Predicting Affective Impressions in Human-Computer Interaction

Author

Daniel Burton Shank, Missouri University of Science and TechnologyFollow
Grace R. Hammond
Fatima Sajjad

Abstract

If affect control theory (ACT) can accurately predict affective impressions of technology, it can be expanded to human-computer interaction. We compared ACT's predicted impressions to actual collected impressions in three studies. Predicted impressions were 5–20% less accurate for technology actors than human actors (Study 1), similarly accurate in evaluation and activity for technological actors and objects after communication behavior (Study 2a), and 17–28% more accurate for technological actors than objects after physical behaviors (Study 2b). Overall predictions were accurate for technology two-thirds to three-fourths of the time, suggesting ACT's utility for modeling human-computer interaction, though there is room for improvement.

Recommended Citation

Shank, D. B., Hammond, G. R., & Sajjad, F. (2025). Predicting Affective Impressions in Human-Computer Interaction. Current Research in Social Psychology, 34 University of Iowa.

Department(s)

Psychological Science

International Standard Serial Number (ISSN)

1088-7423

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2025 University of Iowa, All rights reserved.

Publication Date

01 Jan 2025