Motivational and cognitive determinants of control during conflict processing

Abstract

Previous studies suggest that both reward anticipation and expected or experienced conflicts activate cognitive control. The present study investigated how these factors interact during conflict processing. In two experiments, participants performed a variant of the Stroop task, receiving performance-dependent monetary rewards in some blocks. In addition, we manipulated the level of conflict-triggered reactive and expectancy-driven proactive control: In Experiment 1, we compared the Stroop effect after previously congruent and incongruent trials to examine the conflict adaptation effect (reactive control). We found that the level of motivation did not interact with conflict adaptation. In Experiment 2, we varied the proportion of congruent and incongruent trials to manipulate conflict expectancy (proactive control). The data suggest the effects of motivation to be less pronounced under conditions of high conflict expectancy. We conclude that the interaction of motivation with cognitive determinants of control depends on whether these activate proactive or reactive control processes.

Keywords:

• Reward
• Conflict processing
• Conflict adaptation effect
• Conflict expectancy effect

We thank Klaus Rothermund for his helpful comments on an earlier version of the manuscript.

We thank Klaus Rothermund for his helpful comments on an earlier version of the manuscript.

Notes

¹ We note that the comparison of conflict adaptation effects between low-motivation and high-motivation blocks may potentially be influenced by the general RT speed-up in high-motivation blocks. To control for this potential confound, we re-analysed our data with logarithmised RTs. Importantly, the ANOVA on logarithmised RTs showed no significant Motivation × proactive proactive Previous trial congruency × Current trial congruency interaction, F < 1, p > .38, indicating that the non-significant three-way interaction of the ANOVA on RTs cannot be attributed to the general RT speed-up in the high-motivation condition.

² Similar to Experiment 1, we computed an ANOVA using logarithmised RTs to test whether the significant Motivation × Expectancy × Congruency interaction was affected by the general RT speed-up in high motivation blocks. Because the ANOVA on logarithmised RTs revealed a significant Motivation × Expectancy × Congruency interaction, F(1,15) = 9.72, p < .01, η _p ² = .393, we conclude that the result of this three-way interaction cannot be attributed to the faster RTs in the high motivation condition.

Additional information

Funding

The present research was supported by a grant of the German Research Foundation to T.S. as well as by a grant of CoTeSys [No. 439] to T.S. A.S. was supported by a Ph.D. scholarship of the Bavarian Elite Aid Act (BayEFG).