EEG-based Subject Independent Affective Computing Models

Bozhkov, Lachezar; Georgieva, Pétia; Santos, Isabel M.; Pereira, A.T.; Silva, Carlos Fernandes da

doi:10.1016/j.procs.2015.07.314

Cited by 28 publications

(14 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The literature suggests that early ERP components are sensitive to emotional stimuli, and the right hemisphere plays a critical role in emotion processing. Some local brain regions such as the frontal lobe and the parietal lobe are especially sensitive to emotional pressure [ 14 ]. Recent studies find that the target detection sensitivity for a negative emotional stimulus was higher than that for a neutral stimulus.…”

Section: Introductionmentioning

confidence: 99%

Direct Electrophysiological Mapping of Shape-Induced Affective Perception

Ding

Zhao

et al. 2018

Neural Plasticity

View full text Add to dashboard Cite

Visual information may convey different affective valences and induce our brain into different affective perceptions. Many studies have found that unpleasant stimuli could produce stronger emotional effects than pleasant stimuli could. Although there has been a notion that triangle is perceived as negative and circle as positive, there has been no systematic study to map the degrees of valence of shapes with different affective perceptions. Here, we employed four shapes (ellipse, triangle, and line-drawn happy and angry faces) to investigate the behavior and electrophysiological responses, in order to systematically study shape-induced affective perception. The reaction time delay and the event-related potential (ERP), particularly the early ERP component, were applied to find the associations with different affective perceptions. Our behavioral results showed that reaction time for angry face was significantly shorter than those for the other three types of stimuli (p < 0.05). In the ERP results, P1, N1, P2, and N2 amplitudes for angry face were significantly larger than those for happy face. Similarly, P1, N1, P2, and N2 amplitudes for triangle were significantly larger than those for ellipse. Particularly, P1 amplitude in the parietal lobe for angry face was the strongest, followed by happy face, triangle, and ellipse. Hence, this work found distinct electrophysiological evidence to map the shape-induced affective perception. It supports the hypothesis that affective strain would induce larger amplitude than affective ease does and strong affective stimuli induce larger amplitude than mild affective stimuli do.

show abstract

Section: Introductionmentioning

confidence: 99%

Direct Electrophysiological Mapping of Shape-Induced Affective Perception

Ding

Zhao

et al. 2018

Neural Plasticity

View full text Add to dashboard Cite

show abstract

“…A modification of the ESN is proposed in [14] to fix this problem, where the reservoir weights are iteratively updated applying intrinsic plasticity (IP) adaptation rule. IP computes the equilibrium states of the reservoir neurons and this adaptation step greatly improves the low dimensional projections of the equilibrium states at the output layer.…”

Section: Echo State Network For Feature Extractionmentioning

confidence: 99%

Factor Analysis for Finding Invariant Neural Descriptors of Human Emotions

et al. 2018

Self Cite

View full text Add to dashboard Cite

A major challenge in decoding human emotions from electroencephalogram (EEG) data is finding representations that are invariant to inter- and intrasubject differences. Most of the previous studies are focused in building an individual discrimination model for every subject (subject dependent model). Building subject-independent models is a harder problem due to the high data variability between different subjects and different experiments with the same subject. This paper explores, for the first time, the Factor Analysis as an efficient technique to extract temporal and spatial EEG features suitable to build brain-computer interface for decoding human emotions across various subjects. Our findings show that early waves (temporal window of 200–400 ms after the stimulus onset) carry more information about the valence of the emotion. Also, spatial location of features, with a stronger impact on the emotional valence, occurs in the parietal and occipital regions of the brain. All discrimination models (NN, SVM, kNN, and RF) demonstrate better discrimination rate of the positive valence. These results match closely experimental psychology hypothesis that, during early periods after the stimulus presentation, the brain response—to images with highly positive valence—is stronger.

show abstract

“…Several methods and techniques can be applied to perform emotion recognition through the use of a couple of hardware devices and software such as: analysis of emotional properties based on two physiological data such as, ECG and EEG [3]; unified system for efficient discrimination of positive and negative emotions based on EEG data [4]; automatic recognizer of the facial expression around the eyes and forehead based on Electrooculography (EOG) data giving support to emotion recognition task [5]; use of GSR and ECG data to develop a study to examine the effectiveness of Matching Pursuit (MP) algorithm in emotion recognition, using mainly PCA to reduce the features dimensionality and Probabilistic Neural Network (PNN) as the recognition technique [6]; emotion recognition system based on physiological data using ECG and respiration (RSP) data, recorded simultaneously by a physiological monitoring device based on wearable sensors [7]; emotions recognition using EEG data and also performed an analyze about the impact of positive and negative emotions using SVM and RBF as the recognition methods [8]; new approach to emotion recognition based on EEG and classification method using Artificial Neural Networks (ANN) with features analysis based on Kernel Density Estimation (KDE) [9]; an application that stores several psychophysiological data based on HR, ECG, SpO2 and GSR, that were acquired while the users watched advertisements about smoking campaigns [10]; experiments based on flight simulator to developed a multimodal sensing architecture to recognize emotions using three different techniques for biosignal acquisitions [11]; multimodal sensing system to identify emotions using different acquisition techniques, based on photo presentation methodology [12]; real-time user interface with emotion recognition that depends on the need for skill development to support a change in the interface paradigm to one that is more human centered [13]; recognize emotions through psychophysiological sensing using a multiple-fusion-layer based on ensemble classifier of stacked auto encoder (MESAE) [14].…”

Section: Introductionmentioning

confidence: 99%

Multimodal Approach for Emotion Recognition Based on Simulated Flight Experiments

Roza

Postolache

2019

Sensors

View full text Add to dashboard Cite

The present work tries to fill part of the gap regarding the pilots' emotions and their bio-reactions during some flight procedures such as, takeoff, climbing, cruising, descent, initial approach, final approach and landing. A sensing architecture and a set of experiments were developed, associating it to several simulated flights (N f lights = 13) using the Microsoft Flight Simulator Steam Edition (FSX-SE). The approach was carried out with eight beginner users on the flight simulator (N pilots = 8). It is shown that it is possible to recognize emotions from different pilots in flight, combining their present and previous emotions. The cardiac system based on Heart Rate (HR), Galvanic Skin Response (GSR) and Electroencephalography (EEG), were used to extract emotions, as well as the intensities of emotions detected from the pilot face. We also considered five main emotions: happy, sad, angry, surprise and scared. The emotion recognition is based on Artificial Neural Networks and Deep Learning techniques. The Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) were the main methods used to measure the quality of the regression output models. The tests of the produced output models showed that the lowest recognition errors were reached when all data were considered or when the GSR datasets were omitted from the model training. It also showed that the emotion surprised was the easiest to recognize, having a mean RMSE of 0.13 and mean MAE of 0.01; while the emotion sad was the hardest to recognize, having a mean RMSE of 0.82 and mean MAE of 0.08. When we considered only the higher emotion intensities by time, the most matches accuracies were between 55% and 100%.

show abstract

EEG-based Subject Independent Affective Computing Models

Cited by 28 publications

References 8 publications

Direct Electrophysiological Mapping of Shape-Induced Affective Perception

Direct Electrophysiological Mapping of Shape-Induced Affective Perception

Factor Analysis for Finding Invariant Neural Descriptors of Human Emotions

Multimodal Approach for Emotion Recognition Based on Simulated Flight Experiments

Contact Info

Product

Resources

About