Please Biofeed the Zombies: Enhancing the Gameplay and Display of a Horror Game Using Biofeedback

This paper, presented at a 2007 DiGRA conference, explores whether low-cost biometric sensing can personalize the emotional experience of a video game without altering its core design. The authors adapted the Wild Divine Lightstone, an inexpensive device under $200 that measures heart rate variability and galvanic skin response, to dynamically modulate the horror level Ravenholm in Half-Life 2 through the Source engine. Shaders, screen shake, enemy spawning, movement speed, and weapon damage were all scaled by multipliers derived from each player's physiological readings relative to a calibration baseline, smoothed over two-second intervals to reduce sensor noise. The stated aim is "affective gaming," a form of emotional personalization, though the authors acknowledge that the connection between biometric signals and actual player emotion remains unvalidated.

The user study involved 14 participants, mostly recruited from a university community and including only three women, who played both a standard and a biometrically enhanced version of the zombie level. Nine of the 14 preferred the biometric-driven version, and 11 believed biometric information could make games more engaging. However, the bulky, solid-state sensors were found to interfere with natural interaction, and engagement depended heavily on whether players already enjoyed horror games. The authors conclude that biometric enhancement can deepen engagement for players who are receptive to the genre, while potentially diminishing it for others, and they recommend less intrusive sensor hardware and closer attention to player learning curves.

The final section looks ahead to multiplayer applications, such as one player's physiological state triggering another player's phobias, or an "imposter" game in which players must control biometrically driven avatars to avoid detection, with the authors suggesting such a system might function as a rudimentary lie detector. These proposals are speculative and are not supported by reported experimental data. The paper is situated within affective computing and research on emotion and enjoyment in games, drawing on work such as Picard's analysis of affective states and Ravaja et al.'s study of phasic emotional responses to game events.