Xiaolian Cheng

Who they are

The paper lists Cheng in the Department of Business Administration, Chung-Ang University, Seoul. With Inwoo Nam, Cheng ran a scenario experiment, published in Frontiers in Psychology in July 2026, with 200 US adults who imagined buying a black dress or a black T-shirt online. Every image was made with the same AI tool; only the caption changed, saying the model was AI-generated or human.

For the dress, the AI caption cut the self-expression score from 4.37 to 1.97. The label raised a sense of eeriness, which raised perceived psychological and performance risk, which lowered self-expression. For the T-shirt the label made no reliable difference. The authors list as limits that the two garments may differ in hedonic value, formality and femininity, that the study used a scenario and female models, and that 70 per cent of the sample were women.

Key papers

Cheng, X., & Nam, I. (2026). "Understanding the influence of AI labels on consumer psychology: the moderating role of product type." Frontiers in Psychology, 17, 1878095. doi:10.3389/fpsyg.2026.1878095.

Key ideas

An AI label hurts products bought for what they say about the owner and leaves functional products alone. See Algorithm Aversion.

---