Identification of emotions triggered by different sourced stimuli can be applied to automatic systems that help, relieve or protect vulnerable groups of population. The selection of the best stimuli allows to train these artificial intelligence-based systems in a more efficient and precise manner in order to discern different risky situations, characterized either by panic or fear emotions, in a clear and accurate way. The presented research study has produced a dataset of audiovisual stimuli (UC3M4Safety database) that triggers a complete range of emotions, with a high level of agreement and with a discrete emotional categorization, as well as quantitative categorization in the Pleasure-Arousal-Dominance Affective space. This database is adequate for the machine learning algorithms contained in these automatic systems. Furthermore, this work analyses the effects of gender in the emotion elicitation under audiovisual stimuli, which can help to better design the final solution. Particularly, the focus is set on emotional responses to audiovisual stimuli reproducing situations experienced by women, such as gender-based violence. A statistical study of gender differences in emotional response was carried out on 1332 participants (811 women and 521 men). The average responses per video is around 84 (SD = 22). Data analysis was carried out with RStudio^®.

Emotion recognition is benefitting from the latest research into physiological monitoring and wireless communications, among other remarkable achievements. These technologies can indeed provide solutions to protect vulnerable people in scenarios such as personal assaults, the abuse of children or the elderly, gender violence or sexual aggression. Cyberphysical systems using smart sensors, artificial intelligence and wearable and inconspicuous devices can serve as bodyguards to detect these risky situations (through fear-related emotion detection) and automatically trigger a protection protocol. As expected, these systems should be trained and customized for each user to ensure the best possible performance, which undoubtedly requires a gender perspective. This paper presents a specialized fear recognition system for women based on a reduced set of physiological signals. The architecture proposed is characterized by the usage of three physiological sensors, lightweight binary classification and the conjunction of linear (temporal and frequency) and non-linear features. Moreover, a binary fear mapping strategy between dimensional and discrete emotional information based on emotional self-report data is implemented to avoid emotional bias. The architecture is evaluated using a public multi-modal physiological dataset with two approaches (subject-dependent and subject-independent models) focusing on the female participants. As a result, the proposal outperforms the state-of-the-art in fear recognition, achieving a recognition rate of up to 96.33% for the subject-dependent model.

Audiovisual communication is greatly contributing to the emerging research field of affective computing. The use of audiovisual stimuli within immersive virtual reality environments is providing very intense emotional reactions, which provoke spontaneous physical and physiological changes that can be assimilated into real responses. In order to ensure high-quality recognition, the artificial intelligence (AI) system must be trained with adequate data sets, including not only those gathered by smart sensors but also the tags related to the elicited emotion. Currently, there are very few techniques available for the labeling of emotions. Among them, the Self-Assessment Manikin (SAM) devised by Lang is one of the most popular. This study shows experimentally that the graphic proposal for the original SAM labelling system, as devised by Lang, is not neutral to gender and contains gender biases in its design and representation. Therefore, a new graphic design has been proposed and tested according to the guidelines of expert judges. The results of the experiment show an overall improvement in the labeling of emotions in the pleasure–arousal–dominance (PAD) affective space, particularly, for women. This research proves the relevance of applying the gender perspective in the validation of tools used throughout the years.

Affective computing through physiological signals monitoring is currently a hot topic in the scientific literature, but also in the industry. Many wearable devices are being developed for health or wellness tracking during daily life or sports activity. Likewise, other applications are being proposed for the early detection of risk situations involving sexual or violent aggressions, with the identification of panic or fear emotions. The use of other sources of information, such as video or audio signals will make multimodal affective computing a more powerful tool for emotion classification, improving the detection capability. There are other biological elements that have not been explored yet and that could provide additional information to better disentangle negative emotions, such as fear or panic. Catecholamines are hormones produced by the adrenal glands, two small glands located above the kidneys. These hormones are released in the body in response to physical or emotional stress. The main catecholamines, namely adrenaline, noradrenaline and dopamine have been analysed, as well as four physiological variables: skin temperature, electrodermal activity, blood volume pulse (to calculate heart rate activity. i.e., beats per minute) and respiration rate. This work presents a comparison of the results provided by the analysis of physiological signals in reference to catecholamine, from an experimental task with 21 female volunteers receiving audiovisual stimuli through an immersive environment in virtual reality. Artificial intelligence algorithms for fear classification with physiological variables and plasma catecholamine concentration levels have been proposed and tested. The best results have been obtained with the features extracted from the physiological variables. Adding catecholamine’s maximum variation during the five minutes after the video clip visualization, as well as adding the five measurements (1-min interval) of these levels, are not providing better performance in the classifiers.

Emotion recognition through artificial intelligence and smart sensing of physical and physiological signals (affective computing) is achieving very interesting results in terms of accuracy, inference times, and user-independent models. In this sense, there are applications related to the safety and well-being of people (sexual assaults, gender-based violence, children and elderly abuse, mental health, etc.) that require even more improvements. Emotion detection should be done with fast, discrete, and non-luxurious systems working in real time and real life (wearable devices, wireless communications, battery-powered). Furthermore, emotional reactions to violence are not equal in all people. Then, large general models cannot be applied to a multi-user system for people protection, and health and social workers and law enforcement agents would welcome customized and lightweight AI models. These semi-personalized models will be applicable to clusters of subjects sharing similarities in their emotional reactions to external stimuli. This customization requires several steps: creating clusters of subjects with similar behaviors, creating AI models for every cluster, continually updating these models with new data, and enrolling new subjects in clusters when required. An initial approach for clustering labeled data compiled (physiological data, together with emotional labels) is presented in this work, as well as the method to ensure the enrollment of new users with unlabeled data once the AI models are generated. The idea is that this complete methodology can be exportable to any other expert systems where unlabeled data are added during in-field operation and different profiles exist in terms of data. Experimental results demonstrate an improvement of 5% in accuracy and 4% in F1 score with respect to our baseline general model, along with a 32% to 58% reduction in variability, respectively.

This paper presents how STEM disciplines and Social Sciences are combined to generate a database of audiovisual stimuli with a gender perspective capable of training an intelligent system for detecting dangerous situations in women who are Victims of Gender Violence. In the methodology of stimulus selection, it has been composed of two procedures: 1) According to research criteria for those related to phylogenetic and associative learning fears; 2) According to criteria based on in-depth interviews with experts in violence and content validation through the technique of expert judges for those related to fears that have to do with traumatic experiences of Gender Violence. The results indicate that the methodological combination improves the selection of audiovisual stimuli that are part of the experiments, better adjusting to the needs of women Victims of Gender violence