Code
knitr::include_graphics("lilly poster.jpg")
Days of Discovery 2026 - Emotional Regulation and Interest: What Drives Belief in Social Misinformation?
Poster
Poster Text
Introduction
The internet is widely used as a tool for finding information about mental and physical health (Chen & Wang, 2021). Additionally, individuals also look for information about relationships, i.e., social information (Zori et al., 2024; Mertens et al., 2024; Lindly et al., 2022).
A review of information on social media found that it tends to be inaccurate, that it spreads quickly, and that it is difficult to distinguish from truth (Aïmeur et al., 2023). If individuals are looking for advice about research online, they are likely to come into contact with misinformation.
Emotions can have an impact and serve as a predictor for believing misinformation. A study performed by Martel and others showed that for nearly every emotion evaluated by the PANAS scale, increased emotionality is associated with increased belief in fake news.
A study performed by Karasavva et al. (2025) examined the popular social media platform TikTok and the viewing of information related to ADHD. The findings indicate that more attention and interaction with the post showed an increase in the belief, even if the content was inaccurate.
The current study examines whether emotional regulation and interest have a role in increasing susceptibility to believing social misinformation.
Hypotheses
We predict that decreased emotional regulation will be associated with higher social misinformation acceptance.
We hypothesized that increased interest in the topic will predict higher belief in social misinformation.
Method
The data was collected from two groups: undergraduate students in an introductory psychology course (n = 58, 32%), as well as participants from the community with no college experience (n = 126, 68%). The participants were predominantly White (n = 130, 71%), Black/African American (n = 19, 0%), or Biracial (n = 15, 8%). Participants were predominantly women (53%) and men (46%), with a few participants identifying as non-binary (1%).
Measures
Participants were randomly assigned to view mock social media posts containing either misinformation or accurate information. After viewing posts, participants rated how truthful they found the information as well as their level of interest, both variables were measured using a 5-point Likert scale. Emotional regulation was later assessed in the survey using the DERS-16 (see sample items below).
Discussion
We found that only one aspect of emotion dysregulation, nonacceptance of emotions (b = -.27, p = .049), is related to misinformation acceptance, and the relationship was not in the expected direction. This inverse relationship indicates that increased nonacceptance may operate as a protective factor against social misinformation. A possibility is that those who struggle to accept their emotions are more analytical with misinformation that is emotionally charged. This finding does not support the prior research and points to a more complex relationship than initially hypothesized.
Additionally, we found that increased interest in misinformation posts (b = .65, p < .001) predicted increased belief in social misinformation. However, increased interest in the accurate information posts was negatively related to social misinformation acceptance (b = -.38, p < .001). This indicates that increased interest in a post predicts increased belief in a post, regardless of its level of accuracy. This may be due to the increased engagement or attention to the misinformation stemming from personal interest. This finding is consistent with prior research that found a connection between attention/interest with a post and increased belief (Karasavva et al., 2025).
The key limitation of this study is the small sample size. Future research with a larger sample and more posts is needed to improve power and generalizability.
References
Aïmeur, E., Amri, S., & Brassard, G. (2023). Fake news, disinformation and misinformation in social media: A review. Social Network Analysis and Mining, 13(1), 30. https://doi.org/10.1007/s13278-023-01028-5
Chen, J., & Wang, Y. (2021). Social Media Use for Health Purposes: Systematic Review. Journal of Medical Internet Research, 23(5), e17917. https://doi.org/10.2196/17917
Karasavva, V., Miller, C., Groves, N., Montiel, A., Canu, W., & Mikami, A. (2025). A double-edged hashtag: Evaluation of #ADHD-related TikTok content and its associations with perceptions of ADHD. PLOS ONE, 20(3), e0319335. https://doi.org/10.1371/journal.pone.0319335
Lindly, O. J., Cabral, J., Mohammed, R., Garber, I., Mistry, K. B., & Kuhlthau, K. A. (2022). “I Don’t Do Much Without Researching Things Myself”: A Mixed Methods Study Exploring the Role of Parent Health Literacy in Autism Services Use for Young Children. Journal of Autism and Developmental Disorders, 52(8), 3598–3611. https://doi.org/10.1007/s10803-021-05240-0
Martel, C., Pennycook, G., & Rand, D. G. (2020). Reliance on emotion promotes belief in fake news. Cognitive Research: Principles and Implications, 5(1), 47. https://doi.org/10.1186/s41235-020-00252-3
Mertens, E., Ye, G., Beuckels, E., & Hudders, L. (2024). Parenting Information on Social Media: Systematic Literature Review. JMIR Pediatrics and Parenting, 7(1), e55372. https://doi.org/10.2196/55372
Zori, G. L., Collins, S. L., & Walker, A. F. (2023). Online Sexual Health Information Seeking of Adolescents: A Content Analysis. American Journal of Sexuality Education, 18(3), 378–400. https://doi.org/10.1080/15546128.2022.2111012
Code
Load Libraries
Code
library(sjPlot)
library(patchwork)
library(ggplot2)
library(haven)
library(psych)Load Data
Code
df <- read_sav("lilly Data.updated.498.sav")Hypothesis 1
H1 was tested using regression analysis. Only nonacceptance was significantly related to misinformation acceptance (b = −.27, p = .049), and in the opposite direction predicted. All other dimensions were not significant (p > .05).
Code
reg1 <- lm(soc_mis ~ soc_acc + condition + DERS_CLARITY + DERS_GOALS + DERS_IMPULSE + DERS_STRATEGIES + DERS_NONACCEPTANCE, data=df)
summary(reg1)
Call:
lm(formula = soc_mis ~ soc_acc + condition + DERS_CLARITY + DERS_GOALS +
DERS_IMPULSE + DERS_STRATEGIES + DERS_NONACCEPTANCE, data = df)
Residuals:
Min 1Q Median 3Q Max
-1.91221 -0.48558 -0.04979 0.51555 1.81447
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.55618 0.39468 3.943 0.000159 ***
soc_acc 0.38605 0.10191 3.788 0.000274 ***
conditionm -0.10323 0.18417 -0.561 0.576507
DERS_CLARITY 0.07910 0.13206 0.599 0.550714
DERS_GOALS -0.10318 0.13598 -0.759 0.449977
DERS_IMPULSE -0.03524 0.14783 -0.238 0.812135
DERS_STRATEGIES 0.23908 0.16953 1.410 0.161923
DERS_NONACCEPTANCE -0.26471 0.13285 -1.992 0.049349 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.8751 on 90 degrees of freedom
(86 observations deleted due to missingness)
Multiple R-squared: 0.213, Adjusted R-squared: 0.1518
F-statistic: 3.479 on 7 and 90 DF, p-value: 0.002405Code
p <- ggplot(df, aes(x = soc_mis_int, y = soc_mis)) +
# Points
geom_jitter(
color = "#486A85",
alpha = 0.65,
size = 3,
shape = 16
) +
# Regression line with confidence interval ribbon
geom_smooth(
method = "lm",
se = TRUE,
color = "#486A85",
fill = "#486A85",
alpha = 0.15,
linewidth = 1.2
) +
# Labels
labs(
title = "Hypothesis 2",
subtitle = "Adj. R² = .42, F(4, 93) = 18.47, p < .001",
x = "Social Misinfo. Interest",
y = "Social Misinfo. Acceptance"
) +
# Clean theme
theme_minimal(base_size = 13) +
theme(
plot.title = element_text(face = "bold", margin = margin(b = 4)),
plot.subtitle = element_text(color = "grey50", margin = margin(b = 12)),
plot.caption = element_text(color = "grey60"),
axis.title = element_text(face = "bold", color = "grey30"),
panel.grid.minor = element_blank(),
panel.grid.major = element_line(color = "grey92"),
plot.background = element_rect(fill = "white", color = NA),
plot.margin = margin(20, 20, 20, 20)
) +
theme(
text = element_text(size = 12))
p
Code
#
# ggsave("interaction_plot1.svg", plot = p, width = 12.18, height = 6.7, bg = "transparent")Hypothesis 2
H2 was tested using regression analysis with interest as the predictors and accurate information acceptance and condition as covariates of social misinformation belief. The overall model was significant, R^2 = .443, F(4,93) = 18.47, p< .001.
Both interest in accurate posts (p < .001) and interest in misinformation posts (p < .001) were related to misinformation acceptance, but the relationships were in opposite directions.
Code
reg3 <- lm(soc_mis ~ soc_acc + condition + soc_mis_int + soc_acc_int, data=df)
summary(reg3)
Call:
lm(formula = soc_mis ~ soc_acc + condition + soc_mis_int + soc_acc_int,
data = df)
Residuals:
Min 1Q Median 3Q Max
-1.77955 -0.51834 -0.02316 0.28955 2.04231
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.50591 0.28584 1.770 0.080016 .
soc_acc 0.47269 0.09372 5.044 2.25e-06 ***
conditionm 0.04541 0.15059 0.302 0.763664
soc_mis_int 0.65385 0.09549 6.848 7.95e-10 ***
soc_acc_int -0.38379 0.09747 -3.937 0.000159 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.7245 on 93 degrees of freedom
(86 observations deleted due to missingness)
Multiple R-squared: 0.4427, Adjusted R-squared: 0.4187
F-statistic: 18.47 on 4 and 93 DF, p-value: 3.366e-11Code
p1 <- plot_model(reg3, type = "pred", terms = "soc_mis_int") +
geom_jitter(data = df, aes(x = soc_mis_int, y = soc_mis),
color = "#486A85", alpha = 0.65, size = 3,
width = 0.05, height = 0.05,
inherit.aes = FALSE) +
labs(
title = NULL,
x = "Social Misinfo. Interest",
y = "Social Misinfo. Acceptance"
) +
theme_minimal(base_size = 13) +
theme(
axis.title = element_text(face = "bold", color = "grey30"),
panel.grid.minor = element_blank(),
panel.grid.major = element_line(color = "grey92"),
plot.background = element_rect(fill = "white", color = NA),
plot.margin = margin(20, 20, 20, 20),
text = element_text(size = 12)
)
p2 <- plot_model(reg3, type = "pred", terms = "soc_acc_int") +
geom_jitter(data = df, aes(x = soc_acc_int, y = soc_mis),
color = "#486A85", alpha = 0.65, size = 3,
width = 0.05, height = 0.05,
inherit.aes = FALSE) +
labs(
title = NULL,
x = "Social Acc. Interest",
y = NULL # drop redundant y label
) +
theme_minimal(base_size = 13) +
theme(
axis.title = element_text(face = "bold", color = "grey30"),
panel.grid.minor = element_blank(),
panel.grid.major = element_line(color = "grey92"),
plot.background = element_rect(fill = "white", color = NA),
plot.margin = margin(20, 20, 20, 20),
text = element_text(size = 12)
)
combined <- (p1 + p2) +
plot_layout(axis = "collect") +
plot_annotation(
title = "Hypothesis 2",
subtitle = "Adj. R² = .42, f(4,93) = 18.47, p < .001",
theme = theme(
plot.title = element_text(face = "bold", size = 14, hjust = 0),
plot.subtitle = element_text(color = "grey50", size = 12, hjust = 0)
)
)
combined