Improving Human-Machine Interaction – A Non Invasive Approach to Detect Emotions in Car Drivers

Abstract

Abstract. As cars become increasingly computerized, automatic emotion detection and affective computing provides a promising basis for future-oriented human-computer interaction (HCI) in cars. However, we are still facing severe problems when trying to detect the users' emotional state reliably. This experimental study investigated grip-strength as a new non-invasive method to detect emotions directly in an automobile context. A positive emotion (happiness) and a negative emotion (anger) were examined regarding their influence on gripstrength applied to the steering wheel. Results confirmed and extended preliminary findings: Drivers' grip-strength slightly increased while driving a car when happiness was experienced and especially decreased when anger was experienced. Implications for further research as well as for praxis are outlined.

Emotions and Affect in Human Factors and Human–Computer Interaction: Taxonomy, Theories, Approaches, and Methods

“…During the past decade, there has been a proliferation of research on emotion/affect detection. In addition, HF/HCI researchers have employed novel methods, such as using body movements (Crane and Gross, 2007), mouse and keyboard inputs (e.g., Zimmermann et al, 2003), steering wheel grip intensity (e.g., Oehl et al, 2007), etc. Researchers have applied a variety of methods to measure and detect users' affective states in each of their domains.…”

Adaptive Multimodal Emotion Detection Architecture for Social Robots

“…The wide study of computer vision, which is a field that deals with how computers can gain high-level understanding from images or videos, has allowed the development of different investigations where the analysis of emotions is taken into account. Research ranging from the analysis of emotions in pedestrians [72], patients [73], [74], car drivers [75], people in restaurants [76], to TV show reactions [77], are examples in which this approach can be also applied.…”

Automatic Detection of a Driver’s Complex Mental States

“…Hu and Zhang [13], Lisetti and Nasoz [19] both investigated drivers' drowsiness. Basic emotions such as happiness and anger were also investigated [23]. A few studies have explored more complex mental states [1,9] but not related to driving.…”

“…Meanwhile, there have been some studies in automatic emotion recognitions in cars using various features, such as speech recognition [14], physiological signals like galvanic skin response and heart beat [16,22], and grip strength [23]. While speech recognition relies on the driver to talk, which can be hard when the driver is alone, obtaining physiological signals can sometimes be invasive and distractive.…”

“…Although there have been many studies looking at stress level [16,12], drowsiness [13,19], and basic emotions [23] of a driver, there is not much work on detection of complex mental states, which were actually found to occur more often in real life [27]. Also, speech recognition [14], physiological signals [16,22] and grip strength [23] have been investigated in classifying drivers' mental states. However, analysing emotions from drivers' facial expressions are not fully explored, partially due to the challenge of lighting conditions during driving and the relatively subtle expressions exhibited while driving.…”