Reward Evokes Visual Perceptual Learning Following Reinforcement Learning Rules

Visual perceptual learning (VPL) is defined as a long-term performance enhancement as a result of visual experiences. A number of studies have demonstrated that reward can evoke VPL. However, the mechanisms of how reward evoke VPL remain unknown. One possible hypothesis is that VPL is obtained through reward related reinforcement processing. If this hypothesis is true, learning can only occur when reward follows the stimulus presentation. Another interpretation is that VPL is acquired through an enhancement of alertness in association with reward. If the alertness hypothesis is true, learning should occur when reward precedes the stimulus presentation. In our study, we tested the plausibility of the two hypotheses by manipulating the order of reward and stimulus presentation. In Experiment 1, we separated participants into two groups. During training, the ‘Before’ group received water reward 400ms prior to the onset of trained orientation stimulus while the ‘After’ group received water reward 400ms subsequent to the onset of trained orientation stimulus. Both groups were trained using the Continuous Flash Suppression paradigm to render the stimulus imperceptible to the participants by the presentation of dynamic noise in the untrained eye. We found training only in the ‘After’ group indicating that reward may evoke learning through reinforcement-like processing. In Experiment 2, we excluded the possibility that alertness may not be sufficient to elicit learning when presented before stimulus. We presented beep sound prior to the onset of stimulus to increase alertness. Our finding demonstrated that alertness is sufficient enough to evoke learning. In conclusion, our study provided evidence that reward can evoke VPL through reinforcement process.