EEG and Eye Movement Maps of Chess Players
Laercio R. Silva Junior, Fabio H. G. Cesar, Fabio T. Rocha and Carlos E. Thomaz
Department of Electrical Engineering, Centro Universitario da FEI,
Av. Humberto de Alencar Castelo Branco 3972, Sao Bernardo do Campo, Sao Paulo, Brazil
Keywords:
Chess, Brain Mapping, Eye Movements Maps.
Abstract:
Due to a number of advantages to work in the chess environment and its cognitive complexity nature, this
game has been used a lot in scientific experiments in order to study the human cognitive process. This article
describes the steps to acquisition and processing of electroencephalography signals (EEG) and eye tracking of
volunteers with different levels of proficiency in chess and, after the application of mathematical and statistical
methods, maps are generated to discuss and verify patterns among the chess players. Results show neural
activations in different brain areas as well as distinct eye movements for the investigated chess questions and
volunteers who participated in this study.
1 INTRODUCTION
Among several games, chess is widely used for sci-
entific experiments from the mid-twentieth century
(Davis et al., 1973) and it has great importance in the
values of a society (Kasparov, 2003). Working within
the chess environment has advantages such as facil-
ity of transferring the information to the mathematical
environment or computer languages, flexibility to ex-
perimental variations, the long history of matches that
can be used for statistical analysis and allows working
with players with different skill levels to analyze the
cognitive process (Gobet, 1998). Because it is a plat-
form that offers great complexity due to the amount
of possibilities to be performed in the game, chess is
used in experiments to cognitive brain mapping (Saar-
iluoma, 1995).
Particularly in the chess game, the acquisition of
knowledge becomes possible through a learning pro-
cess and relevant information coding. Another char-
acteristic is related to the professional chess players,
for example, to become a grandmaster in chess a per-
son must study and practice for at least ten years to
learn and memorise a significant amount of gaming
patterns and not only individual pieces movements
(Hy
¨
otyniemi and Saariluoma, 1999; Amidzic et al.,
2001; Ross, 2006; Calderwood et al., 1988).
Several experiments were performed in the chess
environment using brain mapping techniques such as
PET (Positron Emission Tomography), SPECT (Sin-
gle Photon Emission Computed Tomography), mag-
netoencephalography and MRI (Magnetic Resonance
Imaging) that brought a lot of information about hu-
man cognition regarding the practice of chess (H
¨
anggi
et al., 2014; Duan et al., 2014; Kazemi et al., 2012;
Amidzic et al., 2001; Nichelli et al., 1994). How-
ever, few experiments were performed using elec-
troencephalography as a resource to acquire the elec-
trical variations of the brain (Rocha et al., 2016;
Wright et al., 2013; Volke et al., 2002). Analogously,
studies related to eye movements in chess games have
been published showing differences in the behaviour
between experts and non-experts, on which it was ver-
ified that chess players with greater proficiency fix
their gazes on more important areas of the board and
hence they have superior performance in this type
of task (Sheridan and Reingold, 2015; Reingold and
Sheridan, 2011; Blignaut et al., 2008; Reingold et al.,
2001).
Despite several experiments in this area, there is
no complete domain over the brain functioning and
eye movements to solve problems in this field. One
proposal is that chess can be used as an effective tool
in the development of higher mental skills to improve
the knowledge and educational learning(Rocha et al.,
2016). As well as Bilali
´
c (Bilali
´
c et al., 2011a; Bi-
lali
´
c et al., 2011b), this article aims to acquire brain
signals and eye movements of chess players, besides
studying the relation between the implicit knowledge
inherent in learning and codification of relevant infor-
mation in chess. More specifically, we intend to ac-
quire synchronously, and interpret the eye movements
434
Junior, L., Cesar, F., Rocha, F. and Thomaz, C.
EEG and Eye Movement Maps of Chess Players.
DOI: 10.5220/0006191404340441
In Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods (ICPRAM 2017), pages 434-441
ISBN: 978-989-758-222-6
Copyright
c
 2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

and electroencephalography signals of chess players
with different levels of experience to recognize pos-
sible visual cognitive patterns and brain mapping for
specific plays in a chess match.
This paper is organised as follows. Next, in sec-
tion 2, we describe the equipment, method of data ac-
quisition, proficiency calculation and signal process-
ing techniques for electroencephalography and eye
movement. Results have been explained in section
3. Then, in section 4 we discuss and compare the re-
sults in this work with previous studies. Finally, in
section 5, we conclude the paper, discussing its main
contribution.
2 METHODS AND MATERIALS
Brain electrical signals were obtained through elec-
troencephalograph (EEG) device, OpenBCI. This
equipment has sampling frequency of 250Hz, reso-
lution of 32 bits per channel and it is an open-source
tool. For this experiment, the electrical signals were
acquired using eight channels: Fp1, Fp2, T3, T4, P3,
P4 O1 and O2, these channels were chosen to em-
brace the most part of the brain as possible. Figure
1 shows the 10-20 conventional system which was
used as reference for the positioning of the electrodes
(Teplan, 2002; Jasper, 1958).
Figure 1: The left image shows the placement of electrodes
in the 10-20 system and the right image shows the place-
ment of the electrodes chosen for the experiment.
For eye-tracking, we used the equipment Tobii
TX300, witch has sampling frequency of 300Hz, pro-
cessing latency between 1ms and 3.3ms, and preci-
sion of 0.14
◦
. Along with the eye-tracking, it was
used a monitor 23” at a resolution of 1920 x 1080 pix-
els to display the questions and chess game situations
at a resolution of 800 x 800 pixels, centralized on the
monitor. Figure 2 shows the volunteer positioning for
the eye movement data acquisition.
Figure 2: Volunteer positioning for the acquisition of eye
movement signals.
2.1 Participants
Twenty volunteers participated in the experiment,
consisting mostly of teenagers of school age (15.95
± 3.93). In this group, none of the volunteers had
ELO rating and, to separate them in groups accord-
ing to their proficiency, we used an individual score
calculation method proposed by Volke (Volke et al.,
2002):
H
s
= (N
correct
−
N
2
) ·
RT
m
RT
s
, (1)
where H
s
= proficiency of each volunteer, N
correct
=
total amount of correct answers, N = total amount of
questions, RT
m
= mean response time of all the vol-
unteers in all questions and RT
s
= mean response time
of the volunteer.
This equation proves to be effective for this exper-
iment because it takes into account not only the accu-
racy of the answer, but also the time spent in relation
to provide that answer. Additionally, volunteers that
answer all questions with the same alternative, trying
to reach in the worst case 50 % of accuracy, but at a
minimum possible time, end up with score equals to
zero (Rocha et al., 2016).
2.2 Tasks and Stimuli
Each volunteer answered to thirteen different ques-
tions related to the chess game in CHESSLAB soft-
ware proposed by Cesar et. al. (2015), adapted for
the present work. Taking as basis the previous works
presented by Cesar et al. (2015) and Rocha et al.
(2016), the questions were elaborated to the present
work. Precautions were taken in the test construction
on the level of difficulty of the questions, number of
questions elaborated for each category, balancing af-
firmative and negative responses and understanding of
the questions proposed. Table 1 shows the categories
EEG and Eye Movement Maps of Chess Players
435

Table 1: Question categorization.
Category Description
C1 Object recognition
C2 Checkmate in one move
C3 Checkmate
C4 Rule retrieval
that were used in this experiment, from which the cat-
egories 1, 2 e 3 were proposed by Volke et. al. (2002)
and the category 4 was proposed by Nichelli et. al.
(1994).
Regarding to the data acquisition system, it is es-
tablished two distinct moments of interaction with the
user. At first, the volunteers read a question presented
in a written form on a monitor and must press the
space key when they finish reading, understanding
and memorising the presented question. In the sec-
ond moment, the chessboard is displayed on a moni-
tor with a coherent configuration expected in a chess
game, the users must now press S to answer ”yes” to
the question or the N key to answer ”no” (Cesar et al.,
2015). Figure 3 shows an example of question and its
respective chessboard presented to the users.
Before the beginning of the the test, the volun-
teer had register himself providing information on his
level of education, age, area of study, gender, lateral-
ity and self-assessment of the level of proficiency in
chess, besides to receive verbalized instructions that
he had free time to read, understand and memorise the
questions, being informed that after the chessboard
was presented he could not read the question again,
and the time elapsed in this stage was not counted.
2.3 Brain Signal Processing
The original data obtained by OpenBCI do not have
any signal pre-processing. So, first it was imple-
mented a high pass filter with cut-off frequency of
0.5Hz to remove the presented DC level of the elec-
trical signal due to the electronic components. There-
after, the signal passed through a low pass filter with
cut-off frequency of 50Hz. This is usually the maxi-
mum frequency of operation of the brain (Gazzaniga
et al., 2009). Finally, it was implemented a band-stop
filter with cut-off frequency of 60Hz to remove possi-
ble noise regarding the frequency of the national grid.
For all stages it were used Butterworth filters, chang-
ing only the settings and cut-off frequency for each
case (Teplan, 2002).
After the electroencephalogram signals have been
filtered, the processing of EEG signals was conducted
by the method proposed by Rocha et al. (Rocha
et al., 2005), which synthesizes, every 2 seconds
prior to decision-making, the communication among
Figure 3: Question example and its respective board.
the specialized neural agents in the solutions of the
moves through the variation of the electrical activity
recorded by each of the EEG electrodes. Figure 4 il-
lustrates an example of an EEG summarisation of the
previous 2 seconds immediately before the decision
making.
After the steps of pre-processing were performed,
we calculated the linear correlation coefficients of the
electric amplitude of values recorded by each of the
electrodes and all the other ones. To this end, it was
adopted here the Pearson correlation for being a para-
metric test and using the absolute value of the re-
sults. After the correlation calculation, the results are
used to perform entropy calculation among channels.
Equation (2) shows how the calculation is done based
ICPRAM 2017 - 6th International Conference on Pattern Recognition Applications and Methods
436

Figure 4: Example of an EEG summarisation channels.
on the Shannon entropy formula (Shannon, 1949):
h(c
i, j
) = −c
i, j
log
2
c
i, j
− (1 − c
i, j
)log
2
(1 − c
i, j
), (2)
where c
i, j
= correlation of two distinct channels.
Equation (2) shows that, if the correlation between
two channels is equal to 1 or equal to 0 the entropy
will be 0. On the other hand, if the correlation is equal
to 0.5 the entropy will be maximum, equals to 1, in-
dicating the possibility of electrical activity of each
electrode being associated with the electrical activity
of each other.
h(c
i, j
) → 1, if c
i, j
→ 0.5
h(c
i, j
) → 0, if c
i, j
→ 1 or c
i, j
→ 0
Analogously the entropy of the average correla-
tion of each electrode can be calculated according to
the equation (3),
h(
c
i
) = −c
i
log
2
(c
i
) − (1 − c
i
)log
2
(1 − c
i
), (3)
where,
c
i
=
1
(n − 1)
n−1
∑
j=1
c
i, j
, (4)
and n is the number of electrodes. In this experiment
n = 8.
The information provided by a single electrode is
given by the sum of the differences between the av-
erage entropy correlation and the entropy of the elec-
trode with the other channels, that is,
h(c
i
) =
n
∑
j=1
(h(c
i
) − h(c
i, j
)). (5)
With the results obtained from the equation (5)
and for the generation of brain maps, we applied
the PCA (Principal Component Analysis) technique
(Johnson and Wichern, 2007) to find a vector basis
that represents the largest existing variance among the
analyzed data. Then, we have used FA (Factor Anal-
ysis), to describe the association between the entropy
values of the electrodes in a non-supervised way. The
main idea behind FA is to disclose the correlation re-
lationships among the original variables using a few
unobservable random ones, called common factors, to
adequately represent the data (Johnson and Wichern,
2007). After that, we applied the varimax rotation al-
gorithm of the principal component calculated from
the correlation matrix of the synthesized data to allow
an interpretation of the EEG brain mappings with no
overlapping. For the results shown in the following
section it was considered only the first factor, the one
with higher self-value of each group of volunteers.
2.4 Eye Movement Signal Processing
The first stage of pre-preprocessing of the acquired
eye movements data is the interpolation of the in-
complete informations, that is to interpolate the miss-
ing values to not affect the subsequent calculations,
since considering the coordinate (0,0) as part of the
eye movement could introduce errors in calculations
to determine the fixations. Equation (6) describes the
linear interpolation formula used for this purpose:
P
i
= P
s
+ i ·
(P
n
− P
s
)
(n − s)
, (6)
where P
i
represents the missing values contiguous in
vector, P
s
is the value of the point immediately pre-
vious to the missing points, P
n
is the value of the
point immediately after to the missing points with
1 < i < n − s − 1.
This interpolation is only applied when the vec-
tor of missing dots is less than 60ms (about 20 sam-
ples in the eye-tracking device used). Interpolation
in a very large array can generate non-existent fixa-
tions. The time of 60ms is very near to the time of
a blink (50 milliseconds on average), thus conserva-
tively, only minor flaws are corrected.
Then, the noises generated by the device itself
as well as micro movements of the eyes or head are
filtered. The existence of these noises can impair the
detection of events such as the fixation of the look, be-
cause of this, it is necessary to detect the inertia in eye
movements, and the noise can cause false movements.
Equation (7) shows the filtering method, a weighted
EEG and Eye Movement Maps of Chess Players
437

moving average.
P
i
=
1
∑
k
n=0
W
n
·
k
∑
n=0
P
i−n
· W
n
, (7)
where P
i
is a value in point vector, W is the weighted
vector, k is the size of the window, P
i−n
is the value of
the umpteenth previous position in the vector (when
n = 0, P
i−n
will be equal to P
i
). In this case k = 20
and W is linearly decreasing.
After the step of pre-processing of the data, it
is used an algorithm that detects the fixations from
the identification of the inertia of the eye movements
in a particular location, called dispersion detection
(Duchowski, 2002; Salvucci and Goldberg, 2000).
This algorithm requires two parameters to define the
fixations: the space threshold indicates the maximum
acceptable dispersion to consider a point as belong-
ing to a fixation; and time threshold that indicates the
minimum time to consider a set of points as a fixation.
These values were set equal to 120 pixels and 120ms,
respectively.
For the generation of eye movement maps, a
squared matrix is created (point matrix) with the same
size of the original image, which receives the sum
of exposure points from the fixations of a group of
participants. Another matrix is generated, which val-
ues are filled by a Gaussian function. The filling of
the point matrix is done by centralized overlay of the
mask at the matrix point of each fixation. The orig-
inal values of the mask points are added together by
the values contained in the mask matrix weighted by
the duration of fixation.
3 RESULTS
Figure 5 summarizes in histograms the test results in
chess of volunteers with higher and lower proficiency,
showing visually how they are distributed in relation
to the number of correct answers, average response
time for question and category regarding to the ques-
tion.
According to the graphs in Figure 5, it is possible
to verify that the answers of the volunteers had differ-
ent percentages of accuracy by question and variation
of the average response time. It is noted that the cate-
gory 1 had a higher percentage of correct answers as
well as the lowest average response time and the cat-
egory 4 had low percentage of correct answers with
high average response time.
Based on the proficiency results described and
aiming to demonstrate the neural behaviour and
discriminant eye movement among volunteers with
Figure 5: The top graph shows the amount and percentage
of accuracy of each question. The bottom graph shows the
average response time and category of each question. The
categories are the same ones presented in Table 1.
higher and lower proficiency, cognitive maps and vi-
sual attention maps for the following two groups were
generated: more proficient group that scored higher
than 34 (Hs > 34); less proficient group that scored
less than 11 (Hs < 11), through a division by quar-
tiles (Bussab and Morettin, 2010).
Figure 6 shows the brain maps on a color scale
that starts in the red color, indicating low influence of
the brain area for the task solution, passing through
orange, yellow, green, light blue and dark blue color
that represents the factor loadings greater than 0.7 on
a scale ranging from 0 to 1 (Rocha et al., 2016; Rocha
et al., 2014). It is noted that the map of the category 1
for group of volunteers who had a performance higher
than 34 points showed an association among connec-
tivity measured from the electrodes O1, O2, P3, P4
and T4 while the group with the a performance lower
than 11 points showed an association among connec-
tivity measured from the electrodes O1, O2 and P4.
For the category 4 the most proficient group presented
a neural circuit among the electrodes O1, P3 and P4
and the least proficient group presented association
among F1, F2 and P3.
Figure 7 shows two questions: question 12 of cat-
egory 1 showing 100% accuracy of responses and the
lowest average response time; question 13 of category
ICPRAM 2017 - 6th International Conference on Pattern Recognition Applications and Methods
438

Figure 6: Brain maps.
4 for which there was 50% accuracy of responses and
high average time response, as depicted in Figure 5.
The visual attention maps show the areas of the board
where each group of volunteers focused their gazes
to solve the problems presented by a color scale that
begins in green, indicating that there was little fixa-
tions in that spot, through the colors yellow, orange
and red, color that indicates several recordings in that
spot. Furthermore, the color scale is related to the
intra-group information, that is, we considered only
the data of the volunteers who were on the same pro-
ficiency group.
4 DISCUSSION
Through the methods described above we generated
brain maps and visual attention maps shown in Fig-
ures 6 and 7 respectively, evidencing differences in
neural activation patterns and eye movement between
groups of volunteers with higher and lower profi-
ciency in chess.
Observing the brain map in Figure 6 for the most
proficient group, it is found that for category 1 there
was a greater association in the occipital-parietal re-
gion and temporal region of the right hemisphere,
these regions are related to visuospatial processing
Figure 7: Visual attention maps.
(Vanlierde et al., 2003), which shows consistency
with the stimulus presented, as the category involves
the recognition of parts on the board. The group with
the lowest proficiency had a similar brain map in the
occipital area which is the area related to primary vi-
sion processing (Martin, 2014).
For this same category, analyzing the visual atten-
tion maps of Figure 7, the highest proficiency group
had a greater attention on positions of the board where
there were no pieces, despite all the volunteers of this
group answered correctly the questions. The main
difference is found in the elapsed time for the solu-
tion of the questions, as the average response time for
this one was 3.8s and 5.1s for groups of higher and
lower proficiency respectively.
Analyzing the Figure 6 for category 4 it is noted
that the group with the lowest proficiency had a
stronger connection in the area of the frontal lobes,
this area is related to the planning and memory (Mar-
tin, 2014; Gazzaniga et al., 2009), while the group
with the highest proficiency had an association in the
parietal region and occipital region in the left hemi-
sphere, unlike what is found in the literature (Rocha
et al., 2016; H
¨
anggi et al., 2014). This difference
found in relation to other studies may be attributed
to the proficiency between groups of participants be-
cause spite of the differences, any of the groups is
considered experienced in chess.
EEG and Eye Movement Maps of Chess Players
439

Regarding the map of eye movement for category
4 in Figure 7 it is verified that in both groups there is a
greater focus on the localization of the pieces that are
part of the solution for the question presented, and it is
noted that the most proficient group fixed their gazes
for a longer time in this region of interest, which can
contribute to the best performance in this question,
with four correct answers to the highest proficiency
group and one corrected answer to the lowest profi-
ciency group.
5 CONCLUSION
We have carried out a computational experiment that
involves acquisition and processing of EEG signals
and eye movements in chess, generating as final re-
sult cognitive maps that show the brain areas that were
more activated for the solution of the presented stim-
uli and visual attention maps that highlight regions of
preferred fixations of volunteers.
Our results have disclosed differences in the pat-
terns of brain activation and eye movements among
chess players with higher and lower proficiencies,
analogously to other works in the literature (Rocha
et al., 2016; Sheridan and Reingold, 2015; Wright
et al., 2013; Reingold and Sheridan, 2011; Reingold
et al., 2001). In short, chess players with lower pro-
ficiency presented higher dispersion of attention and
visual fixations on non-relevant parts of the stimuli,
demanding more time to analyze and answer the ques-
tions as well as major brain activations in the occipital
areas rather than in the frontal ones.
As future work, we intend to extend the frame-
work proposed increasing the number of volunteers,
especially considering volunteers with ELO rating,
the number of questions and categories to analyze
other discriminant characteristics on chess, and the
EEG spatial resolution.
ACKNOWLEDGEMENTS
The authors would like to thank the financial support
provided by FEI, CAPES and CNPq (444964/2014-
2).
REFERENCES
Amidzic, O., Riehle, H. J., Fehr, T., Wienbruch, C., and
Elbert, T. (2001). Pattern of focal γ-bursts in chess
players. Nature, 412(6847):603–603.
Bilali
´
c, M., Kiesel, A., Pohl, C., Erb, M., and Grodd, W.
(2011a). It takes two-skilled recognition of objects
engages lateral areas in both hemispheres. PLoS One,
6(1):e16202.
Bilali
´
c, M., Langner, R., Ulrich, R., and Grodd, W. (2011b).
Many faces of expertise: fusiform face area in chess
experts and novices. The Journal of Neuroscience,
31(28):10206–10214.
Blignaut, P. J., Beelders, T. R., and So, C. (2008). The vi-
sual span of chess players. In Proceedings of the 2008
symposium on Eye tracking research & applications,
pages 165–171. ACM.
Bussab, W. d. O. and Morettin, P. A. (2010). Estat
´
ıstica
b
´
asica. Saraiva.
Calderwood, R., Klein, G. A., and Crandall, B. W. (1988).
Time pressure, skill, and move quality in chess. The
American journal of psychology, pages 481–493.
Cesar, F. H. G., Rocha, F. T., and Thomaz, C. E.
(2015). Chesslab: Um arcabouc¸o computacional para
aquisic¸
˜
ao e processamento de sinais de rastreamento
ocular em jogadas de xadrez. In 2015 28th SIBGRAPI
Conference on Graphics, Patterns and Images, page
4pp.
Davis, M. D., Morgenstern, O., Hegenberg, L., and
da Mota, O. S. (1973). Teoria dos jogos: uma
introduc¸
˜
ao n
˜
ao-t
´
ecnica.
Duan, X., Long, Z., Chen, H., Liang, D., Qiu, L., Huang,
X., Liu, T. C.-Y., and Gong, Q. (2014). Func-
tional organization of intrinsic connectivity networks
in chinese-chess experts. Brain research, 1558:33–43.
Duchowski, A. T. (2002). A breadth-first survey of eye-
tracking applications. Behavior Research Methods,
Instruments, & Computers, 34(4):455–470.
Gazzaniga, M., Ivry, R., and Mangun, G. (2009). Cognitive
Neuroscience: The Biology of the Mind. Norton.
Gobet, F. (1998). Expert memory: A comparison of four
theories. Cognition, 66(2):115–152.
H
¨
anggi, J., Br
¨
utsch, K., Siegel, A. M., and J
¨
ancke, L.
(2014). The architecture of the chess player s brain.
Neuropsychologia, 62:152–162.
Hy
¨
otyniemi, H. and Saariluoma, P. (1999). Chess-beyond
the rules. Finnish Artificial Intelligence Society.
Jasper, H. H. (1958). The ten twenty electrode system of
the international federation. Electroencephalography
and clinical neurophysiology, 10:371–375.
Johnson, R. and Wichern, D. (2007). Applied Multivariate
Statistical Analysis. Applied Multivariate Statistical
Analysis. Pearson Prentice Hall.
Kasparov, G. (2003). My Great Predecessors, Parts 1-5,
volume 1857444043.
Kazemi, F., Yektayar, M., and Abad, A. M. B. (2012).
Investigation the impact of chess play on develop-
ing meta-cognitive ability and math problem-solving
power of students at different levels of education.
Procedia-Social and Behavioral Sciences, 32:372–
379.
Martin, J. (2014). Neuroanatomia - 4ed: Texto e Atlas.
AMGH Editora.
ICPRAM 2017 - 6th International Conference on Pattern Recognition Applications and Methods
440

Nichelli, P., Grafman, J., Pietrini, P., Alway, D., Carton,
J. C., and Miletich, R. (1994). Brain activity in chess
playing. Nature.
Reingold, E. M., Charness, N., Pomplun, M., and Stampe,
D. M. (2001). Visual span in expert chess players: Ev-
idence from eye movements. Psychological Science,
12(1):48–55.
Reingold, E. M. and Sheridan, H. (2011). Eye movements
and visual expertise in chess and medicine. Oxford
handbook on eye movements, pages 528–550.
Rocha, F. T., de Magalhaes Junior, R. G., Horta, T. O.,
Cesar, F. H. G., D’Israel, D. M., da Silva, W., and
Thomaz, C. E. (2016). Eeg acquisition and pro-
cessing for cognitive brain mapping during chess
problem solving. IEEE Latin America Transactions,
14(3):1129–1134.
Rocha, F. T., Rocha, A. F., Massad, E., and Menezes, R.
(2005). Brain mappings of the arithmetic process-
ing in children and adults. Cognitive Brain Research,
22(3):359–372.
Rocha, F. T., Thomaz, C. E., da Rocha, A. F., and Mas-
sad, E. (2014). Brain mapping and interpretation of
reading processing in children using eeg and multi-
variate statistical analysis. In 2014 27th SIBGRAPI
Conference on Graphics, Patterns and Images, pages
251–258. IEEE.
Ross, P. E. (2006). The expert mind. Scientific American,
295(2):64–71.
Saariluoma, P. (1995). Chess players’ thinking: A cognitive
psychological approach. Psychology Press.
Salvucci, D. D. and Goldberg, J. H. (2000). Identifying
fixations and saccades in eye-tracking protocols. In
Proceedings of the 2000 symposium on Eye tracking
research & applications, pages 71–78. ACM.
Shannon, C. E. (1949). Communication theory of secrecy
systems. Bell system technical journal, 28(4):656–
715.
Sheridan, H. and Reingold, E. M. (2015). Expert vs. novice
differences in the detection of relevant information
during a chess game: evidence from eye movements.
Psychological perspectives on expertise, page 11.
Teplan, M. (2002). Fundamentals of eeg measurement.
Measurement science review, 2(2):1–11.
Vanlierde, A., De Volder, A. G., Wanet-Defalque, M.-C.,
and Veraart, C. (2003). Occipito-parietal cortex acti-
vation during visuo-spatial imagery in early blind hu-
mans. Neuroimage, 19(3):698–709.
Volke, H.-J., Dettmar, P., Richter, P., Rudolf, M., and Buhss,
U. (2002). On-coupling and off-coupling of neocor-
tical areas in chess experts and novices as revealed
by evoked EEG coherence measures and factor-based
topological analysis–a pilot study. Journal of Psy-
chophysiology, 16(1):23.
Wright, M. J., Gobet, F., Chassy, P., and Ramchandani, P. N.
(2013). Erp to chess stimuli reveal expert-novice dif-
ferences in the amplitudes of n2 and p3 components.
Psychophysiology, 50(10):1023–1033.
EEG and Eye Movement Maps of Chess Players
441