Machine in the middle: Exploring dark patterns of emotional human-computer integration through media art

Dickinson, Rod; Semertzidis, Nathan; Mueller, Florian Floyd

Abstract

As our relationship with machines becomes evermore intimate, we observe increasing efforts in the quantification of human emotion, which has historically generated unintended consequences. We acknowledge an amplification of this trend through recent technological developments that aim toward human-computer integration, and explore the dark patterns that may arise for integrating emotions with machinic processes through "machine_in_the_middle". Machine_in_the_middle is an interactive system in which a participant wears an electroencephalographic headset, and their neural activity is analysed to ascertain an approximation of their emotional state. Using electrical muscle stimulation their face is animated into an expression that corresponds with the output of the emotional recognition system. Through our work, we contribute the insight of three possible dark patterns that might emerge from emotional integration, including: reductionism of human emotion, disruptions of agency, and parasitic symbiosis. We hope that these insights inspire researchers and practitioners to approach human-computer integration more cautiously.

• Presentation Conference Type: Conference Paper (published)
• Conference Name: CHI 2022 - Conference on Human Factors in Computing Systems - Proceedings
• Start Date: Apr 29, 2022
• End Date: May 5, 2022
• Acceptance Date: Apr 27, 2022
• Publication Date: Apr 27, 2022
• Deposit Date: Sep 25, 2022
• Publicly Available Date: Sep 27, 2022
• Publisher: Association for Computing Machinery (ACM)
• Book Title: CHI EA '22: Extended Abstracts of the 2022 CHI Conference on Human Factors in Computing Systems
• ISBN: 9781450391566
• DOI: https://doi.org/10.1145/3491101.3503555
• Keywords: Media Art, Brain-computer interface, EEG, Electrical Muscle Stimulation, Emotion detection
• Public URL: https://uwe-repository.worktribe.com/output/9953723
• Publisher URL: https://dl.acm.org/doi/10.1145/3491101.3503555
• Related Public URLs: https://dl.acm.org/doi/proceedings/10.1145/3491101; ; https://dl.acm.org/sig/sigchi