Enhancing mis- and disinformation detection and understanding its influence: leveraging communication accommodation theory and information manipulation theory

Addressing the detection and spread of mis- and disinformation presents a complex challenge. Using the Linguistic Inquiry and Word Count (LIWC) software, we examined the defining linguistic features that distinguish fake from authentic COVID-19 news. Drawing from Communication Accommodation Theory (CAT) and Information Manipulation Theory (IMT), we analysed 70 fake and 70 true news stories. CAT proved more effective than IMT in predicting linguistic cues, with fake news found to contain moralising language, interpersonal conflict, and informal speech markers, including swearing, netspeak, assent words, and exclamation marks – patterns consistent with hate speech. In line with IMT, fake news was found to exhibit higher frequencies of causation, future focus, and certitude words. True news unexpectedly contained more discrepancy terms, markers of clout, analytical thinking, positive emotion, and tentative language which might be explained by journalistic conventions that differ from dyadic interactions. Drawing from Social Identity Theory (SIT), we make the case that fake news stories may position themselves against ‘others’ rather than aligning with a group. This study offers a novel framework for future researchers in the science of detecting mis– and disinformation, and to further understand how fake news deceives at scale and influences unwanted behaviours.