The Effects of Anthropomorphism on How People Evaluate Algorithm-Written News

Abstract

Based on the Modality-Agency-Interactivity-Navigability model and the anthropomorphism theory, this study examines whether embedding human-like characteristics in algorithms increases the persuasiveness of algorithm-written news. This study further investigates how different types of relationships (servant or friend) that human writers form with algorithms determine the persuasiveness of algorithm-generated news. Experiment 1 demonstrated that participants who read the human- and humanized algorithm-written news showed greater emotional involvement in the stories and liked the articles more than those who read the algorithm-written news. The participants also reported that news written by humans, humanized algorithms, and algorithms, had equal news credibility. Experiment 2 further showed that the participants perceived the news as more credible and experienced greater degrees of emotional involvement when human writers formed a partner-to-friend relationship with humanized algorithms while generating news rather than a servant-to-master relationship.

Keywords:

• Computational journalism
• anthropomorphism theory
• algorithm
• emotional involvement
• news evaluation

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1 Adopted from Wikipedia (https://en.wikipedia.org/wiki/Automated_journalism)

Additional information

Funding

This paper was supported by Samsung Research Fund, Sungkyunkwan University, 2019.