Augmented Web Survey with enhanced response UI for Touch-based Psychological State Estimation

In this paper, we propose a method for estimating psychological states such as hesitation and interest during survey responses based on touch operation behavior (touch log). In addition, we propose two enhanced web survey RUIs (Response User Interfaces), (a) a slidebar type RUI and (b) a magnifier type RUI, to expand the operational differences due to hesitation and interest for more accurate estimation. To evaluate the proposed RUIs, we implemented Augmented Web Survey (AWS) as an extension of LimeSurvey and conducted two experiments. Responses were collected from 8 subjects in the experiment to evaluate (a) a slidebar type RUI and from 21 subjects to evaluate (b) a magnifier type RUI. We measured the degree of hesitation by evaluating the subjective hesitation to the questions on a 5-point scale. For the measurement of the degree of interest, the participants were asked to select their top two favorite images among the six images, and the true value was taken as the top two. As a result of comparing the operation logs of the conventional RUI and the slidebar type RUI, it was confirmed that there was a significant difference depending on the degree of hesitation. In addition, more magnification operations were obtained for the magnifier type RUI than the standard pinch-out operation suggesting that it may help estimate psychological states such as hesitation.