%A ZHANG Kuo, ZHAO Ying, WANG Jingxin, %T Optimization and asymmetry effects of reward and punishment on control attention: Evidence from eye movements %0 Journal Article %D 2019 %J Acta Psychologica Sinica %R 10.3724/SP.J.1041.2019.01207 %P 1207-1219 %V 51 %N 11 %U {https://journal.psych.ac.cn/xlxb/CN/abstract/article_4555.shtml} %8 2019-11-25 %X

A wealth of research shows that positive and negative reinforcement critically influence behavior. While it is well established that rewards and penalties can strongly influence mechanisms of executive control, it is unclear whether these two factors exert symmetric or qualitatively distinct behavioral effects. In the current research, we conducted two eye-movement experiments to investigate the influence of monetary reward or punishment on attentional control. We employed these cues in Pro/Anti-saccade tasks in Experiment 1 and Go/No-go tasks in Experiment 2. Crucially, we investigated how either a reward (also referred to as “gain”) or penalty (also referred to as “loss”) influenced inhibitory control in the following trial.

In Experiment 1, participants were instructed to produce simple pro-saccades or more difficult anti-saccades, in conditions in which they received a reward for correct responses or a punishment for incorrect responses or either a reward or punishment. The results showed that, while the accuracy of the pro-saccades was facilitated by reward, the accuracy of the anti-saccades was facilitated by punishment. And the velocity of pro-saccades and anti-saccades were significantly improved by both reward and punishment. In Experiment 2, we further adopted the Go/No-go tasks to explore how reward and punishment affect attentional control via exogenous parafoveal visual cues. This showed essentially the same pattern of effects as Experiment 1. For the Go task, saccade latency significantly decreased when rewards were given relative to punishment or no motivation conditions. And for the No-go task, accuracy increased more in the punishment condition compared to the reward or no motivation conditions. An increase in saccade velocity was observed in the no motivation condition, similarly to in Experiment 1.

In sum, the overall results suggest that both reward and punishment can facilitate the oculomotor control, although the findings reveal a striking asymmetry in the effects of the reward and punishment on behavior. Specifically, positive reinforcement appears to improve approach behaviors, while punishment influences inhibitory behavior. These findings suggest that the two forms of reinforcement are distinct in their influence on behavior.