Interaction Matters: The Effect of Touching the Social Robot PARO on Pain and Stress is Stronger When Turned ON vs. OFF

Abstract

Social touch between humans, as well as between humans and animals, was previously found to reduce pain and stress. We previously reported that touching a social robot can also induce a reduction in pain ratings. However, it is unclear if the effect that touching a robot has on pain perception is due to its appearance and its pleasant touch, or due to its ability to socially interact with humans. In the current experiment, we aimed to assess the contribution of the interactive quality to pain perception. We assessed the effect of touching the social robot PARO on mild and strong pain ratings and on stress perception, on a total of 60 healthy young participants. The robot either interacted with participants (ON group, n = 30) or was turned off (OFF group, n = 30). Touching the robot induced a decrease in mild pain ratings (compared to baseline) only in the ON group while strong pain ratings decreased similarly in both the ON and the OFF groups. The decrease in mild pain ratings in the ON group was significantly greater in participants with a higher positive perception of the interaction with PARO. We conclude that part of the effect that touching the robot has on pain stems from its interactive features.

Keywords: human-robot interaction; interactivity; pain; sociality; stress; touch.

FIGURE 1

An illustration of the experimental setup. The participant (on the left) has the thermal stimulator placed on her non-dominant arm, which is placed on the table. The experimenter (on the right) administers the accurate heat stimuli, and tracks them on the screen. (A) Baseline condition; PARO is not present. The participant uses the custom-made VAS application to rate pain levels. In order to rate the intensity of pain, the participants were asked to slide their finger from left to right on the screen of the tablet. Sliding the finger revealed a red area that expanded as the participant slid her finger further to the right. The leftmost edge of the screen was defined to have a value of zero (no pain), and the rightmost edge was defined to have a value of 10 (the most intense pain). (B) Touch condition; PARO is placed on the table next to the participant, who touches it during the administration of the heat stimuli.

FIGURE 2

Flow chart of the experimental design.

FIGURE 3

The change in perceived stress between BASELINE (T1) and TOUCH (T2) in the ON and OFF groups. Only in the ON group perceived stress decreased from T1 to T2 (*p < 0.05). Values denote mean visual-analogue scale (VAS) ratings (0–10) ± SEM. Results from the ON group are marked in blue, and results from the OFF group are marked in red. T1—before encountering PARO, T2—while touching PARO.

FIGURE 4

Changes in pain perception between BASELINE (T1) and TOUCH (T2). (A) Mild pain: The change in pain ratings (marked with a Δ) from T1 to T2 was significantly different between the ON (blue) and OFF (red) groups (**p < 0.01). Only in the ON group, pain ratings decreased significantly at T2 compared to T1 (***p < 0.001). (B) Strong pain: In both the ON and the OFF groups, pain ratings decreased at T2 compared to T1 (***p < 0.0001). Values denote mean visual-analogue scale (VAS) ratings (0–10) ± SEM. T1—before encountering PARO, T2—while touching PARO.

FIGURE 5

Differences in strong-pain ratings between participants who gave high vs. low ratings to the pleasantness of touching PARO. (A) Within the ON group there was a significant decrease in pain ratings from T1 to T2 both in HPT (***p < 0.001, purple) and LPT (**p < 0.01, green). (B) Within the OFF group there was a significant decrease in pain ratings from T1 to T2 only in HPT (***p < 0.001). The change in pain ratings from T1 to T2 (marked with a Δ) was significantly different between the HPT and LPT sub-groups (**p < 0.01). Values denote mean visual-analogue scale (VAS) ratings (0–10) ± SEM. T1—before encountering PARO, T2—while touching PARO. HPT—participants who gave above-median ratings to the pleasantness of touching PARO. LPT—participants who gave below-median ratings to the pleasantness of touching PARO.

FIGURE 6

Differences in mild-pain ratings between high communicators (HC) and low communicators (LC). Pain ratings decreased at T2 compared to T1 in both the HC (*p < 0.05, gray) and the LC groups (*p < 0.05, blue). However, at T2 pain ratings by HC were significantly lower than ratings by LC (**p < 0.01). Values denote mean visual-analogue scale (VAS) ratings (0–10) ± SEM. T1—before encountering PARO, T2—while touching PARO. HC—participants who gave above-median ratings to their ability to communicate with PARO. LC—participants who gave below-median ratings to their ability to communicate with PARO.

FIGURE 7

Differences in pain ratings between HMA and LMA. Only in the HMA sub-group (brown) there was a significant decrease in pain ratings at T2 compared to T1 (***p < 0.001). The change in pain ratings from T1 to T2 (marked with a Δ) was significantly different between the HMA and the LMA (magenta) sub-groups (**p < 0.01). Values denote mean visual-analogue scale (VAS) ratings (0–10) ± SEM. T1—before encountering PARO, T2—while touching PARO. HMA—participants who gave above-median ratings to their willingness to meet PARO again. LMA—participants who gave below-median ratings to their willingness to meet PARO again.