Attractive Things Do Work Better: A Meta-Analysis on Visual Aesthetics and User Performance – Online Supplementary Materials

This repository contains the online supplementary materials for the study entitled "Attractive Things Do Work Better: A Meta-Analysis on Visual Aesthetics and User Performance." Aesthetics is a central construct in human–computer interaction (HCI), shaping users’ perceptions and attitudes toward interfaces. However, evidence on whether aesthetics improves objective performance remains mixed. This preregistered meta-analysis examines whether users perform better (e.g., are faster, more accurate, or more efficient) with aesthetically appealing versus unappealing interfaces. Eligible studies required successful aesthetic manipulations, demonstrated by significant differences in subjective aesthetic ratings. A systematic literature search yielded 31 studies with 234 effect sizes and 18,794 participants. The meta-analytic model revealed a small to medium positive effect of aesthetics on performance (g = 0.29), though with high unexplained heterogeneity. Significant moderators were interaction device and typography manipulation, while one control variable, namely a posteriori-defined aesthetic conditions, suggested a confounding role of prior user experience (e.g., experienced usability). We discuss methodological limitations, stress the need for higher-quality research, and provide concrete guidelines for future studies and systematic reviews alongside practical implications for HCI. Contents of the Online Supplementary Materials Supplementary Material A – Details on Literature Search & Study Selection A1. List of Study Eligibility Criteria A2. Database Search Criteria A3. Keyword- and Content-Based Screening of Title and Abstracts A4. Hierarchical Structure of Exclusion Criteria Supplementary Material B – Details on Coding Procedure B1. Coding Manual for Study Variables B2. Contextual Questions to the Study Design and Implementation Assessment Device (Study DIAD; Valentine & Cooper, 2008) Supplementary Material C – Details on Meta-Analytic Results C1. Forest Plot for all Included Effect Sizes and Study Quality Ratings C2. Associations Between Moderator Variables Indicated by Cramer's V Values Supplementary Material D – Extended Discussion D1. Internal and External Validity of Included Studies (Study DIAD; Valentine & Cooper, 2008) D2. Recommendations for Future Systematic Reviews In addition, this repository provides the raw data used for the meta-analysis (i.e., coding of primary studies) and the R scripts for (1) preparing the data and (2) performing the statistical analysis.