Behavioural Models of Risk-Taking in Human-Robot Tactile Interactions

Abstract

Touch can have a strong effect on interactions between people, and as such, it is expected to be important to the interactions people have with robots. In an earlier work, we showed that the intensity of tactile interaction with a robot can change how much people are willing to take risks. This study further develops our understanding of the relationship between human risk-taking behaviour, the physiological responses by the user, and the intensity of the tactile interaction with a social robot. We used data collected with physiological sensors during the playing of a risk-taking game (the Balloon Analogue Risk Task, or BART). The results of a mixed-effects model were used as a baseline to predict risk-taking propensity from physiological measures, and these results were further improved through the use of two machine learning techniques-support vector regression (SVR) and multi-input convolutional multihead attention (MCMA)-to achieve low-latency risk-taking behaviour prediction during human-robot tactile interaction. The performance of the models was evaluated based on mean absolute error (MAE), root mean squared error (RMSE), and R squared score (R2), which obtained the optimal result with MCMA yielding an MAE of 3.17, an RMSE of 4.38, and an R2 of 0.93 compared with the baseline of 10.97 MAE, 14.73 RMSE, and 0.30 R2. The results of this study offer new insights into the interplay between physiological data and the intensity of risk-taking behaviour in predicting human risk-taking behaviour during human-robot tactile interactions. This work illustrates that physiological activation and the intensity of tactile interaction play a prominent role in risk processing during human-robot tactile interaction and demonstrates that it is feasible to use human physiological data and behavioural data to predict risk-taking behaviour in human-robot tactile interaction.

Keywords: behaviour model; human–robot tactile interaction; non-verbal interaction; risk-taking behaviour.

Figure 1

The four interaction conditions used in the study [12].

Figure 2

Model diagnostic plots, which consist of four subplots, with the top-left plot titled “Residuals vs. Fitted” used to evaluate the assumption of a linear relationship. The top-right plot titled “Normal Q-Q” was used to test the normality of the residuals. The bottom-left plot titled “Scale-Location (or Spread-Location)” was employed to examine the homogeneity of variance of the residuals. Finally, the bottom-right plot titled “Residuals vs. Leverage” was used to identify influential cases that may significantly impact the regression results when included or excluded from the analysis.

Figure 3

Lasso coefficient path visualization. The red and blue vertical lines on the plot are the values of the minimum lambda and 1-standard-error lambda, respectively, and were obtained from a lasso regression model that had undergone cross-validation.

Figure 4

Data flow of mixed effects model.

Figure 5

Mixed effects model: actual risk-taking behaviour vs. predicted risk-taking behaviour.

Figure 6

Data flow for SVR model.

Figure 7

SVR model: actual risk-taking behaviour vs. predicted risk-taking behaviour.

Figure 8

The proposed multi-input convolutional multihead attention (MCMA) model.

Figure 9

The convolutional block in the proposed MCMA model.