Skip to main navigation Skip to search Skip to main content

Beyond Force: Grip Dynamics as Behavioral Indicators of Stress in Interactions with a Handheld Social Robot

Research output: Working paperPreprint

Abstract

Stress-related arousal is commonly studied through physiological sensing, but it may also be reflected in the dynamics of motor behavior during interaction. This paper investigates whether grip-pressure signals from a pocket-sized tactile social robot can indicate stress-related changes during active handheld use. We conducted a within-subjects A-B-A study with 26 participants, yielding 22 complete datasets, in which participants interacted with the same handheld robot across baseline, stress/arousal, and return-to-baseline conditions. Grip pressure was measured through two side-mounted pressure sensors and analyzed as a combined grip signal. Rather than treating grip pressure only as a measure of force magnitude, we examined three complementary features: peak amplitude, grip variability, and release speed following detected grip peaks. Peak amplitude did not significantly distinguish the stress-inducing condition from the initial baseline, indicating that stress was not reliably expressed as stronger squeezing. Instead, when participants experienced the stress-inducing condition, their grip pressure became less stable and their pressure release after grip peaks became faster. Grip variability increased significantly relative to baseline conditions ((p = .036) and (p < .001)), while release speed showed the strongest observed effect ((p < .001) for both baseline comparisons). These findings suggest that stress-related arousal in handheld robot interaction is better characterized by reduced motor stability and increased release reactivity than by increased grip strength alone.
Original languageEnglish
Number of pages8
Publication statusPublished - Aug 2026

Keywords

  • Affective computing
  • Human-robot interaction
  • Grip-pressure signals
  • Stress arousal
  • Tactile sensing

Fingerprint

Dive into the research topics of 'Beyond Force: Grip Dynamics as Behavioral Indicators of Stress in Interactions with a Handheld Social Robot'. Together they form a unique fingerprint.

Cite this