Artificial Neural Networks to Assess Emotional States from Brain-Computer Interface

Roberto Sánchez-Reolid, Arturo S. García, Miguel Á. Vicente-Querol, Luz Fernández‐Aguilar, Marı́a Teresa López, Antonio Fernández‐Caballero, Pascual González

Electronics · 2018 · 47 citations · 21 references

DOIFull text

Open access

Concepts

TL;DR

Emotion estimation is crucial for modern brain‑computer interface devices such as the Emotiv EPOC+ headset. The study evaluates the classification accuracy of emotional states reported by the Emotiv EPOC+ API. The experiment presented 16 participants with International Affective Picture System images while wearing the headset, compared self‑assessed emotions to IAPS values, and then applied multilayer perceptron neural networks to evaluate the API’s classification accuracy. An ANN with three hidden layers (30, 8, 3 neurons) achieved 85 % accuracy, demonstrating that the Emotiv EPOC+ API’s emotional estimates are reliable enough for real‑time applications without additional EEG processing.

Abstract

Estimation of human emotions plays an important role in the development of modern brain-computer interface devices like the Emotiv EPOC+ headset. In this paper, we present an experiment to assess the classification accuracy of the emotional states provided by the headset’s application programming interface (API). In this experiment, several sets of images selected from the International Affective Picture System (IAPS) dataset are shown to sixteen participants wearing the headset. Firstly, the participants’ responses in form of a self-assessment manikin questionnaire to the emotions elicited are compared with the validated IAPS predefined valence, arousal and dominance values. After statistically demonstrating that the responses are highly correlated with the IAPS values, several artificial neural networks (ANNs) based on the multilayer perceptron architecture are tested to calculate the classification accuracy of the Emotiv EPOC+ API emotional outcomes. The best result is obtained for an ANN configuration with three hidden layers, and 30, 8 and 3 neurons for layers 1, 2 and 3, respectively. This configuration offers 85% classification accuracy, which means that the emotional estimation provided by the headset can be used with high confidence in real-time applications that are based on users’ emotional states. Thus the emotional states given by the headset’s API may be used with no further processing of the electroencephalogram signals acquired from the scalp, which would add a level of difficulty.

References

21