Assessing the Effect on Cognitive Workload Index,
EEG band ratio’s,band frequencies using bandpower
& implementing ML Classification
Abstract— Cognitive resource theory combines individual
differences in attentional resources and task demands to
predict variance in task performance. Mind wandering, on the
other hand, refers to the occurrence of thoughts that are both
stimulus independent and task unrelated. In this paper, the
dataset collected from publicly available online sources and
consists of 36 subjects performing mental arithmetic tasks
(series of subtraction). Based on their performance on the
subtraction task subjects are divided into two groups: “Bad”
and “Good”. Each group is divided into two parts: before the
mental arithmetic task of 180 seconds (resting state), and while
doing the mental arithmetic task of 60 seconds (response). The
band power calculated using wavedec function used to assess
the performance of subjects by analyzing band ratio’s,
cognitive workload index, individual band frequencies,
prefrontal electrode. Binary classification with various
classifier used, three classifiers obtained the same results:
SVM, Gaussian NB, Logistic Regression 73%.
Keywords— Mind Wandering, CWI (Cognitive Workload
index), Binary Classification, SVM, Gaussian NB (Naive Bayes),
Logistic Regression.
I. INTRODUCTION
The human brain could either be consumed by memories
from the past or preoccupied with the long term, whereas
paying attention on the present is critical for healthy lifestyle.
Even when at rest, the brain is constantly busy and interrupts
itself to attend to a specific task. The portion of mental effort
expended by everyone in response to a particular cognitive
task is referred to as the cognitive workload [1].
Cognitive workload based on electroencephalography
(EEG) is an important marker of brain activity for workload
analysis (mental effort). Cognitive workload is a
measurement of the load placed on memory and other
executive functions that are used to demonstrate cognitive
abilities. Memory load was investigated using EEG signals
during cognitive task inducing seven levels of workload
using arithmetic tasks [2,3]. The variations in brain lobes for
different load levels are investigated [4].
The total electrical oscillations in layer potentials created
by the interconnection of the principal inhibitory and
excitatory neurons are represented by the EEG signal [5,6].
Mind wandering (MW) episodes, like concern, are
characterised by the appearance of task-unrelated feelings
and ideas that divert attention away from the current task [7].
Mind Wandering can happen when you're doing something
else, and it shows up as you think about something else [8].
Prospection and future planning [9], creativity [10], and
mental breaks, which can help you get out of a bad mood
[11] have all been linked to Mind Wandering. Others have
consistently viewed Mind Wandering as a condition of
diminished working memory and attentional control
[12,13]and as a predictor of performance errors, in addition
to its relationship with these more positive processes. Mind
Wandering has also been linked to a reduction in attention
and focus [14,15]. Increase in EEG theta band power and
decrease in EEG beta band power were linked to a state of
Mind Wandering in a proof-of-concept study,[16] reported
that a condition of Mind Wandering was associated with high
EEG theta band power and low EEG beta band power. When
you're awake and close your eyes, your alpha waves are
higher than when you're awake and open your eyes [17].
Power has been found to predict various aspects of
cognitive performance on other tasks. Many such research
has found that good memory performance is linked to higher
resting power [18], and that a higher ERD is correlated with
better effectiveness on a semantic search task [19]. External
stimulus, such as task-relevant visual pictures or warning
signs, suppresses alpha waves during cognitive activities
[20]. During brief waiting times between task attempts
[21,22] and mental imagery, which demands inward-directed
thought [23], alpha waves are increased. Decreases in
perceptual discrimination have been linked to enhanced alpha
activity prior to an incoming stimulus [24]. The task-related
power variations show that alpha activity is sensitive to
cognitive demands: it is repressed in response to sensory
stimulation and boosted during periods of no stimulation
[25].
The role of alpha power in the suppression of sensory
processes to protect thinking from continuing distraction has
been demonstrated in numerous studies [26,27], and it is a
promising metric to describe Mind Wandering [28].
Although some studies have observed the contrary tendency,
our findings are consistent with multiple earlier findings that
connect greater alpha power with Mind Wandering in frontal,
occipital, and posterior parts of the brain [29,30,31]. When
compared to episodes of focused attention, a high in theta
and a low in beta power have been reported during Mind
Wandering [32].
In [33] researchers have calculated different EEG band
ratio’s for evaluating the cognitive load indices of the
subjects in pre and post meditation.
The key contribution of the proposed work is to analyse
the band ratio’s, cognitive load index, individual band
frequencies, prefrontal electrode of the subjects, and finally
applied the binary classification with different classifier like
SVM, KNN, Decision Tree, Random Forest, Gaussian NB,
Logistic Regression. The paper is arranged in four sections.
Section 2 includes materials and methods, Section 3 includes
experiments and results, Section 4 includes conclusion.
II. MATERIALS AND METHODS
a) Recording and Selection of the EEG signals
The EEG signals were recorded from Neurocom
Monopolar EEG 23-channel system (Ukraine, XAI-
MEDICA). All 23 electrodes are mounted on the scalp of the
subject according to the International 10-20 scheme. The
electrodes mounted on the scalp recorded the signals from
different parts of the brain, where it is divided into different
lobes like frontal, temporal, parietal, occipital. The electrodes
used for measuring frontal, temporal, parietal, occipital and
central of the brain are given [33]. The interconnected ear
reference electrodes are used to reference all electrodes. The
inter-electrode impedance was kept below 5k, and the
sampling rate was 500Hz. A 0.5Hz cut-off frequency HPF, a
45Hz cut-off frequency LPF, and a 50Hz power line notch
filter is used.
Recording of each EEG signals includes artifacts-free
EEG segments of 180 sec (resting state) and 60 sec (mental
counting). Out of 66 participants 30 were removed from the
database due to imprudent artifacts present with eyes and
muscle. So overall we found 36 subjects best.
b) Attribute of Subjects Participation in the protocol
The subject’s participation in the protocol includes series
of subtraction or called with Number of subtractions. It
includes the process of first taking the difference between the
four-digit number and the result of subtracting, divided by
the subtrahend. More information can be found in the paper
[34].

III. EXPRIMENTS AND RESULTS

1) Individual Band Frequencies:

The nature of EEG signal has a time varying property, in

order to convert the time domain signal to frequency domain
we used wavelet transform. Wavelet transform can provide
the time and frequency signals simultaneously. By using
wavedec function in MATLAB, given the sampling
frequency of 500Hz and obtained 5 sets of frequency bands
like: delta, theta, alpha, beta & gamma. The obtained
frequency bands are analysed to acquire the band power.
The subjects are divided into 2 groups: “Good” & “Bad”,
the task involves before mental arithmetic task, while doing
mental arithmetic task. Individual band frequencies used to
assess the performance of the subject by considering the task
in both, before and while doing.

Fig. 1. Analysis of “Good” subjects performing before and while doing

task(a)Delta band, (b)Theta band, (c)Alpha band, (d)Beta band, (e)Gamma
band.

Fig. 2. Analysis of “Bad” subjects performing before and while doing
task(a)Delta band, (b)Theta band, (c)Alpha band, (d)Beta band, (e)Gamma
band.
*Note: All the graphs in this paper are log scaled(x=y) for easy prediction
and red line used to show the separation among the performance of the task.
The subjects above the line considered as while doing
task of 60 sec (1 min-response), while subjects below the line
considered as before the task of 180 sec (3min-resting state).
We compared both groups (i.e., "Good" and "Bad") and
cannot conclude for Delta band in Fig. 1.(a) and Fig. 2.(a).
When individual groups performance wise, we found that
"Good" participants were more (line below) and felt sleepy in
the moments before the task (resting state), but "Bad"
subjects were less (line below) and did not feel drowsy in the
moments before the task (resting state). In while doing task
(line above) of both groups, we found less number of “Good”
subjects (did not feel sleep) and more number of “Bad”
subjects (due to sleepy).
In Fig 1(b), Fig 2(b) shows the response of the “Good”
and “Bad” subjects for Theta band. When individual group
performance wise we found an equal number (line above and
below) of “Good” subjects, which is due to relaxation in both
the task. Whereas less number (line above) of “Bad” subjects
due to difficulty in while doing the task and more number
(line above) due to relaxation in before the task (resting
state).
In Fig 1(c), Fig 2(c) show the response of the "Good" and
"Bad" subjects for the Alpha band, we compared both groups
(line above), more "Good" subjects and less "Bad" subjects
and vice versa for line below. Individual group performance
wise, we found more (line above) "Good" subjects due to
awake and relaxed while doing the task, and less (line above)
"Bad" subjects due to difficulty while doing the task. In line
below, we see less "Good" subjects awakened and relaxed,
and more "Bad" subjects relaxed.
We compared both groups in line above and cannot
conclude for the beta band in Fig 1(d), Fig 2(d). When
individual group performance we found more number (line
above) of "Good" and "Bad" subjects due to consciousness in
while doing task. Subjects of both groups in the line below
are less conscious in before the task.
In Fig 1(e), Fig 2(e), we compare the responses of the
"Good" and "Bad" subjects for Gamma band, and we cannot
conclude for gamma band (due to equal distribution of
subjects among line below and above). Individual group
performance-wise, we have more "Good" subjects in the line
above, indicating concentration. We are unable to reach a
conclusion for "Bad" subjects (due to equal distribution in
before and while doing task).
2) Cognitive Load indices (EEG band ratios):
EEG cognitive load indices which represent the different
EEG band ratios are analysed using mean and standard
deviation are shown in TABLE I.
The load on memory increases when task difficulty
increases. The performance enhancement in “Bad” subjects
refers to increase in alpha frequency which indicates mind
wandering, in while doing task the same alpha frequency has
decreased which shows less mind wandering, as alpha
frequency increases when eyes closed whereas decreases
when eyes opened(relaxation).The same performance of
alpha frequency is observed for “Good” subjects increased
before the task(more relaxation) and decreased while doing
task(less relaxation) due to eyes closing and opening.
The arousal index represents the excitement in the
subjects, we can see an increase in arousal index for “Bad”
subjects from before task to while doing task, while decrease
in arousal index from before to while doing task for “Good”
subjects, this represents that “Good” subjects showed less
excitement when compared to “Bad” subjects.
The neural activity represents improvement in cognitive
skills, increase in neural activity for “Bad” subjects from
before to while doing, while decrease in before to while
doing task for “Good” subjects, shows “Good” subjects
stressed to task and showed decrement.
The engagement activity, if increases in ‘β’ show alert &
focused, while decrease in ‘θ’ show cognitive load, from
above table we see increase in engagement before task for
“Bad” subjects and decrease in while doing task for “Good”
subjects, decrease in ‘θ’ represents the cognitive load had
effect on “Good” subjects (increased β, less θ). Load index
represents the stress, increment observed in both “Bad” &
“Good” subjects, stress is observed in while doing task.
There is a decrease in alertness activity from before task to
while doing task in “Bad” subjects, whereas decrease to
increase in “Good” subjects, which showed alertness in while
doing task.
The CWI (Cognitive Workload Index) ratio based on
EEG band power [35] obtained using mean and standard
deviation of the band power. It shows decrease to increase in
“Bad” subjects, while increase to decrease in “Good”
subjects. Load showed effect on “Bad” subjects in while
doing the task which decreased the performance, whereas
“Good” subjects did not felt difficulty in while doing the
task.

Fig. 3. CWI for Subjects performing before, while doing task, (a) “Good”
(b) “Bad”

CWI performance for "Good" and "Bad" subjects with

band power is shown in Fig. 3.(a)(b). CWI represents the
load on subjects in which there are more (line above) "Bad"
subjects than "Good" subjects (line above). Due to the load,
the "Bad" subjects had difficulty in performing the task
(while doing task), whereas the "Good" subjects were more
conscious, concentrated, and relaxed (while doing task).
3) Prefrontal Electrode analysis
The prefrontal electrodes used to assess the effect of load
on the skull, we examined the individual data of Fp1, Fp2
electrodes for the subjects under before and while doing task
shown in Fig. 4. Various cognitive skills can be analysed but
we evaluated only cognitive load on subjects performing the
task. Band powers of Fp1 and Fp2 for “Good” and “Bad”
subjects are more in while doing task (line above), whereas
less in before the task (line below).
Fig. 4. Fp1, Fp2 for subjects performing before and while doing task, (a)
(b) “Good”, (c)(d) “Bad”
4) Binary Classification
Different methods used to obtain the binary classification
results like SVM, Knn, Decision Tree, Random Forest,
Gaussian NB, Logistic Regression. The binary classification
is done by taking mean of individual (Fp1, Fp2, F3, F4, Fz,
F7, F8, C3, C4, Cz, P3, P4, Pz, O1, O2, T3, T4, T5, T6)
electrodes & calculated overall mean of 19 electrodes for one
subject, this process repeated for all subjects and finally
obtained mean results are divided into 2 parts: 1.“Bad”
before mental arithmetic(3min) and while doing mental
arithmetic task(1min), 2.“Good” Subjects before mental
arithmetic(3min) and while doing mental arithmetic
task(1min).The obtained mean of all the subjects considered
& divided among “Bad” , “Good”. The subjects mean
merged (before, while doing mental arithmetic task of “Bad”)
and labelled as ‘0’(zero). The same repeated for “Good”
subjects and labelled as ‘1’(one). Finally testing and training [8] Mason MF, Norton MI, Van Horn JD, Wegner DM, Grafton ST,
Macrae CN. Wandering minds: the default network and stimulus-
are done and applied Different
classification techniques. Different
Classifier Outputs independent thought. Science. 2007 Jan 19;315(5810):393-5.
classifier 80
results
73
can be seen in Figure 5, SVM, 73
Gaussian
73
NB, [9] Baumeister RF, Masicampo EJ, Vohs KD. Do conscious thoughts
Logistic Regression
70 obtained
67 the 73%. cause behavior?. Annual review of psychology. 2011 Jan 10;62:331-
60 60 61.
60
[10] Baird B, Smallwood J, Mrazek MD, Kam JW, Franklin MS, Schooler
50 JW. Inspired by distraction: Mind wandering facilitates creative
Percentage

incubation. Psychological science. 2012 Oct;23(10):1117-22.

40
30
20
10
0
SVM KNN DT RT GNB LR
Classifier
Fig. 5. Different classifier outputs-SVM (Support Vector Machine), KNN
(K-Nearest Neighbor), DT (Decision Tree), RT (Random Forest Tree),
Gaussian NB (Naive Bayes), Logistic Regression.
IV. CONCLUSION
In this paper, we presented a framework in four
experiments, in first experiment we have analysed the
individual band frequencies with band power, “Good”
subjects showed better performance in while doing task of
alpha and beta band, while “Bad” subjects showed better
performance in delta and beta band, second experiment
includes the performance of EEG band ratios obtained by
mean and standard deviation of “Bad” and “Good” subjects
(before mental arithmetic(3min) and while doing mental
arithmetic task(1min)), the variations among the bands
resulted to increase and decrease in different activity indices,
third experiment we considered prefrontal electrodes to
know the effect of load on the task, fourth includes binary
classification with mean using different classifiers.
Acknowledgement:
The work was supported by funds from the All India
Council for Technical Education (AICTE), New Delhi, under
Research Promotion Scheme Grant
No.8-71/FDC/RPS(POLICY-1)/2019-20.
REFERENCES
[1] Gupta SS, Manthalkar RR, Gajre SS. Mindfulness intervention for
improving cognitive abilities using eeg signal. Biomedical Signal
Processing and Control. 2021 Sep 1;70:103072.
[2] Wang S, Gwizdka J, Chaovalitwongse WA. Using wireless EEG
signals to assess memory workload in the $ n $-back task. IEEE
Transactions on Human-Machine Systems. 2015 Sep 28;46(3):424-
35.
[3] Zarjam P, Epps J, Lovell NH. Beyond subjective self-rating: EEG
signal classification of cognitive workload. IEEE Transactions on
Autonomous Mental Development. 2015 Jun 4;7(4):301-10.
[4] Zander TO, Kothe C. Towards passive brain–computer interfaces:
applying brain–computer interface technology to human–machine
systems in general. Journal of neural engineering. 2011 Mar
24;8(2):025005.
[5] Gordon N. REVIEW Cognitive functions and epileptic activity.
Seizure. 2000 Apr 1;9(3):184-8.
[6] Nunez PL, Cutillo BA. Neocortical dynamics and human EEG
rhythms. Oxford University Press, USA; 1995.
[7] Smallwood J, Schooler JW. The restless mind. Psychological bulletin.
2006 Nov;132(6):946.
[11] Ruby FJ, Smallwood J, Engen H, Singer T. How self-generated
thought shapes mood—the relation between mind-wandering and
mood depends on the socio-temporal content of thoughts. PloS one.
2013 Oct 23;8(10):e77554.
[12] McVay JC, Kane MJ. Conducting the train of thought: working
memory capacity, goal neglect, and mind wandering in an executive-
control task. Journal of Experimental Psychology: Learning, Memory,
and Cognition. 2009 Jan;35(1):196.
[13] Unsworth N, McMillan BD. Similarities and differences between
mind-wandering and external distraction: A latent variable analysis of
lapses of attention and their relation to cognitive abilities. Acta
psychologica. 2014 Jul 1;150:14-25.
[14] Smallwood J, Nind L, O’Connor RC. When is your head at? An
exploration of the factors associated with the temporal focus of the
wandering mind. Consciousness and cognition. 2009 Mar
1;18(1):118-25.
[15] Stawarczyk D, Majerus S, Catale C, D'Argembeau A. Relationships
between mind-wandering and attentional control abilities in young
adults and adolescents. Acta psychologica. 2014 May 1;148:25-36.
[16] Braboszcz C, Delorme A. Lost in thoughts: neural markers of low
alertness during mind wandering. Neuroimage. 2011 Feb
14;54(4):3040-7.
[17] Adrian ED, Matthews BH. The Berger rhythm: potential changes
from the occipital lobes in man. Brain. 1934 Dec 1;57(4):355-85.
[18] Klimesch W, Vogt F, Doppelmayr M. Interindividual differences in
alpha and theta power reflect memory performance. Intelligence. 1999
Dec 1;27(4):347-62.
[19] Doppelmayr M, Klimesch W, Hödlmoser K, Sauseng P, Gruber W.
Intelligence related upper alpha desynchronization in a semantic
memory task. Brain Research Bulletin. 2005 Jul 30;66(2):171-7.
[20] Thut G, Nietzel A, Brandt SA, Pascual-Leone A. α-Band
electroencephalographic activity over occipital cortex indexes
visuospatial attention bias and predicts visual target detection. Journal
of Neuroscience. 2006 Sep 13;26(37):9494-502.
[21] Carp J, Compton RJ. Alpha power is influenced by performance
errors. Psychophysiology. 2009 Mar;46(2):336-43.
[22] Compton RJ, Arnstein D, Freedman G, Dainer‐Best J, Liss A.
Cognitive control in the intertrial interval: evidence from EEG alpha
power. Psychophysiology. 2011 May;48(5):583-90.
[23] Cooper NR, Croft RJ, Dominey SJ, Burgess AP, Gruzelier JH.
Paradox lost? Exploring the role of alpha oscillations during
externally vs. internally directed attention and the implications for
idling and inhibition hypotheses. International journal of
psychophysiology. 2003 Jan 1;47(1):65-74.
[24] Van Dijk H, Schoffelen JM, Oostenveld R, Jensen O. Prestimulus
oscillatory activity in the alpha band predicts visual discrimination
ability. Journal of Neuroscience. 2008 Feb 20;28(8):1816-23.
[25] Compton RJ, Gearinger D, Wild H. The wandering mind oscillates:
EEG alpha power is enhanced during moments of mind-wandering.
Cognitive, Affective, & Behavioral Neuroscience. 2019
Oct;19(5):1184-91.
[26] Benedek M. Internally directed attention in creative cognition.
[27] Klimesch W. Alpha-band oscillations, attention, and controlled access
to stored information. Trends in cognitive sciences. 2012 Dec
1;16(12):606-17.
[28] Arnau S, Löffler C, Rummel J, Hagemann D, Wascher E, Schubert
AL. Inter‐trial alpha power indicates mind wandering.
Psychophysiology. 2020 Jun;57(6):e13581.
[29] Baldwin CL, Roberts DM, Barragan D, Lee JD, Lerner N, Higgins JS.
Detecting and quantifying mind wandering during simulated driving.
Frontiers in human neuroscience. 2017 Aug 8;11:406.
[30] Macdonald JS, Mathan S, Yeung N. Trial-by-trial variations in
subjective attentional state are reflected in ongoing prestimulus EEG
alpha oscillations. Frontiers in psychology. 2011 May 10;2:82.
[31] Baird B, Smallwood J, Lutz A, Schooler JW. The decoupled mind:
mind-wandering disrupts cortical phase-locking to perceptual events.
Journal of Cognitive Neuroscience. 2014 Nov 1;26(11):2596-607.
[32] van Son D, de Rover M, De Blasio FM, van der Does W, Barry RJ,
Putman P. Electroencephalography theta/beta ratio covaries with mind
wandering and functional connectivity in the executive control
network. Annals of the New York Academy of Sciences. 2019
Sep;1452(1):52-64.
[33] Jadhav N, Manthalkar R, Joshi Y. Assessing effect of meditation on
cognitive workload using EEG signals. InSecond International
Workshop on Pattern Recognition 2017 Jun 19 (Vol. 10443, pp. 269-
273). SPIE.
[34] Zyma I, Tukaev S, Seleznov I, Kiyono K, Popov A, Chernykh M,
Shpenkov O. Electroencephalograms during mental arithmetic task
performance. Data. 2019 Mar;4(1):14.
[35] Choi MK, Lee SM, Ha JS, Seong PH. Development of an EEG-based
workload measurement method in nuclear power plants. Annals of
Nuclear Energy. 2018 Jan 1;111:595-607.

TABLE I. COMPARISON OF COGNITIVE LOAD INDEX BETWEEN BAD AND GOOD SUBJECTS PERFORMING BEFORE(3MIN) & WHILE DOING(1MIN) TASK
USING MEAN.

EEG Bad(3min) Bad(1min) Good(3min) Good(1min)

band Activity Index
ratio Before While doing Before While doing
α 2.99±4.09 2.17±3.69 1.55±2.14 1.49±0.90
Performance Enhancement
θ
β 0.72±1.34 1.22±0.62 1.94±2.07 1.16±1.23
Arousal Index
α
β 2.17±5.51 2.65±2.31 3.02±4.45 1.74±1.12
Neural Activity
θ
β 0.54±1.08 0.83±0.49 1.18±1.41 0.69±0.59
Engagement
(α +θ)
θ 0.33±0.24 0.46±0.27 0.64±0.46 0.66±1.10
Load Index
α
(α +θ) 1.83±0.92 1.19±2.03 0.84±0.70 1.43±1.69
Alertness
β
(β+γ) 1.21±2.19 1.64±1.03 2.08±2.44 1.32±1.12
CWI
(α +θ)