Media Effects Research Lab - Research Archive
Effects of Modality of Misinformation and Debunking on Message Credibility and Memory Recall
Student Researcher(s)
Eunchae Jang (Ph.D Candidate);
Tim Prestby (Ph.D Candidate);
Andrew Pacheco (Ph.D Candidate);
Faculty Supervisor
Introduction
This study investigates how different forms of misinformation (i.e., text versus text + image, text + map) affect people’s perceived credibility and recall of the misinformation. We are especially interested in how satellite maps may be perceived as more credible and foster greater recall compared to other forms if misinformation. It also explores the effectiveness of debunking in the form of fact-checking on calibrating people’s credibility judgements and influencing their memory of the misinformation.
Research Questions + Hypotheses
H1: Text + Satellite map image conditions will have higher message credibility than Text + Ground image conditions.
H2: Text + Satellite map image and Text + Ground image conditions will have higher message credibility than Text-only conditions.
H3: Realism heuristic will mediate the relationship between modality and message credibility.
H4: Authority heuristic will mediate the relationship between modality and message credibility.
H5: Individuals who read debunking news will show lower levels of message credibility than people who do not read debunking news.
H6: Text + Ground image and Text + Satellite map image conditions will show higher memory recall than Text-only conditions.
RQ1: What is the relationship between exposure to debunking news and memory recall?
RQ2: When individuals are exposed to debunking news compared to no exposure to debunking news, what is the relationship between modality and credibility?
RQ3: When individuals are exposed to debunking news compared to no exposure to debunking news, what is the relationship between modality and memory recall?
Method
We conducted a 3 (modality: text vs. text + ground image vs. text + satellite map) × 2 (debunking: no debunking vs. debunking) between-subjects factorial experiment for pre-test. A four to seven days after the pre-test, a post-test was conducted with the same sample and measured dependent variables (message credibility and memory recall). Participants (N = 317) were recruited from CloudConnect.
Participants were randomly assigned to view a misinformation news article about a fire in Central park in text format, text + ground image format, or text + map format. Participants were also randomly assigned to view a second article that debunked the first news article or not.
Figure 1
Stimuli
Note. Stimuli for modality conditions: misinformation news article of a Central Park fire (left). Debunking article stimuli (right)
Participants then answered questions about perceived credibility, mediating variables including the authority and realism heuristic, and recall. Participants then answered questions on demographics and control variables. We conducted ANCOVA, repeated measure ANCOVA, multinomial logistic regression, and Hayes Process Macro to test our hypotheses and research questions.
Results
First, the results showed that there was no main effect of modality on message credibility. There also was not an interaction effect between modality and debunking, and mediation effects of realism heuristic and authority heuristic on the relationship between modality and message credibility were not significant. However, we found a significant main effect of debunking on message credibility immediately after exposure to debunking and over time. That is, exposure to debunking significantly reduced message credibility immediately and over time. Moreover, exposure to debunking news significantly increased the likelihood of accurately remembering correct information compared to no exposure to debunking news. Interestingly, when people did not read debunking news, visual modalities conditions (ground image and satellite map) showed a decrease in message credibility over time compared to text modality condition, while it did not differ when people read debunking news.
Conclusion/Discussion
Our results contrast prior studies as we did not find a significant effect of modality of misinformation on recall or credibility. However, these results may be a product of the single topic we tested and/or the unrealistic news consumption environment of our experiment. Our work builds on prior findings demonstrating that debunking, specifically, fact-checking, can be effective in lowering people’s perceived credibility of misinformation. Interestingly, we found that perceived credibility declined over time for the no-debunking conditions, only for visual modalities. This may be due to the visual modalities cuing heuristic processing that leads to less strong memory retention compared to text and no debunking failing to reinforce the information. Our findings may expand on the theory of the MAIN Model in that that the effects of realism heuristic may be weakened if the message is too sensational. Moreover, we provide guidance on how to effectively debunk misinformation with factchecking.
For more details regarding the study contact
Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173