Sharif Neuroscience Symposium 2023

Symposium Information and Abstracts Booklet

March 2023
Contents

Preface iv

Sponsors v

I Essential Information 1
Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Iran Standard Time (IRST) . . . . . . . . . . . . . . . . . . . . . . . 3
Eastern Standard Time (EST) . . . . . . . . . . . . . . . . . . . . . . 4
Greenwich Mean Time (GMT) . . . . . . . . . . . . . . . . . . . . . . 5
Central European Time (CET) . . . . . . . . . . . . . . . . . . . . . 6
Symposium Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Invited Speakers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
At a Glance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

II Invited Speeches 9
Thursday Talks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Scalar reward signals for vectorial, multi-component choice options in mon-
key orbitofrontal cortex (Wolfram Schultz ) . . . . . . . . . . . . . . . 10
Flexibility in face of uncertainty: behavioural and neural adjustments to
uncertainty across species (Alireza Soltani ) . . . . . . . . . . . . . . . 10
Climate Change and the Social Brain (Michael L. Platt) . . . . . . . . . . 11
What Art can tell us about the Brain (Margaret S. Livingstone) . . . . . . 11
Understanding Opioid Reward (Margolis Elyssa) . . . . . . . . . . . . . . . 12
Friday Talks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Rhythm processing in the cerebellum and basal ganglia (Masaki Tanaka) . 13
How activity correlations between neurons shape encoding and readout of
sensory information (Stefano Panzeri ) . . . . . . . . . . . . . . . . . 13
Causality for Neuroscience and beyond (Konrad Kording) . . . . . . . . . 14
Neural mechanisms of reward-guided decision-making and affect (Peter
Rudebeck ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Dopamine and Value over Time and Space (Joshua Berke) . . . . . . . . . 14

III Accepted Abstracts 15

Oral Presentations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

i
Investigating the relationship between hippocampal ‘reward-cells’ and learn-
ing (Mohammad Yaghoubi, Andrés Nieto-Posadas, Émmanuel Wil-
son, Thomas Gisiger, Sylvain Williams, Mark P. Brandon) . . . . . . 16
Prefrontal cortex simultaneously encodes interactive adaptive processes
with slow and fast dynamics in value learning ( Reza Hashemirad,
Mojtaba Abbaszadeh, Ali Ghazizadeh ) . . . . . . . . . . . . . . . . . 17
Applications of wavelet transform in classification of local field potential
recorded from the rat brain in conditioned place preference paradigm
( AmirAli Kalbasi, Shole Jamali, Mahdi Aliyari Shoorehdeli Abbas
Haghparast ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Motor Planning Determined Lateralization more than Sensory Information
( Tahmineh A Koosha, Abdol-Hosssein Vahabie2, Babak Najar Araabi ) 19
Design, synthesis, and biological evaluation of novel indanone derivatives
as cholinesterase inhibitors for potential use in Alzheimer’s disease
( Aysan Etemadi, Salar Hemmati, Mohammad Shahrivar-Gargari,
Yasaman Tamaddon Abibiglue, Ahad Bavili, Siavoush Dastmalchi ) . 20
Locomotion is associated with straighter neural trajectories for natural
movies in mouse visual cortex ( Xingyu Zheng1, Maxwell Ruckstuhl,
Mohammad Yaghoubi ) . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Introduction of a phase decoder for decoding the content of working mem-
ory ( Shakoor Mohammadi, Zahra Bahmani ) . . . . . . . . . . . . . 22
Cortical Correlates of Discounted Cumulative Reward ( Nikan Amirkhani,
Tandis Salem, Aryan Zoroufi, Marc Coutanche ) . . . . . . . . . . . . 23
Poster Presentations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Mental representation of regular past tense forms in first language and
second language processing ( Ebrahim Safaie ) . . . . . . . . . . . . . 24
Detection of Patients with Multiple Sclerosis Disorder Based on EEG Sig-
nal ( Ali Abedi, Gholamreza Moradi ) . . . . . . . . . . . . . . . . . . 25
Understanding how individuals’ risk strategy evolves in light of hierarchical
Gaussian filtering ( AmirHossein Tehrani Safa, Reza Ghaderi, Atiye
Sarabi-Jamab ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
The evaluation of excitatory and inhibitory connections in developmental
in vitro cell cultures ( Leila Rezayat, Mohammad-Reza A. Dehaqani,
Ehsan Rezayat, Amir M Sodagar ) . . . . . . . . . . . . . . . . . . . 27
Low-Intensity Transcranial Magnetic Stimulation shows potential for in-
tervention in ADHD ( Elahe Khosrowabadi, Firouze Mahjoub Navaz,
Jalil Gholizadeh Soltani, Farhad Soleymani ) . . . . . . . . . . . . . . 28
Changes in phase amplitude coupling of EEG as a potential mechanism
to perceive emotional contents of the music ( Zohreh Rahimi, Zhaleh
Mohammad Alipour, Reza Khosrowabadi ) . . . . . . . . . . . . . . . 29
Facial keypoint detection by self-supervised multi-task learning in thermal
images ( Poorya Aghaomidi, Zahra Bahmani, Sadaf Aram ) . . . . . 30
Reduction of mental fatigue by electrical stimulation of pilot’s brain (
Fahimeh Nadalizadeh, Zahra Bahmani ) . . . . . . . . . . . . . . . . 31
The Effect of Feature Normalization on Motor Imagery Task Classification
( Hamed Nazemi, Mohammad Amin Ebrahimzadeh, Alireza Taheri ) . 32

ii
 The effect of repeated and increasing injection of dexamethasone on mor-
phological changes and apoptosis rate in mouse cerebellar cortex (
Nooshin Goudarzi, Seyed amir Hosseini, Farzad Rajaie ) . . . . . . . 33
The way to the light: What are the pathways that mediate blindsight?
A systematic review ( MR.Balali, S.M.Hosseini, M. Khanboloki, M.
Mohammadi, AM.Khorshidvand, MT.Joghataei ) . . . . . . . . . . . 34

IV People 35
Scientific Committee 36
Scientific Chairs: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Reviewers: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Executive Team 37

iii
Preface

The fifth Sharif Neuroscience Symposium (SNS 2023) which is being held on 2nd-3rd
of March 2023 (11-12th of Esfand 1401) is an annual event that covers key advances in
cognitive, computational and systems neuroscience as well as neuro-engineering by inviting
experts in the fields from across the world to share their new findings and perspectives in
a collegial atmosphere.
This year, the symposium consists of a series of 10 invited talks by internationally
renowned experts across various research areas including cognitive neuroscience, learning
and memory, decision making, sensory processing, vision sciences, motor control and com-
putational neuroscience. In addition, the symposium received abstract submissions from
various research groups in Iran and internationally. Following, peer review by the SNS
2023 scientific committee, 19 abstracts were selected and 8 of them were invited for oral
presentations.
In addition to the main symposium, three satellite workshops covering concepts and
techniques in fMRI, Artificial Intelligence and Deep Learning, and Introduction to Net-
works, Neurons, and Learning were scheduled and held in the weeks leading up to or
following the main symposium. The symposium and workshops were warmly received by
the neuroscience community and were attended by over 400 participants overall.
This booklet provides comprehensive information about SNS 2023, including the de-
tailed meeting schedule, list of symposium sponsors, abstracts of invited speakers, and
all accepted abstracts. We would like to extend our gratitude to the executive commit-
tee, comprising students from the electrical engineering department at Sharif University
of Technology, for their tireless efforts leading up to the symposium, which ensured its
success. This year, the symposium is being held in a hybrid format, making it easily
accessible to the global research community. We hope that the symposium will further
ignite enthusiasm for neuroscience among researchers both locally and internationally, and
look forward to seeing its positive scientific impact resonate for months and years to come
among students and faculty in the field.

Ali Ghazizadeh, PhD

Symposium Chair, SNS 2023
Faculty, Electrical Engineering Department, Sharif University of Technology
Dean, School of Cognitive Sciences, IPM

iv
Sponsors

v
 Part I

Essential Information

1
Schedule

2
Iran Standard Time (IRST)

3
Eastern Standard Time (EST)

4
Greenwich Mean Time (GMT)

5
Central European Time (CET)

6
Symposium Links
The fifth symposium will be held in a hybrid format. The remote section consists of a
main hall in which all invited speeches and selected oral presentations take place. Other
oral presentations take place in two parallel online rooms. Panels take place in panel
rooms separately after the talk of each invitee. All classes can be accessed using the same
username and password provided to each registered participant. Please note that each
username is only permitted to login from only one device at a time.

• Main Hall:
https://vc.sharif.edu/ch/sns2023-main

• Panel:
https://vc.sharif.edu/ch/sns2023-panel

• Parallel Room I:
https://vc.sharif.edu/ch/sns2023-parallel-1

• Parallel Room II:

https://vc.sharif.edu/ch/sns2023-parallel-2

7
Invited Speakers
• Joshua Berke, Professor, Department of Neurology and Psychiatry, UCSF
• Margolis Elyssa, Professor, UCSF Weill Institute for Neurosciences
• Konrad Kording, Professor, University of Pennsylvania
• Margaret S. Livingstone, Professor, Harvard Medical School
• Stefano Panzeri, Professor, Department of Excellence for Neural Information Pro-
cessing, University Medical Center Hamburg-Eppendorf
• Michael L. Platt, Professor, University of Pennsylvania
• Peter Rudebeck, Professor, Icahn School of Medicine at Mount Sinai
• Wolfram Schultz, Professor, University of Cambridge
• Alireza Soltani, Professor, Dartmouth College, NH
• Masaki Tanaka, Professor, Hokkaido University

At a Glance

8
 Part II

Invited Speeches

9
Thursday Talks

Scalar reward signals for vectorial, multi-component choice

options in monkey orbitofrontal cortex March 2nd
10:30-11:30
Wolfram Schultz (IRST),
University of Cambridge Main Hall

Rewards contain multiple components. The question arises how such vectorial re-
wards are represented by single-dimensional, scalar neuronal signals that are suitable for
economic decision-making. Revealed Preference Theory provides formalisms for inves-
tigating choices between such vectorial rewards in a well-structured manner, including
convenient graphs. During stochastic choice between multi-component rewards containing
the same two juice components with different quantities, we identified neuronal signals
for vectorial, multi-component rewards in the orbitofrontal cortex (OFC) of monkeys. A
scalar signal integrated the values from all reward components in the structured manner of
the Theory. The signal followed the behavioral indifference curves within their confidence
limits, was indistinguishable between differently composed but equally preferred rewards,
predicted choice, demonstrated reward-specific satiety, and complied with an optimality
axiom. Further, distinct signals in specific OFC neurons coded the reward components
separately but followed indifference curves as a population. These data demonstrate how
scalar neural signals can represent vectorial, multi-component rewards.

Flexibility in face of uncertainty: behavioural and neural

adjustments to uncertainty across species March 2nd
14:00-15:00
Alireza Soltani (IRST),
Dartmouth College, NH Main Hall

Survival depends on the ability of an animal to adapt and adjust to changes in the
environment. This is an inherently challenging task because the environment and thus
the feedback it provides to the animal (e.g., reward feedback) can change in many ways
and at different timescales. These include fluctuations due to the probabilistic nature of
reward feedback, changes in reward probabilities over time, and changes in how reward
can be obtained by selecting certain actions or choice options, creating different forms of
uncertainty each requiring a different type of response and adjustment. In this talk, I
present recent methods we have developed for quantifying and studying behavioural and
neural adjustments to different types of uncertainty in the reward environment. I show
how these methods can be used to reveal computational and neural mechanisms underlying
flexibility in learning and decision making across mammals.

10
 Climate Change and the Social Brain March 2nd
15:30-16:30
Michael L. Platt (IRST),
University of Pennsylvania Main Hall

Deeper and more numerous social connections promote health, well-being, survival,
and even financial success. By the same token, social exclusion and the loss of social
partners result in feelings similar to physical pain. In my talk, I will discuss our work
aimed at defining the biological mechanisms that mediate our ability and desire to con-
nect and the impact of these capacities on resilience. We leverage a unique 17-year field
study of thousands of free- ranging rhesus macaques and a biobank including genomic,
neuroanatomical, and brain transcriptomic data, collected before and after a major cy-
clone, and parallel neurophysiological and pharmacological work in the laboratory. We
find that monkeys who have more friends are more successful and show increased gene
expression and structural connectivity within the social brain network, particularly the
middle superior temporal sulcus (mSTS). Neurophysiological studies in our lab show mSTS
neurons encode a rich array of information necessary to guide dynamic, evolving social
interactions. Pharmacologically inactivating mSTS impairs the sophistication of social
interactions. In the field, monkeys who lived through the cyclone showed upregulation
of aging-related genes in immune and inflammatory pathways, accompanied by physical
decline. Behaviorally, monkeys responded to the acute and chronic stress of the cyclone
by becoming more social and less aggressive, and monkeys who made more social con-
nections had higher survivorship after the hurricane, as did their offspring. Our studies
data indicate that biological predispositions for social connections shape resilience to the
stresses of life, including catastrophic disruptions due to climate change.

What Art can tell us about the Brain March 2nd

17:00-18:00
Margaret S. Livingstone (IRST),
Harvard Medical School Main Hall

Artists have been doing experiments on vision longer than neurobiologists. Some
major works of art have provided insights as to how we see; some of these insights are
so fundamental that they can be understood in terms of the underlying neurobiology.
For example, artists have long realized that color and luminance can play independent
roles in visual perception. Picasso said, “Colors are only symbols. Reality is to be found
in luminance alone.” This observation has a parallel in the functional subdivision of our
visual systems, where color and luminance are processed by the evolutionarily newer,
primate-specific What system, and the older, colorblind, Where (or How) system. Many
techniques developed over the centuries by artists can be understood in terms of the
parallel organization of our visual systems. I will explore how the segregation of color and
luminance processing are the basis for why some Impressionist paintings seem to shimmer,
why some op art paintings seem to move, some principles of Matisse’s use of color, and
how the Impressionists painted “air”. Central and peripheral vision are distinct, and I will
show how the differences in resolution across our visual field make the Mona Lisa’s smile
elusive, and produce a dynamic illusion in Pointillist paintings, Chuck Close paintings,
and photomosaics. I will explore how artists have figured out important features about
how our brains extract relevant information about faces and objects, and I will discuss
why learning disabilities may be associated with artistic talent.

11
 Understanding Opioid Reward March 2nd
18:30-19:30
Margolis Elyssa (IRST),
UCSF Weill Institute for Neurosciences Main Hall

Increased dopamine release from the axon terminals of ventral tegmental area (VTA)
neurons in the nucleus accumbens (NAc) is reinforcing and increases motivated behaviors,
while decreases in NAc dopamine diminish motivation. Opioid drugs such as morphine
and fentanyl primarily activate the mu opioid receptor (MOR) to produce their rewarding
effects. Activation of MORs in the VTA is both sufficient to produce reward and required
for the rewarding effects of systemically administered opioids, making it a key brain region
for understanding opioid reward. While dopamine has been the major focus of investiga-
tions into the function of the VTA, in opioid-naı̈ve animals VTA mediated opioid reward
is dopamine-independent. These observations raise the question: which neural circuits un-
derly the motivational effects of VTA opioid receptor activation? In the canonical model
MOR activation excites dopamine neurons indirectly through decreasing the activity of
inhibitory inputs. However, we found direct MOR postsynaptic excitation in a subset of
VTA neurons, including both dopamine and non-dopamine neurons. These postsynaptic
actions are more potent than disinhibition and vary depending upon the projection target
of the VTA neuron. We have also found that kappa opioid receptors, whose activation is
aversive in the VTA, also control subpopulations of VTA dopamine neurons with specific
projection targets. Together these findings show that opioids differentially control parallel
neural circuits through the VTA that impact different aspects of behavior.

12
Friday Talks

Rhythm processing in the cerebellum and basal ganglia March 3rd

8:30-9:30
Masaki Tanaka (IRST),
Hokkaido University Main Hall

Events occurring in regular cycles cause rhythm perception, often accompanied by syn-
chronized movements such as tapping or dancing. When perceiving rhythm, we anticipate
the occurrence of the next stimulus, focus our attention on that moment, and in some
cases prepare for movement. Previous studies have shown that the cerebellum and basal
ganglia are involved in temporal information processing, but little is known about their
specific roles in rhythm perception. We examined neuronal activity in the cerebellar nuclei
and the striatum in monkeys trained in behavioral tasks that require sensory prediction of
regularly repeated visual stimuli, or movements in synchrony with the rhythm. Our results
suggest that in addition to motor control, the cerebellum generates internal models that
predict upcoming sensory events, while the striatum may represent internalized rhythms
as periodic motor preparations.

How activity correlations between neurons shape encoding

and readout of sensory information March 3rd
12:30-13:30
Stefano Panzeri (IRST),
Department of Excellence for Neural Information Processing, University Medical Center Main Hall
Hamburg-Eppendorf

The collective activity of a population of neurons is critical for many brain functions.
A fundamental question is how activity correlations between neurons affect how neural
populations process information. We present a theory, built on the analyses of simultane-
ous recordings of activity of populations on neurons either in sensory or posterior parietal
cortical areas, of how such correlations serve multiple functions performed by neural pop-
ulations, including shaping the encoding of information in population codes, generating
codes across multiple timescales, and facilitating information transmission to and readout
by downstream brain areas to guide behavior. Here, we review this theory and we further
present ideas on how to combine large-scale simultaneous recordings of neural populations,
computational models, analyses of behavior, optogenetics, and anatomy to unravel how
the structure of correlations might be optimized to serve multiple functions.

13
 Causality for Neuroscience and beyond March 3rd
16:30-15:30
Konrad Kording (IRST),
University of Pennsylvania Main Hall

As scientists, we often ask how something works. What we usually mean with that is
that we want to know how one aspect of the world (say, one neuron) affects another aspect
of the world (say, another neuron). Dr. Kording will give an intuition of the relevant prob-
lems and approaches. Focusing on quasi-experimental approaches and machine learning,
he will give an overview of how to broaden the scope of meaningful causal techniques.

Neural mechanisms of reward-guided decision-making and

affect March 3rd
16:30-17:30
Peter Rudebeck (IRST),
Icahn School of Medicine at Mount Sinai Main Hall

How do decide what to pursue? How do the outcomes of our choices influence our
emotional state? In our daily lives, our brains are constantly having to learn and update
the costs and benefits associated with different available courses of action in order to
guide our decisions and control our affective state. In the first part of my seminar, I will
discuss work where we have investigated how the brain computes the costs and benefits of
different options based on the probability and type of outcomes that can be received. Using
a combination of functional neuroimaging, pathway-specific chemogenetics, and neural
recordings in macaques, this work has pin-pointed two parts of the prefrontal cortex, the
ventrolateral and orbital prefrontal cortex, and their connections with the amygdala as
being critical for these functions. In the second part of my seminar, I will detail a series of
experiments where we have investigated the functions of the subcallosal anterior cingulate
cortex (ACC), an area that is heavily implicated in the pathophysiology of depression.
In this work we show that the subcallosal ACC signals rewards with temporally specific
codes and deep brain stimulation of this area modulates brain-wide circuits that control
affect. This work has begun to reveal the specific brain circuits and mechanisms involved
in decision-making and affect.

Dopamine and Value over Time and Space March 3rd

18:30-19:30
Joshua Berke (IRST),
Department of Neurology and Psychiatry, UCSF Main Hall

I will describe results from several recent experiments designed to better understand
dopamine signals and their roles in learning and motivation. I will show how dopamine
responses to reward-predictive cues signals reward prediction errors (RPEs) – consistent
with conventional theory – but that the underlying reward prediction differs between target
regions in a manner consistent with distinct scales of temporal processing. Furthermore,
dopamine ramps up as animals approach reward, as if encoding reward predictions (values),
rather than RPE. I will present results from a novel dynamic maze task, revealing how
dopamine value signals are updated using both internal knowledge and propagation across
space. Finally I will show our latest findings on how this dopamine release interacts with,
is locally sculpted by, local acetylcholine release within striatal microcircuits.

14
 Part III

Accepted Abstracts

15
Oral Presentations

Investigating the relationship between hippocampal

‘reward-cells’ and learning March 2nd,
9:00 IRST,
Mohammad Yaghoubi1,2 , Andrés Nieto-Posadas1 , Émmanuel Wilson1 , Thomas Gisiger1 , Parallel room
Sylvain Williams1,2 , Mark P. Brandon1,2 1
1 Department of Psychiatry, Douglas Hospital Research Centre, McGill University, 6875
Boulevard LaSalle, Verdun, QC H4H 1R3, Canada
2 Integrated Program in Neuroscience, McGill University, 1033 Pine Ave. W., Montreal,

QC H3A 1A1, Canada

The hippocampus is known to form and maintain a place code of the environment
for flexible navigation [O’Keefe and Dostrovsky 1971]. Several studies have shown that
the structure of this spatial code over-represents rewarded locations and supports goal-
directed navigation [Markus et al 1995, Dupret et al 2010]. However, these studies have
been conducted mostly in well-trained animals and have focused primarily on hippocampal
coding before and after learning the rewarded location. Here, we record the activity of
hundreds of hippocampal neurons as mice learn to perform a touchscreen-based memory-
dependent navigation task from the early days of training up to the point when the mouse
masters the task. In the task, a white square is presented in one of five positions on a
touchscreen. A nose poke to this square makes it disappear and starts a delay period (2-8
sec). Following the delay, two white squares are displayed, and the mouse must choose
the non-matching square to receive a reward. Our analysis shows that the proportion
of recruited reward cells is increased in the early days of learning followed by a plateau
towards later days, reflecting the contribution of reward coding during the early stages of
learning. Additionally, we observe that the activity of reward cells is not limited to reward
consumption, and is expanded to other parts of the task. Reward cells carry less spatial
tuning compared to the rest of the hippocampal population; instead, they carry more
task-related information. Altogether, hippocampal reward cells may provide an efficient
representation to downstream cortical areas that could facilitate learning.

16
 Prefrontal cortex simultaneously encodes interactive
adaptive processes with slow and fast dynamics in value
learning March 2nd,
9:00 IRST,
Reza Hashemirad2 , Mojtaba Abbaszadeh2 , Ali Ghazizadeh1,2 Parallel
1 Bio-intelligence
Research Unit, Sharif Brain Center, Electrical Engineering Department, Room 2
Sharif University of Technology, Tehran, Iran
2 School of Cognitive Sciences, Institute for Research in Fundamental Sciences (IPM),

Tehran, Iran

Many objects around us have stable values but every so often some values can change from
high to low and vice versa requiring flexible decision making. Previous works suggested a
segregated network for stable and flexible value processes in the caudate nucleus [Kim and
Hikosaka 2013] but the neural circuitry underlying the interaction of these two processes
is not known. Given the role of ventrolateral prefrontal cortex (vlPFC) in value learning
and its connectivity with the caudate, we recorded single-unit activity in this region in two
macaque monkeys who performed a paradigm in which stable object values underwent an
abrupt reversal using an object reward learning task. A free-viewing task used to assess
value dependent gaze bias revealed a gaze bias switch immediately after reversal training.
Interestingly, this switch was abolished after ¿15 minutes testing and reverted back toward
the initial stable bias. This snap-back phenomena can be predicted by assuming two
interactive processes one with fast learning and forgetting and the other with slow learning
and forgetting. Notably, vlPFC neurons firing encoded these slow and fast processes
multiplexed in early and late component of its firing after object onset, respectively. While
the early part of firing consistently signaled the old object values, the late component
showed a change after reversal which faded in the later testing. These results implicate
vlPFC as the plexus for interaction of stable and flexible object values in the brain and
reveals temporal multiplexing of information as the mechanism of such interaction.

17
 Applications of wavelet transform in classification of local
field potential recorded from the rat brain in conditioned
place preference paradigm March 2nd,
9:30 IRST,
AmirAli Kalbasi1 , Shole Jamali2 , Mahdi Aliyari Shoorehdeli3 Abbas Haghparast4 Parallel
1 Department of Mechatronics, Faculty of Electrical Engineering, K. N. Toosi University Room 1
of Technology, Tehran, Iran
2 Neuroscience Research Center, School of Medicine, Shahid Beheshti University of

Medical Sciences, Tehran, Iran

This study aimed to investigate the multi-label classification of Local Field Potential (LFP)
recordings from the hippocampus (HIP) and nucleus accumbens (NAc) of the rat brain
in response to different types of reward using the conditioned place preference (CPP)
paradigm. Rats were conditioned to saline, morphine, and food rewards and LFP record-
ings were performed simultaneously from HIP and NAc in pre- and post-test of CPP. LFP
data were labeled into four categories: treatment groups (saline, morphine, and food),
test phases (pre and post), recording channels (HIP and NAc), and position of the ani-
mal in CPP chambers. Features were extracted from LFP signal using continuous wavelet
transform (CWT), wavelet coherence and wavelet scattering, and data were classified using
decision trees, multilayer perceptron and support vector machines. HIP- or NAc-CWT and
HIP-NAc WCOH based-classification results showed that in the Food, HIP and HIP-NAc
features had the best accuracy for classifying data into specific CPP chambers, whereas,
in Morphine, the accuracy of NAc features was the best. With wavelet scattering-based
features, data is classified into treatment groups, test phases, and recording channels with
80% accuracy. Classifying LFP data in the Food -post-test- HIP (99.75%) and Morphine-
post-test- NAc (99.58%) had the best accuracy among all classification categories. This
study shows that NAc activity is more important for the induction of CPP in morphine
reward, while HIP and HIP-NAc connectivity are more essential for inducing food re-
ward. Proposed approach offers a unique window into properly classifying LFP data for
interpreting neural circuit activity underlying behavior.

18
 Motor Planning Determined Lateralization more than
Sensory Information March 2nd,
9:30 IRST,
Tahmineh A Koosha1 , Abdol-Hosssein Vahabie22 , Babak Najar Araabi2 Parallel
1 School of psychology, university of Tehran, Computational and machine learning Room 2
laboratory
2 school of ECE, University of Tehran, Computational and machine learning laboratory

The hemispheric lateralization shows differences in the global and local stimuli process-
ing. Studies have shown different performances by controlling the visual fields. However,
since each hemisphere controlled the opposite side of the body, we supposed that the
hand response might affect lateralization. As far as we know, there is no report of the
hand response in the local and global processing. We examined these predictions on a
large sample of left- and right-handers (n = 123). 85 right-handers (49 women) and 39
left-handers (15 women) completed combinatorial blocks (432 trails) of the Navon task.
Participants were required to respond with different hands. In global processing, the faster
reaction time is in the right-hand response. Also, there is a trend of faster performance
by controlling the left visual field. This evidence supports the priority of motor plan-
ning over sensory information in global processing. In the local processing, there is no
significant effect of the hand response, but results are shown faster performance in the
left visual field. Surprisingly, there was no significant interplay of handedness and gender
with the results. Indeed, there was no difference in the motor planning of global and local
processing between left and right-handers and men and women.

19
 Design, synthesis, and biological evaluation of novel
indanone derivatives as cholinesterase inhibitors for
potential use in Alzheimer’s disease March 3rd,
9:30 IRST,
Aysan Etemadi1,2 , Salar Hemmati3 , Mohammad Shahrivar-Gargari1,4 , Yasaman Parallel
Tamaddon Abibiglue2 , Ahad Bavili2 , Siavoush Dastmalchi1,2,5 Room 1
1 Biotechnology Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
2 Department of Medicinal Chemistry, School of Pharmacy, Tabriz University of Medical
Sciences, Tabriz, Iran
3 Drug Applied Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
4 Pharmaceutical Analysis Research Center and Faculty of Pharmacy, Tabriz University

of Medical Sciences, Tabriz, Iran

5 Faculty of Pharmacy, Near East University, POBOX: 99138, Nicosia, North Cyprus,

Mersin 10, Turkey

As the Cholinesterases are one of the important targets in Alzheimer’s disease, a novel se-
ries of indanone derivatives containing meta/para-substituted aminopropoxy benzyl/benzylidene
moieties were designed based on the structures of donepezil and ebselen analogues. The
designed compounds were synthesized and their acetylcholinesterase (AChE) and butyryl-
cholinesterase (BChE) inhibitory activities were measured. Inhibitory potencies (IC50
values) for the synthesized compounds ranged from 0.12 to 11.92 µM and 0.04 to 24.36
µM against AChE and BChE, respectively. Compound 5c showed the highest AChE in-
hibitory potency with IC50 value of 0.12 µM, whereas the highest BChE inhibition was
achieved by structure 7b (IC50= 0.04 µM). Structure-activity relationship (SAR) analysis
revealed that there is no significant difference between meta and para-substituted deriva-
tives in AChE and BChE inhibition. However, the most potent AChE inhibitor 5c belongs
to meta-substituted compounds, while the most active BChE inhibitor is para-substituted
derivative 7b. The order of enzyme inhibition potency based on the substituted amine
group is dimethyl amine ¿ piperidine ¿ morpholine. Compounds containing C=C linkage
are more potent AChE inhibitors than the corresponding saturated structures. Molecu-
lar docking studies indicated that 5c interacts with AChE in a very similar way to that
observed experimentally for donepezil. Moreover, the experimental results revealed that
these structures could interact with Fe2+ and Cu2+ and may have metal-chelating ac-
tivity. Finally, one may suggest that the introduced indanone-aminopropoxy benzylidene
scaffold could be used for further lead optimization in drug-discovery development against
Alzheimer’s disease.

20
Locomotion is associated with straighter neural trajectories
for natural movies in mouse visual cortex March 3rd,
9:30 IRST,
Xingyu Zheng11 , Maxwell Ruckstuhl2 , Mohammad Yaghoubi3 Parallel
1 Simons Center for Quantitative Biology, Cold Spring Harbor Laboratory, One Room 2
Bungtown Road, Cold Spring Harbor, USA
2 Neuroscience Graduate Program, University of Michigan, 204 Washtenaw Ave., Ann

Arbor, USA
3 Integrated Program in Neuroscience, McGill University, 1033 Pine Ave. W., H3A 1A1.

Montreal, Canada

Sensory systems transform inputs from the environment into internal representations that
capture aspects of the information useful in downstream computation and behavior. These
representations are not rigid, static reflections of the external world but are dynamically
modulated by fluctuating cortical and behavioral states. Locomotion, for example, is as-
sociated with facilitative changes in stimulus representation in the mouse visual cortex
[Vinck et al., 2015]. How does sensory processing in the visual system support down-
stream behavior? ”Straightening” hypotheses posit that an important role of the visual
system is to straighten representations across time and state space, enabling prediction
through simple linear extrapolation [H´enaff et al., 2019] and discrimination between ob-
jects. Whether this phenomenon depends on behavioral states has not been tested. In
this work, we leverage publicly available data and infer the neural trajectories of a nat-
ural movie stimulus in the 2D embedding of the stable representation similarity space in
locomoting and stationary mice. We show that locomotion influences not only encoding
quality but also the structure of the encoding of natural videos, resulting in straighter
neural trajectories in multiple visual areas. This observed effect is not merely the result
of increased response reliability; it also holds true for single-trial analysis. Moreover, we
find that this effect is greater in the ventral visual area VISl than in the dorsal area VISal,
supporting the two-stream hypothesis that attributes object manifold untangling to ven-
tral processing [DiCarlo and Cox, 2007]. We demonstrate a powerful instance where an
etiologically important behavior modulates the transformation of complex, naturalistic
stimuli to yield a more accurate and predictive encoding

21
Introduction of a phase decoder for decoding the content of
working memory March 3rd,
10:00 IRST,
Shakoor Mohammadi1 , Zahra Bahmani1 Parallel
1 Department of Electrical and Computer Engineering, Tarbiat Modares University, Jalal Room 1
AleAhmad, Tehran, Iran

Working memory (WM) is a cognitive ability to temporarily maintain information and

influences our goal-directed behaviors. It is an important component of higher cognitive
functions such as attention and decision making. As stated by [Bahmani, Zahra, et al.
(2018)], WM is not coded in the form of rate coding in sensory areas. The relations
between the timing of spikes and the phases of ongoing αβ oscillations have an important
role in coding of this information. Consequently, we introduced a phase decoder based
on the machine learning algorithms to be able to decode the content of WM. The spiking
activity and local field potentials (LFPs) in V4 and frontal eye field (FEF) were recorded
using a 16-channel linear array electrode from monkeys’ brains. The LFP signal of each
channel, normalized and band-pass filtered in 24 frequency bands. Then, we extracted
spike-phase-locking (SPL) of each spike-LFP pairs, phase-phase-locking (PPL) of each
V4-FEF paired channel, instantaneous phases of V4 and FEF at the spike times and their
combinations as our phase-based features and applicated various classification algorithms
on them to extract content of WM. The classification accuracy in the memory period was
significantly higher than chance level using phase-based features at αβ band, while the
accuracy using rate-based feature set was around chance level. The best features included
FEF instantaneous phases, FEF’s SPLs, FEF-V4 instantaneous phase differences and V4-
FEF PPLs had 75% of accuracy. Phase decoding is a neuronal mechanism that corresponds
to the brain’s biology and its rapid information coding functionality.

22
 Cortical Correlates of Discounted Cumulative Reward March 3rd,
10:00 IRST,
Nikan Amirkhani1 , Tandis Salem2 , Aryan Zoroufi3 , Marc Coutanche4 Parallel
1 School
of Medicine, Tehran University of Medical Sciences, Tehran, Iran Room 2
2 Department of Cell and Molecular Biology, Faculty of Life Science and Biotechnology,
Shahid Beheshti University, Tehran, Iran
3 Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology,

Cambridge, MA, USA

4 Department of Psychology, University of Pittsburgh, Pittsburgh, USA

Rewards are not only valued according to their magnitude but also relative to other re-
wards [Cromwell et al. 2005]. Single-neuron recording has demonstrated the importance
of the orbitofrontal cortex in relative reward processing [Tremblay and Schultz 1999], but
whole-brain analyses are still lacking. Here, we use fMRI data from a subset of the Hu-
man Connectome Project’s gambling task [Barch et al. 2013] to ask which cortical regions
respond preferentially to a reward stimulus in the presence of low-value rewards in close
temporal vicinity compared with high-value rewards in close temporal vicinity. This was
quantified using the Discounted Cumulative Gain measure (DCGt = ti=0 logReward
P t
2 (t−i+2)
) to
relativize according to the subject’s net worth, but also to overemphasize more recent
feedback [Järvelin and Kekäläinen 2002]. The data was collected during a gambling task,
in which subjects guessed whether the value of a hidden card was larger or smaller than
5 and received feedback accordingly. We use a generalized linear model [Nelder and Wed-
derburn 1972] to compare how ”win” feedback after a streak of good or bad luck predicts
activity for each separate cortical region. We found high intersubject variability, with a
maximal cortical correlation of r = 0.6 in single-subject analyses falling to r = 0.15 in
pooled analyses. Cortical areas with the highest correlation included the combined ante-
rior cingulate cortex and medial prefrontal cortex, paracentral and mid-cingulate cortex,
and dorsolateral prefrontal cortex. Interestingly, orbitofrontal cortex activity showed very
little association. Future research should utilize data from a hypothesis-specific task to
explore the contributions of subcortical components to relative reward processing.

23
Poster Presentations
Mental representation of regular past tense forms in first
language and second language processing
Ebrahim Safaie1
1 Department of English Language, Islamic Azad University, Lamerd Branch, Lamerd,
IRAN

There is a substantial debate over the mental representation of regular past tense forms in
both first language (L1) and second language (L2) processing [Clahsen & Neubauer, 2010;
Pinker, 1999; Safaie, 2020; 2021; Silva & Clahsen, 2008; Ullman, 2001; Ullman, 2005]. For
the L1 processing, dual-mechanism models [Ullman, 2001; 2005] suggest two processing
routes for the English past tense inflection: a rule application route for regular verb forms
and a lexical look-up route for irregular verb forms. For the L2 processing, Ullman [2001;
2005] proposed that both verb types are retrieved via the look-up route. By virtue of this
assumption, L2 grammar should suffer from a compositionality deficiency such that L2ers
should treat both regular and irregular past tense forms similarly. A compositionality
deficiency refers to the impairment of the rule-based capacity for composing regular verbs
with the ed suffix (but not irregular verbs which have rather idiosyncratic mappings).
To address this issue, this study investigates regular and irregular past tense forms in
intermediate-to-advanced Persian-English learners and native English speakers (NSs) in
two online experiments. These experiments used Speeded Acceptability Judgment tasks
involving past tense forms in plausible sentences. The results of both the accuracy and
reaction time data show a pattern of convergence between NSs and L2ers. Particularly, in
experiment 1 testing both regular and irregular past tense forms, the accuracy data pro-
vided new evidence showing that L2ers converge on NSs in showing distinct processes for
regulars and irregulars. This native-like profile indicates that L2ers use compositionality
for the regulars and whole-word access for the irregulars. Further, in experiment 2 testing
the frequency effect on the processing of regular past tense, regular verbs yielded a reverse
or an anti-frequency effect in both groups, I explain these results from the perspective of
dual-mechanism models.

24
 Detection of Patients with Multiple Sclerosis Disorder
Based on EEG Signal
Ali Abedi1 , Gholamreza Moradi1
1 Electrical Engineering Department, Amirkabir University of Technology, Tehran, Iran

Background: multiple sclerosis (MS) is a disease of the central nervous system character-
ized by a variety of clinical problems that arise from demyelination and inflammation of
axonal pathways. The disorder commonly appears in patients between the ages of 20 and
40. The signs and symptoms of MS are monocular blindness, motor weakness or paral-
ysis, abnormal somatic sensations, double vision, and dizziness. Objective: to explore
the capabilities of applying a SVM and KNN method for classification of EEG signal, to
standard neurological examination of MS progression. Methods: In this study we have
used two different approaches, which are called the k-nearest neighbors classifier (KNN)
and the support vector machine (SVM), to classify healthy people and MS patients. KNN
algorithm classifies a new sample by finding the k nearest neighbors to that sample based
on a similarity or distance measure. In this pilot study we formulate a two-class ”healthy
control” and ”having MS” classification problem. A dataset of electromyographic signals
from 71 individuals (31 MS patients and 40 healthy controls) and electroencephlaographic
signal from 67 (31 MS patients and 36 healthy controls) patients. Results: different fea-
tures of the signals were extracted in order to train and validate a classification model.
finally, a support vector machine classifier was obtained giving Accuracy = 99, 95% and
K-nearest neighbor classifier was obtained giving Accuracy = 98, 27%.

25
 Understanding how individuals’ risk strategy evolves in
light of hierarchical Gaussian filtering
AmirHossein Tehrani Safa1,2 , Reza Ghaderi1 , Atiye Sarabi-Jamab2
1 Institutefor Cognitive and Brain Sciences, Shahid Beheshti University, G.C., Evin Sq.,
P. O. Box 19839-63113, Tehran, Iran
2 School of Cognitive Sciences, Institute for Research in Fundamental Sciences (IPM), P.

O. Box 19395-5746, Tehran, Iran

An example of social reinforcement is learning others’ risk preferences. In this study,

we developed a hierarchical Gaussian filter, a famous Bayesian computational learning
algorithm, which is capable of inferring others’ potentially time-varying risk attitudes.
To evaluate our proposed computational learning model, we designed scenarios in which
we had to recover the true values of risk attitude. Simulations were conducted using a
classical economic paradigm, in which repeated choices were required between probabilistic
high-reward gambles and non-probabilistic low-payoff payments.
Risk attitude is assumed to be stable in the first scenario, which occurs frequently in
reality. For instance, an individual may maintain a risk-averse attitude towards investing
in the stock market, regardless of the potential reward. In the next scenario, risk attitudes
slowly change. For example, as an individual becomes more familiar with the stock market,
their risk attitude may slowly shift from being risk-averse to risk-seeking. In the last
scenario the risk attitudes change unexpectedly. For example, the risk attitude fluctuates,
depending on the current level of market volatility. This allows individuals to adjust their
risk attitude according to current market conditions, increasing risk-taking in times of
high volatility, and decreasing risk-taking in times of low volatility.
Simulation data were also fitted to the Rescorla-Wagner, Sutton K1 dynamic Rein-
forcement Learner, and non-hierarchical Bayesian learner. Our proposed learning model
outperformed in all scenarios (Exceedance probability = 1) except for the first scenario,
where the non-hierarchical Bayesian learner was sufficient due to a lack of volatility.

26
 The evaluation of excitatory and inhibitory connections in
developmental in vitro cell cultures
Leila Rezayat1 , Mohammad-Reza A. Dehaqani1 , Ehsan Rezayat1 , Amir M Sodagar1
1 K. N. Toosi University of Technology Faculty of Electrical Engineering

Excitation and inhibition activity in the neural circuitry is a balance. This balance is gen-
erated during the development of neural circuitry. Inhibitory connection is a major player
in regulating neural network dynamics, and regulation of excitatory-inhibitory balance is
critical for optimal circuit function. The dynamics of neurons in different stages of devel-
opment can be investigated using the experimental model of neural networks cultivated in
laboratory conditions together with MEA. After about 4 DIV of culture, the neurons start
to connect to each other and have spontaneous electrophysiological activity. The patterns
of collective rhythmic activity change in time spontaneously during in vitro development.
The dynamics of networks have been investigated from various aspects, but despite the
importance of inhibition, studies of functional connectivity are often neglected due to the
difficulty of its identification. In this work, we followed the dynamics of this type of com-
munication over time by examining mutual behavior and solidarity. Our studies show that
inhibitory connections formed later than excitatory connections and are also weaker, and
the network reaches excitatory-inhibitory balance after about 21 DIVs.

27
 Low-Intensity Transcranial Magnetic Stimulation shows
potential for intervention in ADHD
Elahe Khosrowabadi1 , Firouze Mahjoub Navaz1 , Jalil Gholizadeh Soltani2 , Farhad
Soleymani3
1 Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran,
2 Alborz University of Medical Sciences, Karaj, Iran,
3 Institute for cognitive science studies, Tehran, Iran

Response inhibition is an impaired cognitive function in ADHD individuals which is ob-

served even in the minimal demanding cognitive tasks. Studies have shown that transcra-
nial magnetic stimulation (TMS) specially on dorsolateral prefrontal cortex (DLPFC) can
improve the response inhibition. Nevertheless, the TMS has a low spatial resolution and
its effect may not be observed in a single-session intervention. Studies showed that low-
intensity TMS has higher spatial resolution. Therefore, in this study we aimed to evaluate
effectiveness of this method for intervention of response inhibition in ADHD individuals.
In a double-blind paradigm, performance of the adult ADHDs while executing a Stroop
color and word test (SCWT) were measured during a sham or a real stimulation of the
DLPFC. Subsequently, response inhibitions of the participants were measured before and
after the stimulation. Number of correct, wrong and missed answers to 96 computer-
ized trials, and response times of the answers were measured. In addition, changes in
electro-cortical activities during the rest phase before and after the stimulation were also
evaluated.
After checking for data normality, paired t-tests between behavioral data showed that
low-intensity magnetic stimulation of the DLPFC can improve response inhibition (reduce
errors) even in a single-session intervention of ADHD individuals. Having said that, the
answering times did not change significantly. The behavioral changes were associated with
significant changes in power of EEG in delta and beta frequency bands at the frontal areas.
The proposed stimulation protocol with low-intensity TMS had a fair effect on the
response inhibition in adult ADHDs. Therefore, it potentially could be suggested as a
treatment protocol on the response inhibition in ADHD individuals.

28
Changes in phase amplitude coupling of EEG as a potential
mechanism to perceive emotional contents of the music
Zohreh Rahimi1 , Zhaleh Mohammad Alipour2 , Reza Khosrowabadi3
1 1Institute of Medical Science and Technology, Shahid Beheshti University, Tehran, Iran,
2 Brain and Mind Institute, Department of Neuroscience, Schulich School of Medicine
and Dentistry, Western University, London Ontario, Canada,
1 Institute for Cognitive and Brain Sciences, Shahid Beheshti University, Tehran, Iran

Investigation of the neural correlates of music-evoked emotions has raised interest during
the last decades. Nevertheless, the cognitive process and underlying neural mechanism
involved in the perception of musical rhythm are not well understood. In this study, we
hypothesized that changes in musical rhythmic patterns alter the related perceived emo-
tions. These emotional changes are associated with a phase-amplitude coupling (PAC)
between the phase of the low-frequency bands such as theta (4–8 Hz), and the amplitude
of the high-frequency bands such as alpha (8–14 Hz) and beta (14–30 Hz). Therefore, 18
boys aged from 10 to 14 years were recruited, and their brain activities were recorded using
a 32-channel EEG recorder while they were exposed to 12 musical excerpts. We changed
musical rhythmic patterns by manipulating note values in beats while keeping other mu-
sical elements in a fixed state. A 2-dimensional self-assessment manikin questionnaire
was used to assess the experienced emotions. According to behavioral data, perceived
valence and arousal levels were significantly enhanced by increasing the complexity of mu-
sical rhythmic patterns. In addition, the PAC results based on the mean vector length
algorithm showed that increasing complexity of musical rhythmic patterns significantly
enhances the PAC between theta phase and alpha and beta amplitudes at the frontal,
fronto-central, central, and parietal brain regions. Moreover, behavioral changes were
significantly correlated with the PAC changes at the frontal, fronto-central, and centro-
parietal regions. These findings may improve our knowledge about the brain mechanism
involved in the emotional processing of music.

29
 Facial keypoint detection by self-supervised multi-task
learning in thermal images
Poorya Aghaomidi1 , Zahra Bahmani2 , Sadaf Aram1
1 Department of Electrical and Computer Engineering, Tarbiat Modares University,
Master Student, Tehran, Iran
2 Department of Electrical and Computer Engineering, Tarbiat Modares University,

Assistant Professor, Tehran, Iran

Keypoint detection has been gaining momentum for various functions such as face recog-
nition, facial region tracking and facial expression recognition. While most of the research
is conducted in the visual spectrum, the potential of infrared imaging remains mainly
untapped due to the lack of large datasets. Physiological indexes which reflect mental
states of humans can be extracted from facial thermal images. An important component
of autonomous mental state detection is face tracker that estimates precise coordinates
of facial components. To this end, an accurate algorithm that detects facial keypoints in
thermal videos was designed. We used the dataset [Kopaczka et al. 2018] that manually
annotated 68 facial keypoints on 94 subjects. The state-of-the-art self-supervised method
was used to compensate for having a small number of samples in the dataset. We pre-
trained our CNN with two pretext tasks. We rotated the images and trained the CNN
to detect the rotation angle and simultaneously classify all 94 subjects. These tasks were
executed with an accuracy of 100%, and 88.89% respectively. Consequently, our network
learned substantial information about the data and extracted salient features of individ-
ual faces. We removed the fully connected part and applied the learned knowledge to
our downstream task which was keypoint detection. The precision of the proposed algo-
rithm was significantly increased for facial keypoints detection. The NME (Normalized
Mean Error) achieved in this method was 1.45 which is impressively lower than the NME
achieved in a fully supervised manner (NME=4.43). This result indicates the superiority
of self-supervised learning over supervised learning.

30
 Reduction of mental fatigue by electrical stimulation of
pilot’s brain
Fahimeh Nadalizadeh1 , Zahra Bahmani1
1 Department of Electrical and Computer Engineering, Tarbiat Modares, Jalal
AleAhmad, Tehran, Iran

Human augmentation is among emerging technologies which enhances cognitive abilities of

people. Pilots are encountered with unpredictable and long work hours, circadian disrup-
tion, and insufficient sleep which make them strong candidates of mental fatigue. In this
study, we stimulated DLPFC of pilots’ brains using transcranial direct current stimulation
(tDCS) in order to reduce mental fatigue. Mental fatigue was measured by investigating
the reaction time (RT) in a visual task. Images of characters or words added by noise and
displayed for 500 milliseconds on the screen. Subjects should type what they saw in less
than 4 seconds. 200 images of characters and words were displayed at each condition. Each
subject did the task in four conditions: morning (non-fatigue), after one hour of flight (fa-
tigue), fatigue with sham stimulation, fatigue with anodal stimulation. Subjects received
20 minutes of electrical stimulation on DLPFC during the anodal condition. The duration
of stimulation was only 5 seconds in the sham condition. The results demonstrated that
RTs were significantly increased in fatigue condition in comparison to non-fatigue condi-
tion. While the anodal stimulation effectively reduced the RTs of pilots in comparison to
fatigue and sham conditions. The accuracy of responses also showed similar effects. We
also compared the mental fatigue reduction of pilots and non-pilot subjects. The proposed
mechanisms reduced the RTs of pilots significantly more than that of non-pilot subjects.
This study provided a non-invasive and systematic framework for mental fatigue reduction
of pilots. This technology can be used in order to improve safety of flights.

31
 The Effect of Feature Normalization on Motor Imagery
Task Classification
Hamed Nazemi1 , Mohammad Amin Ebrahimzadeh2 , Alireza Taheri3
1 Mechanical Engineering Dept., Sharif University of Technology, Azadi St., Tehran, Iran
2 Center for Cognitive Science, Institute for Convergent Science and Technology (ICST),
Sharif University of Technology, Tehran, Iran

Motor Imagery (MI) hand movement detection has gained the attention of many re-
searchers to design brain-computer interfaces for people with disabilities. Data normal-
ization is one of the pre-processing steps where the features are scaled to contribute to all
features equally. To this end, an experiment has been designed/conducted in which the
EEG signals were recorded from 6 typically developing participants without having previ-
ous experience in EEG data collection; these subjects were all male with an age average
of 24 years. The EEG signals were recorded in the IPM School of Cognitive Science with
a 64-channel EEG cap. After artifact removal and channel interpolation, the recorded
signals were decomposed with ICA. Following this step, the features were extracted after
applying CSP and selected with the mRMR algorithm. In this study, the effect of 12 differ-
ent normalization methods was investigated on the final accuracy of the left and right-arm
movement task. These normalization methods include decimal scaling, hyperbolic tangent,
logistic sigmoid, mean-centered, min-max, median and median absolute, Pareto scaling,
power transformation, tangent hyperbolic, tangent hyperbolic variant, variable stability
scaling, and z-score. In the next step, Naive Bayes, Decision Tree, Support Vector Ma-
chine, and K-Nearest Neighbour classifiers were used to investigate the final accuracy of
different normalization methods. For reporting the final results, the k-fold cross-validation
algorithm was applied with k=5. According to the results, the tangent hyperbolic variant
outperforms other normalization methods (11.4 ± 0.4 % error rate), while logistic sigmoid
normalization has the least efficient error rate (24 ± 1.8 %).

32
 The effect of repeated and increasing injection of
dexamethasone on morphological changes and apoptosis
rate in mouse cerebellar cortex
Nooshin Goudarzi1 , Seyed amir Hosseini2 , Farzad Rajaie3
1 Student Research Committee, Faculty of Medicine, Qazvin University of Medical
Sciences, Qazvin, Iran
2 Department of Anatomical sciences, School of Medicine, Qazvin University of Medical

Sciences, Qazvin, Iran

3 Professor of Histology and Embryology, Cellular and Molecular Research Center,

Qazvin University of Medical Sciences, Qazvin, Iran

Background: Glucocorticoids, despite their widespread use, leave side effects on various
organs of the body, including the brain. In this study, the effect of different doses of
dexamethasone on the morphological changes of the rat cerebellar cortex was studied.
Materials and methods: In this study, 40 adult female mice were randomly divided into 4
groups of 10. Mice in the control group were injected with normal saline, while mice in
the experimental groups were subjected to intraperitoneal injection of dexamethasone with
concentrations of 4, 7 and 10 mg/kg daily for one week. One day after the last injections,
the mice were killed. Microscopic slides were prepared and the number of Purkinje cells,
the size of Purkinje cells, the thickness of the granular layer, the thickness of the cortex
and the amount of apoptosis were determined and analyzed by Oneway test, ANOVA and
Post Hoc Test. Findings: Increasing the dose of dexamethasone significantly decreased
the number of Purkinje cells (p¡0.001). Also, injection of dexamethasone with a dose of 4
mg/kg, unlike the doses of 7 and 10 mg/kg, increased the size of Purkinje cells(p=0.72),
but had no effect on the granular layer. The granular layer was increased in doses of
7 and 10 mg/kg. Also, in each dose, a decrease in the thickness of the cerebellar cortex
(p=0.041) and an increase in apoptosis (p¡0.001) were observed. Conclusion: The frequent
injection of dexamethasone with multiple changes in the structure of the cerebellar cortex
and increased apoptosis can have harmful effects on the developmental nervous system of
female mice.

33
The way to the light: What are the pathways that mediate
blindsight? A systematic review
MR.Balali1 , S.M.Hosseini1 , M. Khanboloki1 , M. Mohammadi1 , AM.Khorshidvand1 ,
MT.Joghataei2
1 School of medicine, Iran University of Medical Sciences, Tehran, Iran
2 Cellular and Molecular Research Center (CMRC), Iran University of Medical Sciences,
Tehran, Ira

Although damage to area V1 makes patients unable to consciously see the visual stim-
uli, few patients retain some of their visual abilities, a phenomenon called blindsight.
Unfortunately there isn’t consensus on its responsible neural mechanisms yet. We gath-
ered different researches about the mechanisms that facilitate blindsight in Humans and
Monkeys. This systematic review gathered 24 articles from online databases focusing on
blindsight mediating pathways. Human studies focusing on non-affective blindsight claim
that lateral geniculate nucleus (LGN) to area MT/v5 pathway, the pathways from supe-
rior colliculus to higher cortical areas and spared v1 islands are behind blindsight ability.
There was also one study suggesting that blindsight can be attributed to the compensatory
functional role of the undamaged hemisphere through transcallosal connections. Affective
blindsight studies suggest that colliculus-pulvinar-amygdala pathway and the pathways
transited from extrastriate cortex through anterior areas ending in amygdala are the main
mechanisms. Animal studies emphasize on the role of pathways from LGN to extrastriate
areas and superior colliculus-pulvinar pathway in mediating blindsight. Altogether spared
v1 islands, pathways from LGN to area mt/v5 and from superior colliculus to higher corti-
cal areas seem to be the main mechanisms that provide blindsight in humans. In monkeys,
the LGN to extrastriate pathway and superior colliculus-pulvinar-extrastriate pathway are
the main responsible pathways. Shedding light on blindsight mediating pathways can lead
to better understanding of the brain visual system and also providing new strategies for
visual rehabilitation in v1-damaged patients.

34
Part IV

People

35
Scientific Committee
Scientific Chairs:
• Dr. Ali Ghazizadeh, Assistant professor, Electrical Engineering Department,
Sharif University of Technology, School of Cognitive Sciences, Institute for
Research in Fundamental Sciences (IPM), Systems and Computational Neu-
roscience
• Dr. Safura Rashid Shomali, Resident Researcher, Institute for Research in
Fundamental Sciences (IPM), School of Cognitive Sciences, Cognitive Neuro-
science

Reviewers:
• Dr. Saeed Behzadipour, Associate professor, Mechanical Engineering Depart-
ment, Sharif University of Technology, Neuro-Rehabilitation
• Dr. Mohammadreza Daliri, Professor, Electrical Engineering Department,
Iran University of Science and Technology (IUST), Cognitive neuroscience
• Dr. Reza Ebrahimpour, Professor of Cognitive Neuroscience, Shahid Rajaee
University (SRTTU), School of Cognitive Sciences, Institute for Research in
Fundamental Sciences (IPM), Cognitive Neuroscience
• Dr. Alireza Taheri, Assistant Professor, Mechanical Engineering Department,
Sharif University of Technology, Social and Cognitive Robotics
• Dr. Ali Shahbazi, Associate Professor, Department of Neuroscience, Iran Uni-
versity of Medical Sciences, Advanced Technologies in Medicine

36
Executive Team

SNS is organized by the students in the electrical engineering department at Sharif Uni-
versity of Technology. Below is the list of undergraduate and graduate students who
organized the forth symposium (SNS 2023):

• Executive Heads:
Arsalan Firoozi, Armin Panjehpoor, MohammadAmin Alamalhoda
• Workshops:
Omid Sharafi
• Graphic Designer:
Emad Memarzadeh, Alireza Rafiei
• Branding:
Amirreza Hatamipour, Parnian Taheri, Heliya Shakeri, Shervin Mehrtash,Yasamin
Medghalchi
• Online support team:
Negin Esmailzadeh, Tina Halimi

37