Emotion Measurements Through the Touch of Materials Surfaces

Abstract

The emotion generated by the touch of materials is studied via a protocol based on blind assessment of various stimuli. The human emotional reaction felt toward a material is estimated through (i) explicit measurements, using a questionnaire collecting valence and intensity, and (ii) implicit measurements of the activity of the autonomic nervous system, via a pupillometry equipment. A panel of 25 university students (13 women, 12 men), aged from 18 to 27, tested blind twelve materials such as polymers, sandpapers, wood, velvet and fur, randomly ordered. After measuring the initial pupil diameter, taken as a reference, its variation during the tactile exploration was recorded. After each touch, the participants were asked to quantify the emotional value of the material. The results show that the pupil size variation follows the emotional intensity. It is significantly larger during the touch of materials considered as pleasant or unpleasant, than with the touch of neutral materials. Moreover, after a time period of about 0.5 s following the stimulus, the results reveal significant differences between pleasant and unpleasant stimuli, as well as differences according to gender, i.e., higher pupil dilatation of women than men. These results suggest (i) that the autonomic nervous system is initially sensitive to high arousing stimulation, and (ii) that, after a certain period, the pupil size changes according to the cognitive interest induced and the emotional regulation adopted. This research shows the interest of the emotional characterization of materials for product design.

Pupil Reactions to Tactile Stimulation: A Systematic Review

“…All studies indicated a positive correlation between stimuli perception and pupil dilation ( Table 2 ). As shown in Figure 3 , while Bertheaux et al ( 2020 ) found that neutral materials lead to a lower pupil dilation than materials with high emotional intensity, independently of their valence (being pleasant or unpleasant), Van Hooijdonk et al ( 2019 ) associated the difference in the pupil dilation only with the stimulus intensity, not subjective pleasantness. The four go/no-go tasks (Lee and Margolis, 2016 ; Schriver et al, 2018 , 2020 ; Mückschel et al, 2020 ) and the 2AFC task (Ganea et al, 2020 ) found a significantly broader pupil dilation for go trials than no-go ones.…”

“…As Table 2 and Figure 3 show, the nine studies were conducted between 1996 and 2020. Three used human subjects (two with adults and one with adolescents) (Van Hooijdonk et al, 2019 ; Bertheaux et al, 2020 ; Mückschel et al, 2020 ), one wild mice (Lee and Margolis, 2016 ), two with C57BL6J mice (Ganea et al, 2020 ; Lee et al, 2020 ), two albino rats (Schriver et al, 2018 , 2020 ), and one juvenile male Japanese monkeys ( Macaca fuscata ) (Iriki et al, 1996 ). Four studies used go/no-go tasks (Lee and Margolis, 2016 ; Schriver et al, 2018 , 2020 ; Mückschel et al, 2020 ), one a vibration detection task (Lee et al, 2020 ), one a two-alternative forced choice (2AFC) task (Ganea et al, 2020 ), one used a passive task (Iriki et al, 1996 ), and two discrimination of affective touch tasks (Van Hooijdonk et al, 2019 ; Bertheaux et al, 2020 ).…”

“…Iriki et al ( 1996 ) did not report the equipment other than saying it was a motor-only shot (MOS) camera. The reported equipment were a 10-Hz in-house system (Schriver et al, 2018 , 2020 ), 30-Hz EyeTribe tracker (Van Hooijdonk et al, 2019 ), Point Gray Chameleon3 camera (Ganea et al, 2020 ), DMK22BUC03 (Lee et al, 2020 ), 50-Hz Allied Vision Technologies Pike F-032 camera (Lee and Margolis, 2016 ), 60-Hz ISCAN-ETL-100H (Bertheaux et al, 2020 ), and 500-Hz RED 500 (Mückschel et al, 2020 ). All studies had controlled luminance and either excluded the trials where there were blinks or interpolated the data.…”

“…Bertheaux et al ( 2020 ) presented 12 different stimulus sources classified as follows: three unpleasant (different sandpaper roughness levels), seven neutral (three 100% polylactic material and four mixes of 50% polylactic material and 50% other materials), and two pleasant (velvet and synthetic fur). The participants should actively touch the materials with their dominant hand in three rounds of 15 s; the first one was not considered on data analysis to avoid the surprise element.…”

“…While the 30-Hz systems (Van Hooijdonk et al, 2019 ; Ganea et al, 2020 ; Lee et al, 2020 ) are the slowest commercially available, there was an in-house-developed system that was only 10 Hz (Schriver et al, 2018 , 2020 ). Typically used are 50 and 60 Hz (Lee and Margolis, 2016 ; Bertheaux et al, 2020 ) since these are the most common frequencies in camera technology for a long time. The 500 Hz (Mückschel et al, 2020 ) started to be used in 2000 and, from the studies found in this research, is the best sampling rate.…”