14:00   Oral Session 6-GZ – Affective Computing: Tools & Applications 2
Chair: Rana el Kaliouby
14:00
25 mins 		Therapy Progress Indicator (TPI): Combining speech parameters and the subjective unit of distress
Egon L. van den Broek, Frans van der Sluis, Ton Dijkstra
Abstract: A posttraumatic stress disorder (PTSD) is a severe handicap in daily life and its treatment is complex. To evaluate the success of treatments, an objective and unobtrusive expert system was envisioned: an therapy progress indicator (TPI). Speech was considered as an excellent candidate for providing an objective, unobtrusive emotion measure. Speech of 26 PTSD patients was recorded while they participated in two reliving sessions: re-experiencing their last panic attack and their last joyful occasion. As a subjective measure, the subjective unit of distress was determined, which enabled the validation of derived speech features. A set of parameters of the speech features: signal, power, zero crossing ratio, and pitch, was found to discriminate between the two sessions. A regression model involving these parameters was able to distinguish between positive and negative distress. This model lays the foundation for an TPI for patients with PTSD, which enables objective and unobtrusive evaluations of therapies.
14:25
25 mins 		Emotional Responses to Haptic Stimuli in Laboratory versus Travelling by Bus Contexts
Katri Salminen, Veikko Surakka, Jussi Rantala, Jani Lylykangas, Pauli Laitinen, Roope Raisamo
Abstract: The present aim was to study if emotional ratings of haptic stimuli are affected by different contexts. A touchscreen device was used to produce 8 stimuli in laboratory and travelling by bus conditions. The stimuli were varied by amplitude, burst number and rise time. 10 participants were to rate the stimuli using four emotion related bipolar scales: pleasantness, arousal, approachability, and dominance. The results showed that the stimuli were rated as more pleasant, less arousing, and less dominant in the bus than in the laboratory. The stimuli with three bursts and high amplitudes were rated as significantly more unpleasant, arousing, and dominating than the low amplitude stimuli with one burst. The results suggest that simple haptic stimulation can evoke different emotional responses. The results can be utilized when designing, for example, pleasant or dominating haptic feedback.
14:50
25 mins 		Analyzing the Impact of Camera Viewpoint on Player Psychophysiology
Hector Perez Martinez, Arnav Jhala, Georgios Yannakakis
Abstract: Information about interactive virtual environments, such as games, is perceived by users through a virtual camera. While most interactive applications let the users control the camera, in complex navigation tasks within 3D environments users often get frustrated with the interaction. In this paper, we motivate for the inclusion of camera control as a vital component of affective adaptive interaction in games and investigate the impact of camera viewpoints on psychophysiology of players through an evaluation game survey experiment. The statistical analysis presented demonstrates that emotional responses and physiological indexes are affected by camera settings.
15:15
25 mins 		AffectButton: Towards a Standard for Dynamic Affective User Feedback
Joost Broekens, Willem-Paul Brinkman
Abstract: Emotions are an important aspect of life. Persons have emotions while using products and technology. It is becoming more and more important to be able to assess these emotions for multiple reasons: (a) to develop better products, (b) to better understand how the user interacts with products, and (c) because the affective state of the user is of importance to the product itself (e.g., in the case of social software, persuasive computing, recommendation). In general there are two ways of extracting affective user feedback: explicit and implicit. Here we present a new interface component that enables users to give explicit affective feedback in a flexible and dynamic way. We call this component the AffectButton. Based on statistical analysis of affective user input gathered with the AffectButton in three user studies, we present evidence that users can use the button effectively to enter affective feedback. Furthermore, the feedback is reliable and valid.
15:40
25 mins 		The emotion slider: a self-report device for the continuous measurement of emotion
Gaël Laurans, P.M.A. Desmet, P.P.M. Hekkert
Abstract: The emotion slider is a device developed to collect self-reports of the valence of users’ experiences with interactive systems based on recent theories on the embodiment of emotion and basic approach/avoidance behavioral tendencies. To test it, participants (N = 51) watched 10 positive and 10 negative slides from the International affective picture system while using the emotion slider in two different ways: pushing the handle to report positive feelings and pulling it for negative feelings in one condition (incongruent condition) and pushing the handle to report negative feelings and pulling it for positive feelings in the other (congruent condition). Response times were significantly different between the two usage conditions but the direction of this difference did not conform to the prediction. Shorter response time was associated with fewer errors. The conclusion describes some implications for human-computer interaction research.