Attribution Practices for the Man-Machine Marriage: How Perceived Human Intervention, Automation Metaphors, and Byline Location Affect the Perceived Bias and Credibility of Purportedly Automated Content

Abstract

Automation is increasingly serving a role in the production of news. Attribution practices that recognize the journalistic contributions of automation, however, have yet to be standardized. Perceptions of human intervention, the metaphor used to describe the automation algorithm, and the location where the byline for automation is located are all attribution decisions that may impact the credibility or perceived bias of the news products purportedly produced via automation. To that end, an online experiment (N = 601) was conducted to test such possibilities using a 2 (source attribution: machine vs. machine and human) × 3 (attribution metaphor: “robot reporter” vs. “news algorithm” vs “no metaphor” control) × 2 (attribution location: start of article vs. end of article) between subjects design. Results revealed that news purportedly written together by human and automated authors is perceived as less biased than news written solely via automation. As for perceived credibility, news that disclosed the role played by automation at the beginning of the news article was perceived as less credible than news when the byline for automation appeared at the end of the article. Theoretical and practical implications of these findings are discussed.