GisPy: A Tool for Measuring Gist Inference Score in Text

Hosseini, Pedram; Schmidtová, Patrícia; Wolfe, Christopher L.; Diab, Mona; Broniatowski, David A.

doi:10.18653/v1/2022.wnu-1.5

Cited by 3 publications

(2 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work also highlights the value of automated techniques, such as GIS (Dandignac & Wolfe, 2020; Wolfe et al, 2019), that can be used to infer gist at scale; see also (Hosseini et al, 2022). The results of Studies 1 and 2 are especially important in this context because they demonstrate a systematic relationship between correlates of gist representation (e.g., cohesion) and sharing online.…”

Section: Discussionmentioning

confidence: 83%

The role of mental representation in sharing misinformation online.

Broniatowski,

Hosseini,

Porter

et al. 2024

Journal of Experimental Psychology: Applied

View full text Add to dashboard Cite

Fuzzy-Trace Theory (FTT) posits that people share misinformation online if it promotes gist mental representations cuing motivationally-relevant values. Most people value the truth. Thus, per FTT, people decide to share messages that they perceive as true. FTT also predicts that messages will be more effective if they communicate a simple gist. We test these predictions by examining the roles of mental representation and epistemic quality in decisions to share misinformative articles on Facebook across two experiments and two correlational studies. In studies 1 and 2, we use Facebook data to test the hypothesis that gist proxies in text are associated with online sharing. In study 3, we experimentally manipulate subjects' exposure to a gist-based intervention that explains why a misinformative article is false, a simple debunk stating only that the article is false (but not explaining why) and a verbatim condition providing relevant detailed information but allowing subjects to draw their own conclusions. We found that the gist condition decreased intentions to share misinformation. Finally, in study 4, we replicated this finding, and show that the gist condition also reduces misinformation endorsement. Results provide support for FTT's predictions regarding reducing sharing and endorsement of misinformation on social media.Impact and implications statement: Evidence supports Fuzzy-Trace Theory's predictions that misinformation expressing a gist is more likely to go viral. Furthermore, providing a message explaining the gist of why misinformation is false can reduce endorsement of that misinformation and online sharing. These findings go beyond existing remedies for correcting misinformation. Gist-based interventions should therefore be added to the arsenal of techniques for combating online misinformation.

show abstract

Section: Discussionmentioning

confidence: 83%

The role of mental representation in sharing misinformation online.

Broniatowski,

Hosseini,

Porter

et al. 2024

Journal of Experimental Psychology: Applied

View full text Add to dashboard Cite

show abstract

“…As a consequence, gist-based interventions can be labor intensive, especially in contrast to techniques such as accuracy nudges (Pennycook et al, 2021), which are attractive, in part, because of their scalability. This highlights the importance of developing automated techniques, such as Gist Inference Scores (Dandignac & Wolfe, 2020; that can be used to infer gist at scale (see also Hosseini et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

The Role of Mental Representation in Sharing Misinformation Online

Broniatowski

Hosseini²,

Porter³

et al. 2023

Preprint

View full text Add to dashboard Cite

Fuzzy-Trace Theory posits that people will be more likely to share misinformation online if it promotes gist mental representations that cue motivationally-relevant values. In this paper, we test these predictions by examining the combined roles of mental representation and valenced affect in decisions to share misinformative articles on Facebook. We present the results of three experiments and two correlational studies providing evidence for Fuzzy-Trace Theory’s predictions. In Studies 1 and 2, we use the largest public dataset of Facebook behavior available to researchers to test the hypothesis that proxies for gist representations in text are associated with to decisions to share information online. Study 3 identifies two mechanisms that are theorized to drive sharing: gist-based mental representation and epistemic valence endorsements – i.e., evaluations about the truth, plausibility, accuracy, etc., of an article. Study 4 confirms that these valence endorsements and endorsement of the correct gist are associated with increased intentions to share misinformation. Finally, in Study 5, we develop a gist-based intervention based on these mechanisms, show that it can reduce intentions to share misinformation, and demonstrate that these findings replicate. Results provide empirical support for Fuzzy-Trace Theory’s predictions regarding how to reduce sharing of misinformation on social media.

show abstract

Straight from the horse’s mouth: communicating the gist of mRNA vaccines

Brace,

Wolfe

2024

Journal of Applied Communication Research

View full text Add to dashboard Cite

GisPy: A Tool for Measuring Gist Inference Score in Text

Cited by 3 publications

References 1 publication

The role of mental representation in sharing misinformation online.

The role of mental representation in sharing misinformation online.

The Role of Mental Representation in Sharing Misinformation Online

Straight from the horse’s mouth: communicating the gist of mRNA vaccines

Contact Info

Product

Resources

About