Effects of interaction-visualizing feedback on heart rate variability and psychological outcomes during online shared eating: a randomized, open-label, controlled preliminary study in healthy adults

Prior studies suggest that sharing biosignals may enhance social presence and connectedness, and that online shared eating may influence heart rate variability (HRV), dyadic physiological covariation, loneliness, and mood. However, the effects of biosignal-sharing feedback during online commensality remain insufficiently examined. This preliminary randomized study investigated whether interaction-visualizing feedback—sharing participants’ physiological information and visualizing interpersonal interactions during online shared eating—affects autonomic and psychological outcomes. This open-label, randomized, controlled preliminary study included healthy adults. Participants ate together online in dyads and were allocated to either an intervention group using HRV-linked interaction-visualizing feedback or a control group using constant illumination. The primary outcome was the change in the standard deviation of normal-to-normal intervals (SDNN) from the pre-meal resting phase to each subsequent experimental phase (ΔSDNN). Secondary outcomes included exploratory within-dyad correlation coefficients of ΔSDNN between dyad members, and pre–post changes in the Three-Item Loneliness Scale (TILS) and UWIST Mood Adjective Checklist (UMACL). Linear mixed models accounted for repeated measurements within the participants and dependency within the dyads. A total of 30 dyads were enrolled. One control dyad was excluded because of insufficient HRV data acquisition, leaving 29 dyads for the analysis. The group-by-phase interaction for ΔSDNN was not statistically significant (F(2, 96.5) = 0.102, p = .903). During the first half of the meal, a significant positive within-dyad correlation in ΔSDNN was observed in the intervention group (Spearman’s ρ = 0.525, p = .045), but not in the control group (ρ = −0.169, p = .563). The between-member absolute difference in ΔSDNN during the first half of the meal was smaller in the intervention group than in the control group (Mann–Whitney U = 45.0, p = .008). Group-by-time interactions were not statistically significant for TILS ( p = .556), UMACL energetic arousal ( p = .953), or UMACL tense arousal ( p = .197). Interaction-visualizing feedback did not significantly affect ΔSDNN or psychological outcomes during online shared eating sessions. Exploratory dyad-level findings suggested possible early phase within-dyad similarity in SDNN change. The University Hospital Medical Information Network Clinical Trials Registry (UMIN), UMIN000051250. Registered on June 3, 2023. https://center6.umin.ac.jbetweengi-open-bin/ctr_e/ctr_view.cgi?recptno=R00005844 https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R00005844 .