Supplementary Materials - An association between biased impression updating and relationship facilitation - A behavioral and fMRI investigation

Aug 25, 2026

Supplementary materials supporting a behavioral and fMRI investigation into how biased impression updating relates to relationship facilitation. The document provides full-sample analyses for Study 1, stimulus pretesting procedures for moral and immoral vignettes, and correlation results linking friends reported, inclusion-of-other-in-self, liking, and time spent. It further reports associations between impression updating and emotion-word use in memory recall, and brain activation during Theory of Mind tasks including belief versus photo comparisons, with reported effect estimates and beta values.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages31

Reading Time1 h 18 min

AccessFree to download

Analyses with the full sample in Study 1

Ordinal regression and closeness-updating effects

Stimuli pretesting in Study 1

Moral/immoral vignette ratings and group comparisons

Correlations between variables

Correlations with IOS, liking, and weekly hours

Correlations among pre-exposure ratings

Correlations between updating and reaction time

Supplementary Section 1-5+ (Study 1 and Study 2 results)

Effects of biased updating, impression updating, and matching across studies

Exploratory survey item descriptions and findings

Supplementary Materials for An association between biased impression updating and relationship facilitation: A behavioral and fMRI investigation Park, B. and Young, L. 1. Analyses with the full sample in Study 1 2. Stimuli pretesting in Study 1 3. Correlations between variables 4. Descriptions and findings with exploratory survey items in Study 1 5. Effect of biased updating on the number of friends after controlling for different variables 6. Impression updating 7. Study 1 analysis matched with Study 2 8. Descriptions and findings with exploratory survey items in Study 2 9. Associations between participants’ impression updating and their use of emotion words in memory recall 10. Activation in response to Theory of Mind task, belief versus photo in Study 2 11. Findings with beta values 12. The effect of how often participants made new friends on how many friends participants reported having in Study 2 13. Association between participants’ RTPJ activity and general trustworthiness updates in Study 2 Supplementary Section 1: Analyses with the full sample in Study 1 We ran the analyses without excluding any participants (N = 125) and found similar patterns as reported in the main paper. First, ordinal regression with the categorized closeness updating for friend while imagining the behaviors of their friend or a stranger (“Negative update”= -1, “No update” = 0, “Positive update” = +1) on the number of friends that participants reported having revealed that those who never updated, and those who positively updated, reported having a directionally greater number of friends compared to those who negatively updated their friend-closeness ratings ( Estimate = .38 , S. E. = .28, Wald = 1.88, Odds ratio = 1.47 , p = . 171, 95% CI for Estimate = [- . 17, .93]) . Similarly, when switching the categorized updates participants showed while they were imagining the behaviors with their changes in closeness ratings between post-exposure evaluation (i.e. , after they imagined all behaviors) and preexposure evaluation (i.e. , before they imagined any behaviors), we found that participants who less negatively updated their friend-closeness ratings reported having more friends ( Estimate = .52, S. E. = .21, Wald = 5.80, Odds ratio = 1.67 , p = .016, 95% CI for Estimate = [ . 10, .93]) . Supplementary Section 2: Stimuli pretesting in Study 1 This pretesting was conducted for a separate fMRI project (Kim, Mende-Siedlecki, Anzellotti, & Young, in prep) . In total, 1,700 vignettes describing moral or immoral behaviors were pre-tested (1,200 from Mende-Siedlecki, Baron, & Todorov, 2013; 500 new generated) in terms of the moral relevance (“How morally relevant is this behavior?” ranging from 0 [not at all relevant] to 8 [extremely relevant]) , perceived frequency (“How many people, out of 100, have done this behavior?”) , arousal (“How emotionally arousing is this behavior?” ranging from 0 [not at all] to 8 [extremely]), and valence (“How positive/negative is this behavior?” ranging from-4 [extremely negative] to 4 [extremely positive]) , along with other items for the purpose of this project. We took absolute values of valence ratings to compare the magnitude of the valence between moral and immoral behaviors. Participants’ pretested ratings of each story were averaged to a single value for each variable. Please see ( Kim et al. , in prep) for further information. A series of independent t-tests on the eight behaviors selected for the present research revealed that the moral and immoral behaviors were not significantly different in moral relevance (moral M = 6. 13, S. E. = .20; immoral M = 6. 15, S. E. = . 10; p = .934), perceived frequency (moral M = 20.67, S. E. = 2.05; immoral M = 22.78, S. E. = 1.41; p = .428), arousal (moral M = 4.52, S. E. = .35; immoral M = 3.89, S. E. = .23; p = . 186) , or absolute valence (moral M = 2.32, S. E.= .29; immoral M = 2.28, S. E. = .27; p = .931) . Supplementary Section 3

FAQs

What statistical approach is used in the full-sample analyses for Study 1?

Ordinal regression is used with categorized closeness updating for friend while participants imagine friend or stranger behaviors. Reported effects compare participants who never updated, positively updated, and negatively updated friend-closeness ratings.

How were the stimuli for Study 1 pretesting created and evaluated?

A total of 1,700 vignettes describing moral or immoral behaviors were pre-tested using ratings of moral relevance, perceived frequency, arousal, and valence, with valence magnitude compared via absolute values.

What correlations are reported between number of friends and relationship-related measures?

The supplementary section reports Pearson correlations between reported number of friends and IOS, liking, and weekly hours with the friend across Study 1 and Study 2, including significance notes for select associations.

Aug 25, 2026

Upload by Valentina