Cognitive regulation of ventromedial prefrontal activity evokes lasting change in the perceived self-relevance of persuasive messaging

Abstract

Persuasive messages can change people's thoughts, feelings, and actions, but these effects depend on how people think about and appraise the meaning of these messages. Drawing from research on the cognitive control of emotion, we used neuroimaging to investigate neural mechanisms underlying cognitive regulation of the affective and persuasive impact of advertisements communicating the risks of binge drinking, a significant public health problem. Using cognitive control to up-regulate (vs. down-regulate) responses to the ads increased: negative affect related to consequences of excessive drinking, perceived ad effectiveness, and ratings of ad self-relevance made after a one-hour delay. Neurally, these effects of cognitive control were mediated by goal-congruent modulation of ventromedial prefrontal cortex and distributed brain patterns associated with negative emotion and subjective valuation. These findings suggest that people can leverage cognitive control resources to deliberately shape responses to persuasive appeals, and identify mechanisms of emotional reactivity and integrative valuation that underlie this ability. Specifically, brain valuation pattern expression mediated the effect of cognitive goals on perceived message self-relevance, suggesting a role for the brain's valuation system in shaping responses to persuasive appeals in a manner that persists over time.

Keywords: cognitive control; emotion; fMRI; persuasion.

© 2019 Wiley Periodicals, Inc.

Figures

Figure 1

(a) Scanner cognitive regulation task and post‐scanner ad re‐exposure task. In the scanner task, participants saw a cue to either look naturally, to up‐regulate their response, or to down‐regulate their response (using an emotion‐focused strategy or a persuasion‐focused strategy). In the post‐scan re‐exposure task, participants viewed images they had seen in the scanner task and rated their current negative affect, perceived ad effectiveness, and ad self‐relevance. (b) Multilevel mediation approach. We used within‐person mediation to ask whether a priori brain regions and patterns of interest (the mediator variables) could explain the effect of experimentally manipulated cognitive up‐ vs. down‐regulation (the predictor variable) on ratings of negative emotion, perceived ad effectiveness, and self‐relevance (the outcome variables) [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

Behavioral results. Cognitive regulation evoked goal‐congruent modulation of (a) negative affect and perceived ad effectiveness ratings made in the scanner task, as well as (b) change in ad self‐relevance ratings made in the re‐exposure task [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 3

Effects of trial type on brain activity for (a) vmPFC, a region identified as showing a whole‐brain corrected omnibus main effect of trial type, and (b) expression of multivariate patterns associated with negative emotion and valuation. (Plotted estimates reflect posterior means with 95%CI) [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 4

Multilevel mediation. Effects of cognitive regulation—that is experimental instructions to up‐regulate versus down‐regulate one's response to an ad—on (a) in‐scanner negative affect, (b) in‐scanner perceived effectiveness, and (c) message self‐relevance at re‐exposure were mediated by cognitively‐driven change in brain responses associated with emotion and value. Path coefficients represent posterior means with 95% credible intervals; paths with 95% intervals that cross zero shown in lighter gray. Within line plots, black lines reflect overall group estimate, light gray lines reflect person‐specific estimates. Visualization of posterior distributions reflect kernel density estimates of MCMC draws, with results from four parallel chains overlaid