Catálogo de publicaciones - libros

Recently, there has been considerable interest in the automated recognition of affect from written and spoken language. In this paper, we investigate how information on a speaker’s affect may be inferred from lexical features using statistical methods. Dictionaries of affect offer great promise to affect sensing since they contain information on the affective qualities of single words or phrases that may be employed to estimate the emotional tone of the corresponding dialogue turn. We investigate to what extent such information may be extracted from general-purpose dictionaries in comparison to specialized dictionaries of affect. In addition, we report on results obtained for a dictionary that was tailored to our corpus.

The largest part of the advertising market is already electronic. This means there is an opportunity for automatizing some of the production processes, such as producing and rendering catchy language expressions. A system is described that produces creative, affective variations of familiar textual expressions and animates them according to the affective contents. The textual variations are based on lexical semantics techniques such as an algorithm that looks for affective similarity, and on concepts such as optimal innovation, realized through assonant substitutions. Animation results from the realization of a scripting language for kinetic typography. The combination of affective variation and automatic affective perceptual rendering of key elements in the resulting expression is meant to guarantee effectiveness in advertising communication.

An investigation into capturing the relation of physiology, beyond heart rate recording, to expressed preferences of entertainment in children’s physical gameplay is presented in this paper. An exploratory survey experiment raises the difficulties of isolating elements derived (solely) from heart rate recordings attributed to reported entertainment and a control experiment for surmounting those difficulties is proposed. Then a survey experiment on a larger scale is devised where more physiological signals (Blood Volume Pulse and Skin Conductance) are collected and analyzed. Given effective data collection a set of numerical features is extracted from the child’s physiological state. A preference learning mechanism based on neuro-evolution is used to construct a function of single physiological features that models the players’ notion of ‘fun’ for the games under investigation. Performance of the model is evaluated by the degree to which the preferences predicted by the model match those expressed by the children. Results indicate that there appears to be increased mental/emotional effort in preferred games of children.

The purpose of this study is to investigate whether users’ subjective interaction states which indicate whether they feel comfortable interacting with artificial agents can be comprehended by means of biological index. As a biological index, we focused on the heart rate variability and used an RSA ratio for the time period when the respiratory sinus arrhythmia (RSA) value is higher than the Mayer wave sinus arrhythmia (MWSA) value divided by the total interaction period. To clarify the effectiveness of this RSA ratio, we conducted two experiments to measure this index among participants who played a simple cooperation game with an artificial agent, which either responded as the participant desired or responded randomly. The results showed that the RSA ratio determined the subjective interaction state which indicated whether they felt comfortable or uncomfortable with their interactions with the agents.

The present study was the first in line of a series of experiments investigating the possibilities of using voluntarily produced physiological signals in computer-assisted therapy. The current aim was to find out whether computer-guided voluntary facial activations have an effect on autonomous nervous system activity. Twenty-seven participants performed a series of voluntary facial muscle activations, while wireless electrocardiography and subjective experiences were recorded. Each task consisted of activating either the muscle (activated when frowning) or the muscle (activated when smiling) at one of three activation intensities (i.e. low, medium, and high). Our results showed a voluntary facial activation control effect (FACE) on psychological (i.e. level of experience) and physiological activity. Different muscle activations produced both task-specific emotional experiences and significant changes in heart rate and heart rate variability. Low intensity activations of both muscles were the most effective, easy to perform, and pleasant. We conclude that the FACE can clearly open the route for regulating involuntary physiological processes.

The present aim was to explore heart rate responses when stimulating participants with technology primarily aimed at the rehabilitation of older adults. Heart rate responses were measured from 31 participants while they listened to emotionally provoking negative, neutral, and positive musical clips. Ratings of emotional experiences were also collected. The results showed that heart rate responses to negative musical stimuli differed significantly from responses to neutral stimuli. The use of emotion-related physiological responses evoked by stimulating devices offers a possibility to enhance, for example, emotionally stimulating or otherwise therapeutic sessions.

This study used self-report, facial EMG, galvanic skin response, and eye tracking to assess users’ participants’ responses to two charity web sites. For each site, page content and colour of presentation (colour versus black & white) were manipulated. Results support the utility and diversity of these measures. Each provided information about users’ responses to web page designs, with a good deal of variation in sensitivity to experimental manipulations. In combination the data from these measures allowed further inferences to be drawn, supporting a multi-dimensional view of user experience and the need for multi-method approaches to evaluation.

A student, supported by his classmates, throws a piece of chalk at the teacher who is writing on the blackboard. The teacher rapidly turns back and promptly catches him in the act. The student blushes and suddenly realizes how bad it was what he did.

Wishful thinking is an affective bias towards more pleasant interpretations/beliefs. We model wishful thinking effects, to some extent, with a belief revision theory: . By modeling tendencies to believe in goals, “disliked situations” appear as “contradicting beliefs” (the target of belief revision). Furthermore, the current set of beliefs is selected according to an assessment of preference that accounts for rational factors, as well as an affective factor: “likeability”.

Recent research has shown that virtual agent able to express empathic emotions enhances human-machine interaction. In this paper, we present the capabilities that virtual agent should have to be empathic toward a user. Moreover, we propose a computational representation of emotions which may be experienced by a user during a human-machine dialog. This semantically grounded formal representation enables a rational dialog agent to identify from a dialogical situation the empathic emotion and its intensity that he should express.