Poster #10 - Interactions Among Performance-Based Incentives and Reward Sensitivity on Executive Functions During Early Childhood

• In Event: 1-115 - Poster Session 03
  In Poster Session: PS 03 Section - Attention, Learning, Memory

Integrative Statement

Executive functions (EFs) are higher-order cognitive processes (e.g., inhibitory control) that predict numerous developmental outcomes, such as academic success (Diamond, 2013). Adult research has demonstrated that small-to-moderate rewards enhance EF by directing attention toward task-relevant information (Botvinick & Braver, 2015). Furthermore, motivation-EF associations during adulthood are moderated by reward sensitivity, which encompasses positive affect and arousal in response to rewards (Padmala & Pessoa, 2011), as measured by the Behavior Activation System (BAS) scales (Carver & White, 1994). However, these potential associations during early childhood remain unclear. Research on the effects of reward on children’s EF generally falls under a “Hot-Cool” framework (Zelazo et al., 2010) and utilizes separate tasks, varying in demands, to measure EF in motivationally-salient (“hot”) and neutral (“cool”) contexts (e.g., Brock et al., 2009). Furthermore, it is unknown whether reward sensitivity moderates the effects of incentives on EF in young children.
Using a within-subjects design, we administered two equivalent EF measures to sixty-three 3- to 5-year-olds (30 girls; Mage= 4.61 years, SD= 0.55). Children were presented with a series of sun/moon (Day/Night task; Gerstadt et al., 1994) and elephant/mouse cards (Big/Small task). For the non-Stroop phase of each task, children said the word that “matched” the card. Children then completed the more challenging Stroop phase, saying the opposite (e.g., “night” for sun/“big” for mouse). For one of the two tasks, children received a sticker as a reward for every correct response. Child reward sensitivity was measured using a parent-report version of the BAS (Blair, 2003).
A 2 Phase (non-Stroop/Stroop) × 2 Incentive (no reward/reward) repeated-measures MANCOVA on children’s error rates with reward sensitivity as a covariate revealed a significant Phase × Incentive × Reward Sensitivity interaction, F(1, 61)= 7.03, p= .01. Overall, children made significantly fewer errors on the rewarded condition than the non-rewarded condition during the Stroop phase, but not the non-Stroop phase (see Figure 1) and this effect was moderated by children’s reward sensitivity. This three-way interaction was driven by a significant Incentive × Reward Sensitivity interaction for the Stroop phase only (F= 10.43, p= .002; Non-Stroop: F< 1). Because we are specifically interested in reward-based change in EF, we calculated a difference score between children’s performance on the Stroop phase rewarded and non-rewarded conditions to probe this interaction. This difference score was negatively correlated with children’s reward sensitivity, r(61)= -.38, p= .002 (see Figure 2). Thus, children with higher reward sensitivity either exhibited smaller reward-based improvements, or greater reward-based impairment, on the Stroop phase than children with lower reward sensitivity.
Consistent with adult research, children made fewer errors on the Stroop task when receiving performance-based incentives than while receiving no rewards. This may indicate that small rewards facilitate attentional orienting towards task-relevant stimuli during early childhood. Furthermore, we found that reward sensitivity was inversely related to children’s reward-based improvements in Stroop performance. This suggests that children with high reward-sensitivity may have experienced attentional arousal beyond optimal levels, which has been found to impair cognitive performance in adults (Ariely et al., 2009).

Authors

• Lauren J. Bryant, University of Connecticut
  Presenting Author

• Julie Marino, University of Connecticut
  Non-Presenting Author

• Nicholas Pinnock, University of Connecticut
  Non-Presenting Author

• Michaela Barratt, University of Connecticut
  Non-Presenting Author

• Dilsara Liyanage, University of Connecticut
  Non-Presenting Author

• Kimberly Cuevas, University of Connecticut
  Non-Presenting Author