Evgenia Hristova

Biosignal Based Emotion Analysis of Human-Agent Interactions

A two-phase procedure, based on biosignal recordings, is applied in an attempt to classify the emotion valence content in human-agent interactions. In the first phase, participants are exposed to a sample of pictures with known valence values (taken from IAPS dataset) and classifiers are trained on selected features of the biosignals recorded. During the second phase, biosignals are recorded for each participant while watching video clips with interactions with a female and male ECAs. The classifiers trained in the first phase are applied and a comparison between the two interfaces is carried on based on the classifications of the emotional response from the video clips. The results obtained are promising and are discussed in the paper together with the problems encountered, and the suggestions for possible future improvement.

Investigation of context effects in iterated prisoner's dilemma game

Context effects during Prisoners Dilemma (PD) game playing are investigated. The Cooperation Index (CI) – a quantity computed as a relation between payoffs – defines a cooperativeness scale, along which PD games can be distributed. Context is selected by manipulating the CI range of the games played. It is found that the level of cooperation depends not only on the CI of the current game but also on the CI of the other games in the sequence. The influence of context on the full CI scale is investigated by introducing probe games covering the whole CI range. The results are compared with the prediction of a model that takes into account the current game, the previous play history, and the predicted opponents move. The model is found to be sensitive to both the CI and to the CI of the other games form the game set and a very good agreement between the model and the experimental data was found.