ArticlePDF Available

QEEG-based neural correlates of decision making in a well-trained eight year-old chess player

Authors:
  • Harvard Alumni for Mental Health- Harvard University

Abstract and Figures

The neurocognitive substrates of decision making (DM) in the context of chess has appealed to researchers' interest for decades. Expert and beginner chess players are hypothesized to employ different brain functional networks when involved in episodes of critical DM upon chess. Cognitive capacities including, but not restricted to pattern recognition, visuospatial search, reasoning, planning and DM are perhaps the key determinants of rewarding and judgmental decisions in chess. Meanwhile, the precise neural correlates of DM in this context has largely remained elusive. The quantitative electroencephalography (QEEG) is an investigation tool possessing a proper temporal resolution in the study of neural correlates of cognitive tasks at cortical level. Here, we used a 22-channel EEG setup and digital polygraphy in a well-trained 8 year-old boy while engaged in playing chess against the computer. Quantitative analyses were done to map and source-localize the EEG signals. Our analyses indicated a lower power spectral density (PSD) for higher frequency bands in the right hemisphere upon DM-related epochs. Moreover, the information flow upon DM blocks in this particular case was more of posterior towards anterior brain regions.
Content may be subject to copyright.
UNCORRECTED PROOF
Journal of Integrative Neuroscience 00 (20xx) 1–14 1
DOI 10.3233/JIN-170056
IOS Press
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
QEEG-based neural correlates of decision
making in a well-trained eight year-old chess
player
Abolfazl Alipour a,b, Sahar Seifzadeh c, Hadi Aligholi b,dandMohammadNamib,d,e,
aDepartment of Psychological and Brain Sciences, Indiana University, Bloomington, Indiana, USA
bNeuroscience Laboratory-NSL (Brain, Cognition and Behavior), Department of Neuroscience, School
of Advanced Medical Sciences and Technologies, Shiraz University of Medical Sciences, Shiraz, Iran
cYoung Researchers and Elite Club, Qazvin Branch, Islamic Azad University, Qazvin, Iran
dDepartment of Neuroscience, School of Advanced Medical Sciences and Technologies, Shiraz
University of Medical Sciences, Shiraz, Iran
eClinical Neurology Research Center, Shiraz University of Medical Sciences, Shiraz, Iran
Received 9 September 2017
Accepted 13 October 2017
Abstract. The neurocognitive substrates of decision making (DM) in the context of chess has appealed to researchers’ interest
for decades. Expert and beginner chess players are hypothesized to employ different brain functional networks when involved in
episodes of critical DM upon chess. Cognitive capacities including, but not restricted to pattern recognition, visuospatial search,
reasoning, planning and DM are perhaps the key determinants of rewarding and judgmental decisions in chess. Meanwhile, the
precise neural correlates of DM in this context has largely remained elusive. The quantitative electroencephalography (QEEG)
is an investigation tool possessing a proper temporal resolution in the study of neural correlates of cognitive tasks at cortical
level. Here, we used a 22-channel EEG setup and digital polygraphy in a well-trained 8 year-old boy while engaged in playing
chess against the computer. Quantitative analyses were done to map and source-localize the EEG signals. Our analyses indicated
a lower power spectral density (PSD) for higher frequency bands in the right hemisphere upon DM-related epochs. Moreover,
the information flow upon DM blocks in this particular case was more of posterior towards anterior brain regions.
Keywords: Chess, decision making, QEEG, power spectra, functional connectivity
1. Introduction
Decision making (DM), characterized as the act of selecting the best among different alternatives,
plays a defining role in various aspects of our life. Complex human cognition, such as DM under uncer-
tainty, is represented through a dynamic spatio-temporal activity in the brain [15]. While one’s wise and
informed decisions may contribute to success and satisfaction, ill-advised decisions often lead to failure.
The multifaceted process of decision making is potentially linked to a wide range of variables including
input, process, output and feedback [10,19].
*Corresponding author: Mohammad Nami, Department of Neuroscience, School of Advanced Medical Sciences and
Technologies, Shiraz University of Medical Sciences, Shiraz 71348-14336, Iran. E-mail: torabinami@sums.ac.ir.
[research-article] p. 1/14
0219-6352/17/$35.00 © 2017 – IOS Press and the authors. All rights reserved
UNCORRECTED PROOF
2A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
A popular strategy board game like chess brings a compact and easily controllable task environ-
ment for decision making. As such, many researchers have become interested in using chess as the
task paradigm of choice to assess the neural dynamics of various mental and cognitive skills including
DM [7,9,12,24]. In a landmark study on chess players which was done several decades ago by de Groot,
key findings postulated that expertise in chess was more determined by pattern recognition rather than
search. The general image on the cognitive mechanisms involved in critical DM in the context of chess
has been largely transformed since de Groot’s study [7].
Along these lines, results from other investigations [25,26] argued that intelligence plays key part
both in DM and reasoning. In this way, some studies proposed solid correlations between the IQ scores
and the reasoning capacity of individuals [27]. With regard to the processing speed, when players are
forced to play faster, their ability during the play tends to be less predictive. However, expert players
are shown to perform noticeably better than novice chess players in terms of rapid object recognition
abilities [5,29].
There is a growing trend especially among young adolescents to join chess clubs, however the question
whether playing chess may improve their global DM skills remains elusive. It is then the matter of debate
whether and how playing chess may empower adolescents’ brain in more productive and error free, fast,
wise and rewarding DM.
In case the quantitative EEG (QEEG) provides us with new insights about the cortical brain regions
involved in critical DM among expert chess players new avenues may open for DM-chess related re-
search. Potentially through reverse engineering, areas involved in a grand average QEEG of elite chess
players may define a pathway toward targeting more or less same networks in the brain of novice players
helping them to gain DM, at least in chess, in a faster and more efficient way.
The emergence of neurotechnological tools such as transcranial magnetic stimulation (TMS), tran-
scranial direct current stimulation (tDCS) and neuro-feedback provides novel approaches for cognitive
empowerment through modulation of the involved brain networks [3].
While our working-team pursues its long way to systematically investigate the above, the present case
study highlights the results of our pilot assessments on the neural correlates of DM at cortical level using
a 22 channel-QEEG recording setup in a well-trained eight year-old male chess player when engaged in
playing chess against the computer.
2. Case study
2.1. Subject
Our case study was done on P.A., an 8-year old right-handed male elite chess player with regional and
national recognition in chess competitions. The subject was instructed to sleep well the night before the
experiment and was not under the effect of any medicine or stimulant food or drink. Both the subject
and his legal guardian read and signed the informed consent to take part in the study.
2.2. Experimental procedure
Subject played the Chess Titans on Windows 10 at a self-claimed challenging difficulty level (level 6)
using the mouse and right hand only. Subject played the game for 15 minutes and eventually lost to the
computer.
[research-article] p. 2/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 3
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
2.3. EEG recording and pre-processing
EEG was recorded at two different conditions including the resting state and critical DM blocks while
playing chess. A 22-channel bipolar EEG montage using a 32-channel amplifier system (3840, NR-
SIGN, BC, Canada) was used for QEEG data acquisition. The channel dipoles, based on the interna-
tional 10-20 system, included FP2-F7, FP2-T5, F8-F3, F8-P3, F4-C3, F4-P1, T4-O1, C4-O1, T6-P1,
FP1-FP2, P1-P2, FP2-C4, FP2-P2, F8-T6, F4-T4, F4-O2, T4-P4, C4-P4, T6-O2, O2-P2, O1-O2 where
FP stands for frontopolar, C for central, F for frontal, P for parietal, T for temporal and O for occipi-
tal.
For the resting state condition, subject was sat in the dim-lit recording room (320 lux) and fixated on
a crosshair for ten minutes. Resting state EEG recording continued for 10 minutes.
EEG was recorded throughout the game period and the subject verbally indicated the start of a critical
DM epoch which eventually ended in moving a piece. The reported critical DM epochs (n=6) were
extracted for further analysis (mean duration 14 ±4 seconds).
Other than the EEG recording, subject was attached to a digital sampling unit for autonomic system
polygraphy. By this, real-time galvanic skin conductance (GSC) and heart rate variability (HRV) were
recorded using the Vilistus DSU, UK [21](Fig.1).
All EEG signals were then imported in EEGLAB version 13.0.0 running on MATLAB R2013a. Pre-
processing of the signals included the rejection of visually detectable artifacts and applying a low-pass
filter of 1 Hz and a high-pass filter of 48 Hz at the first step. Afterwards, an Infomax-based independent
Fig. 1. Data acquisition setup. The study subject engaged in chess against the computer. He reported 6 episodes of critical deci-
sion making challenge during which real-time EEG recording was marked and subsequently grand averaged in brain mapping.
The galvanic skin conductance (GSC) and heart rate variability (HRV) were also being monitored during the whole process
with special marking upon challenging decision making episodes.
[research-article] p. 3/14
UNCORRECTED PROOF
4A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
component analysis (ICA) decomposition algorithm (runica.m in EEGLAB) was used in order to find
possible artifacts. Four of the 22 components were identified as artifactual (3 eye movement artifacts
and 1 muscle artifact component). After removal of these artifactual components, the ICA algorithm
was used once again to achieve 22 artifact-free components. The obtained EEG signals were then used
in two different analysis platforms including EEGLAB and NeuroGuide (v.2.3.8, Applied Neuroscience,
USA).
2.4. Topographic mapping of EEG signals
Corresponding to the 6 critical DM epochs, the resting state EEG was segmented into 6 artifact free
epochs for comparison. The NeuroGuide software was used to analyze and plot the topographic maps
and power spectra of the EEG activity during both resting state and critical DM epochs. The abso-
lute and relative powers of the signals in theta, alpha, and beta bands were calculated and compared
between the resting state and critical DM epochs. Moreover, coherence and power ratio in the same
frequency bands were computed and compared between the two states. The Wilcoxon signed-rank test
was used to compare the difference between relative and absolute power spectra, power ratio, and co-
herence between the two states. Pvalues <0.05 were considered as statistically significant throughout
the analyses.
2.5. EEG source connectivity analysis
The DIPFIT2 plug-in of EEGLAB was used to source localize the EEG signals. Size of the head model
was modified in order to match the size of the 8 year-old subject based on the insights from BenAbdelka-
der et al. 2008 [4,8]. In short, the DIPFIT2 plug-in uses ICA components to estimate sources of EEG
activity in the brain. For the 22 components existing in the signals, 22 sources have been calculated.
The Source Information Flow Toolbox (SIFT) [8,20] was also used to estimate the information flow
between sources. In addition, the direct Directed Transfer Function (dDTF) [16] was used as a measure
of connectivity between the 22 components. The obtained number was averaged for all time-points and
all frequencies between 4 to 46 Hz (with 2 Hz intervals) in both resting state and critical DM epochs
to obtain a 22 ×22 ‘overall connectivity matrix’. Overall connectivity matrix of the decision making
epochs was subtracted from overall connectivity matrix in the resting state in order to obtain a connec-
tivity matrix revealing the differences in information flow between the two states. Consequently, major
sink and source nodes were identified.
3. Results
The grand average QEEG amplitude data for DM epochs versus resting state turned to show sig-
nificantly higher absolute power in the range of theta and alpha frequency bands in the right fronto-
central and temporoparietal brain regions upon task-positive brain cortical activity versus resting state.
Furthermore, except for the F4-C3 dipole, beta amplitude was less in DM states compared to resting
state.
Figures 24demonstrate the comparative absolute power values for theta, alpha and beta spectral
bands. As illustrated in Fig. 2, theta amplitude was found to significantly dominate upon DM epochs
grand average versus resting state QEEG in F8-T3 [t(5)=2.1, %95CI =(1.56)(0.03),p=
0.03], F4-C3 [t(5)=2.85, %95CI =(2.42)(0.41),p=0.006], T4-P3 [t(5)=2.02, %95CI =
[research-article] p. 4/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 5
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 2. Spiderweb chart plotting the absolute power values for theta frequency band upon decision making (orange) and resting
state (blue) QEEG data. The analysis revealed higher theta power upon decision making task in F8-T3, F4-C3, T4-P3, C4-P3
and T6-O1 dipoles compared to resting state. Values are presented in μV2.#P<0.05, ##P<0.01 and ###P<0.001. FP:
frontopolar, C: central, F: frontal, P: parietal, T: temporal and O: occipital.
(2.26)(0.03),p=0.04], C4-P3 [t(5)=4.48, %95CI =(0.95)(2.4),p<0.001], and T6-O1
[t(5)=2.76, %95CI =(1.41)(0.005),p=0.01] dipoles.
Figure 3illustrates the difference in absolute power for alpha frequency band in DM grand aver-
age epochs versus resting state in various dipole locations. Alpha amplitude was found to signifi-
cantly dominate upon DM epochs grand average versus resting state QEEG in F4-C3 [t(5)=2.93,
%95CI =(2.36)(0.71),p=0.009]. F8-T3 however, showed a dominant alpha upon resting state
compared to task-positive epochs [t(5)=2.22, %95CI =(1.57)(0.08),p=0.03].
[research-article] p. 5/14
UNCORRECTED PROOF
6A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 3. Spiderweb chart plotting the absolute power values for alpha frequency band upon decision making (orange) and resting
state (blue) QEEG data. The analysis revealed higher alpha power upon decision making task in F4-C3 dipole compared
to resting state, ##P<0.01. The F8-T3 dipole however showed a dominant alpha power upon resting state compared to
task-positive epochs, *P<0.05. Values are presented in μV2. FP: frontopolar, C: central, F: frontal, P: parietal, T: temporal
and O: occipital.
AsshowninFig.4, beta power was diminished in the anterior brain regions in task-positive states
compared to resting QEEG. In other words, unlike the higher amplitude for beta in many derivations
upon resting state, alpha and theta amplitude took over in right hemisphere dipoles in the task-positive
state. Based on our comparative analyses, beta power was found to be predominantly higher in rest-
ing state than DM epochs in F8-T3 [t(5)=2.72, %95CI =(0.034)(2.37),p=0.008], F8-F4
[t(5)=2.93, %95CI =(2.66)(0.47),p=0.006], F4-C4 [t(5)=3.11, %95CI =(0.83)(1.3),
[research-article] p. 6/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 7
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 4. Spiderweb chart plotting the absolute power values for beta frequency band upon decision making (orange) and resting
state (blue) QEEG data. The analysis revealed higher beta power in resting state than DM epochs in F8-T3, F8-F4, F4-C4,
C4-P4, and FP2-F3. *P<0.05, **P<0.01, ##P<0.01. Values are presented in μV2. FP: frontopolar, C: central, F: frontal,
P: parietal, T: temporal and O: occipital.
p=0.004], C4-P4 [t(5)=2.22, %95CI =(1.46)(0.07),p=0.03] and FP2-F3 [t(5)=2.16,
%95CI =(1.54)(0.04),p=0.04].
The FFT relative power analysis revealed an increased alpha band power in the right centroparietal
region in DM versus resting states. In addition, beta relative power was diminished in FP2 region upon
DM epochs. Such a decrease was evident in FP1 across high beta band frequency.
[research-article] p. 7/14
UNCORRECTED PROOF
8A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
In terms of FFT power ratio, delta/theta and delta/alpha ratios were increased in FP1 upon DM epochs.
Moreover, centroparietal regions demonstrated a rise in alpha/beta and alpha/high beta power ratio.
Figure 5demonstrates the spatial distribution and connectivity brain maps for resting states (sec-
tion (A)) versus the grand average for the six DM blocks (section (B)) suggesting lower than expected
Fig. 5. The QEEG topographical spectral brain maps. The analysis demonstrates absolute power values across spectra upon
resting state (section (A), upper panel) and decision making QEEG data (panel (B), upper panel). The resting state QEEG’s
centroparietal activity is compatible with default mode network activity. Grand average for the six decision making blocks
(section (B)) suggests lower than expected beta power in the anterior brain regions rather than occipital and centroparietal areas.
The lower panels in sections (A) and (B) in turn represent resting state and decision making beta coherence maps indicating a
frontal beta hypocoherence upon task-positive rather than resting states.
[research-article] p. 8/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 9
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 5. (Continued.)
beta power in the anterior brain regions rather than occipital and centroparietal areas. This perhaps indi-
cates that DM process in our subject was more of an automatic and less of cognitive nature regulated by
subcortical (corticostrial) networks rather than cortico-cortical pathways. Interestingly, beta coherence
in anterior brain regions diminished upon DM epochs (grand average for 6 blocks, section (B), lower
panel) compared to resting state (Section (A), lower panel) (Fig. 5).
[research-article] p. 9/14
UNCORRECTED PROOF
10 A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 6. Autonomic response upon decision making task. Panels (A)-(C) represent heart rate variability (HRV), pNN50 (number
of pairs of successive NN intervals that differ by more than 50 ms) and GSR (in μSiemens). The yellow shades represent
decision making (DM) blocks of which the EEG data were extracted and analyzed. Despite no significant difference in GSR
in DM and non-DM blocks there was an apparent increase in HRV and PNN50 in DM blocks. This suggests an autonomic
component involved when the subject is engaged in DM task.
With regard to the autonomic response data upon DM blocks, Fig. 6shows an increased HRV and
interbeat group change but not GSR in DM blocks as highlighted in yellow. This suggests an autonomic
component involved when the subject is engaged (even subcortically) in DM task (Fig. 6).
Figure 7represents the EEG source connectivity analysis. As outlined in the methods section, an
Infomax-based ICA decomposition algorithm using the EEGLAB in MATLAB yielded 22 artifact-free
components. Based on our analysis, the source of information flow upon DM block EEG analyses
were localized at components 3, 4, 13, 14, 15 and 16 which were anatomically linked to a left ven-
tromedial prefrontal cortex (vmPFC) as well as right occipital and right medial temporal cortices. On
the other hand, the information flow sink were localized at components 9, 11, 19 and 22 which were
linked to the left orbitofrontal cortex (OFC), left PPC and right intraparietal sulcus (IPS). As such, it
turned out that the information flow upon DM blocks were more of posterior towards anterior brain
regions.
[research-article] p. 10/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 11
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
Fig. 7. Information flow between major sources and sinks upon critical decision making. (A) (left to right): position of informa-
tion sources in sagittal, coronal, and transverse sections. (B): the color matrix for information flow between all signal sources.
Each square represents the information flow difference between Resting and DM states for a specific pair of signal sources. For
instance, the red color in element (4, 3) indicates an increased flow of information and the causal effect from signal source 3 on
signal source 4 during DM epochs. (C): Position of information sources in 3D space. (D) (left to right): position of “information
sinks” in sagittal, coronal, and transverse sections. (E): the reciprocal color matrix (versus (B)) for information flow between
all signal sources. (F): Position of information sinks in 3D space. Sources and sinks were identified by thresholding the amount
of their causal effect. While the sources of information flow upon DM blocks were localized at components 3, 4, 13, 14, 15 and
16 (anatomically linked to the left vmPFC, right occipital and right medial temporal cortices); the information flow sinks were
localized at components 9, 11, 19 and 22 ( anatomically linked to the left OFC, left PPC and right IPS).
4. Discussion
The present case study was an attempt to investigate the plausible cortical networks potentially in-
volved in DM task while a well-trained 8 year-old boy was engaged in chess play against the computer.
Results from the QEEG analysis indicated a lower power spectral density (PSD) for higher frequency
bands power in the right hemisphere upon DM epochs. Further analyses suggested that the information
flow upon DM blocks in this particular case were more of posterior towards anterior brain regions.
The inferior frontal gyrus has also been proposed to be centrally involved in DM processes [28]. In the
present investigation however, we observed predominant theta frequency upon DM epochs in the right
[research-article] p. 11/14
UNCORRECTED PROOF
12 A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
inferior frontal included dipole. The hypothetical explanation for this observation may revolve around
the role of dorsolateral prefrontal cortex and inferior frontal gyrus in the process of DM when cognitive
control plays a central role [6]. This was apparently found to be less employed in this particular case
study when the subject was engaged in critical DM process. The neural correlates of decision making
is known to involve at least the anterior cingulate cortex, middle frontal gyrus, and inferior frontal
gyrus/insula, with recent insights suggesting that decisions may emerge from distributed processes [28,
31].
Some earlier reports have indicated the role of left parietal theta power as a correlate of memory
retrieval and DM [14,31]. Our finding was in agreement with the above results since a major sink of
information flow in our QEEG analysis was found to be the left posterior parietal cortex (PPC). Indeed,
unraveling the neural mechanism of our observation might help to explain some key electrophysiological
determinants of DM. As outlined in the results section, the electrophysiological difference associated
with decision-making epochs mainly corresponded to the distribution and power of theta frequency band
in fronto-central and posterior parietal cortices. Based on our QEEG findings and source connectivity
analysis, we propose a neural model more of an automatic and less of cognitive nature regulating the
process of DM in our tested elite chess player.
Studies have considered chess as a task paradigm to evaluate the players’ brain activity under am-
biguous circumstances and time pressure. For instance, a number of neuroimaging investigations using
the functional magnetic resonance imaging (fMRI) strived to localize the neural activity upon percep-
tual DM. According to such evidence, cortical and subcortical brain regions and structures including
the frontal and parietal cortices, thalamus and striatum were found to be largely involved in modulating
decisions’ accuracy and uncertainty [11,13,22,23]. In other well-designed studies, expert chess play-
ers’ memory recall was compared with that of beginners or less skilled players with the whole-brain
analysis conducted especially on regions of interest such as anterior cingulate cortex (ACC), bilateral
intraparietal sulci (IPS), bilateral ventromedial and dorsolateral prefrontal cortices (vmPFC and dlPFC)
and prefrontal cortices (PFC). Results from such studies on chess player brains corroborated that com-
ponents of the frontoparietal network (FPN) are not only linked to consciousness and attention but also
working memory [9,11,24]. Despite the above, in our case study, the FPN was shown to be less involved
once this well-trained chess player was making critical decisions. Instead centroparietal areas were found
to shower higher amplitude for beta frequency band in QEEG may and the sinks for information flow
(estimated through ICA) turned to be the left PPC and right IPS.
Assessment tools such as EEG are not only less costly and more convenient to administer but also
capable of rendering a proper temporal resolution, hence may be considered appropriate to address the
temporal sequencing of DM signals [1]. For instance, computational model-based approach to EEG data
acquired during simple binary choice task have shown to yield dependable data on the temporal sequence
of information flow in the brain [17,18].
In studies which examined the pattern recognition of four simple conditions in chess, evoked coher-
ences of EEG signals were found to be sensitive to sensory as well as mental activity (theta and beta
coherence, respectively). Meanwhile, beta coherence considerably depended on the task type [30].
When it comes to perceptual DM, the process is more error-prone, especially near the threshold.
Despite the proposed internal noise in neural systems which seems to be responsible for such errors,
it appears that a mixture of bottom-up and top-down sources drive this potential complexity [2]. Such
a complexity in interpreting data should be considered when EEG is used as the method of choice in
neurocognitive studies.
[research-article] p. 12/14
UNCORRECTED PROOF
A. Alipour et al. / Please provide the short title 13
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
5. Conclusion
Our present QEEG findings propose a less cognitive load upon DM task in this particular subject
based on the dominance of right and posterior alpha/theta versus beta frequency bands. As such, the
FPN was shown to be less involved once our subject was involved in critical DM. This data may suggest
that perhaps an early system responds preferentially to outcomes only in order to initiate a fast automatic
alertness response when someone is an expert chess player.
Though our research opens up new avenues for the investigation of the neural system underlying nor-
mal DM, future studies should demonstrate the level of FPN and subcortical nuclei involvement when
elite versus novice chess players engage in critical DM tasks. The combined use of advanced neurotech-
nological tools such as fMRI, functional near infra-red spectroscopy (fNIRS) and magnetoencephalog-
raphy (MEG) would offer novel opportunities for more in-depth investigation of the above.
6. Conflict of interest
None.
Acknowledgement
Authors would like to thank Ms. S. Safaie for her assistance with this report.
References
[1] A. Alipour, A. Mojdehfarahbakhsh, A. Tavakolian, T. Morshedzadeh, M. Asadi, A. Mehdizadeh and M. Nami, Neural
communication through theta-gamma cross-frequency coupling in a bistable motion perception task, Journal of integrative
neuroscience 04 (2016), 539–551. doi:10.1142/S0219635216500291.
[2] S. Amitay, J. Guiraud, E. Sohoglu, O. Zobay, B.A. Edmonds, Y.X. Zhang and D.R. Moore, Human decision making based
on variations in internal noise: An EEG study, PloS one 7(2013), e68928. doi:10.1371/journal.pone.0068928.
[3] N. Ashjazadeh, R. Boostani, H. Ekhtiari, M. Emamghoreishi, M. Farrokhi, A. Ghanizadeh, G. Hatam, H. Hadianfard,
M. Lotfi and S.M.J. Mortazavi, Operationalizing Cognitive Science and Technologies’ Research and Development; The
“Brain and Cognition Study Group (BCSG)” Initiative from Shiraz, Iran, Basic and Clinical Neuroscience 2, 2014, p. 104.
[4] C. BenAbdelkader and Y. Yacoob, Statistical estimation of human anthropometry from a single uncalibrated image, Com-
putational Forensics (2008), 200–220.
[5] B.D. Burns, The effects of speed on skilled chess performance, Psychological Science 7(2004), 442–447. doi:10.1111/j.
0956-7976.2004.00699.x.
[6] E.C. Cieslik, K. Zilles, S. Caspers, C. Roski, T.S. Kellermann, O. Jakobs, R. Langner, A.R. Laird, P.T. Fox and S.B. Eick-
hoff, Is there “one” DLPFC in cognitive action control? Evidence for heterogeneity from co-activation-based parcellation,
Cerebral cortex 11 (2013), 2677–2689. doi:10.1093/cercor/bhs256.
[7] M.H. Connors, B.D. Burns and G. Campitelli, Expertise in complex decision making: The role of search in chess 70 years
after de Groot, Cognitive science 8(2011), 1567–1579. doi:10.1111/j.1551-6709.2011.01196.x.
[8] A. Delorme, T. Mullen, C. Kothe, Z.A. Acar, N. Bigdely-Shamlo, A. Vankov and S. Makeig, EEGLAB, SIFT, NFT,
BCILAB, and ERICA: New tools for advanced EEG processing, Computational intelligence and neuroscience 2011
(2011), 130714.
[9] P. Eisele, Judgment and decision-making: Experts’ and novices’ evaluation of chess positions, Perceptual and motor skills
1(2004), 237–248. doi:10.2466/pms.98.1.237-248.
[10] D.P. Forbes and F.J. Milliken, Cognition and corporate governance: Understanding boards of directors as strategic
decision-making groups, Academy of management review 3(1999), 489–505.
[11] J. Grinband, J. Hirsch and V.P. Ferrera, A neural representation of categorization uncertainty in the human brain, Neuron
5(2006), 757–763. doi:10.1016/j.neuron.2006.01.032.
[research-article] p. 13/14
UNCORRECTED PROOF
14 A. Alipour et al. / Please provide the short title
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
[12] D.H. Gustafson, K. Bosworth, R.P. Hawkins, E.W. Boberg and E. Bricker, CHESS: A computer-based system for provid-
ing information, referrals, decision support and social support to people facing medical and other health-related crises, in:
Proceedings Symposium on Computer Applications in Medical Care, 1992, pp. 161–165.
[13] H.R. Heekeren, S. Marrett, P.A. Bandettini and L.G. Ungerleider, A general mechanism for perceptual decision-making
in the human brain, Nature 7010 (2004), 859–862. doi:10.1038/nature02966.
[14] J. Jacobs, G. Hwang, T. Curran and M.J. Kahana, EEG oscillations and recognition memory: Theta correlates of memory
retrieval and decision making, Neuroimage 2(2006), 978–987. doi:10.1016/j.neuroimage.2006.02.018.
[15] L. Jollans, R. Whelan, L. Venables, O.H. Turnbull, M. Cella and S. Dymond, Computational EEG modelling of decision
making under ambiguity reveals spatio-temporal dynamics of outcome evaluation, Behavioural brain research 321 (2017),
28–35. doi:10.1016/j.bbr.2016.12.033.
[16] A. Korzeniewska, M. Ma´
nczak, M. Kami´
nski, K.J. Blinowska and S. Kasicki, Determination of information flow direction
among brain structures by a modified directed transfer function (dDTF) method, Journal of neuroscience methods 1
(2003), 195–207. doi:10.1016/S0165-0270(03)00052-9.
[17] S.P. Kyathanahally, A. Franco-Watkins, X. Zhang, V.D. Calhoun and G. Deshpande, A realistic framework for investigat-
ing decision making in the brain with high spatiotemporal resolution using simultaneous EEG/fMRI and joint ICA, IEEE
Journal of Biomedical and Health Informatics 3(2017), 814–825. doi:10.1109/JBHI.2016.2590434.
[18] T. Larsen and J.P. O’Doherty, Uncovering the spatio-temporal dynamics of value-based decision-making in the human
brain: A combined fMRI–EEG study, Phil Trans R Soc B 1655 (2014), 20130473. doi:10.1098/rstb.2013.0473.
[19] M. Mahmoud, Y. Liu, H. Hartmann, S. Stewart, T. Wagener, D. Semmens, R. Stewart, H. Gupta, D. Dominguez and
F. Dominguez, A formal framework for scenario development in support of environmental decision-making, Environmen-
tal Modelling & Software 7(2009), 798–808. doi:10.1016/j.envsoft.2008.11.010.
[20] T.R. Mullen, The Dynamic Brain: Modeling Neural Dynamics and Interactions from Human Electrophysiological Record-
ings, University of California, San Diego, 2014.
[21] F. Nazaraghaie, F. Torkamani, B. Kiani and M. Torabi-Nami, EEG-guided meditative training through geometrical ap-
proach: An interim analysis, Avicenna Journal of Phytomedicine (2015).
[22] M.P. Paulus, N. Hozack, L. Frank and G.G. Brown, Error rate and outcome predictability affect neural activation in
prefrontal cortex and anterior cingulate during decision-making, Neuroimage 4(2002), 836–846. doi:10.1006/nimg.2001.
1031.
[23] M.N. Shadlen and W.T. Newsome, Neural basis of a perceptual decision in the parietal cortex (area LIP) of the rhesus
monkey, Journal of neurophysiology 4(2001), 1916–1936.
[24] M. Sigman, P. Etchemendy, D.F. Slezak and G.A. Cecchi, Response time distributions in rapid chess: A large-scale
decision making experiment, Frontiers in neuroscience 4(2010), 60.
[25] K. Stanovich, R. West and R. Hertwig, Individual differences in reasoning: Implications for the rationality debate? – Open
Peer Commentary – The questionable utility of cognitive ability in explaining cognitive illusions, 2000.
[26] K.E. Stanovich and R.F. West, Individual differences in reasoning: Implications for the rationality debate?, Behavioral
and brain sciences 5(2000), 645–665. doi:10.1017/S0140525X00003435.
[27] R.J. Sternberg, The ability is not general, and neither are the conclusions, Behavioral and Brain Sciences 5(2000), 697–
698. doi:10.1017/S0140525X00573439.
[28] A. Thielscher and L. Pessoa, Neural correlates of perceptual choice and decision making during fear-disgust discrimi-
nation, The Journal of neuroscience: The official journal of the Society for Neuroscience 11 (2007), 2908–2917. doi:10.
1523/JNEUROSCI.3024-06.2007.
[29] H.L. Van Der Maas and E.-J. Wagenmakers, A psychometric analysis of chess expertise, The American journal of psy-
chology 118 (2005), 29–60.
[30] H.-J. Volke, P. Dettmar, P. Richter, M. Rudolf and U. Buhss, On-coupling and off-coupling of neocortical areas in chess
experts and novices as revealed by evoked EEG coherence measures and factor-based topological analysis – a pilot study,
Journal of Psychophysiology 16 (2002), 23–36. doi:10.1027//0269-8803.16.1.23.
[31] D. Zhang, R. Gu, L.S. Broster, Y. Jiang, W. Luo, J. Zhang and Y.J. Luo, Linking brain electrical signals elicited by current
outcomes with future risk decision-making, Frontiers in behavioral neuroscience 8(2014), 84.
[research-article] p. 14/14
... Subsequently, to find spectral signatures of top-down modulation in visual cortex, we looked at the changes in the power of beta band rhythms (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) in electrodes close to primary visual areas (i.e. O1, O2, P7, P8) when perception of binocular rivalry is matched with prerivalry stream motion direction. ...
... Additionally, in order to observe changes in interaction between higher and primary cortical areas, we exported the ICA pruned data from EEGLab and imported it to Neuroguide QEEG software [also explained in (16)] and calculated the phase lag between channels in each hemisphere at beta band. Then, we used a paired t-test (ttest.m on Matlab) to see if the phase lag between intra-hemispheric electrode pairs changes during successful predictions. ...
Preprint
Full-text available
Predictive coding theory suggests that conscious perception results from the interaction between top-down and bottom-up signals in the brain. However, the electrophysiological signatures of top-down predictions are not clear yet. Here, we cued subjects to expect a certain perceptual state in a binocular rivalry task and quantitatively analyzed their EEG signals during the cue period and prior to binocular rivalry task. We found that when predictions can successfully influence the perception of the rivalrous stimuli, the power of beta band rhythms increases in primary visual cortices and beta band phase lag in a frontal-visual homologous channel pair was notably diminished. Building upon earlier works, our findings suggest that beta rhythm is potentially considered as a signature of top-down communication in the brain.
... In 2017, Alipour et al. published a case report on the use of quantitative electroencephalography in an eight-year-old boy with substantial chess training while he was playing chess against a computer. The authors observed a lower power spectral density in higher frequency bands in the right hemisphere in decision making during the game 30 . ...
... In 2017, Alipour et al. published a case report on the use of quantitative electroencephalography in an eight-year-old boy with substantial chess training while he was playing chess against a computer. The authors observed a lower power spectral density in higher frequency bands in the right hemisphere in decision making during the game 30 . ...
Article
Full-text available
The chess game comprises different domains of cognitive function, demands great concentration and attention and is present in many cultures as an instrument of literacy, learning and entertainment. Over the years, many effects of the game on the brain have been studied. Seen that, we reviewed the current literature to analyze the influence of chess on cognitive performance, decision-making process, linking to historical neurological and psychiatric disorders as we describe different diseases related to renowned chess players throughout history, discussing the influences of chess on the brain and behavior.
Article
Full-text available
Functional significance of the neural oscillations has been debated since long. In particular, oscillations have been suggested to play a major role in formation of communication channels between brain regions. It has been previously suggested that gamma coherence increases during communication between hemispheres when subjects perceive a horizontal motion in Stroboscopic Alternative Motion (SAM) stimulus. In addition, disruption of this coherence may change the horizontal perception of SAM. In this study, we investigated the changes of Cross-Frequency Coupling (CFC) in EEG signals from parietal and occipital cortices during horizontal and vertical perception of SAM. Our results suggested that while the strength of CFC in parietal electrodes showed no significant change, CFC in P3-P4 electrode-pair demonstrated a significant correlation during horizontal perception of SAM. Therefore, the CFC between theta- and gamma-band oscillations seems to be correlated with changes in functional interactions between brain regions. Accordingly, we propose that in addition to gamma coherence, CFC is perhaps another neurophysiological mechanism involved in neural communication.
Thesis
Full-text available
“The mind is the music that neural networks play.” This quote from computational neurobiologist T.J. Sejnowski underscores a growing scientific consensus that studying the structure and function of vast networks of connections between brain regions is essential to understanding cognitive and affective state maintenance, sensorimotor information processing and control, etiologies and remedies for numerous neuropathologies, as well as a host of other facets of our conscious (and non-conscious) experience. Towards this goal, an ongoing challenge lies in identifying – in vivo in humans – spatiotemporal cortical network dynamics, at the level of individuals and groups, across experimental task conditions, and at the level of single trials. In this dissertation, I first introduce the Source Information Flow Toolbox (SIFT), a novel open-source software package for identification of neuronal dynamics and causal interactions in electrophysiological source and sensor data. The software integrates with the widely used EEGLAB analysis suite, addressing a need for robust tools for identifying single- and multi-trial multivariate brain network dynamics across time, frequency, anatomical source location, and subjects. I then introduce and assess two new methods (Measure Projection Analysis and Multi-view Hierarchical Bayesian Learning) for statistical analysis of source-level dynamics (including connectivity) across groups of subjects in the presence of missing data. The remaining chapters focus on applications of dynamical modeling approaches in SIFT to open problems within the fields of cognitive neuroscience, clinical neuroscience and neuroengineering. I first present three studies examining single-trial time-varying spatiotemporal network dynamics underlying generation and maintenance of epileptic seizures. Next I present a case study examining the effect of visual feedback on an occipito-parietal-motor network in freezing-of-gait in patients with Parkinson’s disease. The final chapters focus on new directions in neuroengineering and brain-computer interfaces (BCI) leveraging neural dynamical system identification. We first review the history and state of the BCI field and summarize important new directions in BCI design. I then present a novel system for real-time mobile brain imaging, artifact rejection, neuronal system identification, and cognitive state prediction, and demonstrate its application in predicting response error commission from cortical network dynamics using a new high-density mobile dry EEG system. http://search.proquest.com/docview/1619637939
Article
Full-text available
While there is a growing body of functional magnetic resonance imaging (fMRI) evidence implicating a corpus of brain regions in value-based decision-making in humans, the limited temporal resolution of fMRI cannot address the relative temporal precedence of different brain regions in decision-making. To address this question, we adopted a computational model-based approach to electroencephalography (EEG) data acquired during a simple binary choice task. fMRI data were also acquired from the same participants for source localization. Post-decision value signals emerged 200 ms post-stimulus in a predominantly posterior source in the vicinity of the intraparietal sulcus and posterior temporal lobe cortex, alongside a weaker anterior locus. The signal then shifted to a predominantly anterior locus 850 ms following the trial onset, localized to the ventromedial prefrontal cortex and lateral prefrontal cortex. Comparison signals between unchosen and chosen options emerged late in the trial at 1050 ms in dorsomedial prefrontal cortex, suggesting that such comparison signals may not be directly associated with the decision itself but rather may play a role in post-decision action selection. Taken together, these results provide us new insights into the temporal dynamics of decision-making in the brain, suggesting that for a simple binary choice task, decisions may be encoded predominantly in posterior areas such as intraparietal sulcus, before shifting anteriorly.
Article
Full-text available
4.6 Effect of Electromagnetic Radiation on Cognitive Functions Cognitive Functions One of the other areas of interest in BCSG-Shiraz is the study of the possible detrimental effects of ionizing and non-ionizing radiation in microwave frequency range on brain health and cognitive functions. Over the past years, the ‘Ionizing and Non-ionizing Radiation Protection Research Center’ at SUMS has focused on studying the health effects of exposure of laboratory animals and humans to some common and/or occupational sources of electromagnetic fields such as mobile phones (Mortazavi, Rouintan et al. 2012, Mortazavi 2013, Mortazavi, Mosleh-Shirazi et al. 2013, Mortazavi, Taeb et al. 2013). While radar equipments deal with high-power radiofrequency waves by producing high-voltage and high frequency alternating electrical current, some people are routinely exposed to the pulsed high- frequency electromagnetic fields. Critical job holders in navigation, aviation, national defense, and weather forecasting are shown to be at increased risk (Mortazavi, Taeb et al. 2013). However, with the popular use of mobile phones and MRI (Mortazavi, Mahbudi et al. 2011), the possible hazards of the electromagnetic radiation is not limited to the above groups. Recent investigations on the health concerns following occupational exposure to radiofrequency radiation has not only supported the induction of toxic effects at cellular level (Deng, Zhang et al. 2013) but also suggested adverse effect of such exposures on short-term memory and reaction time as exemplary domains of cognitive functions (Foroozandeh, Derakhshan-Barjoei et al. 2013, Mortazavi, Taeb et al. 2013). Findings have also indicated that the visual reaction time (VRT) of university students is significantly affected by a 10 min exposure to electromagnetic fields emitted by a mobile phone (Mortazavi, Rouintan et al. 2012). Furthermore, it has been shown that occupational exposures to microwave radar radiations may decrease reaction time in radar workers (Mortazavi, Taeb et al. 2013). BCSG-Shiraz would pursue continued research on further possible effects of electromagnetic radiation on brain and cognitive functions (http://crrs.sums.ac.ir/ en/index.html).
Article
Full-text available
The experience of current outcomes influences future decisions in various ways. The neural mechanism of this phenomenon may help to clarify the determinants of decision-making. In this study, thirty-nine young adults finished a risky gambling task by choosing between a high- and a low-risk option in each trial during electroencephalographic data collection. We found that risk-taking strategies significantly modulated mean amplitudes of the event-related potential (ERP) component P3, particularly at the central scalp. The event-related spectral perturbation and the inter-trial coherence measurements of the independent component analysis (ICA) data indicated that the "stay" vs. "switch" electrophysiological difference associated with subsequent decision-making was mainly due to fronto-central theta and left/right mu independent components. Event-related cross-coherence results suggested that the neural information of action monitoring and updating emerged in the fronto-central cortex and propagated to sensorimotor area for further behavior adjustment. Based on these findings of ERP and event-related oscillation (ERO) measures, we propose a neural model of the influence of current outcomes on future decisions.
Article
We recorded the activity of single neurons in the posterior parietal cortex (area LIP) of two rhesus monkeys while they discriminated the direction of motion in random-dot visual stimuli. The visual task was similar to a motion discrimination task that has been used in previous investigations of motion-sensitive regions of the extrastriate cortex. The monkeys were trained to decide whether the direction of motion was toward one of two choice targets that appeared on either side of the random-dot stimulus. At the end of the trial, the monkeys reported their direction judgment by making an eye movement to the appropriate target. We studied neurons in LIP that exhibited spatially selective persistent activity during delayed saccadic eye movement tasks. These neurons are thought to carry high-level signals appropriate for identifying salient visual targets and for guiding saccadic eye movements. We arranged the motion discrimination task so that one of the choice targets was in the LIP neuron's response field (RF) while the other target was positioned well away from the RF. During motion viewing, neurons in LIP altered their firing rate in a manner that predicted the saccadic eye movement that the monkey would make at the end of the trial. The activity thus predicted the monkey's judgment of motion direction. This predictive activity began early in the motion-viewing period and became increasingly reliable as the monkey viewed the random-dot motion. The neural activity predicted the monkey's direction judgment on both easy and difficult trials (strong and weak motion), whether or not the judgment was correct. In addition, the timing and magnitude of the response was affected by the strength of the motion signal in the stimulus. When the direction of motion was toward the RF, stronger motion led to larger neural responses earlier in the motion-viewing period. When motion was away from the RF, stronger motion led to greater suppression of ongoing activity. Thus the activity of single neurons in area LIP reflects both the direction of an impending gaze shift and the quality of the sensory information that instructs such a response. The time course of the neural response suggests that LIP accumulates sensory signals relevant to the selection of a target for an eye movement.
Article
Complex human cognition, such as decision-making under ambiguity, is reflected in dynamic spatio-temporal activity in the brain. Here, we combined event-related potentials with computational modelling of the time course of decision-making and outcome evaluation during the Iowa Gambling Task. Measures of choice probability generated using the Prospect Valence Learning Delta (PVL-Delta) model, in addition to objective trial outcomes (outcome magnitude and valence), were applied as regressors in a general linear model of the EEG signal. The resulting three-dimensional spatio-temporal characterization of task-related neural dynamics demonstrated that outcome valence, outcome magnitude, and PVL-Delta choice probability were expressed in distinctly separate event related potentials. Our findings showed that the P3 component was associated with an experience-based measure of outcome expectancy.
Article
Human decision-making is a multidimensional construct, driven by a complex interplay between external factors, internal biases and computational capacity constraints. Here we propose a layered approach to experimental design in which multiple tasks - from simple to complex - with additional layers of complexity introduced at each stage, are incorporated for investigating decision-making. This is demonstrated using tasks involving intertemporal choice between immediate and future prospects. Previous functional magnetic resonance imaging (fMRI) and electroencephalographic (EEG) studies have separately investigated the spatial and temporal neural substrates, respectively, of specific factors underlying decision making. In contrast, we performed simultaneous acquisition of EEG/fMRI data and fusion of both modalities using joint independent component analysis (jICA) such that: (i) the native temporal/spatial resolutions of either modality is not compromised and (ii) fast temporal dynamics of decision-making as well as involved deeper striatal structures can be characterized. We show that spatio-temporal neural substrates underlying our proposed complex intertemporal task simultaneously incorporating rewards, costs and uncertainty of future outcomes can be predicted (using a linear model) from neural substrates of each of these factors, which were separately obtained by simpler tasks. This was not the case for spatial and temporal features obtained separately from fMRI and EEG respectively. However, certain prefrontal activations in the complex task could not be predicted from activations in simpler tasks, indicating that the assumption of pure insertion has limited validity. Overall, our approach provides a realistic and novel framework for investigating the neural substrates of decision making with high spatio-temporal resolution.
Article
UI - 21904911DA - 20020321IS - 1053-8119LA - engPT - Journal ArticleCY - United StatesSB - IM