IEEE Systems, Man and Cybernetics Magazine - July 2020 - 36

36	

N

N

C

TS
C
Fg
M
D
M
FB
TS
C
FB

SP
Fg
M
D
M

C

FB

C

SP

Accuracy (%)

mental state classification, but it is only when many trainfour RGCs (FgMDM, TSC, and the two new variants proing trials are available (around 700 in our study), which is
posed here: FBFgMDM and FBTSC), and CNN.
not always possible. However, other factors also differ
The first results we highlight are the CNN classification
between both data sets studied and could also explain difperformances we obtained across the different conditions
ferences in CNN performances,
and data sets. This algorithm has
including the EEG epoch length (2 s
a higher mean accuracy (although
for workload and 60 s for emotions),
insignificantly so) than the origiDeeper analyses
and the nature of the mental states
nal authors' results; the baseline
studied (workload versus emotions).
CSP + LDA; and, more importantare needed to
Emotions are thought to originate
ly, both FBCSP and Riemannian
fully disentangle
from deep brain areas [5] and thus
methods, with both subject-spethese factors, by
are known to be difficult to estimate
cific and subject-independent califrom EEG. In the future, deeper analbrations on the workload data set.
systematically
yses are needed to fully disentangle
Moreover, obtaining reasonable
varying the mental
these factors, by systematically varyperformances in a subject-indeing the mental states studied, the
pendent calibration from only 2 s
states studied, the
EEG epoch length, and the number
of EEG data and 21 users for caliEEG epoch length,
of training trials.
br at ion m a ke s t he CNN pa rAnother relevant result is the
t i c u larly interesting to design
and the number of
promising classification perforcalibration-free neuroadaptive
training trials.
mances of the proposed RGCs.
technologies in the future. In conFBTSC and FBFgMDM outpertrast, this algorithm significantly
formed the results from the data
underperformed subject-specific
sets' authors in most conditions/data sets. Moreover, FBFgand subject-independent calibrations on both the valence
MDM with subject-specific calibration, and FBFgMDM and
and arousal data sets. All algorithms indeed outperformed
FBTSC with subject-independent calibration, reached highthis CNN in all conditions on the emotion data sets.
er mean accuracies than all other algorithms, except the
Multiple factors could explain the observed algorithm
CNN on the workload data set. More interestingly, the small
performances. First, the number of trials that are used for
number of trials in the emotion data sets did not seem to
training models is important. In [33], the authors tested the
affect their performances since they also reached the highshallow ConvNet on multiple motor-imagery data sets
est mean accuracies on both the emotion-valence and emo(from 288 to 1,168 trials) and often obtained significantly
tion-arousal data sets, both with subject-specific
better performances with the CNN than with FBCSP. In
calibration. These promising results compared to standard
our study, the workload data set contained 720 training triRGCs (TSC and FgMDM) are probably due to the extra
als whereas both valence and arousal data sets contained
spectral information extracted with the filter bank, and our
39 training trials only (with cross-validation calibration).
study enabled us to quantify this gain.
This might suggest that the CNN could be useful for
Finally, FBCSP + LDA obtained a higher mean accuracy
than CSP + LDA, although not significantly so, in all condi100
tions/data sets, and the higher overall mean accuracy for
valence classification with subject-independent calibra80
tion. However, it did not produce higher mean accuracies
than others in any other condition. It should be noted that
60
such results reflect the performances obtained in offline
evaluation. As such, they are likely to be similar to those
40
secured in offline or open-loop mental state monitoring,
e.g., for noeuroergonomics (such as mental workload moni20
toring) or neuromarketing (such as emotion monitoring).
The performances are likely to change in closed-loop
0
applications with neuroadaptive technologies and will
thus need to be evaluated in this context as well.
Such results enable us to suggest guidelines about
Algorithms
which algorithm to use for mental states classification
from EEG. First, the CNN is recommended for mental
Subject Specific
Subject Independent
workload classification with both subject-specific and subject-independent calibration, although it seems to need a
Figure 4. The balanced classification accuracies on
lot of training trials (at least several hundreds) so it should
the emotion-arousal data set.
IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE Ju ly 2020



IEEE Systems, Man and Cybernetics Magazine - July 2020

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2020

Contents
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover1
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover2
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Contents
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 2
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 3
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 4
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 5
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 6
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 7
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 8
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 9
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 10
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 11
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 12
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 13
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 14
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 15
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 16
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 17
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 18
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 19
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 20
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 21
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 22
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 23
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 24
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 25
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 26
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 27
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 28
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 29
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 30
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 31
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 32
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 33
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 34
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 35
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 36
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 37
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 38
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 39
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 40
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 41
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 42
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 43
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 44
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 45
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 46
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 47
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 48
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 49
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 50
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 51
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 52
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 53
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 54
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 55
IEEE Systems, Man and Cybernetics Magazine - July 2020 - 56
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover3
IEEE Systems, Man and Cybernetics Magazine - July 2020 - Cover4
https://www.nxtbook.com/nxtbooks/ieee/smc_202110
https://www.nxtbook.com/nxtbooks/ieee/smc_202107
https://www.nxtbook.com/nxtbooks/ieee/smc_202104
https://www.nxtbook.com/nxtbooks/ieee/smc_202101
https://www.nxtbook.com/nxtbooks/ieee/smc_202010
https://www.nxtbook.com/nxtbooks/ieee/smc_202007
https://www.nxtbook.com/nxtbooks/ieee/smc_202004
https://www.nxtbook.com/nxtbooks/ieee/smc_202001
https://www.nxtbook.com/nxtbooks/ieee/smc_201910
https://www.nxtbook.com/nxtbooks/ieee/smc_201907
https://www.nxtbook.com/nxtbooks/ieee/smc_201904
https://www.nxtbook.com/nxtbooks/ieee/smc_201901
https://www.nxtbook.com/nxtbooks/ieee/smc_201810
https://www.nxtbook.com/nxtbooks/ieee/smc_201807
https://www.nxtbook.com/nxtbooks/ieee/smc_201804
https://www.nxtbook.com/nxtbooks/ieee/smc_201801
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1017
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0717
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0417
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0117
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1016
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0716
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0416
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0116
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1015
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0715
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0415
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0115
https://www.nxtbookmedia.com