Computational EEG Analysis Techniques When Playing Video Games: A Systematic Review †
2. Systematic Review
2.1. Searching Terms
2.2. Eligibility Criteria
- Impact: based on citations per year, to ensure certain quality. The criterion is to have a minimum of four citation per year from the date of publication.
- Publication year: we limited the search to publications from 1998 that was the year in which the popularity of this subject started to grow.
- Language: This criterion is to select the works understandable by the majority of the readers, being English the most standardized language for scientific publications.
2.3. Document Collection
2.4. Manual Screening
- First phase consisted in determining if the articles were using or not video games while recording the EEG data. This was made because there were some articles that used traditional games and there were some others that did not record the data while using a video game. At the end of this phase, the number of documents were 36.
- In the Second phase we selected the most interesting and relevant documents from the previous phase. After this phase we got the final list of 14 papers.
2.5.1. Initial Search String
2.5.2. Final Selected Documents
- Fast Fourier Transform (FFT) . FFT is an efficient algorithm that implements Discrete Fourier Transform (DFT), which transforms a signal from the time domain to the frequency domain.
- Discrete Wavelet Transform (DWT). DWT refers to a family of transformation techniques that decompose signals obtaining information about time and frequency domains, in contrast to FFT that only decomposes into the frequency domain.
- Morlet Wavelet Transform . Morlet Wavelet Transform is a DWT that uses a complex exponential with a Gaussian envelop.
- Hamming Window . Hamming Window is a window function which is a mathematical function used to avoid discontinuity at the beginning and at the end of a signal. It is a preprocessing technique useful when working with FFT in order to get better results.
- Bandpass Filter (Low-pass filter or high-pass filter). It filters the signal removing the parts over a low-pass frequency and bellow a high-pass frequency which are specified explicitly.
- Laplacian Filter (Spatial Laplacian Derivation, SLD). This preprocessing technique is used for improving the spatial resolution of EEG signals by using the Laplace operator.
- Independent Component Analysis (ICA) . ICA is a feature extraction method that transforms multivariate random signal into a signal having components that are mutually independent.
- Principal Component Analysis (PCA) . PCA is a dimensionality reduction technique by creating new uncorrelated features.
- Canonical Analysis (Canonical Variance Analysis). Canonical Analysis is a statistical technique that captures a relationship between a set of predictor variables and a set of criterion variables. It is commonly used in feature selection.
- Quadratic Discriminant Analysis (QDA). QDA is a classifier related to LDA but it is more flexible than LDA to classify.
- Support Vector Machine (SVM) . SVM is a supervised machine learning technique that allows classification and regression (SVR). It works by finding linear separators between classes, but it can perform non-linear classification by using the kernel trick.
- Naive Bayes Classifier . This method classifies by giving to an element (defined by its features) a probability to belong to each of the classes. It can be used either for classification or regression.
- Artificial Neural Networks (ANN). Artificial Neural Networks are a powerful machine learning technique inspired by biological neural networks that can do either classification and regression.
- Bhattacharyya Distance . The Bhattacharyya distance is a measure of the similarity between two probability distributions and can be used to know the degree of separability of classes in classification.
- Mahalanobis Distance . The Mahalanobis distance is a measure of the distance between two multidimensional random variables that haves into account the correlation between each variable. It is used in some classifier algorithms as the distance. This is a particular case of Bhattacharyya Distance.
- Event Related Potential (ERP) . ERP is an EEG analysis technique that consists in measuring the EEG of a signal that is result of a certain event or stimulus. At a given moment, the brain is working in many tasks and a single recording has usually a lot of noise, so, in order to deal with it, the ERP technique captures more than one recording of an event and averages them.
- Higuchi Fractal Dimension (HFD . This algorithm returns a number that is related to the fractal dimension of a signal. The fractal dimension is a mathematical concept used to describe fractal objects. Fractal dimension is usually applied in the context of signal processing by assuming that a signal is a fractal object. In EEG, fractal dimension is a feature that indicates the complexity of a signal.
- Box-Counting Dimension . This is another algorithm to compute fractal dimension that uses a different approach than HFD, but returns similar results.
- Sample Entropy . Sample Entropy is a feature that measures the complexity of a signal and it is calculated as the negative logarithm of the number of subseries of length m that have distance < r divided by the number of subseries of length m + 1 that also have distance < r.
- Lempel-Ziv Complexity (LZC). LZC is a feature that measures the complexity of a signal. It can be calculated by discretizing the signal, then counting the number of different sequences (c) and finally dividing c between the maximum possible number of different sequences in order to normalize it.
- Averaging. A common used method of creating features is by averaging another features, signals data or transformed signals between two limits.
- Cross-Correlation Coefficient (CCC). CCC is a method to determine the degree to which two signals or numerical series are correlated.
- Dynamic Time Warping (DTW) . DTW is an algorithm for measuring similarity between two temporal sequences.
- Phase Locking Value (PLV) . PLV is a feature that can be used to detect the synchronization of two signals.
- sLORETA . sLORETA is a method that computes images of electric neuronal activity from EEG.
- ANOVA. ANOVA is a collection of statistical models and their associated procedures used to analyse the differences among more than two groups.
- Least Significant Difference (LSD). LSD is a statistical technique that consists in creating a value that indicates if two groups are different between them when the difference between their means is lesser than this value.
- Genetic Algorithms (GAs). GAs are a family of optimization algorithms that are inspired by the evolution theory.
3. Conclusions and Future Work
- Chanel, G.; Rebetez, C.; Bétrancourt, M.; Pun, T. Emotion Assessment from Physiological Signals for Adaptation of Game Difficulty. IEEE Trans. Syst. Man Cybern. A Syst. Hum. 2011, 41, 1052–1063. [Google Scholar] [CrossRef]
- Lalor, E.; Kelly, S.; Finucane, C.; Burke, R.; Smith, R.; Reilly, R.; McDarby, G. Steady-State VEP-Based Brain-Computer Interface Control in an Immersive 3D Gaming Environment. Eurasip J. Appl. Signal Process. 2005, 2005, 3156–3164. [Google Scholar] [CrossRef]
- Huang, D.; Qian, K.; Fei, D.Y.; Jia, W.; Chen, X.; Bai, O. Electroencephalography (EEG)-Based Brain-Computer Interface (BCI): A 2-D Virtual Wheelchair Control Based on Event-Related Desynchronization/ Synchronization and State Control. IEEE Trans. Neural Syst. Rehabil. Eng. 2012, 20, 379–388. [Google Scholar] [CrossRef] [PubMed]
- Scherer, R.; Lee, F.; Schlögl, A.; Leeb, R.; Bischof, H.; Pfurtscheller, G. Toward Self-Paced Brain-Computer Communication: Navigation through Virtual Worlds. IEEE Trans. Biomed. Eng. 2008, 55, 675–682. [Google Scholar] [CrossRef]
- Finke, A.; Lenhardt, A.; Ritter, H. The MindGame: A P300-Based Brain-Computer Interface Game. Neural Netw. 2009, 22, 1329–1333. [Google Scholar] [CrossRef] [PubMed]
- Russoniello, C.; O’Brien, K.; Parks, J. The Effectiveness of Casual Video Games in Improving Mood and Decreasing Stress. J. Cyber Ther. Rehabil. 2009, 2, 53–66. [Google Scholar]
- Wang, Q.; Sourina, O.; Nguyen, M. Fractal Dimension Based Neurofeedback in Serious Games. Vis. Comput. 2011, 27, 299–309. [Google Scholar] [CrossRef]
- Berta, R.; Bellotti, F.; De, G.; Pranantha, D.; Schatten, C. Electroencephalogram and Physiological Signal Analysis for Assessing Flow in Games. IEEE Trans. Comput. Intell. AI Games 2013, 5, 164–175. [Google Scholar] [CrossRef]
- Millán, J.; Ferrez, P.; Galán, F.; Lew, E.; Chavarriaga, R. Non-Invasive Brain-Machine Interaction. Int. J. Pattern Recogn. Artif. Intell. 2008, 22, 959–972. [Google Scholar] [CrossRef]
- Zhang, C.; Wang, H.; Wu, M.H. EEG-Based Expert System Using Complexity Measures and Probability Density Function Control in Alpha Sub-Band. Integr. Comput.-Aided Eng. 2013, 20, 391–405. [Google Scholar] [CrossRef]
- Mu, Y.; Guo, C.; Han, S. Oxytocin Enhances Inter-Brain Synchrony during Social Coordination in Male Adults. Soc. Cognit. Affect. Neurosci. 2016, 11, 1882–1893. [Google Scholar] [CrossRef] [PubMed]
- Bai, O.; Lin, P.; Huang, D.; Fei, D.Y.; Floeter, M. Towards a User-Friendly Brain-Computer Interface: Initial Tests in ALS and PLS Patients. Clin. Neurophysiol. 2010, 121, 1293–1303. [Google Scholar] [CrossRef] [PubMed]
- Johnny, C.; Tan, D. Using a Low-Cost Electroencephalograph for Task Classification in HCI Research; ACM: New York, NY, USA, 2006; pp. 81–90. [Google Scholar]
- Mondéjar, T.; Hervás, R.; Johnson, E.; Gutierrez, C.; Latorre, J. Correlation between Videogame Mechanics and Executive Functions through EEG Analysis. J. Biomed. Inform. 2016, 63, 131–140. [Google Scholar] [CrossRef]
- Cooley, J.W.; Tukey, J.W. An Algorithm for the Machine Calculation of Complex Fourier Series. Math. Comput. 1965, 19, 297–301. [Google Scholar] [CrossRef]
- Grossmann, A.; Morlet, J. Decomposition of Hardy Functions into Square Integrable Wavelets of Constant Shape. SIAM J. Math. Anal. 1984, 15, 723–736. [Google Scholar] [CrossRef]
- Blackman, R.B.R.B.; Tukey, J.W.J.W. The Measurement of Power Spectra from the Point of View of Communications Engineering; Dover Publications: New York, NY, USA, 1959. [Google Scholar]
- Comon, P. Independent Component Analysis, A New Concept? Signal Process. 1994, 36, 287–314. [Google Scholar] [CrossRef]
- F.R.S; K.P., LIII. On Lines and Planes of Closest Fit to Systems of Points in Space. Lond. Edinb. Dublin Philos. Mag. J. Sci. 1901, 2, 559–572. [Google Scholar] [CrossRef]
- Pregenzer, M.; Flotzinger, D.; Pfurtscheller, G. Distinction Sensitive Learning Vector Quantisation—A New Noise-Insensitive Classification Method. In Proceedings of the 1994 IEEE International Conference on Neural Networks (ICNN’94), Orlando, FL, USA, 28 June–2 July 1994; Volume 5, pp. 2890–2894. [Google Scholar]
- Kohonen, T. Learning Vector Quantization. In Self-Organizing Maps; Springer Series in Information Sciences; Springer: Berlin/Heidelberg, Germany, 1995; pp. 175–189. [Google Scholar]
- Fisher, R.A. The Use of Multiple Measurements in Taxonomic Problems. Ann. Eugen. 1936, 7, 179–188. [Google Scholar] [CrossRef]
- Rao, C.R. The Utilization of Multiple Measurements in Problems of Biological Classification. J. R. Stat. Soc. Ser. B (Methodol.) 1948, 10, 159–203. [Google Scholar] [CrossRef]
- Cortes, C.; Vapnik, V. Support-Vector Networks. Mach. Learn. 1995, 20, 273–297. [Google Scholar] [CrossRef]
- Zhang, H. The Optimality of Naïve Bayes. In Proceedings of the FLAIRS2004 Conference, Miami Beach, FL, USA, 17–19 May 2004. [Google Scholar]
- Bhattacharyya, A. On a Measure of Divergence between Two Multinomial Populations. Sankhyā Indian J. Stat. (1933–1960) 1946, 7, 401–406. [Google Scholar]
- Mahalanobis, P. On the Generalised Distance in Statistics; National Institute of Science of India: Karnataka, India, 1936; Volume 2, pp. 49–55. [Google Scholar]
- Luck, S. An Introduction to The Event-Related Potential Technique; MIT Press: Cambridge, MA, USA, 2005. [Google Scholar]
- Higuchi, T. Approach to an Irregular Time Series on the Basis of the Fractal Theory. Phys. D Nonlinear Phenom. 1988, 31, 277–283. [Google Scholar] [CrossRef]
- B. Mandelbrot, B. The Fractal Geometry of Nature. Am. J. Phys. 1983, 51, 468. [Google Scholar] [CrossRef]
- Raghavendra, B.; Dutt, D. Computing Fractal Dimension of Signals Using Multiresolution Box-Counting Method. World Acad. Sci. Eng. Technol. 2010, 37, 1266–1281. [Google Scholar]
- Richman, J.S.; Moorman, J.R. Physiological Time-Series Analysis Using Approximate Entropy and Sample Entropy. Am. J. Physiol.-Heart Circ. Physiol. 2000, 278, H2039–H2049. [Google Scholar] [CrossRef] [PubMed]
- Lempel, A.; Ziv, J. On the Complexity of Finite Sequences. IEEE Trans. Inf. Theory 1976, 22, 75–81. [Google Scholar] [CrossRef]
- Dynamic Time Warping. In Information Retrieval for Music and Motion; Springer: Berlin/Heidelberg, Germany, 2007; pp. 69–84. [CrossRef]
- Aydore, S.; Pantazis, D.; Leahy, R.M. A Note on the Phase Locking Value and Its Properties. Neuroimage 2013, 74, 231–244. [Google Scholar] [CrossRef]
- Pascual-Marqui, R.D. Standardized Low-Resolution Brain Electromagnetic Tomography (sLORETA): Technical Details. Methods Find Exp. Clin. Pharmacol. 2002, 24 (Suppl. D), 5–12. [Google Scholar]
|chanel et al. ||In this article they find different difficulty levels in games affects players’s emotion and that playing several times in the same difficulty gives rise to boredom. Also they train classifiers to detect three emotional classes that they identify.||Laplacian filter, bandpass filter, LDA, QDA, SVM, EEG_W|
|Lalor et al. ||This paper presents a BCI for binary control with high precision in a 3D game.||Squared FFT, FFT of autocorrelation LDA|
|Huang et al. ||This article shows BCI for a 2D virtual wheelchair game with a multiclass control and with high hit rate.||Laplacian Filter, Hamming window, FFT, Mahalanobis distance|
|Scherer et al. ||It presents a BCI control (of three classes) for a 3D video game.||Bandpass, averaging, LDA, DSLVQ|
|Finke et al. ||They present a BCI for controlling a character on a three-dimensional game board by detecting P300 events.||PCA, LDA, ERP|
|Russoniello et al. ||In this paper, they study the effects of Casual Video Games (CVGs) on mood and stress through EEG and they find that CVGs increase mood and reduce stress.||FFT, LSD (Least Significant|
|Wang et al. ||It proposes a non linear fractal dimension based method quantify concentration level in neurofeedback games.||HFD, Box-Counting, ROC|
|Berta et al. ||They performed a spectral characterization of the video-gaming experience using a a four-electrode EEG.||PCA, FFT, ANOVA, SVM|
|Millan et al. ||This paper reviews the field of BCI, they develop three BCI systems and discuss about current research directions.||Canonical analysis, Gaussian|
|Zhang et al. ||They propose EEG-based expert system by using complexity features. The EEG-based expert system is tested by using a video game.||DWT, ICA, sample Entropy, Lempel-Ziv complexity, sLoreta, neural networks|
|Mu et al. ||They research about oxytocin (OT) effects on an individual’s brain activity during social interaction using a real-time coordination game and recording EEG in order to examine the OT effects.||Morlet Wavelet Transform,|
Phase Locking Value
|Bai et al. ||In the article they develop a BCI system suitable for ALS and PLS patients that does not require too much training. They develop a video game in order to test the system.||Laplacian Filter, Bhattacharyya distance, Mahalanobis distance, Genetic algorithm, ROC curve, SVM|
Tan et al. 
|This article describes two experiments in which they explore task classification through EEG. In the second experiment they use tasks that involve playing a video game.||FFT, averaging different values, phase coherence, Correlation based Feature Selection, Bayesian network classifiers|
|Mondejar et al. ||This paper explores cognitive skills and their training through the uses of serious games by using EEG to assess that relation.||Cross-Correlation Coefficient,|
Dynamic Time Warping
|Preprocessing||Dimensionality reduction||ICA, PCA, canonical analysis, DSLVQ, LDA, QDA, GA|
|Filtering||Hamming window, bandpass filter, Laplacian filter, ICA|
|Transformations||FFT, DWT, Morlet Wavelet Transform|
|Feature extraction||One Channel||HFD, Box-Counting, sample entropy, LZC, Averaging, ERP|
|Multi-channel||CCC, DTW, PLV, ERP|
|Classification||LDA, QDA, SVM, Naive-Bayes Classifier, ANN|
|Statistical analysis||ANOVA, LSD|
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Cabañero-Gómez, L.; Hervas, R.; Bravo, J.; Rodriguez-Benitez, L. Computational EEG Analysis Techniques When Playing Video Games: A Systematic Review. Proceedings 2018, 2, 483. https://doi.org/10.3390/proceedings2190483
Cabañero-Gómez L, Hervas R, Bravo J, Rodriguez-Benitez L. Computational EEG Analysis Techniques When Playing Video Games: A Systematic Review. Proceedings. 2018; 2(19):483. https://doi.org/10.3390/proceedings2190483Chicago/Turabian Style
Cabañero-Gómez, Luis, Ramon Hervas, Jose Bravo, and Luis Rodriguez-Benitez. 2018. "Computational EEG Analysis Techniques When Playing Video Games: A Systematic Review" Proceedings 2, no. 19: 483. https://doi.org/10.3390/proceedings2190483