Search TESS:


Self-Affirmation and Implementation Intentions: The Moderating Role of Discrete Emotion

Download data and study materials

Download proposal


Principal Investigator(s):

Rebecca Ferrer
National Cancer Institute

William M. P. Klein
National Cancer Institute

Sample size: 747
Field period: 07/23/2014-02/24/2015



This experiment examines whether emotion moderates the effect of self-affirmation on intentions to engage in proactive behavior following a message about a health threat. Specifically, we examine whether self-affirmation – a process by which individuals reflect on cherished personal values – differentially affects the persuasiveness of a message about the link between alcohol and breast cancer depending on whether individuals are in a particular emotional state. Previous evidence suggests that self-affirmation may reduce defensiveness to threatening health information, increasing openness to the message and resulting in increased disease risk perceptions, disease-related worry, and intentions to engage in preventive behavior. However, self-affirmation may be differentially effective depending on the prior emotional state of the individual. Women who report having consumed one or more alcoholic beverage in the past month were randomly assigned to write about an emotional event (something that made them happy, fearful, angry, or surprised) or to a neutral emotion condition. Then, they were randomly assigned to self-affirm (write about why a particular value is important to them) or not (write about why a particular value might be important to someone else). Following the emotion task and self-affirmation, subjects read about the link between alcohol and breast cancer. Finally, they were asked questions about their intentions to reduce drinking, their perceived risk of breast cancer, and their worry about breast cancer.


We hypothesized that self-affirmation would be most effective for those asked to recall a happy or angry experience, and least effective for those asked to recall a fearful or surprising experience.

Experimental Manipulations:

Self-affirmation vs. no affirmation
Autobiographical emotion induction (anger/ fear/ happiness/ surprise) vs. neutral

Key Dependent Variables:

Intentions to quit drinking
Specific plans to quit drinking
Risk perceptions for breast cancer
Worry about breast cancer

Summary of Findings:

Self-affirmation and emotion interacted such that self-affirmation improved behavior change plans among neutral emotion participants, but resulted in less specific behavior change plans among those in angry states. Moreover, anger seemed to disrupt the self-affirmation process; essays were less self-affirming in the anger condition. Within the self-affirmation condition, mediation analyses suggest the negative effects anger on behavior change plans may be at least partially mediated by this difference in affirmational essay content.


Ferrer, R. A., Klein, W. M. P., & Graff, K. (under review). Self- affirmation increases defensiveness toward threatening information among angry (and sad) individuals: Results from two national samples

Ferrer, R. A., & Klein, W. M. P. (2016, January). Self-affirmations increase defensiveness against risk information among angry individuals: Results from a national sample. Paper presented at the annual meeting of the Society for Social and Personality Psychology, San Diego CA.

Ferrer, R. A., Klein, W. M. P., & Graff, K. A. (2017, February 16). Self-Affirmation Increases Defensiveness Toward Health Risk Information Among Those Experiencing Negative Emotions: Results From Two National Samples. Health Psychology. Advance online publication.


Copyright © 2014, TESS