Session IIId on Cognitive Modeling included three interesting presentations, all of which were related to modeling some aspect of human behavior. The first presentation was given by Tibor Bosse, and addressed the paper ‘Effects of Virtual Training on Emotional Response’. The presentation started with a description of the STRESS project of which the paper is part. This project aims to study the added value of Virtual Reality-based training for professionals that have to make difficult decisions under stress, such as police officers or public transport employees. Next, the speaker described an experiment that was performed to investigate the impact of different types of virtual training on participants’ reported emotional response when looking at affective images. The main results of the experiment indicated that a training based on reappraisal caused participants to rate the pictures as significantly less intense, whereas a content-based training resulted in significantly higher ratings. Moreover, a second experiment, performed with the same participants six months later, indicated that these effects are fairly persistent over time, and that they transfer to different pictures with similar characteristics.
The second presentation was given by Ruurdje Procee, and addressed the paper ‘A Formal Model of Procrastination’. The speaker started by presenting the goal of the research, namely to develop a formal model of procrastination (i.e., the practice of carrying out less urgent tasks in preference to more urgent ones) based on factors that are found in the psychological literature. As she pointed out, such a model can be useful for two reasons: first, it can be used to gain a better understanding of the factors involved in the process of procrastination (as well as their interactions). And second, it can be implemented within e-coaching applications that offer personalized support to individuals in their struggle with procrastination. After that, some details of the model were presented, both at a conceptual level (in terms of qualitative causal influences between the relevant concepts, which include task-related, personality-related and other factors) and at a formal level (in terms of a mathematical specification that can serve as basis for a computational model). The presentation was concluded with a discussion on the authors’ ongoing work to evaluate the model, which will be based on simulations as well as empirical data from a psychological experiment.
The third presentation was given by Janneke van der Zwaan, and addressed the paper ‘A Qualitative Evaluation of Social Support by an Empathic Agent’. This presentation started with a description of the authors’ research goal to explore to what extent embodied agents can provide social support to, among others, victims of cyberbullying. To this end, they implemented a prototype of an empathic virtual buddy that uses both verbal and nonverbal behavior to comfort users. The buddy includes a conversation model (that specifies the structure and contents of conversations) as well as an emotional model (that determines when the virtual buddy expresses sympathy, compliments and encouragements). Next, to understand what aspects of interaction with the virtual buddy are perceived as being supportive, they organized a focus group discussion among pedagogical experts. To illustrate the benefit of this discussion, the speaker cited a number of the recommendations and other comments provided by the participants. She concluded by summarizing the results of the study, which indicate that the experts felt supported by interaction with the virtual buddy, and that the focus group discussion turned out a useful approach to evaluate embodied agents in a social context.
