The influence of performance contingent and non-contingent rewards on children’s cognitive control engagement

• In Event: Poster Session 11
  In Poster Session: PS 11 Topic - Attention, Learning, Memory

Abstract

Provision of external rewards has been suggested to enhance cognitive control specifically by promoting preparatory, proactive control engagement in both children and adults. Offering external rewards has also been widely used for motivating children at school and at home. However, how children may respond to different rewarding manners has not been elucidated. In adults, performance contingent and non-contingent rewards act differently: contingent rewards promote proactive engagement, whereas the effect of non-contingent rewards is alike to positive emotions which may impede proactive control (e.g. Fröber and Dreisbach, 2014, 2016). Nevertheless, our recent study (Jin, Auyeung, and Chevalier, 2020) found that both reward and positive emotion could increase proactive control in childhood, especially for younger children who tend to rely mostly on reactive control (only respond to occurred events). Therefore, it is likely that, unlike adults, children might engage proactive control in response to both contingent and non-contingent rewards.

The current study investigated how performance contingent and non-contingent rewards affect cognitive control engagement in childhood, using an AX-CPT task (Fig. 1) with 5- to 6-year-olds (M = 6.26 years, SD = 0.44, n = 61), 9- to 10-year-olds (M = 9.86 years, SD = 0.58, n = 60) and young adults (M = 19.25 years, SD = 1.57, n = 57). In the AX-CPT, contextual cue primes can be either sustainably maintained before or reactively retrieved after the onset of target probes, which allows the dissociation between proactive (i.e., cue preparation) and reactive (i.e., target-evoked response) processes. All participants first completed a baseline phase in which no reward information was provided, followed by a reward phase in which they could win reward points. Specifically, the contingent group could collect a reward point after making a correct response in the reward-cued trial, whereas the non-contingent group could receive the reward point after every reward-cued trial even for errors.

The contingent group responded more accurately than the non-contingent group in the reward phase, suggesting that rewards have a greater overall influence when contingent on performance. However, compared to the baseline phase, external reward, regardless of performance-contingency, led to better performance (i.e., greater accuracy and faster responses) in trials where proactive control was efficient, suggesting more proactive engagement even when the reward was non-contingent (Fig. 2a-2b). Both children and adults had higher accuracy for reward trials than non-reward trials (Fig. 2c). Moreover, while older children and adults kept their responding pace constant across trials, younger children in the contingent group speeded up in reward trials than in non-reward trials (Fig. 2d).

In summary, these findings suggest that, as hypothesised, children are sensitive to both performance contingent and non-contingent rewards. Both types of reward enhanced proactive control engagement in all three age groups. These findings not only shed light on the effect of external rewards on children’s cognitive processes but also have important implications for educational fields, suggesting that non-contingent rewards seem to be sufficient for encouraging mature forms of control in children but contingent rewards may be more effective for generally promoting cognitive performance.

Authors

• Xiaoyu Jin, University of Edinburgh
  Presenting Author

• Bonnie Auyeung, University of Edinburgh
  Non-Presenting Author

• Nicolas Chevalier, University of Edinburgh
  Non-Presenting Author