An exciting possibility for compensating for loss of sensory function is to augment deficient senses by conveying missing information through an intact sense. Here we present an overview of techniques that have been developed for sensory substitution (SS) for the blind, through both touch and audition, with special emphasis on the importance of training for the use of such devices, while highlighting potential pitfalls in their design. One example of a pitfall is how conveying extra information about the environment risks sensory overload. Related to this, the limits of attentional capacity make it important to focus on key information and avoid redundancies. Also, differences in processing characteristics and bandwidth between sensory systems severely constrain the information that can be conveyed. Furthermore, perception is a continuous process and does not involve a snapshot of the environment. Design of sensory substitution devices therefore requires assessment of the nature of spatiotemporal continuity for the different senses. Basic psychophysical and neuroscientific research into representations of the environment and the most effective ways of conveying information should lead to better design of sensory substitution systems. Sensory substitution devices should emphasize usability, and should not interfere with other inter- or intramodal perceptual function. Devices should be task-focused since in many cases it may be impractical to convey too many aspects of the environment. Evidence for multisensory integration in the representation of the environment suggests that researchers should not limit themselves to a single modality in their design. Finally, we recommend active training on devices, especially since it allows for externalization, where proximal sensory stimulation is attributed to a distinct exterior object.

There has been steady progress in the field of affective computing over the last two decades that has integrated artificial intelligence techniques in the construction of computational models of emotion. Having, as a purpose, the development of a system for treating phobias that would automatically determine fear levels and adapt exposure intensity based on the user’s current affective state, we propose a comparative study between various machine and deep learning techniques (four deep neural network models, a stochastic configuration network, Support Vector Machine, Linear Discriminant Analysis, Random Forest and k-Nearest Neighbors), with and without feature selection, for recognizing and classifying fear levels based on the electroencephalogram (EEG) and peripheral data from the DEAP (Database for Emotion Analysis using Physiological signals) database. Fear was considered an emotion eliciting low valence, high arousal and low dominance. By dividing the ratings of valence/arousal/dominance emotion dimensions, we propose two paradigms for fear level estimation—the two-level (0—no fear and 1—fear) and the four-level (0—no fear, 1—low fear, 2—medium fear, 3—high fear) paradigms. Although all the methods provide good classification accuracies, the highest F scores have been obtained using the Random Forest Classifier—89.96% and 85.33% for the two-level and four-level fear evaluation modality.

In this paper, we investigate various machine learning classifiers used in our Virtual Reality (VR) system for treating acrophobia. The system automatically estimates fear level based on multimodal sensory data and a self-reported emotion assessment. There are two modalities of expressing fear ratings: the 2-choice scale, where 0 represents relaxation and 1 stands for fear; and the 4-choice scale, with the following correspondence: 0—relaxation, 1—low fear, 2—medium fear and 3—high fear. A set of features was extracted from the sensory signals using various metrics that quantify brain (electroencephalogram—EEG) and physiological linear and non-linear dynamics (Heart Rate—HR and Galvanic Skin Response—GSR). The novelty consists in the automatic adaptation of exposure scenario according to the subject’s affective state. We acquired data from acrophobic subjects who had undergone an in vivo pre-therapy exposure session, followed by a Virtual Reality therapy and an in vivo evaluation procedure. Various machine and deep learning classifiers were implemented and tested, with and without feature selection, in both a user-dependent and user-independent fashion. The results showed a very high cross-validation accuracy on the training set and good test accuracies, ranging from 42.5% to 89.5%. The most important features of fear level classification were GSR, HR and the values of the EEG in the beta frequency range. For determining the next exposure scenario, a dominant role was played by the target fear level, a parameter computed by taking into account the patient’s estimated fear level.

Emotions constitute an indispensable component of our everyday life. They consist of conscious mental reactions towards objects or situations and are associated with various physiological, behavioral, and cognitive changes. In this paper, we propose a comparative analysis between different machine learning and deep learning techniques, with and without feature selection, for binarily classifying the six basic emotions, namely anger, disgust, fear, joy, sadness, and surprise, into two symmetrical categorical classes (emotion and no emotion), using the physiological recordings and subjective ratings of valence, arousal, and dominance from the DEAP (Dataset for Emotion Analysis using EEG, Physiological and Video Signals) database. The results showed that the maximum classification accuracies for each emotion were: anger: 98.02%, joy:100%, surprise: 96%, disgust: 95%, fear: 90.75%, and sadness: 90.08%. In the case of four emotions (anger, disgust, fear, and sadness), the classification accuracies were higher without feature selection. Our approach to emotion classification has future applicability in the field of affective computing, which includes all the methods used for the automatic assessment of emotions and their applications in healthcare, education, marketing, website personalization, recommender systems, video games, and social media.In the 1970s, Paul Ekman identified six basic emotions [2], namely anger, disgust, fear, joy, sadness, and surprise. Russell and Mehrabian proposed a dimensional approach [3] which states that any emotion is represented relative to three fundamental dimensions, namely valence (positive/pleasurable or negative/unpleasurable), arousal (engaged or not engaged), and dominance (degree of control that a person has over their affective states).Joy or happiness is a pleasant emotional state, synonymous with contentment, satisfaction and well-being. Sadness is the opposite of happiness, being characterized by grief, disappointment, and distress. Fear emerges in the presence of a stressful or dangerous stimulus perceived by the sensory organs. When the fight or flight response appears, heart rate and respiration rate increase. Also, the muscles become more tense in order to contend with threats in the environment. Anger is defined by fury, frustration, and resentment towards others. Surprise is triggered by an unexpected outcome to a situation, ranging from amazement to shock, whereas disgust is synonymous with dislike, distaste, or repugnance, being the most visceral of all six emotions.The DEAP database [4] was created with the purpose of developing a music video recommendation system based on the users' emotional responses. The biophysical signals of 32 subjects have been recorded while they were watching 40 one-minute long excerpts of music videos eliciting various emotions. The participants rated each video in terms of valence, arousal, dominance, like/dislike and familiarity on a scale from one to nine. The physiological signals were: galvanic skin response (GSR), plethysmograph (PP...

For people who suffer from a visual impairment, navigating in an unknown environment represents a considerable challenge. To help them develop spatial orientation and mobility skills, researchers have employed audio-based computer games as a practical, interactive and user-centered learning approach. The highly immersive and attractive nature of audio games enables blind people to create a spatial representation of their environment that can be assigned to aid performing real-life navigation tasks. This paper reviews the most notable navigational audio-based games in terms of their conceptual and technological approach, accessibility and user interaction efficiency.

The Sound of Vision project involves developing a sensory substitution device that is aimed at creating and conveying a rich auditory representation of the surrounding environment to the visually impaired. However, the feasibility of such an approach is strongly constrained by neural flexibility, possibilities of sensory substitution and adaptation to changed sensory input. We review evidence for such flexibility from various perspectives. We discuss neuroplasticity of the adult brain with an emphasis on functional changes in the visually impaired compared to sighted people. We discuss effects of adaptation on brain activity, in particular short-term and long-term effects of repeated exposure to particular stimuli. We then discuss evidence for sensory substitution such as Sound of Vision involves, while finally discussing evidence for adaptation to changes in the auditory environment. We conclude that sensory substitution enterprises such as Sound of Vision are quite feasible in light of the available evidence, which is encouraging regarding such projects.

No abstract

No abstract