood. In a previous psychophysical study, subjects were repeatedly exposed to a task-irrelevant global motion display that induced the perception of not only the local motions but also a global motion moving in the direction of the spatiotemporal average of the local motion vectors. As a result, subjects enhanced their sensitivity only to the local moving directions, suggesting that early visual areas (V1/V2) that process local motions are involved in task-irrelevant VPL. However, this hypothesis has never been examined by directly examining the involvement of early visual areas (V1/V2). Here, we employed a decoded neurofeedback technique (DecNef) using functional magnetic resonance imaging. During the DecNef training, subjects were trained to induce the activity patterns in V1/V2 that were similar to those evoked by the actual presentation of the global motion display. The DecNef training was conducted with neither the actual presentation of the display nor the subjects’ awareness of the purpose of the experiment. As a result, subjects increased the sensitivity to the local motion directions but not specifically to the global motion direction. The training effect was strictly confined to V1/V2. Moreover, subjects reported that they neither perceived nor imagined any motion during the DecNef training. These results together suggest that that V1/V2 are sufficient for exposure-based task-irrelevant VPL to occur unconsciously.Significance StatementWhile numerous studies have shown that visual perceptual learning (VPL) occurs as a result of exposure to a visual feature in a task-irrelevant manner, the underlying neural mechanism is poorly understood. Previous psychophysical experiments suggest that early visual areas (V1/V2) are involved in task-irrelevant VPL. However, this hypothesis has never been examined by directly examining the involvement of early visual areas (V1/V2). Here, using decoded fMRI neurofeedback, the activity patterns similar to those evoked by the presentation of a complex motion display were repeatedly induced only in early visual areas. The training sensitized only the local motion directions and not the global motion direction, suggesting that V1/V2 are involved in task-irrelevant VPL.