Cognitive attention behaviour detection systems using Electroencephalograph (EEG) signals


Keshava Murthy G.N.1,2, Zaved Ahmed Khan1*

1Medical Biotechnology Division, School of Bio Sciences and Technology, VIT University, Vellore, India

2Siddaganga Institute of Technology, Tumkur-572103, Karnataka, India

*Corresponding Author E-mail:




Using EEG signals to estimate cognitive intellectual state has pinched increasing attention in recent years, especially in the framework of brain-computer interface (BCI) design. Nevertheless, this goal is extremely difficult because, in addition to the multifaceted relationships between the cognitive state and EEG signals that yields the non-stationarity of the features extracted from EEG signals, there are artefacts bring together by eye blinks and head and body motion. In this paper, we present a review of such Cognitive Attention behaviour estimation system, which can estimate the subject’s cognitive state from the measured EEG signals. In most of the systems, a mutual information based method is employed to diminish the dimensionality of the features as well as to increase the robustness of the system. Classifiers were implemented and the results are taken to be the final decisions. The results of an introductory test with data from freely moving subjects performing numerous tasks as opposed to the strictly controlled experimental set-ups of BCI provide strong support for this approach.


KEYWORDS: Brain connectivity, divergent thinking, electroencephalogram (EEG), Cognitive behaviour, Brain-Computer Interface.




The cognitive neuroscience literature has focused on inferring the brain bases of decision-making, such as calculative decision making and decision-making under different conditions (e.g., uncertainty, risk, and ambiguity), decisions related to rewards and utility, and the equivalent intentions to engage in decision-related tasks. Researchers have proposed different models of a neuron which were able to explain few fundamental neural dynamics and signal processing underlying a neuronal activity. However, modelling of how information is acquired, processed and stored in brain is still in its phase of infancy. Understanding the dynamics and computation of a single neuron and its role in large neural network is crucial in the study of neuroscience because brain function relies on the interplay of millions of these neurons. Many aspects of brain functions and behaviour of human body can only be discussed based on how information is transmitted from neuron to neuron. All cognitive processes inside a brain are carried out with neuronal commotion comprising of synapsing and spike generation in a network of neurons.


This neural activity is driven either by the mechanism within a neuron or by the feedback interaction between a population of neurons resulting in a neural oscillation. The synchronized activity of neurons in these networks gives rise to macroscopic electric fields which results in voltage that can be measured in the scalp via EEG. Scientists have achieved some success in investigating brain activities based on recordings from EEG. For example, they have categorized the depth of sleep of a human being based on the delta wave (less than 4Hz) in EEG signals.


The language of communication between various nervous cells is through electrical signal. Electroencephalography (EEG) is a tool for measuring electrical activity [1] [2] generated in the brain, which opens a window for exploring neural activity and brain functioning. The study of the brain electrical movement, through the electroencephalographic records, is one of the most important tools for the diagnosis of neurological diseases [3]. The EEG signal is measured using electrodes placed on the scalp [4], which record the electrical field generated by the nerve cells [5] [6]. 


The millisecond time-based resolution of EEG allows scientists to inspect not only vacillations of EEG activity as a function of task demand or subject samples but also to differentiate between functional inhibitory and excitatory actions [7]. Changes in the brain’s electrical activity occur very quickly, and extremely high time resolution is mandatory to determine the precise moments at which these electrical events take place [8]. EEG provides good observational data of variability in mental status because of its high temporal resolution [9]. EEG waveform depends on the conscious level of the person. Large amounts of data are generated by EEG monitoring systems for EEG changes, and their complete visual analysis is not routinely possible [10] [11]. Computers have long been recommended to solve this problem and thus, automated systems to recognize EEG changes have been under study for several years [12].


EEG also has an important clinical use in diagnosing and treating patients with epilepsy [13]. Epileptic seizures are characterized by a burst of electrical activity usually originating from a particular area within the brain [14]. EEG is an important method for studying the transient dynamics of the human brain’s large-scale neuronal circuits [15]. EEG shows a good correlation with the mental stress in terms of suppression of alpha waves and improvement of theta waves. Alpha waves are more active in occipital and frontal regions of the brain [16]. These waves are associated with idleness of the brain. So in no stress condition, when the brain is doing no activity, alpha waves are dominant. In stressful situations, the power of alpha waves falls down showing the change in response under stress [17].


Beta waves show erratic behaviour in different frequencies in different parts of the brain and power in theta waves increases under stress or mental tasks. The accurate classification of electrical activity in a particular state of human brain helps in neurological diagnosis and also for inaugurating standards for instrumentation development. This classification also helps in the brain computer interfacing which has been gaining wide attraction in the research industry [18, 19]. Many researchers have incorporated EEG assessment in observing, understanding and explaining various cognitive aspects, some of which are discussed in the following sections.


Figure 1. Architecture of Cognition Model


An approach to model the cognitive functions using the features of stimuli and corresponding EEG responses features, to categorize cognitive model parameters by processing the stimuli (input) and responses (output) features and thus turning the modelling task into system  identification problem. Various descriptors could be taken out from stimuli such as image, speech, PST (Psychomotor Vigilance Task) or movement.



2.1. Memory- Memory is attributed to encoding, storage and retrieval of information from the brain. Fragments of the brain like the amygdala, hippocampus and striatum are involved in the specific types of memory. Going subterranean into the physiology, learning and memory are concomitant with changes in neuronal synapses. The interplay of theta (3.5-7.5Hz) and alpha (7.5-13Hz) oscillation in human EEG reflects the transfer of information in working and long term memory system [20]. Brain wave oscillations at different frequency band have been perceived during information processing in association with memory. Increased alpha commotion and theta commotion was found during the retention of memory task [21].


2.2. Emotion Valance- Frontal EEG asymmetry has been the issue of discussion in transforming EEG to metrics that can provide inferences about the emotional state of a person [22]. A greater left frontal brain activity to be associated with the positive valance of mind whereas the greater right frontal brain activity to be related to negative valance [23]. Furthermore, excessive reassurance seeking is often related to depression [24].


2.3. Fatigue- Fatigue is a state of decrease of performance which may be due to various whys and wherefores like deprivation of sleep or from continuous physical or mental workload. Different brain waves (theta, alpha and beta) have been analysed to investigate pattern at different brain sites and also on entire brain average to assess Drivers fatigue [25] [ 26].


2.4. Distraction- Distraction is a mental process that causes shift of attention. An experimental analysis was performed on cortical responses on expectancy of probable pain stimulus where the subject had to mentally perform the arithmetic task in pain + cognition condition [27].


2.5. Arousal- Arousal is a state of mind of being awake and highly alert to respond to the stimuli. Researchers have established linkage between the specific EEG spectra with the state of mind. Analogous to the blood pressure, temperature, heart rate taken as preliminary indicators to measure corresponding body activations, the brain rate can be used to measure the alertness of mind. The weighted mean frequency of EEG spectrum could be analyzed to predict the degree of mental arousal [28].



The contribution of our work is to make a review of how physiological measurements change with various human cognitive states. For this purpose, the set of variables well thought-out as input are Electrocardiogram (ECG), Electroencephalogram (EEG) and Blood Pressure (BP). Different affective states were created by giving several instructions and training among to the subjects. Consequently the variation in ECG, EEG and BP is evaluated for various states. But it is very difficult to ensure that a Person is in one mental state at any particular time as human being has the supernatural power of doing parallel activities. So it has to assume that a subject is in one condition at a particular time due to the heterogeneity of physiological signal. Another significant problem is that, physiological signals are affected by motion artifact. Again, the results can be varied from man to man with their age, gender, weight, height etc. The fundamental application of this variability measurement is that, it will help us to develop modern computer that have the ability to interface with human.


3.  AIM:

The objective of this review of Cognitive Attention behaviour estimation system, which can estimate the subject’s cognitive state from the measured EEG signals.



The edifice steps of this paper are as follows. The introductory section ends with a brief review of EEG analysis and cognitive state aspects estimation From EEG.

In Section 4.1. We address the contemporary solution  provide by researchers for quantifying cognitive state from EEG, this section will be divided into some subsections for taking points from different researches.


Section 4.2. We present the stress index development in EEG signals.


Section 4.3. Addresses some other researches performed on brain behaviour using EEG signals using various technologies.


In Section 4.4. We present the review on divergence thinking based several neuroimaging studies on EEG signals, given the central importance of this most extraordinary capacity of the human mind, one would think that the underlying neurocognitive mechanisms of creative thinking are the subject of  intense research efforts in the behavioural and brain sciences.


4.1. Techniques for quantifying cognitive state from EEG signals

Some attention has been paid to bio-signals for identifying physiological states in previous work [5]. In the very earlier stage for mental state recognition main emphasis was given in audio visual signal like facial expression, speech etc [6].


In some research work the physiological signals were recorded at the same time when the subject was listening to different music [6]. Again some researchers considered ECG and Blood Pressure as input parameters to determine the variation in physiological signals with mental states [7]. In some previous works, researchers used neural network [1] [2], Hilbert Huang transform etc. for analysis purpose. Authors in [10] discussed the mental loads on muscle tendon, blink rate and blood pressure. However, this work is used to determine the EEG Electronics and Vision along with ECG and BP.


There are few methods which covers the problem domain, in a research related to mental states estimation with the Variation of  Physiological Signals, in its author pronounce the consequences of mental circumstances due to the difference of electrocardiogram, electroencephalogram and blood pressure using BIOPAC system. It was found that the effect of motor action and thought states are mainly on blood pressure while memory related and emotion state mainly affect the ECG measurement. The EEG mainly detects the signal of task performed by the specific brain region where the electrodes are placed. Here the electrodes are placed in occipital lobe region which gives mainly the variation in alpha amplitude of EEG with eyes closed and eyes opened. Alpha wave amplitudes vary with the subjects attention to mental tasks performed with eyes closed [29].


Another method is the feature based approach for estimating cognitive state using EEG signals. The state of brain and its rapid transition from one state to the other is responsible for various activities and cognitive functions. These brain states are the result of balanced coordination between integrating and segregating activities of different lobes through rhythmic oscillations. Such coordination has been studied in recent times through synchronization of EEG signals generated from different lobes. In a research [30], the authors have considered Synchronization Likelihood (SL) to measure the synchronization or integration between the lobes. The synchronization information is stored in SL matrix and the principal components of an SL matrix have been used to represent the state of brain at any instant. Finally, the time series of weight vectors corresponding to the principal components of SL matrices at each time point has been used to classify different states of brain at different stages of a sleep deprived experiment.


One way to quantifying the behaviour is Dependence Measures; this study hypothesizes and demonstrates that it is possible to automatically discriminate face processing from processing of a simple control stimulus based on processed EEGs in an online fashion with high temporal resolution using measures of statistical dependence applied on steady-state visual evoked potentials. Correlation, mutual information, and a novel measure of association, referred to as generalized measure of association (GMA), were applied on filtered current source density data [31].


Table 1. Analyzed results of EEG signals from [29]


Dependences between channel locations [31] were assessed for two separate conditions elicited by distinct pictures (a face and a Gabor grating) flickering at a rate of 17.5 Hz. Filter settings were chosen to minimize the distortion produced by band passing parameters on dependence estimation [31].


Figure 2. Suggested Methodology for System Identification in [31]


Many cognitive architectures have been proposed with an aim of simulating and understanding human cognition. A new approach were given in the methodology [32] to model cognitive functions by relating the varying event related potential, brain waves, spectral density and latency in EEG outcomes are then related with the stimuli features to predict the cognitive state of mind.



Figure 3. Using outright correlation to weight graph connections for channel 72. First two subplots (a) and (b) show interpolated correlation measures over right and left (R and L) head surface for face condition (F) and subsequent subplots (c) and (d) exhibit the same for Gabor patch condition (G).


A research also focuses on how a presentation affects the cognitive interference. Cognitive task for longer duration increases subject’s psychological loadings because of extracting ordered answer from visual stimuli. This study investigated how intermittent odorant presentation during cognitive task contributes to cognitive function and statement of psychological loadings [33]. Spectral Centroids techniques, is one of the better technique to extract human stress features. A research presents the combination of electroencephalogram (EEG) power spectrum ratio and spectral centroids techniques to extract unique features for human stress from EEG signals. The combination of these techniques was able to improve the k-NN (k-Nearest Neighbour) classifier accuracy to detect and classify human stress from two cognitive states; Close-eye (CE) and Open-eye (OE) [34].


Figure 4. The subject is seated in the car simulator with the EEG, EOG, and EMG electrodes. Inset:  Illustration of visual stimuli corresponding to the virtual roadway environment in [35].


Recognition of driver’s intention from electroencephalogram (EEG) can be helpful in developing an in-car brain computer interface (BCI) systems for intelligent cars. This could be beneficial in enhancing the quality of interaction between the driver and the car to provide the response of the intelligent cars in line with driver’s intention. The experimental protocol is a variant of the contingent negative variation (CNV) paradigm with Go and No-go conditions in driving framework. Offline results using QDA and LDA classifiers show the feasibility of recognizing these signals in single-trial. This information can be exploited by in-car BCI systems that monitor the driver’s brain state. Experimental setup in the research is shown in figure 4.  The summary of this section and the papers located is found in Table 3.


The author propose a new approach, Over-complete Spectral Regression, that combines several potentially advantageous attributes and empirically demonstrate its superior performance on these data compared to the ten other CSA methods tested. Author discusses results from computational, neuroscience and experimentation points of view [36]. The summary of the methods and results in determining stress, depression and workload level are shown in Table 4.



The k-NN classifier was used to classify stress features due to its robustness and simplest supervised machine learning algorithm [37] [38]. The stress index is created based on the stress features in EEG signals and classification of the features. Researchers had introduced various methods to identify stress features from EEG signal. For instance [39] had identified and confirmed the stress features using k-NN classifier with 88.9% accuracy. Among the stress features were Spectral Centroid, Energy ratio, Bandwidth, Statistical methods and Zero Crossing rate [40] [ 41].


The author in [42] used Kernel Density Estimation (KDE) to extract stress feature from EEG signal. Sample entropy is used to detect the difference in EEG signals due stress and fatigue condition [43]. The combination features from time-domain (cross-correlation) and frequency-domain (power spectral density) generate high classification accuracy at 97% using k-NN [44].


4.3. Analysis of other Brain behaviour using EEG signals based on Training and Sensing

Recent sensor technology and analysis advances in signal processing and machine learning make it possible to  noninvasively monitor brain signals and derive from them useful aspects of a person‘s cognitive state in near real time. It is now becoming feasible to integrate this technology into real-world, real-time systems to enhance human-machine interaction across a wide range of application domains including clinical, industrial, military and gaming. 


The analysis of EEG behaviour can be done using Neurofeedback training analysis. Neurofeedback (NF) training has revealed it’s therapeutically effects to treat a variety of neurological and psychological disorders, and has demonstrated its feasibility to improve certain cognitive aptitudes in healthy users.  A research [45] presents a NF training aimed at improving working memory performance in healthy users by the enhancement of upper alpha band.


Figure 5. Increase in the working memory test score (final minus initial scores) for the control and the NF training group. On each box, the central mark is the median, the edges of the box are the   25th and 75th   percents.



Insight in Resting-state Brain Activity can be done using Functional Network Analysis. Insight occurs when problem solutions arise suddenly and is associated with an ‘‘Aha!’’ experience. In a work [46], author mainly focused on gamma-band (30 – 40 Hz) activity, as it has been predominantly implicated in a wide variety of cognitive processes, and more importantly associated with insight solutions. Motion sickness occurs when the brain receives conflicting sensory information from body, inner ear and eyes. In some cases, a decreased ability to actively control the body’s postural motion also causes motion sickness.


Figure 6.  Insight-related gamma-band functional network in resting-state. Each node refers to an EEG electrode while each    line      corresponds to the channel-pair whose coherences differed significantly high-insight and low-insight


A good approach is From EEG data, features associated with left motor brain area, parietal brain area and occipital midline brain area which predicted MSL were extracted by an optimal classifier implemented by an Inheritable Bi-objective Combinatorial Genetic Algorithm (IBCGA) with support vector machine [47]. The system flow chart proposed in [47] is shown in figure 7. The summary of the methods and results in analysis other brain behaviour using EEG signals based on training and sensing are also shown in Table 5.


Figure 7. System flowchart of the proposed motion sickness assessment system in [47]

Table 2.  Summary of number of features contained in the selected best feature set, and the estimation accuracy of each feature set








Number of selected features







Training Accuracy







Test Accuracy








4.4. Divergence thinking based several Neuroimaging studies on EEG signals

EEG is a non-invasive measure of electrical brain activity. It is a record of electric field potentials, represented as changes in potential difference between different points on the scalp, which arise primarily from excitatory and inhibitory postsynaptic potentials. Although the EEG signal has been used for decades to limn the neurophysiological changes that accompany mental processes, it was not used to study aspects of creativity until the late 1990s. Several EEG parameters are relevant to this review. EEG data are reported in frequency ranges. At the low end of the scale is delta activity, which is a regular, low-amplitude wave of 1–5Hz. This frequency band reflects a low neuronal firing rate and is mostly associated with deep sleep. Theta activity is a medium-amplitude, medium-frequency rhythm of 5–8Hz. A person exhibiting this rhythm reports feeling drowsy. Alpha activity is a fairly regular pattern between 8 and 12Hz. The alpha band is prominent when a person is minimally aroused—awake but relaxed. Beta activity, which is an irregular pattern between 12 and 30Hz, occurs mostly during alertness and active thinking. Finally, there is the gamma rhythm, which represents oscillations around the 40Hz mark that are associated with the binding of perceptual information. Power changes in time locked events are known as event-related synchronization (ERS) and event-related desynchronization (ERD) [9]. Synchrony is used to indicate synchronous activity between pairs of electrodes that is independent of spectral power or amplitude.


Event related potential (ERP) is EEG recorded in response to external stimuli. Stimulus locked ERPs are usually much smaller in amplitude than EEG and are described in terms of their characteristic scalp distribution, polarity, and latency. The findings from different studies are difficult to compare because (a) Investigators have used a host of divergent thinking tests, including original measures that have not been replicated and for which standardization and validation data are lacking; (b) These divergent thinking tasks, as diverse as they already are, are then compared with a number of different control conditions; and (c) Only a small fraction of the studies in this category use high-density EEG. These limitations should be kept in mind as conclusions are drawn from these data.  The data are summarized, for the sake of consistency, according to categories dominant in the EEG literature. This yields three main themes: (a) laterality, (b) changes to the alpha band, and (c) changes to all other frequencies.


The idea that creativity is a function, primarily or exclusively, of the right brain is surely the most popular theory on the neural basis of creativity in the wider public. However, duly sharpened versions of this idea have been seriously entertained in neuroscience as well [48] [49].  EEG studies on divergent thinking do not confirm this contention. There are a few studies that can be recruited to support a special role for the right hemisphere in divergent thinking [50] [51] [52]. It should be noted, however, that the authors responsible for three of the seven articles cited above do not themselves interpret their data in this manner [50], and they do not believe the data should be used for this purpose [50].  Whereas only one study could possibly be drafted to claim the opposite, a left-brain theory of creativity as it were, the majority of the investigations (14 in total), including those from authors cited above for possibly supporting right-brain dominance, lend no support to the In Sum the EEG data on divergent thinking fail to substantiate the notion of lateralization in creativity for right brain theory [53] [54] [55].


In sum, the EEG data on divergent thinking fail to substantiate the notation of lateralization in creativity for either cerebral hemisphere. Given this overall hit-and-miss pattern of EEG results for the divergent thinking paradigm, delving deeply into the possible functional meaning of creativity-related EEG changes is challenging, to say the least. Anyone wishing to attempt this speculative exercise would have to, as a first measure, explain away most of the existing evidence, as there are as many, if not more, data against any position one cares to take. 


Alternatively, a more in-depth understanding of the cognitive processes underlying a particular divergent thinking task could go a long way in disentangling the contradictory data. The real hope for the future of this endeavour undoubtedly lies in such an approach



Table 3. Summary of methods and results for method described in section 4.1




Tamanna Tabassum Khan Munia, (2012)

Mental States Estimation with the Variation of Physiological Signals

pronounce the consequences of mental circumstances due to the difference of electrocardiogram, electroencephalogram and blood pressure using BIOPAC system

Aurobinda Routray, (2013)

Classification of brain states using principal components analysis of cortical EEG synchronization and hmm

Synchronization Likelihood (SL) to measure the synchronization or integration between the lobes.

Bilal Fadlallah, (2012)

Quantifying Cognitive State From EEG Using Dependence Measures

Automatically discriminate face processing from processing of a simple control stimulus based on processed EEGs in an online fashion with high temporal resolution using measures of statistical dependence applied on steady-state visual evoked potentials.

Nabaraj Dahal, (2011)

Modelling of Cognition using EEG: A Review and a New Approach

Model cognitive functions by relating the varying event related potential, brain waves, spectral density and latency in EEG outcomes are then related with the stimuli features to predict the cognitive state of mind.

Takayuki Koike, (2011)

Effect of odorant presentation on changes in cognitive interference and brain activity during counting Stroop task

Investigated how intermittent odorant presentation during cognitive task contributes to

 cognitive function and statement of psychological loadings

Norizam Sulaiman, (2011)

Development of EEG-Based Stress Index

The combination of electroencephalogram (EEG) power spectrum ratio and Spectral Centroids techniques to extract unique features for human stress from EEG signals.

Zahra Khaliliardali, (2012)

Detection of Anticipatory Brain Potentials during Car Driving

Anticipation as the cognitive state leading to specific actions during car driving. The experimental protocol is a variant of the contingent negative variation (CNV) paradigm with Go and No-go conditions in driving framework.


Table 4. Summary of methods and results for techniques described in section 4.2., and 4.3.




H. Hinrikus et al. (2009)  

Relative difference in EEG Power Spectrum at two frequency bands.

Procedure index for depression – SASI (EEG Spectral Asymmetry Index).

Nassef et al. (2009)

Ratio of Theta power at Frontal area with Alpha power at Parietal site

Produce Task Load Index. 

R.K. Sinha, (2001)

EEG Power Spectra and Neural Network

Produce 3 indexes; Acute, Chronic and Normal

H.C Shih and S.H Fang, (2008)

Use Heart Rate Interval and PPG

Produce Physiology Index (PI)

Handri et al.(20080

Use EEG, ECG and Skin Temperature and k-NN Classification

Produce 2 level of Mental Stress (Low and High)

K.S. Park et al.(2011)

Use EEG Relative Power of all frequency bands

Produce index for Fear, Sadness, Peace and Happy.

M.S. Avidan et al. (2008)

Use EEG Power Spectrum

Produce Bi-spectral Index (BIS) with scale from 0 – 100.

M. Teplan. (2006)

Use Dynamic Bayesian Network (DBN) 

Produce model to estimate human stress level

Y. Tran et al, (2007)

Use EEG Power Spectrum and Entropy

Produce Indicator for fatigue and stress based on Entropy

Carlos Escolano et al. (2011)

EEG-based Upper Alpha Neurofeedback Training Improves Working Memory Performance

Presents a Nero-feedback training aimed at improving working memory performance in healthy users by the enhancement of upper alpha band.

John Kounios et al. (2011)

Functional Network Analysis of Insight in Resting-state Brain Activity

Work focused on gamma-band (30 – 40 Hz) activity, as it has been predominantly implicated in a wide variety of cognitive processes, and more importantly associated with insight solutions.

Chin-Teng Lin et al. (2012)

Motion Sickness  Estimation System

Predicted MSL were extracted by an optimal classifier implemented by an inheritable bi-objective combinatorial genetic algorithm (IBCGA) with support vector machine.


Table 5.  Summary of methods and results for techniques described in section 4.4.



Type of Creativity Test

Type of Design

     Main finding                       

Danko et al. (2009)

DTT: a creative task (overcoming a stereotype); CTT: a memory task

Between-task comparison and task vs. rest

The creative task produced a marked increase in EEG power in the Beta 2 and gamma bands. Induction had a much stronger effect on the cortex than the creative task.

Razumnikova et al. (2009)

DTT: to give original solution to one figural and to one verbal task; CTT: same tasks but S was to give any solution

Between-task comparison and task vs. rest

There was higher activation in RH, but that effect was independent of gender, test, and creative instruction. Desynchronization of Alpha 1, 2 and Beta 2 rhythms. Some changes to specific frequency were task dependent

Fink et al.


DDT: (a) AUT, (b) invent names belonging to abbreviations; CTT: (c) Think of characteristics of normal objects, (d) WE Task

Between-task comparison and task vs. rest

For AUT, there was strong alpha ERS in frontal regions, and high originality was associated with alpha ERS in posterior brain regions, especially in RH. Low originality showed no hemispheric differences. Creative cognition was associated with frontal alpha synchronization.


Bazanova and Aftanas, (2008)

DDT: TTCT nonverbal

Within-task comparison and task vs. rest

The maximum alpha activity peak frequency was not significantly correlated with creativity score. Originality showed a trend with lowest values for individual alpha peak frequencies.

Grabner et al. (2007)

DTT: unusual situations that need an explanation taken from the TTCT

Within-task comparison and task vs. rest

More original ideas elicited a stronger alpha ERS and higher phase coupling in RH. Originality was indexed in lower alpha.

Shemyakina and Danko, (2007)

DTT: possible definitions differing in meaning of (a) emotionally positive, (b) negative, and (c) neutral nouns

Within-task comparison and task vs. rest

The creative task without emotional induction led to a local decrease in beta power in the left frontotemporal area and a coherence decrease in most cortical zones. Emotional induction had a much stronger effect on the state of the cortex than the creative task did.

Razumnikova, (2007)


Within-task comparison and task vs. rest

The RAT showed higher power and coherence in beta, increased theta power at frontal sites, and increased alpha ERDs over posterior and prefrontal leads. Originality was positively correlated with more coherence focused in the fronto-parietal regions of both hemispheres in beta and in left parietotemporal loci for Alpha

Fink and Neubauer, (2006b)

DTT: Unusual problems requiring creative solutions— (a) Insight task, (b) utopian situations, and (c) the AUT; CTT: WE Task

Covariate (2 weeks DTT training), between-task comparison, and task vs. rest

The training group had (a) higher originality sores for DTT and worse scores for the CTT, (b) higher alpha ERS, and (c) higher power increases in anterior cortices and in right temporal and parietal sites. Divergent thinking was linked to low alpha power, which could reflect hypofrontality needed to produce novel ideas.

Krug et al. (2003)

DTT: unique consequences of a hypothetical situation; CTT: logical thinking and mental arithmetic

Covariate (menopausal women: placebo estrogen and testosterone), between-task comparison, and task vs. rest

For placebo, CTT showed lower alpha activity, and DTT showed lower beta power at central and parietal leads. Estrogen impaired DTT, enhanced CTT, and was accompanied by less dimensional complexity over right posterior regions. Testosterone effects were opposite: They increased performance and dimensional complexity. Estrogen induced a shift from a divergent to a convergent mode of processing.



NOTE: Terminologies description

ACC = Anterior cingulated cortex;  AUT = Alternative Uses Task; BA = Brodmann’s area; BOLD = Blood oxygenation level dependent; CBF = Cerebral blood flow; CFT = Creative Functioning Test; CTT = Convergent Thinking Task; D2 = Dopamine 2 receptor; DLPFC = Dorsolateral prefrontal cortex; DTI = Diffusion tensor imaging;  DTT = Divergent Thinking Task; EEG = Electroencephalography;  ERD = Event-related desynchronization; ERS = Event-related synchronization; HC = High-creative group; LC = Low-creative group;  LFT = Letter Fluency Test; LH = left hemisphere;  NAA = N-acetylaspartate;  NIRS = Near-infrared spectroscopy; PFC = Prefrontal cortex; RAT = Remote Associates Test; RH = Right hemisphere; S = Subjects; SAT = Simple Associates Task; SMA = Supplementary motor area; SPECT = Single photon emission computed tomography; SPQ = Schizotypal Personality Questionnaire; TAT = Thematic Apperception Test; TTCT = Torrance Test of Creative Thinking; VLPFC = Ventrolateral prefrontal cortex; WAIS = Wechsler Adult Intelligence Scale;  WE = Word End.



In conclusion, varying parameters extracted from EEG as a response to different stimuli could be used as a metric in the study of cognitive modelling research. The collaborative work with neurologist, mathematician, psychologist, engineers can produce more refined model in the future which can explain how people organize knowledge and produce intelligent behaviour. As the research continues, we hope, a more elaborative model could emerge explaining all the details of the way we perceive think about and make decision in this world. Brain functioning is based on the firing of millions of neurons in a synchrony. Analysing and explaining their complex signalling, dynamics, network loops, transmission pathways stands as a major challenge in the study. Apart from this, the random spiking times by individual neuron is continually adding noise in the brain signalling and we considering it as a challenge for the research.



1.       G. Buzsaki, C.A. Anastassiou, and C. Koch, “The origin of extracellular fields and currents - EEG, ECG, LFP and spikes”, Nature Reviews Neuroscience, Vol. 13, pp. 407-420, 2012.

2.       P.K. Sinha, and T. Koshy, “Monitoring Devices for Measuring the Depth of Anaesthesia - An Overview”, Indian Journal of Anaesthesia, Vol. 51, No. 5, pp. 365-38, 2007.

3.       Sheikhani, H. Behnam, M.R. Mohammadi, M. Noroozian, P.
Golabi, “Connectivity analysis of quantitative Electroencephalogram background activity in Autism disorders with short time Fourier transform and Coherence values”, Congress on Image and Signal Processing, pp. 207-212, 2008.

4.       S.J. Lupien, F. Maheu, M.T.A Fiocco, T.E. Schramek, “The eects of stress and stress hormones on human cognition: Implications for the field of brain and cognition”, Brain and Cognition, Vol. 65, No. 3, pp. 209–237, 2007.

5.       Guler, E. Derya, and Ubeyli.  “Adaptive neuro-fuzzy inference system for classification of EEG signals using wavelet coefficients”, Journal of Neuroscience Methods, Vol. 148, No. 2, pp. 113–121, 2005.

6.       A.S. AlMejrad, “Human Emotions Detection using Brain Wave Signals: A Challenging”, European Journal of Scientific Research, Vol. 44, No.4, pp. 640-659, 2010.

7.       S. Makeiga, K. Gramanna, T.P. Junga, T.J. Sejnowskib, H. Poiznerb,  “Linking brain, mind and behaviour”, International Journal of Psychophysiology, Vol. 73, No. 2, pp. 95–100, 2009.

8.       K.J. Blinowska, “Methods for localization of time-frequency specific activity and estimation of information transfer in brain.” International Journal of Bio electromagnetism, Vol. 10, No.1, pp. 2 -16, 2008.

9.       B. Graimann, B. Allison, and G. Pfurtscheller, “Brain–Computer Interfaces: A Gentle Introduction”, Brain-Computer Interfaces, the Frontiers Collection, pp. 1-27, 2010.

10.     Subhani Rauf and Xia Likun and Malik, A. Saeed, “EEG SIGNALS TO MEASURE MENTAL STRESS”, In preceding of 2nd International Conference on Behavioral, Cognitive and Psychological Sciences, pp. 6-10, 2011.

11.     N.F. Gulera, E.D. Ubeylib, and N. Gulera, “Recurrent neural networks employing Lyapunov exponents for EEG signals classification”, Expert Systems with Applications, Vol. 29, No. 3, pp. 506–514, 2005.

12.     N.O.S. Ba-Karait, S.M. Shamsuddin, and R. Sudirman,  “Swarm Negative Selection Algorithm for Electroencephalogram Signals Classification”, Journal of Computer Science, Vol. 5, No. 12, pp. 995-1002, 2009.

13.     J. Olsen, M.G. Baker, T.  Freund, M.D. Luca, J. Mendlewicz, I. Ragan, and M. Westphal, “Consensus document on European brain research”, Journal of Neurol Neurosurg Psychiatry, Vol. 77, pp. i1- i49, 2006.

14.     R. Harikumar, M. Balasubramani, T. Vijayakumar, “Performance Analysis of Patient Specific Epilepsy Risk Level Classifications from EEG Signals Using Two Tier Hybrid (Fuzzy, Soft Decision Trees Models and MLP Neural Networks) Classifiers”, International Journal of Soft Computing and Engineering, Vol. 2, No2, 2012.

15.     M.D. Serruyaa, M.J. Kahana, “Techniques and devices to restore cognition”, Behavioural Brain Research, Vol. 192, No. 2, pp. 149–165, 2008.

16.     T. Banaschewski, D. Brandeis,  “Annotation: What electrical brain activity tells us about brain function that other techniques cannot tell us - a child psychiatric perspective”, Journal of Child Psychology and Psychiatry, Vol. 48, No. 5,  pp 415–435, 2007.

17.     B.L. Liang Lu, J. Shin, M. Ichikawa, “Massively Parallel Classification of EEG Signals Using Min-Max Modular Neural Networks”, IEEE Transactions on Biomedical Engineering, Vol. 51 , No.3, pp. 551- 558, 2004. 

18.     K.  Lee,  “Evaluation of Attention and Relaxation Levels of Archers in Shooting Process using Brain Wave Signal Analysis Algorithms”, Korean Journal of the Science of Emotion and Sensibility, Vol. 12, No 3, pp.341-350, 2009.

19.     M.K. Ahirwal, N. Dlondhe, “Power Spectrum Analysis of EEG Signals for Estimating Visual Attention”, International Journal of Computer Applications, Vol. 42, No.15, pp. 22-25, 2012.

20.     P.  Sauseng, W. Klimesch, W. Gruber,  M. Doppelmayr, W. Stadler,  M. Schabus, “The Interplay between Theta and Alpha Oscillations in the Human Electroencephalogram Reflects the Transfer of Information between Memory Systems," Neuroscience Letters, vol. 324, no. 2, pp. 121-124, 2002.

21.     C. Herrmann, D. Sensowski, S. Rottger, “Phase-Locking and Amplitude Modulations of EEG Alpha: Two Measures Reflect Different Cognitive Processes in a Working Memory Task." Experimental Psychology, vol. 51, no. 4, p. 311, 2004.

22.     J. Coan, J. Allen,   “Frontal EEG asymmetry as a Moderator and Mediator of Emotion” Biological Psychology, Vol. 67, no. 1-2, pp. 7- 50, 2004.

23.     E.H. Jones, “Clarifying the Emotive Functions of Asymmetrical Frontal Cortical Activity," Psychophysiology, vol. 40, no. 6, pp. 838- 848, 2003.

24.     J. Minnix, J. Kline, G. Blackhart, J. Pettit, M. Perez, T. Joiner, "Relative Left-Frontal Activity is Associated with Increased Depression in High Reassurance-Seekers," Biological Psychology, vol. 67, no. 1-2, pp. 145-155, 2004.

25.     B. Jap, S. Lal,  Fischer, E. Bekiaris, "Using EEG Spectral Components to Assess Algorithms for Detecting Fatigue," Expert Systems with Applications, vol. 36, no. 2, p p. 2352-2359, 2009.

26.     H. Eoh, M. Chung, S. Kim,  “Electroencephalographic Study of Drowsiness in Simulated Driving with Sleep Deprivation," International Journal of Industrial Ergonomics, vol. 35, no. 4, pp. 307-320, 2005.

27.     M. Tlauka, C.R. Clark, P. Liu, Conway.  “Encoding Modality and Spatial Memory Retrieval," Brain and Cognition, vol. 70, no. 1, pp. 116-122, 2009.

28.     J. Allen, J. Coan, M. Nazarian,  “Issues and Assumptions on the Road from Raw Signals to Metrics of Frontal EEG Asymmetry in Emotion," Biological Psychology, vol. 67, no. 1-2, pp.183-218, 2004.

29.     T.T.K. Munia, “Mental States Estimation with the Variation of Physiological Signals”, IEEE/OSA/IAPR International Conference on Infonnatics, Electronics and Vision, 2012.

30.     Routray, “Classification of brain states using principal components analysis of cortical eeg synchronization and hmm”, IEEE, 2013.

31.     Fadlallah, “Quantifying Cognitive State from EEG Using Dependence Measures”, IEEE Transactions on Biomedical Engineering, Vol. 59, NO. 10, 2012.

32.     N. Dahal, School of Electrical and Information Engineering University of South Australia Adelaide, SA, Australia, “Modeling of Cognition using EEG: A Review and a New Approach”, IEEE, 2011.

33.     T.  Koike, “Effect of odorant presentation on changes in cognitive interference and brain activity during counting Stroop task”, International Conference on Biometrics and Kansei Engineering, 2011.

34.     N. Sulaiman, “Development of EEG-Based Stress Index”, International Conference on Biomedical Engineering , 27-28 February, Penang, 2012.

35.     Z. Khaliliardali, R. Chavarriaga, L.A. Gheorghe, J.R. Mill´an, “Detection of Anticipatory Brain Potentials during Car Driving”, 34th Annual International Conference of the IEEE EMBS San Diego, California USA, 28 August - 1 September, 2012.

36.     C.A. Kothe, S. Makeig, “Estimation of Task Workload from EEG Data: New and Current Tools and Perspectives”, 33rd Annual International Conference of the IEEE EMBS Boston, Massachusetts USA, August 30 - September 3, 2011.

37.     S. Handri, S. Nomura, K. Nakamura, “Detecting Stress Based on the Schedule of an Intermittent Mental Workload Using Physiological Sensor,” in Proc. 11th International Conference on Humans and Computers, pp. 123-126, 2008.

38.     M. Murugappan, N. Ramachandran, Y. Sazali,  “Classification of human emotion from EEG using discrete wavelet transform,” Journal of Biomedical Science and Engineering, vol. 3, pp. 390-396, 2010.

39.     N. Sulaiman, M.N. Taib, S.A.M. Aris, N.H.A. Hamid, S. Lias, Z.H.Murat,  “EEG-based Stress Features Using Spectral Centroids Technique and k-Nearest Neighbor Classifier,” in Proc. 13th International Conference on Modelling and Simulation (UKSim), pp. 69-74, 2011.

40.     S. Sun, C.  Zhang, “Assessing Features for Electroencephalographic Signal Categorization,” in Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 417-420, 2005.

41.     M.L. Massar, M. Fickus, E. Bryan, D.T. Petkie, A.J.T.Jr, “Fast Computation of Spectral Centroids,” Journal of Advanced Computing Mathematics, June, 2010.

42.     K.S. Rahnuma,  A. Wahab,  N. Kamaruddin, H. Majid,   “EEG Analysis for Understanding Stress Based on Affective Model Basis Function,” in Proc. 15th IEEE International Symposium on Consumer Electronics, pp. 592-597, 2011.

43.     Y.R. Tran, A. Thuraisingham, N. Wijesuriya, H.T. Nguyen, A. Craig,  “Detecting Neural Change during Stress and Fatigue effectively: a comparison of Spectral Analysis and Sample Entropy,” in Proc. IEEE EMBS Conference on Neural Engineering, pp. 350-353, 2007.

44.     Z. Iscan, Z. Dokur, T. Demilrap,  “Using Classification of Electroencephalogram signals with combined time and frequency features”, International Journal of Expert System with Application, vol. 38, pp. 10499-10505, 2011.

45.     C. Escolano, M. Aguilar, J. Minguez, “EEG-based Upper Alpha Neurofeedback Training Improves Working Memory Performance”, 33rd Annual International Conference of the IEEE EMBS Boston, Massachusetts USA, August 30 - September 3, 2011.

46.     J. Kounios, M. Beeman, H. Liang, Senior Member, “Functional Network Analysis of Insight in Resting-state Brain Activity”, Fourth International Workshop on Advanced Computational Intelligence Wuhan, Hubei, China; October 19-21, 2011.

47.     C.T. Lin, “Motion Sickness Estimation System”, WCCI, IEEE World Congress on Computational Intelligence June, 10-15, Brisbane, Australia, 2012.

48.     M. Jung-Beeman, E. Bowden, J. Haberman, J. Frymiare, S. Aramber-Liu, R. Greenblatt, J. Kounios, Neural activity when people solve problems with insight. PloS Biology, 2, pp. 500–510, 2004.

49.     J. Kounios, J. Fleck, D. Green, L. Payne, J. Stevenson, E. Bowden,  M. Jung-Beeman,  The origins of insight in resting-state brain activity. Neuropsychologia, 46, 281–291, 2008.

50.     Fink, R. Grabner, M. Benedek, A.  Neubauer, Short communication: Divergent thinking training is related to frontal electroen-cephalogram alpha synchronization. European Journal of Neuroscience, 23, 2241–2246, 2006. 

51.     R. Grabner, A. Fink, A. Neubauer, Brain correlates of self-rated originality of ideas: Evidence from event-related power and phase locking changes in the EEG. Behavioral Neuroscience, 121, 224–230, 2007.

52.     N. Jausˇovec, K. Jausˇovec, EEG activity during the performance of complex mental problems. International Journal of Psycho- physiology, 36, 73–88, 2000.

53.     O.M. Bazanova, L.I. Aftanas, Individual measures of electro- encephalogram alpha activity and non-verbal creativity. Neurosciences and Behavioral Physiology, 42, 227–235, 2008.

54.     S. Danko, N.V. Shemyakina, Z.V. Nagornova, M. Starchenko, M., Comparison of the effects of the subjective e complexity and verbal creativity on the EEG spectral power parameters. Human Physiology, 35, 381–383, 2009.

55.     R. Krug, M.  Mölle, C.  Dodt, H.  Fehm, J. Born, Acute influences of estrogen and testosterone on divergent and convergent thinking in postmenopausal women. Neuropsychopharmacology, 28, 1538–1545, 2003.




Received on 22.10.2013       Modified on 18.12.2013

Accepted on 14.01.2014      © RJPT All right reserved

Research J. Pharm. and Tech. 7(2): Feb. 2014; Page 238-247