Misinformation

In  this section, we describe a range of projects  designed to combat the misinformation crisis. Some are designed to reduce the spread of misinformation, some  are design to help mitigate its harmful effects, some are designed to detect it and others are designed to better develop best practices.

Developing Best Practice

To be maximise their effectiveness, researchers need to agree how the misinformation crisis should be addressed and then coordinate their efforts. As a first step, using a crowdsourcing method, we have elicited perceptions of best practice from an international group of experts. This information will be used to help facilitate the Future of Disinformation Seminar organised by Melbourne Connect.

  • Kruger, A, Saletta, M, Ahmad, A, Howe, P (2024) Structured expert elicitation on disinformation, misinformation, and malign influence: Barriers, strategies, and opportunities. Harvard Kennedy School of Misinformation Review. [PDF]

Reducing the Spread of Disinformation on Social Media

Almost half of Australians use social media as their primary news source. Because it lacks editorial oversight, it is a primary spreader of misinformation. Developing ways of reducing the spread of misinformation on social media, while maintaining  freedom of speech, is a key goal of the lab. In the publications below, we describe a particularly effective way of doing this.

  • Howe, P.D.L., Perfors, A., Ransom, K.J., Walker, B., Fay, N., Kashima, Y., Saletta, M., Dong, S. (2024). Self-Certification: A novel method for increasing sharing discernment on social media. PloS One, 19(6), e0303025. [PDF]
  • Howe, P, Perfors, A, Ransom, K, Walker, B, Fay, N, Kashima, Y, Saletta, M, Dong, S. (2023). Self-censorship appears to be an effective way of reducing the spread of misinformation on social media. Proceedings of the Annual Meeting of the Cognitive Science Society, 45(45). [PDF]

Effect of Public Corrections

One popular way of countering misinformation is to publicly correct it. How effective is this?  Actually, not always as effective as you might think...

  • Mickleberg, A, Walker, B, Ecker, U, Howe, P, Perfors, A, Fay, N (in press). Did he or didn’t he? Mixed evidence for the continued influence of retracted misinformation on person impressions.

...though sometimes it does work

  • Mickleberg, A, Walker, B, Ecker, U, Howe, P, Perfors, A, Fay, N (2024). Does mud always stick? No evidence for the continued influence of misinformation on person impressions. Collabra, 10(1). [PDF].

Body of Statements

In our experiments, we often need to present statements to participants to see if they believe them, share them etc. But we can't just present any old set of statements; they need to be carefully constructed to be balanced over a range of properties. This paper describes how we do that.

  • Mickelberg, B., Walker, B., Ecker, U.K.H., Howe, P.D.L., Perfors, A. & Fay, N. (2022). Impression formation stimuli: A corpus of behaviour statements rated on morality, competence, informativeness, and believability. PLOS ONE,17(6), e0269393. [PDF]

Modelling Work

Mathematically modelling can be really  helpful in drawing attention to our implicit assumptions and using our intuitions to make precise predictions that we can then test empirically. Below is a paper that describes some of our work in this area.

  • Howe, PDL, Perfors, A, Ransom, K (2021). What interventions can decrease or increase belief polarisation in a population of rational agents? Proceedings of the Annual Meeting of the Cognitive Science Society, 43. [PDF]