Misinformation and Influence in Cyberspace: The Importance of Individual Differences and Communication Channel
Files
(Thesis)
Date
2021
Authors
Parsons, Kathryn Marie
Editors
Advisors
Butavicius, Marcus
Delfabbro, Paul
Delfabbro, Paul
Journal Title
Journal ISSN
Volume Title
Type:
Thesis
Citation
Statement of Responsibility
Conference Name
Abstract
This thesis examines the role of Cialdini’s (2009) six principles of influence (i.e., authority, consistency, liking, reciprocity, scarcity and social proof) in three contexts that feature a high prevalence of misleading, deceptive and biased information: phishing emails, online news and social media. These contexts were chosen because of their relevance to defence and national security and because of their role in alleged political interference and broader financial, health and safety consequences. Aspects of communication relating to the message, channel, receiver and context are examined to better understand, a) the situations in which information is more persuasive, and b) the characteristics of people who are more susceptible to persuasion. The thesis begins with an overview of relevant literature relating to the importance of the topic; the principal theoretical frameworks examined; and, the principal research gaps and aims which the thesis sought to address. This is followed by a series of chapters or papers that summarise the findings from specific research investigations and a final discussion and summary of the findings and future directions. Paper 1 evaluated the prevalence of Cialdini’s (2009) influence principles in social media posts of scientific research and the effect of these on social media engagement. The study involved 100 social media news stories from Australia’s five top news sites. These news stories were evaluated for the presence of six principles of influence and for the accuracy with which the social media post summarised the news report. The measure of social media engagement used incorporated the number of people who liked, shared or commented on the post. The results showed that social media posts were most likely to emphasise the authority of the reported research, with posts that emphasised authority found to receive significantly more social media engagement than posts without this principle. By contrast, social media posts that used the consistency and social proof principles were significantly less popular than those without these influence tactics. Paper 2 used the context of online science news reporting to examine several factors that may affect how people process a news report. The study examined the influence of misinformation, source authority, and whether the news report was first viewed in a social media format. It also examined how are these relationships are affected by differences between individuals. A total of 258 participants took part in an online experiment with a 2 (low misinformation vs. high misinformation) x 2 (low authority vs. high authority) x 2 (news report only vs. social media post and news report) between-subjects factorial design. Participants were asked to read a news report, which differed based on the described manipulations (e.g., low or high on authority). They also completed measures of individual differences, such as susceptibility to authority, trust in news and impulsivity. The results showed that individual differences (i.e., receiver characteristics) and mode of communication (i.e., channel) played an important role in understanding how people process online news. For example, people were more influenced by perceived rather than objective accuracy and authority. In addition, viewing a news report with a misleading headline on social media tended to reduce a reader’s memory and understanding of the associated news report, but also increased the intended behaviour change evoked by the news report. Paper 3 examined how readers evaluate online news encountered on social media. We investigated further components of McGuire’s (1989) Communication Persuasion Matrix Model, including aspects associated with the message, such as source authority and social proof, and also aspects of the receiver and the context. A 2 (social proof: low vs. high) x 2 (source authority: low vs. high) x 2 (information processing task order: before vs. after) between-subjects randomised design was used, and 736 participants were presented with a social media post of online news. The results showed that the influence of the authority of the source was dependent on the level of social proof evident in the post (i.e., the number of likes, shares and comments on a message). High authority sources were more likely to influence people than low authority sources when social proof was low. However, such authority effects were not present when social proof was high. The results also revealed that aspects of the receiver (e.g., susceptibility to persuasion, decision-making preference) and the context (e.g., what task someone was doing before viewing the post) play an important role in understanding how people are influenced by news on social media. In Paper 4, we presented a role-play scenario-based study to examine online social influence in phishing emails. A total of 985 participants were presented with 14 emails, which included 7 genuine and 7 phishing emails. For both genuine and phishing emails, this consisted of one email with no social influence principle, and one with each of Cialdini's (2009) six social influence principles. The results indicated that phishing emails utilising the scarcity and social proof principles were least successful, whereas those applying consistency and reciprocity principles were most successful. In support of the importance of individual differences, the principles that an individual considered to be most persuasive corresponded with the phishing emails to which that individual was most susceptible. Paper 5 reported on the factors that predict the ability to resist phishing attacks, which we term phishing resilience. In this study, we measured a larger number of potential contributing individual, cultural, organisational and interventional factors than any previous study. A total of 607 participants completed two stages of an online experiment, with the first stage consisting of measures of information security awareness and other individual difference measures, and the second stage consisting of a phishing study. We found that people were better at resisting phishing attacks when they: had higher information security awareness; were older; were less susceptible to social pressures; had a rational rather than impulsive decision-making preference; and were less likely to trust in information technology safeguards. In conclusion, this thesis highlighted the importance of considering individual and contextual factors when understanding whether an individual will be influenced rather than just the specific influence attempt. Persuasion in different online environments was found to be more about individual differences (e.g., susceptibility to persuasion), channel differences (e.g., social media vs. news) and context differences (e.g., what task an individual is doing beforehand) than it is about the persuasion principle that was used. For example, although our results suggest that, on average, people are able to detect and deflect influence attempts in the online realm that use the social proof principle, our findings also highlight that this is not true of all individuals. We demonstrated that an individual’s susceptibility to specific influence attempts can be measured using a simple test that predicts behaviour, at least in an experimental setting. It is only when these factors are all considered that there will be a chance of understanding, and ultimately reducing unwanted influence attempts.
School/Discipline
School of Psychology
Dissertation Note
Thesis (Ph.D.) -- University of Adelaide, School of Psychology, 2021
Provenance
This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legals