Algorithmic or Human Source? Examining Relative Hostile Media Effect With a Transformer-Based Framework

Media and Communication

article

Chenyan Jia¹  Ruibo Liu² 

[1] Moody College of Communication, The University of Texas at Austin;Department of Computer Science, Dartmouth College

关键词: algorithms;    automated journalism;    computational method;    hostile media effect;    source credibility;   

DOI  :  10.17645/mac.v9i4.4164

学科分类：医学（综合）

来源: Cogitatio Press

【 摘 要 】

The relative hostile media effect suggests that partisans tend to perceive the bias of slanted news differently depend‐ing on whether the news is slanted in favor of or against their sides. To explore the effect of an algorithmic vs. humansource on hostile media perceptions, this study conducts a 3 (author attribution: human, algorithm, or human‐assistedalgorithm) × 3 (news attitude: pro‐issue, neutral, or anti‐issue) mixed factorial design online experiment (N = 511). Thisstudy uses a transformer‐based adversarial network to auto‐generate comparable news headlines. The framework wastrained with a dataset of 364,986 news stories from 22 mainstream media outlets. The results show that the relative hos‐tile media effect occurs when people read news headlines attributed to all types of authors. News attributed to a solehuman source is perceived as more credible than news attributed to two algorithm‐related sources. For anti‐Trump newsheadlines, there exists an interaction effect between author attribution and issue partisanship while controlling for peo‐ple’s prior belief in machine heuristics. The difference of hostile media perceptions between the two partisan groups wasrelatively larger in anti‐Trump news headlines compared with pro‐Trump news headlines.

【 授权许可】

CC BY