'Specially For You' - Examining the Barnum Effect's Influence on the Perceived Quality of System Recommendations
Pang Suwanaposee, Carl Gutwin, Zhe Chen, Andy Cockburn
Abstract
The ‘Barnum effect’ is a psychological phenomenon under which people assign higher quality ratings to personality descriptions developed ‘specially for you’ than the same descriptions described as ‘generally true of people.’ This effect suggests that recommender interfaces could elevate the perceived quality of recommendations simply by indicating that they are explicitly personalised. We therefore conducted a crowd-sourced experiment (n=492) that examined the perceived quality of personalised versus non-personalised movie recommendations for good and bad movies – importantly, the actual recommendations were identical, and were merely presented as being either personalised or not. Contrary to the Barnum effect, results showed numerically lower mean quality scores for personalised recommendations, but with no significant difference. Our findings suggest that Barnum-like effects of personalisation have at most a small influence on perceived quality, and that designers should not rely on this effect to improve user experience (despite online design guidance suggesting the opposite).
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Cited by top-tier papers1
Ask how each one uses itRelated papers
- User Trust in Recommendation Systems: A comparison of Content-Based, Collaborative and Demographic FilteringMengqi Liao, S. Shyam Sundar, Joseph B. WaltherCHI 2022 · 53 citations
- Vis Ex Machina: An Analysis of Trust in Human versus Algorithmically Generated Visualization RecommendationsRachael Zehrung, Astha Singhal, Michael Correll, Leilani BattleCHI 2021 · 19 citations
- On the Negative Perception of Cross-domain Recommendations and ExplanationsDenis Kotkov, Alan Medlar, Yang Liu, Dorota GlowackaSIGIR 2024 · 5 citations
- Trading Personalization for Accuracy: Data Debugging in Collaborative FilteringLong Chen, Yuan Yao, Feng Xu, Miao Xu et al.NeurIPS 2020 · 8 citations
- Post Processing Recommender Systems with Knowledge Graphs for Recency, Popularity, and Diversity of ExplanationsGiacomo Balloccu, Ludovico Boratto, Gianni Fenu, Mirko MarrasSIGIR 2022 · 48 citations
