Electrophysiological Recording Techniques 2nd Edition Robert P Vertes Editor

Electrophysiological Recording Techniques 2nd
Edition Robert P Vertes Editor download
https://ebookbell.com/product/electrophysiological-recording-
techniques-2nd-edition-robert-p-vertes-editor-46163204
Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.
Electrophysiological Recording Techniques 1st Edition L Stan Leung
Auth
https://ebookbell.com/product/electrophysiological-recording-
techniques-1st-edition-l-stan-leung-auth-2527278
Electrophysiological Disorders Of The Heart Expert Consult Online And
Print 2e 2nd Edition Sanjeev Saksena Mbbs Md Facc Fesc Fhrs Faha
https://ebookbell.com/product/electrophysiological-disorders-of-the-
heart-expert-consult-online-and-print-2e-2nd-edition-sanjeev-saksena-
mbbs-md-facc-fesc-fhrs-faha-2521884
Electrophysiological Analysis Of Synaptic Transmission 2nd Edition
Nicholas Graziane
https://ebookbell.com/product/electrophysiological-analysis-of-
synaptic-transmission-2nd-edition-nicholas-graziane-44649546
Electrophysiological Maneuvers For Arrhythmia Analysis 1st Edition
George J Klein
https://ebookbell.com/product/electrophysiological-maneuvers-for-
arrhythmia-analysis-1st-edition-george-j-klein-4938600

Electrophysiological Analysis Of Synaptic Transmission Yan Dong
https://ebookbell.com/product/electrophysiological-analysis-of-
synaptic-transmission-yan-dong-5258190
Electrophysiological Foundations Of Cardiac Arrhythmias A Bridge
Between Basic Mechanisms And Clinical Electrophysiology Andrew L Wit
https://ebookbell.com/product/electrophysiological-foundations-of-
cardiac-arrhythmias-a-bridge-between-basic-mechanisms-and-clinical-
electrophysiology-andrew-l-wit-6685000
Electrophysiological Disorders Of The Heart 1st Edition Sanjeev
Saksena
https://ebookbell.com/product/electrophysiological-disorders-of-the-
heart-1st-edition-sanjeev-saksena-1867684
Basic Electrophysiological Methods 1st Edition Ellen Covey Matt Carter
https://ebookbell.com/product/basic-electrophysiological-methods-1st-
edition-ellen-covey-matt-carter-4983598
Introduction To Electrophysiological Methods And Instrumentation 1st
Edition Franklin Bretschneider
https://ebookbell.com/product/introduction-to-electrophysiological-
methods-and-instrumentation-1st-edition-franklin-bretschneider-2154364

Electro physiological
Recording
Techniques
Robert P. Vertes
Timothy A. Allen Editors
Neuromethods 192
Second Edition

NEUROMETHODS
Series Editor
Wolfgang Walz
University of Saskatchewan
Saskatoon, SK, Canada
For further volumes:
http://www.springer.com/series/7657

Neuromethodspublishes cutting-edge methods and protocols in all areas of neuroscience as
well as translational neurological and mental research. Each volume in the series offers tested
laboratory protocols, step-by-step methods for reproducible lab experiments and addresses
methodological controversies and pitfalls in order to aid neuroscientists in experimentation.
Neuromethodsfocuses on traditional and emerging topics with wide-ranging implicationsto
b
rain function, such as electrophysiology, neuroimaging, behavioral analysis, genomics,
neurodegeneration, translational research and clinical trials.Neuromethodsprovides investi-
gators and trainees with highly useful compendiums of key strategies and approaches for
successful research in animal and human brain function including translational “bench to
bedside” approaches to mental and neurological diseases.

ElectrophysiologicalRecording
Techniques
Second Edition
Edited by
Robert P. Vertes
Ctr for Complex Systems & Brain Sci, Florida Atlantic University, Boca Raton, FL, USA
Timothy A. Allen
Department of Psychology, Florida International University, Miami, FL, USA

Editors
Robert P. Vertes
Ctr for Complex Systems & Brain Sci
Florida Atlantic University
Boca Raton, FL, USA
Timothy A. Allen
Department of Psychology
Florida International University
Miami, FL, USA
ISSN 0893-2336 ISSN1940-6045(electronic)
Neuromethods ISBN978-1-0716-2630-6 ISBN 978-1-0716-2631-3 (eBook)
©Springer Science+Business Media, LLC, part of Springer Nature 2022
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations.
This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.
https://doi.org/10.1007/978-1-0716-2631-3

Preface to the Series
Experimental life sciences have two basic foundations: concepts and tools. The Neuro-
methods series focuses on the tools and techniques unique to the investigation of the
nervous system and excitable cells. It will not, however, shortchange the concept side of
things as care has been taken to integrate these tools within the context of the concepts and
questions under investigation. In this way, the series is unique in that it not only collects
protocols but also includes theoretical background information and critiques which led to
the methods and their development. Thus, it gives the reader a better understanding of the
origin of the techniques and their potential future development. The Neuromethods
publishing program strikes a balance between recent and exciting developments like those
concerning new animal models of disease, imaging, in vivo methods, and more established
techniques, including, for example, immunocytochemistry and electrophysiological tech-
nologies. New trainees in neurosciences still need a sound footing in these older methods in
order to apply a critical approach to their results.
Under the guidance of its founders, Alan Boulton and Glen Baker, the Neuromethods
series has been a success since its ﬁrst volume published through Humana Press in 1985. The
series continues to ﬂourish through many changes over the years. It is now published under
the umbrella of Springer Protocols. While methods involving brain research have changed a
lot since the series started, the publishing environment and technology have changed even
more radically. Neuromethods has the distinct layout and style of the Springer Protocols
program, designed speciﬁcally for readability and ease of reference in a laboratory setting.
The careful application of methods is potentially the most important step in the process
of scientiﬁc inquiry. In the past, new methodologies led the way in developing new dis-
ciplines in the biological and medical sciences. For example, physiology emerged out of
anatomy in the nineteenth century by harnessing new methods based on the newly discov-
ered phenomenon of electricity. Nowadays, the relationships between disciplines and meth-
ods are more complex. Methods are now widely shared between disciplines and research
areas. New developments in electronic publishing make it possible for scientists that
encounter new methods to quickly ﬁnd sources of information electronically. The design
of individual volumes and chapters in this series takes this new access technology into
account. Springer Protocols makes it possible to download single protocols separately. In
addition, Springer makes its print-on-demand technology available globally. A print copy
can therefore be acquired quickly and for a competitive price anywhere in the world.
Saskatoon, SK,
Canada W olfgang Walz
v

Preface
The present volume is a continuation of the previous edition ofElectrophysiological Record-
ing Techniques(2011) edited by Robert Vertes and Robert Stackman, Jr. As with the
previous edition, this volume explores various topics, incorporating state-of-the-art electro-
physiological and anatomical methods and their application to the study of several systems of
the brain involved in a range of functions. This edition includes several new chapters and
some chapters with extensive updates which complement and extend the content of the ﬁrst
edition. For example, in this edition, new chapters emphasize the value and difﬁculty of
multi-site recordings using depth or surface electrodes. This latter point takes center stage
when considering the innumerable interactions between the many regions of the brain to
support behavior, and the many species needed to identify fundamental neurophysiological
mechanisms. The new addition also contains several chapters addressing the different
electrophysiological techniques used in non-human animals and humans, and discusses
the different advantages and the types of questions each can answer.
First, the new chapter by Choi and Hwang provides a comprehensive account of
electroencephalography (EEG) recordings applied to mouse research with an approach
that mimics the coverage from arrays more typical of a human psychiatric session. Although
EEG has been a feature of clinical neurology for almost 100 years, the authors introduced
several unresolved issues including the source of EEG signals and the biological basis of their
variation in behavior and disease that can be resolved in rodent research models. The authors
discuss sources of noise such as pink noise (1/f) and cable noise that can differentially affect
frequency speciﬁc parts of EEG recordings, and describe methods such as the type of
electrode, contacts, spacing, and surgical implant molds that are capable of reducing such
confounds. A procedural approach to EEG analysis is presented including data selection,
artifact removals, and the creation of spatiotemporal topologies.
The new chapter by Ito focuses on methods for high-density multisite thalamic and
cortical recordings allowing measures of interregional synchrony that lend themselves to
examining the neurophysiological mechanisms of complex brain-behavior relationships. As
an example, the chapter presents goal-directed spatial navigation in rodents as a model of
complex information processing in the brain. That is, this cognitive task requires sensation,
spatial representation, memory formation, goal representation, memory retrieval, decision-
making, and action selection and thus involves the interactions of several brain regions. The
author provides detailed and practical laboratory methods for building multisite hyperdrives
capable of targeting two structures in freely behaving rodents and simultaneously recording
the activity of high-density neuronal ensembles and local ﬁeld potentials (LFPs). The author
further addresses several analytical challenges in multisite recordings with solutions includ-
ing spectral coherence analysis, spike-phase analysis, spike-ﬁeld coherence, and the compar-
isons of measures across observed behavior conditions.
The chapter by Vivar and van Praag describes procedures for identifying inputs to adult-
born granule cells of the dentate gyrus and for characterizing the properties of these afferent
(projecting) neurons. Speciﬁcally, they provide a detailed description of techniques for
labeling adult-born neurons (starter cells) through dual viral injections and the retrograde
trans-synaptic transport of rabies virus from starter cells to monosynaptic inputs (traced
vii

cells) to them. By recod
neurochemical identity
new-born granule cells.
viral injections into start
immature versus mature
examination of the poss
states (epilepsy) on the
ing from “tagged” traced cells (in slices), they could determine the
and electrophysiological characteristics of neurons projecting to
Further, by adjusting the time interval between the ﬁrst and second
er cells (between 3 and 90 days), they were able to deﬁne inputs to
granule cells. Finally, as was pointed out, the method allows for an
ible effects of various conditions (exercise, learning) or pathological
organization of afferents to new-born granule cells.
viii Preface
The chapter by Vertes, Linley, and Viena examines the nucleus reuniens (RE) of the
ventral midline thalamus, describing RE circuitry, functional properties, electrophysiological
characteristics, and involvement in CNS disorders. With respect to function, nucleus
reuniens has been shown to serve a critical role in cognitive, affective, and executive
functions. Speciﬁcally, alterations of RE signiﬁcantly disrupt spatial working memory, atten-
tion/attentional set, reversal learning, and fear conditioning/fear extinction. Regarding
electrophysiological characteristics, subsets of RE neurons display spatial properties (head
direction cells, place cells), exhibit trajectory-dependent ﬁring (predicting directional
choices), and discharge at signiﬁcantly higher rates in waking and/or REM sleep compared
to slow wave sleep. The latter ﬁnding suggests that RE neurons may exert arousal or
attentional effects on target structures during waking or REM sleep. Finally, RE has been
linked to the CNS disorders of schizophrenia (SZ) and epilepsy. Disruptions of nucleus
reuniens produce abnormal oscillations between the hippocampus and medial prefrontal
cortex, and associated SZ-like cognitive deﬁcits, whereas the over activation of reuniens
produces pronounced epileptic seizure activity.
In an extension of his chapter in the original volume, Bressler provides a detailed
overview of the neural mechanism responsible for event-related potentials (ERPs). As was
described, ERPs are triggered by external sensory and motor events or by “internal”
spontaneous or self-generated states, and represent dynamical changes in the brain to
internal/external events. Various methods were described for analyzing cortical ERPs, or
extracting their non-varying features from the general ﬁeld activity. These include time
domain analysis, frequency domain analysis, spatial analysis, and interdependency analy-
sis—with each reﬂecting distinct aspects of the triggered ERPs. The ERP was compared to
its magnetic correlate, the event-related ﬁeld (ERF), and as pointed out, for some analyses,
both methods are preferable to imaging techniques, as the time course of changes more
closely mimic the underlying neural events.
The chapter by Pittman-Polletta and Kocsis examines various methods used to assess the
cross-frequency coupling of electrophysiological signals (EEG and LFPs) such as amplitude-
amplitude coupling (AAC) and phase-phase coupling (PPC), but it describes in detail phase-
amplitude coupling (PAC). The PAC involves ﬂuctuations in the amplitude of fast rhythms
dependent on the phase of slower rhythms. The authors provide a detailed description of the
signal processing techniques used to examine and quantify PACs—from the preprocessing
of data through to statistically determining the strength of PAC coupling. As was pointed
out, PAC analysis is a valuable tool for quantifying the relationship between simultaneously
occurring events, reﬂecting functional interactions between structures (or networks) exhi-
biting oscillations at different frequencies. The authors give examples of the application of
PAC analysis from their work showing interactions between delta/theta and high-frequency
oscillations (HFO) of the cortex and hippocampus as well as between high gamma activity in
the hippocampus (driver) and theta in the cortex (receiver).
The new chapter by Jayachandran and Allen looks at the neurophysiological mechanisms
that encode time and approaches to studying this fundamental aspect of cognition and

behavior. Studying the
unique issues for rigorou
isolated from other co
fundamental design con
Jayachandran and Allen
ual processing levels rela
speciﬁc behavioral meth
in vivo electrophysiolo
scientiﬁc recommendat
activity (SUA) recording
cells—each of which co
types might combine
3D-printable dual-site s
examine neural ensembl
be critical to encoding t
neurobiological underpinnings of temporal content raises several
s science. For example, time cannot be generated in the lab or easily
nfounding time-varying covariates. Thus, the authors enumerate
siderations to aid rigorous studies of time in a laboratory setting.
introduce a simple taxonomy for time, and emphasize how individ-
te to actual versus reconstructed time. They detail methods for two
ods that can be used to study time in memory in conjunction with
gical recordings, and emphasize how these designs satisfy their
ions. Next, theauthors
present results from several single-unit
studies that have discovered ramping cells, time cells, and sequence
ntribute to temporal processing—and they discuss how these cell
to support memory for time. Lastly, the authors present a
ilicon probe that can be chronically implanted in rats and used to
es and LFPs across the large system of regions in the brain known to
emporal content.
Preface ix
The new chapterby Grifﬁn further focuses on multisite recordings emphasizing the
reason for using such difﬁcult techniques, and the barriers to entry for new researchers interested in large-scale brain networks. Methods for motivating rats are presented with
procedures presented for food restriction and the choice of food rewards, and advice is given on the care needing to be taken when working with rats in motivated tasks that occur outside
their homecage and with an experimenter. Grifﬁn provides detailed information related to
surgically implanting rats and how to go about their ﬁrst on-head plugin event, and the time
needed to carefully drive tetrodes. Localizing neurophysiological signatures of hippocampal layers, such as sharp wave ripples (SWRs), is discussed. The chapter tackles many practical
issues related to the construction and implementation of multisite recordings, and the
choices that can be made in their design related to the speciﬁc research question. Lastly,
the author discusses the need to integrate neural recordings with more interventionist
approaches such as lesions, inactivation, chemogenetics, and optogenetics—each providing
their own unique beneﬁts to testing causal contributions to neurophysiological research.
The new chapter by Mattfeld addresses issues in behavior methodologies in neurosci-
ence, a critical topic for this book since the purpose of the brain is to advantage behavior.
The focus is on comparative approaches that have been, and should be, used in neurophysi-
ological studies in the ﬁeld of behavioral and cognitive neuroscience. Mattfeld highlights the
fact that comparative approaches are frequently touted but rarely executed. The chapter
addresses a wide range of tasks but focuses on one with great comparative success called
conditional associative learning. The author details species-similar and species-speciﬁc con-
siderations in implementation and interpretation of this task. For example, the type of cue or
reward used may be different in each species but the underlying process held constant. The
author then addresses comparative loss of function approaches such as animal lesions and
human brain damage, and consideration needed in interpreting different approaches. The
chapter makes a comparison of the many electrophysiological approaches used in animals
and humans from single neuron recordings to LFPs and juxtaposes this with a detailed
account of functional magnetic resonance imaging (fMRI) during conditional associative
learning, in addition to EEG, magnetoencephalography (MEG), and positron emission
tomography (PET).
In the ﬁnal chapter of this edition, Helfrich provides a comprehensive account of
intracranial electroencephalography (iEEG) performed in patients with drug-resistant epi-
lepsy. The author discusses the advantages from historical approaches and compares

intracranial recording a
humans (SUA, LFP, an
fMRI, and PET) highl
analysis, and that the tr
intracranial recordings,
spatial and temporal reso
iEEG signal and address
author then provides pr
in the clinic. The chapt
and event-related pote
meaningful conclusions
that arise in iEEG work
pproaches in non-human primates (SUA, LFP and iEEG) and
d iEEG) to non-invasive approaches used in humans (EEG, MEG,
ighting that each provides information about particular levels of
ansfer functions between levels are largely unknown. Unlike most
iEEGs tend to cover large areas of the brain, and have exceptional
lution. The author gets into the details of the physiological basis of
es other approaches including adding arrays for unit recordings. The
actical methodological and equipment information for using iEEG
er presentsanalytical
approaches such as frequency decomposition
ntials, and the use of various statistical approaches for drawing
. Lastly, Helfrich provides several notes for consideration of issues
in the clinic.
x Preface
In conclusion,as a continuation of the previous volume, the present edition introduces
and details several recently developed techniques currently used to examine the electrophys-
iological properties of several CNS systems, which affect behavior across a range of species
from rodents to humans. It is our expectation that the present edition would have wide
applicability for basic researchers as well as for clinicians in utilizing some of the principles
and research designs described herein in their research programs.
Boca Raton, FL, USA Robert P. Vertes
Miami, FL, USA Timothy A. Allen

Contents
Preface to the Series........................................................... v
Preface . .................................................................... vii
Contributors.................................................................xiii
1High-Density Electroencephalography in Freely Moving Mice........ ........ 1
Jee Hyun Choi and Eunjin Hwang
2Multisite Recording for the Analysis of Information Flow
Between Thalamocortical Regions.......... ....... ........ ....... ........
15
Hiroshi T. Ito
3Rabies Virus Tracing of Monosynaptic Inputs
to Adult-Born Granule Cells........ ....... ....... ........ ....... ........
37
Carmen Vivar and Henriette van Praag
4Nucleus Reuniens: Circuitry, Function, and Dysfunction............ .....
... 55
Robert P. Vertes, Stephanie B. Linley,
and Tatiana D. Viena
5Event-Related Potentials of the Cerebral Cortex............. ....... ........
103
StevenL. Bressler
6Assessing Neural Circuit Interactions and Dynamics
with Phase-Amplitude Coupling............ ....... ........ ....... ........
125
BenR. Pittman-Polletta and Bernat Kocsis
7Candidate Neural Activity for the Encoding of Temporal
Content in Memory........ ........ ....... ....... ........ ....... .
.
..... .
147
MaanasaJayachandran and Timothy A. Allen
8Multisite Recording During Memory-Guided Behavior....... ....... ........
183
AmyLynn Grifﬁn
9Comparative Tasks for Comparative Neurophysiology........ ....... ........
193
Aaron T. Mattfeld
10Human Intracranial Cognitive Neurophysiology............. ....... ........
221
RandolphF. Helfrich
Index . . .................................................................... 247
xi

TIMOTHYA. ALLEN•Cognitive Neuroscience Program, Department of Psychology, Florida
S
TEVENL. BRESSLER•Center for Complex Systems & Brain Sciences, Florida Atlantic
J
EEHYUNCHOI•Center for Neural Science, Korea Institute of Science and Technology,
A
MYLYNNGRIFFIN•Department of Psychological and Brain Sciences, University of
R
ANDOLPHF. HELFRICH•Hertie Institute for Clinical Brain Research, Center for
E
UNJINHWANG •Center for Neural Science, Korea Institute of Science and Technology, Seoul,
H
IROSHIT. ITO•Max Planck Institute for Brain Research, Frankfurt am Main, Germany
M
AANASAJAYACHANDRAN •Cognitive Neuroscience Program, Department of Psychology,
B
ERNATKOCSIS•Department of Psychiatry, Beth Israel Deaconess Medical Center/Harvard
S
TEPHANIEB. LINLEY•Center for Complex Systems and Brain Sciences, Florida Atlantic
A
ARONT. MATTFELD •Cognitive Neuroscience Program, Florida International University,
B
ENR. PITTMAN-POLLETTA •Department of Mathematics & Statistics, Boston University,
H
ENRIETTE VANPRAAG•Department of Biomedical Science, Charles E. Schmidt College of
R
OBERTP. VERTES•Center for Complex Systems and Brain Sciences, Florida Atlantic
T
ATIANAD. VIENA•Department of Psychology, Florida International University, Miami,
C
ARMENVIVAR•Laboratory of Neurogenesis and Neuroplasticity, Department of Physiology,
Contributors
International University, Miami, FL, USA; Department of Environmental Health
Sciences, Robert Stempel College of Public
Health, Florida International University,
Miami, FL, USA
University, Boca Raton, FL, USA
Seoul, South Korea
Delaware, Newark, DE, USA
Neurology, University of Tu
¨bingen, Tu¨
bingen, Germany
South
Korea
Florida International University, Miami, FL, USA
Medical School, Boston,MA,
USA
University, Boca Raton, FL,USA;
Departmentof
Psychology, Florida Atlantic University,
Boca Raton, FL, USA; Department of Psychological Science, University of North Georgia,
Dahlonega, GA, USA
Miami, FL, USA
Boston, MA, USA
Medicine, and Brain Institute,Florida
Atlantic University, Jupiter, FL, USA
University, Boca Raton,FL,
USA; Department of Psychology, Florida Atlantic University,
Boca Raton, FL, USA
FL, USA
Biophysics and
Neuroscience, Center for Research and Advanced Studies of the National
Polytechnic Institute, Mexico City, Mexico
xiii

Chapter1
High-Density Electroencephalography in Freely
Moving Mice
Jee Hyun Choi and Eunjin Hwang
Abstract
Electroencephalography (EEG), an electrophysiological monitoring method to record the electrical activity
of the brain, has been one of the most important noninvasive brain imaging tools in psychiatry, but
surprisingly little is known about how the neural correlates of various EEG features are linked to cognition.
Recent advances in neuroscience and related technologies make this an ideal time for new discoveries about
the origin and signiﬁcance of the contents of EEG. In particular, understanding the molecular and cellular
mechanisms underlying diverse EEG features has been facilitated using mouse models under genetic,
pharmacological, or optogenetic manipulations. A core challenge in mouse EEG was to obtain topographi-
cal neuroimaging that can be compared to human EEG. To overcome this challenge, we have developed a
high-density EEG using a polyimide-based microarray that ﬁts to the mouse skull and can be applied to
various studies with a high spatiotemporal accuracy in free behavioral states. The beneﬁts of mouse high-
density EEG are not only that it provides cross-species neuroimaging data comparable to human EEG but
also that it helps in dissecting enigmatic brain activity by probing the neural substrates of cognition when
combined with optogenetics. The aim of this chapter is to introduce the methodological aspects of high-
density EEG in mice. We explain the electrodes, surgery, recording, and analysis procedures and present
applications in studying the origin of EEG signals. In addition, we point to potential areas where this
technique will provide mechanical insight into circuit dysfunction in major psychiatric conditions.
Key wordsElectroencephalography (EEG), Mouse, Microelectrode, Topography, Freely moving
mice, Preclinical studies
1 Introduction
Electroencephalography (EEG) and related measures have been
one of the most important neuroimaging tools in human neurosci-
ence and the clinic by providing a topographical mapping of
cortical activities. Topographical features of regional power spectral
density or networks of connectivity strength across different corti-
cal regions have been studied with EEG, and tremendous effort has
been made to understand where EEG signals come from and what
brain function they represent. Recent advances in noninvasive EEG
promoted the understanding of the neurological origin of
Robert P. Vertes and Timothy A. Allen (eds.),Electrophysiological Recording Techniques, Neuromethods, vol. 192,
https://doi.org/10.1007/978-1-0716-2631-3_1,©Springer Science+Business Media, LLC, part of Springer Nature 2022
1

particular EEG features [1]. However, investigation of the molecu-
lar and cellular levels of perturbations in humans is greatly limited
due to its invasiveness and safety concerns.
2 Jee Hyun Choi and Eunjin Hwang
In recent years, a variety of genetic mouse models have allowed
more comprehensive studies on the origin and meaning of neuronal
synchrony and oscillations featured in EEG signals. A thorough
investigation using optogenetic, electrophysiological, or neurolog-
ical intervention has subdued the controversy that EEG rhythms or
particular waveforms are not simple epiphenomena but have neu-
rological origins associated with cognitive or physiological func-
tions [2 –5]. However, conventional EEG in mice using only one or
two electrodes is not ideally suited as a translational tool because of
its lack of spatial information. Most EEG phenotypes in human
patients are expected to be reproduced in mouse disease models. In
addition, recent studies found that a composite of neuronal oscilla-
tions at different frequency bands in different cortical regions is
crucial in orchestrating the massive parallel computing of neuronal
units [6]. Therefore, the regionally distinctive oscillatory patterns,
in conjunction with behaviors, remain to be described in mouse
models. The lack of spatial information is a hindrance for mouse
researchers in interpreting the data at the network level or associat-
ing it with the ﬁndings from human EEG.
A decade ago, we introduced high-density EEG for freely
moving mice using a polyimide-based microarray to the neurosci-
ence community [7 ,8]. The primary purpose of this tool is to
signify EEG as a cross-species neuroimaging tool for comparative
study of neuronal activities in conjunction with cognition and
behaviors. The microarrays were speciﬁcally designed to ﬁt the
extracranial surface of the mouse, accommodating 40–60 channels,
and had a thickness of 8–10μm. The overall proﬁle is shaped like a
pinnate leaf, allowing a secondary probe or optical ﬁbers to intrude
into the deep brain, enabling simultaneous EEG/spike recording
or combination of EEG and optogenetics.
Despite t
onceptual advance, there has been a limited
expansion in the research community in using high-density EEG
in mice. There are multiple reasons that researchers hesitate to
adapt high-density EEG in their studies. The major challenge is
the difﬁculty of analyzing and interpreting complex EEG data. The
montage of mouse high-density EEG is different from that of
humans, and mice have more movement artifacts than humans.
These technical issues require the researcher to put more effort
into EEG preprocessing before using conventional EEG neuroim-
aging platforms. The nonstationary nature of EEG requires an
experimental paradigm to produce equivalent epochs to be aver-
aged to extract only the test-relevant signals, which is more chal-
lenging in mice. In humans, most cognitive paradigms are designed
to make all trials equivalent; however, these repetitive test para-
digms are rare in mice. Most importantly, although mice and

humans have core nervous systems and share central operational
mechanisms, evident differences exist, particularly in higher-level
cognition. Nonetheless, many fundamental mechanisms of neural
communication are universal across different species, and under-
standing these mechanisms from mice is likely to expand knowl-
edge of the human brain. The purpose of this chapter is to review
the methodology of high-density EEG and note several technical
considerations associated with it. We have been both developing
and recording high-densityEEG
for over 10 years, and we have
written this chapter based on our experience with the aim of
providing information to investigators who are interested in
obtaining high-density EEG in freely moving mice.
High-Density Electroencephalography in Freely Moving Mice 3
2 Materials
2.1 Technical
Challenges in EEG
Electrodes for Freely
Moving Mice Acquiring high-density EEG from mice in freely moving states
poses a signiﬁcant technical challenge since the small size of the
mouse brain (~0.5 cm
3
) limits the space for implantation of many
electrodes and the agile movements of the animal produce signiﬁ-
cant movement artifacts. In the last decades, tremendous advances
in chronic neural recordings have been achieved, particularly in the
ﬁelds of implantable silicon probe arrays [9], accelerated by devel-
opment of wireless recording systems. Expanding these technolo-
gies to acquire high-density EEG can take several technical factors
into account but is not straightforward due to the following dis-
tinct features of high-density EEG. First, high-density EEG
requests to collect information from a broad region of cortex.
This implies that the electrodes should be distributed over the
surface of a semi-inﬁnite medium, and the interface condition for
the contact should be equivalent for all EEG electrodes. The most
appropriate surface to apply the electrodes in mice is the skull
surface. This involves building a ﬂexible, dry electrode that can ﬁt
onto the curved skull surface. Second, EEG activities often appear
as oscillations whose frequencies are arranged from low (~1 Hz) to
high (~150 Hz). The serious issue in this broadband signal is the
abundance in the nonneural signal sources – some from physiologi-
cal origins such as heartbeat, respiration, vasomotion, etc., and
some from environmental noise, such as ventilation, acoustic reso-
nance of buildings, line noise, etc. Moreover, 1/f noise, which is
inversely proportional to the frequency, is common in biological
systems affecting the low-frequency regime of EEG. Therefore,
consistent contact/attachment to the skull with high mechanical
stability and effective electrical shielding is needed.

4 Jee Hyun Choi and Eunjin Hwang
2.2 Polyimide-Based
Microarray for Mouse
High-Density EEG To resolve the above issues, we applied a polyimide as the carrier
substrate and fabricated a microarray as depicted in Fig.1
[7,8]. Polyimide has been shown to have superior biocompatibility
and good mechanical properties [10] and has been widely used as a
substrate for ﬂexible neural probes [11]. As the skull surface is
ellipsoidal, the overall structure has a backbone line with multiple
branches on both sides. The thickness is crucial for stable adher-
ence. There is a trade-off in the thickness of microarray: in the
thinner, the yielding rate of production drops, resulting in more
cost per array, whereas in the thicker, it is more easily detached from
the skull. We tested different thicknesses and found that the most
appropriate thickness was 7–8μm with an acceptable level of up to
10μm. Figure1bshows different electrode conﬁgurations for
40 and 64 channels. In most cases, the 40-channel conﬁguration
is enough to generate topographical mapping. The 64-channel
conﬁguration is appropriate for somatosensory research, presenting
14 channels on top of the somatosensory cortex. For the electrical
contacts and interconnection lines, we used Pt [7], but later studies
showed that an Ag/AgCl electrode presents similar levels of signal
quality. A series of tests revealed that the most relevant features in
the electrode conﬁguration are the size of electrical contact, the
width of the interconnection line, and the interline spacing. We
found that the adequate impedance for EEG measurements was
obtained with an electrode contact size of ~0.2 mm
2
. The width of
interconnection lines and the interline spacing are key factors in the
design because they determine the overall size. We found that there
is no signiﬁcant generation of induced noise or cross-talk for widths
>15μm and interline spacing>20μm. With the best conﬁgura-
tion, the impedance ranges from 10 to 300 kΩat 30 Hz when
measured on the skull.
Fig. 1Polyimide-based microarray for mouse high-density EEG. (a ) Picture of a 40-channel microarray. (b )
Conﬁguration of 40-channel (left) and 60-channel (right) microarrays. Inset images are actual microarrays
positioned on a mouse skull. (c ) Electrode locations overlaid on a cortical region map

High-Density Electroencephalography in Freely Moving Mice 5
3 Methods
3.1 Fixation of
Microarray For a favorable surgery outcome, we suggest using bone cement as
a ﬁxation material. Different types of commercially available bone
cements are applicable, but an ideal cementing material should be
noncorrosive, tightly ﬁlling, and not affected by moisture and
should have long-term stability. A typical cause of failure is the
detachment of the cement from the skull. A couple of anchoring
microscrews on the skull helps in preventing this detachment. Also,
when one applies cement, any debris on the skull surface should be
cleaned. We use a tap-water-soaked cotton swab for the cleaning,
which is used for pressing the microarray as well. We make the
surface wet with tap water when we place and press the microarray
and wait until the surface is fully dry. Placing cement is extremely
important for preventing its detachment from the skull. We suggest
using low-viscosity cement and applying the cement with a brush.
One should check the cement is fully dried before placing the next
cement.
3.2 Recording The measurement principle and instrumentation for mouse EEG
are identical to those for human EEG. The basic principle of EEG
recording is to measure the potential difference between two points
as the direct consequence of the existing electric dipole created by a
synchronously generated postsynaptic potential in the cortex.
These potentials can be measured by a voltage ampliﬁer. Typically,
the microarray electrode has sub-Ω-level resistance, but on the
skull, the impedance level is approximately 200 kΩwhen measured
at 30 Hz. Determination of the interface impedance is not trivial
because the interface between the electrode and the skull constitu-
tes a signiﬁcant resistance and capacitance. In human scalp EEG,
the interface impedance is at the level of ~50 kΩ, and in general, the
difference of impedance between channels does not affect the read-
ings. However, when variation of the interface impedances between
channels is high, we recommend normalizing the signal levels.
Typically, we perform normalization of EEG by dividing the spec-
tral power above 200 Hz in all channels using EEG acquired in the
quiescent moment of mouse. For comparisons between different
mice, normalization of EEG is required. Additionally, an electrode
with an impedance level above 500 kΩshould be excluded from
further analysis.
3.3 Artifacts and
Noises Recording extracranial EEG from freely moving mice is subject to
more environmental noise than is present in other electrophysio-
logical modalities. The environmental noise may arise from a variety
of sources, such as the mouse’s own body, instrumentation, or the
experimenter. The waveforms of noises depend on their noise types,
and unfortunately, many noises have waveforms similar to the EEG

r
such as sleep waves, epilepsy, oscillations, etc. Therefore, it is of
utmost importance to create perfect conditions to minimize the
noise. First, it is critical to keep the animal within a closed equipo-
tential surface so that the net electric ﬂux outside the brain is close
to zero. This is performed by shielding the recoding place with a
conducting shell (Faraday cage) and connecting all the electrical
devices, experimenters, and animals to this cage. In chronic record-
ing, a metal ﬂoor is
preferred to a glass or plastic ﬂoor. In acute or
head-ﬁxed recording, the earbars or ﬁxers should be wired to the
Faraday cage to maintain the body and cage equipotential. To
discharge any electric surge, grounding the Faraday cage to the
Earth is helpful, but in a noise-rich environment (e.g., near subway
or in the hospital), a ﬂoating ground could be better than earth
ground. Another serious contamination in chronic recording of
mouse EEG is the swing noise generated when the wire bundle
swings as the headmoves.
We found that the noise comes mostly
from the connection between the wires and microarray connector,
and hence it is suggested to this part rigid to avoid any artifact due
to connector movement. Despite all the above efforts, an ideal
recording without artifacts is pragmatically unachievable. There-
fore, EEG researchers normally perform ofﬂine postprocessing to
remove the contamination.
6 Jee Hyun Choi and Eunjin Hwang
3.4 Analyzing EEG The surface potential produced by a mouse brain has an order of a
fewμV, which is a comparable level to the human brain. Basically,
EEG signals are nonstationary data and the operations are done in a
time domain or frequency domain. In the case of averaging multi-
ple events in the time domain, an ergodic property of the signals is
assumed, i.e., all events are equivalent, which may be violated in the
real world. Hence, prior to the data analysis, several experimental
considerations are critical to obtain consistent signals across trials.
For example, cognitively or behaviorally equivalent trials, precise
timing, and uncorrelated noise across different trials are all impor-
tant conditions for a good EEG. In the case of a study of spontane-
ous activities, spectral power or phase-amplitude coupling
properties are analyzed in the frequency domain. In either case,
preprocessing is critical to uncorrelate the noise from the signal.
3.4.1 Preprocessing One of the ﬁrst steps in EEG analysis is to identify and remove
potentially contaminated signals or artifacts. Mice produce signiﬁ-
cant levels of artifacts, especially in a freely mobile condition, and in
many situations, these artifacts are not stereotypical and are con-
fused with real brain activity, such as epileptic activity. During the
experiment, an accelerometer recording the movement [12]o
videotaping helps in distinguishing the movement artifacts from
the signals. In our experience, laboratory mice, even the wild type,
produce epileptic events more often than humans. In most cases,
the epileptic events occur during a quiescent state and do not occur
in all channels like movement artifacts do, and this characteristic

allows the investigators to distinguish the events. It is recom-
mended to separate the epileptic periods prior to further analysis.
High-Density Electroencephalography in Freely Moving Mice 7
The other commonly monitored artifact is the electrocardiog-
raphy (ECG), i.e., the electrical activity of the heart. ECG is often
observed in acute recording and has a stereotypic waveform. The
artifact with a well-deﬁned temporal signature like ECG could be
easily identiﬁed with an independent component analysis (ICA),
which is a blind source separation method, decomposing a signal
into statistically independent components from each other
[13]. ICA could also be used to separate neural signals which are
independent from other signals in the time and space domain. An
open-source algorithm of ICA is available from EEGLAB, a
MATLAB-based toolbox released at the Swartz Center for Compu-
tational Neuroscience [14] (downloadable athttps://sccn.ucsd.
edu/eeglab/).
In the case of mouse EEG, muscle activities are not detected,
probably due to cementing. Instead, a concern arose in comparing
rhythmical activities across a long period of time because the mouse
brain is more vulnerable to variation of pink noise. Pink noise
appears as a low-frequency component with an inverse function of
frequency in the frequency domain. It is also known to be ubiqui-
tous. The strength of pink noise ﬂuctuates over time to a signiﬁcant
degree in the mobile condition. Hence, investigators should be
cautious when they compare spectral power across time. It is
recommended to eliminate the pink noise or study the power in a
ratiometric manner.
3.4.2 EEG TopographyOne of the advantages of high-density EEG in mice is the ability to
investigate the spatiotemporal change of cortical activation with
topographical analysis. EEG topography is a neuroimaging tech-
nique for mapping the spatial distributions of cortical activity.
Basically, EEG topography is a contour plot on the top surface of
the mouse brain. First, we create a mesh (set of imaginary points) to
represent a cortical surface. Typically, the stereotaxic resolution is
used for the mesh size so that any EEG channel can be correspon-
dent to a mesh. Second, we limit the surface to an ellipsoidal
boundary. Third, the potential value of the mesh is calculated
from an interpolation method. According to our tests, different
types of interpolation do not affect the results to a noticeable
degree. We normally use cubic spline interpolation. Figure2
s
hows a c
ine-mode presentation of EEG topographies, while the
ventral posterior medial nucleus of the thalamus (VPM) of a Thy1-
ChR2-EYFP mouse was optogenetically stimulated. The outer sur-
face of the mouse neocortex was rendered by aspm_surffunction in
the SPM8 open-source toolbox (Wellcome Trust Centre for Neu-
roimaging UCL, London, UK) based on the mouse MRI down-
loaded from the open database of the magnetic resonance
microimaging neurological atlas group [15].

8 Jee Hyun Choi and Eunjin Hwang
Fig. 2Topographical mapping of cortical signal propagation in response to optogenetic stimulation of VPM. (a )
Schematics of the signal propagation pathway (upper) and site of stimulation (lower). (b ) Cine-mode
presentation of EEG topography
4 Note on Applications
4.1 Simultaneous
Recordings of High-
Density EEG and
Subcortical Brain
Activities EEG is a widely used method to monitor human brain activities due
to its noninvasiveness and high temporal resolution. However,
since human EEG measures the voltage changes from the scalp,
an investigation of subcortical origins or their interplay with corti-
cal activity is highly limited in human EEG. In this section, we
introduce a method to combine high-density EEG with a conven-
tional local ﬁeld potential (LFP) to simultaneously monitor brain
activities in the subcortical region and cortex in mice. By combin-
ing EEG and LFP, we are able to investigate the functional connec-
tivity between the subcortical region and cortex. Here, we show an
example of combined EEG and LFP in a study of spike-and-wave
discharge (SWD) in the thalamus and cortex, simultaneously. The
SWDs are known to be the EEG hallmark of absence seizure [16]
and are often associated with loss of consciousness [17 ]. We moni-
tored the thalamic activation with cortical activation and the func-
tional connectivity between the two regions when the mouse
experienced absence seizure. The origin of such connectivity is
still under investigation, but an emergent phenomenon of syn-
chrony between the thalamus and cortex has been observed previ-
ously. The synchrony analysis of LFP and EEG quantitatively
assesses the functional connectivity of the thalamus and cortex,
and a calculation of time lag of thalamic and cortical SWDs leads
us to infer the causation of the signal.

High-Density Electroencephalography in Freely Moving Mice 9
4.1.1 Methodological
Considerations of
Simultaneous Recording
of
High-Density EEG and LFPMost of the procedures of implementing LFP together with EEG
are the same as implementation of EEG only. The only difference is
that making a hole for insertion of the LFP electrode and the
ﬁxation of the LFP electrode takes place after the EEG microarray
has been attached on the skull. In the case where the subcortical
region of interest is along the midline, where the interconnection
lines of EEG channels are located, the hole can be moved laterally
byxin the same coronal plane, and the depth of electrode insertion,
d, is calculated by the Pythagorean theorem (x
2
+z
2
¼d
2
), in which
zis the dorsoventral coordinate. The angle of the electrode inser-
tion equals the tangent ofz/x. The position of the LFP electrode
should be conﬁrmed by brain histology after all experiments are
conducted. Typically, it is recommended to match the impedances
of the reference and active electrode at similar levels [18], but
sharing the same reference and ground electrodes for the LFP
electrode delivers similar signal quality. Moreover, it is important
to reduce the number of electrodes for the well-being of the
animals in long-term chronic recordings.
4.1.2 Comparison of EEG
and LFP Signals During
SWD One of the most interesting questions to explore about seizures is
the seizure initiation factors. Here, we exemplify the application of
high-density EEG to pharmacologically induced absence seizure.
To produce SWDs in the mouse, we injected gamma-butyrolactone
(GBL, 100 mg/kg, i.p.) in awake mice and monitored them during
the freely moving state. A previous study of simultaneous record-
ings of intracellular potentials and EEG showed that the spike of
SWD in EEG corresponds to synchronous burst ﬁrings activated by
low-voltage activated channels [19]
hich consequently recruit
the neurons to activate synchronously. By comparing the time
lags between spike peaks, the initiation or propagation of seizure
can be studied. Figure3ashows an exemplary EEG and LFP during
SWD. In each channel, the ﬁrst positive spikes were identiﬁed and
marked with inverted triangles. The polarity of the depth electrode
is the opposite to that of the surface electrode; thus, it was inverted
in the plot. In the particular event, the ﬁrst spike was monitored in
the frontal cortex including the prefrontal and motor cortices. The
level of seizure was calculated by total power in the period of SWD
(Fig.3c)
nd the functional connectivity between different cortical
regions (Fig.3d) was represented as thickness of the lines. In this
case, the functional connectivity was calculated by cross-correlation
coefﬁcient, which assesses the similarity of two signals. Other vari-
ables, such as the phase synchronization index, could also be used
to gauge the synchronization level of the two regions [20]. Most of
the connections were between the two sites located symmetrically
with respect to the midline. Likewise, the functional connectivity
between the thalamus and cortex (Fig.3e) was calculated by the
cross-correlation coefﬁcient and mapped as a color map. The tem-
poral relationship between spikes was assessed with peak latency
and mapped for the ﬁrst and second peaks, respectively (Fig.3f, g).

10 Jee Hyun Choi and Eunjin Hwang
Fig. 3Simultaneous recording of EEGs and LFPs during GBL-induced SWD. (a) Exemplary traces of SWD. (b)
Cortical high-density EEG montage and relative location of LFP electrode in VB (green dot). (c ) Power map of
SWD. Total power from 1 Hz to 60 Hz was used for topographical representation. (d ) Cortico-cortical functional
connectivity during SWD. Functional connectivity was measured by computing the cross-correlation between
channels. Connections stronger than 0.95 were shown as links between channels. (e ) Functional connectivity
between the cortex and thalamus. (f) Latency map of ﬁrst spikes with respect to the ﬁrst spike of the FP1
channel. (g ) Latency map of second spikes with respect to the second spike of the FP1 channel
4.2 Use of High-
Densi
ty EEG in
Preclinical StudiesDespite much investigations by pharmaceutical companies, neuro-
psychiatry or neurological diseases notoriously have a low clinical
trial success rate (6.2% without biomarker and 8.3% with bio-
marker, data from 2005 to 2015 [21]). One compelling explana-
tion is a failure in selecting the right animal models and relevant
biomarkers for the disease. Among major diseases, there are chal-
lenges and limitations in the validated and predictive animal models
for neuropsychiatry or neurological diseases because most cognitive
dysfunctions in human patients are not directly assessable in animal
models. Therefore, it is critical to develop cross-species assays for
comparing equivalent test scores in human patients and animal

models. In this respect, high-density EEG could be pivotal to
bridge the translational gap to the clinic by presenting the compar-
ative neuromarkers of cognitive dysfunctions.
High-Density Electroencephalography in Freely Moving Mice 11
One suggestion is the auditory steady-state response (ASSR)
test battery. The ASSR is an electrophysiological response entrained
to periodic sound pulses and, in particular, the ASSR appears as a
resonant response to stimuli in the gamma frequency band
(30–50 Hz) in humans [22] and rodents [23]. Gamma-band
ASSR has been suggested as a neuromarker for the detection of
psychosis. For instance, abnormal gamma-band ASSR has been
robustly observed in patients with ﬁrst-episode or chronic schizo-
phrenia [24–26]. The ASSR test could be effective and useful in
screening the schizophrenia mouse model. Although it is known
that more than 100 genes are associated with schizophrenia [27],
schizophrenia mouse models are not validated by the genetic fac-
tors alone since the gene-environment interactions also play a
crucial role for the manifestation of schizophrenia symptoms
[28]. Therefore, most validations of schizophrenic mouse models
have been based on the behavioral tests such as the prepulse inhibi-
tion evaluating sensorimotor gating [29], locomotor activity
reﬂecting psychomotor agitation [30], social withdrawal [31 ], cog-
nitive deﬁcits in working memory, learning, and attention [30],
which deliver neither direct evidence nor comparative results. Con-
trary to the behavioral assays above, the gamma-band ASSR is
observed in both humans and animals [32]. Additionally, it is a
cross-species comparative marker because the generator for
gamma-band oscillations has a cellular origin rather than a circuitry
origin [33,34]. An example of an impaired ASSR and auditory
evoked potential was reported for a schizophrenia mouse model
[35]. In many cases, knockout mouse models might carry more
phenotypes than patients. We believe that a combination of ﬁne
tuning of genetic modiﬁcation and cross-species comparative high-
density EEG could be used to effectively identify the patient-like
mouse model.
In sum, we believe that more precise selection of a mouse
model of disease could be achieved with high-density EEG valida-
tion, which will enhance the translational value of the mouse model
and maximize the synergy of preclinical and clinical studies.
References
1.Cohen MX (2017) Where does EEG come
from and what does it mean?
Trends
Neurosci
40(4):208–218.https://doi.org/10.1016/j.
tins.2017.02.004
2.Bragin A, Jando G, Nadasdy Z, Hetke J,
Wise K, Buzsaki G (1995) Gamma (40-100
Hz) oscillation in the hippocampus of the
behaving rat. J Neurosci 15(1 Pt 1):47–60
3.Buzsaki G, Wang XJ (2012) Mechanisms of
gamma oscillations. Annu Rev Neurosci 35:
203–225.https://doi.org/10.1146/annurev-
neuro-062111-150444
4.O’Keefe J
ecce ML (1993) Phase relation-
ship between hippocampal place units and the
EEG theta rhythm. Hippocampus 3(3):

12 Jee Hyun Choi and Eunjin Hwang
317–330.https://doi.org/10.1002/hipo.
450030307
5.Tort AB, Kramer MA, Thorn C, Gibson DJ,
Kubota Y, Graybiel AM, Kopell NJ (2008)
Dynamic cross-frequency couplings of local
ﬁeld potential oscillations in rat striatum and
hippocampus during per
formance of a T-maze
task. Proc Natl Acad Sci U S A 105(51):
20517–20522.https://doi.org/10.1073/
pnas.0810524105
6.Singer W (2013) Cortical dynamics revisited.
Trends Cogn Sci 17(12):616–626.https://
doi.org/10.1016/j.tics.2013.09.006
7.Choi JH, Koch KP, Poppendieck W, Lee M,
Shin HS (2010) High resolution electroen-
cephalography in freely moving mice. J Neuro-
physiol 104(3):1825–1834.https://doi.org/
10.1152/jn.00188.2010
8.Lee M, Kim D, Shin HS, Sung HG, Choi JH
(2011) High-density EEG recordings of the
freely moving mice using polyimide-based
microelectrode. J Vis Exp (47):2562.https://
doi.org/10.3791/2562
9.Buzsaki G, Stark E, Berenyi A, Khodagholy D,
Kipke DR, Yoon E, Wise KD (2015) Tools for
probing local circuits: high-density silicon
probes combined with optogenetics. Neuron
86(1):92–105.https://doi.org/10.1016/j.
neuron.2015.01.028
10.Richardson RR Jr, Miller JA, Reichert WM
(1993) Polyimides as biomaterials: preliminary
biocompatibility testing. Biomaterials 14(8):
627–635
11.Viventi J, Kim DH, Vigeland L, Frechette ES,
Blanco JA, Kim YS, Avrin AE, Tiruvadi VR,
Hwang SW, Vanleer AC, Wulsin DF, Davis K,
Gelber CE, Palmer L, Van der Spiegel J, Wu J,
Xiao J, Huang Y, Contreras D, Rogers JA, Litt
B (2011) Flexible, foldable, actively multi-
plexed, high-density electrode array for
mapping brain activity in vivo. Nat Neurosci
14(12):1599–1605.https://doi.org/10.
1038/nn.2973
12.Hwang E, Kim S, Han K, Choi JH (2012)
Characterization of phase transition in the tha-
lamocortical system during anesthesia-induced
loss of consciousness. PLoS One 7(12):
e50580.https://doi.org/10.1371/journal.
pone.0050580
13.Bell AJ, Sejnowski TJ (1995) An information-
maximization approach to blind separation and
blind deconvolution. Neural Comput 7(6):
1129–1159
14.Delorme A, Makeig S (2004) EEGLAB: an
open source toolbox for analysis of single-trial
EEG dynamics including independent compo-
nent analysis. J Neurosci Methods 134(1):
9–21.https://doi.org/10.1016/j.jneumeth.
2003.10.009
15.Ma Y, Smith D, Hof PR, Foerster B,
Hamilton S, Blackband SJ, Yu M, Benveniste
H (2008) In vivo 3D digital atlas database of
the adult C57BL/6J mouse brain by magnetic
resonance microscopy. Front Neuroanat 2:1.
https://doi.org/10.3389/neuro.05.001.
2008
16.Pinault D, O’Brien TJ (2005) Cellular and
network mechanisms of genetically-
determined absence seizures. Thalamus Relat
Syst 3(3):181–203.https://doi.org/10.
1017/S1472928807000209
17.Blumenfeld H (2005) Consciousness and epi-
lepsy: why are patients with absence seizures
absent? Prog Brain Res 150:271–286.
https://doi.org/10.1016/S0079-6123(05)
50020-7
18.Schomer DL, da Silva FL (2011) Nieder-
meyer’s electroencephalography: basic princi-
ples, clinical applications, and related ﬁelds,
6th edn. Lippincott Williams & Wilkins,
Philadelphia
19.Pinault D, Leresche N, Charpier S, Deniau JM,
Marescaux C, Vergnes M, Crunelli V (1998)
Intracellular recordings in thalamic neurones
during spontaneous spike and wave discharges
in rats with absence epilepsy. J Physiol 509
(Pt 2):449–456
20.Stam CJ (2005) Nonlinear dynamical analysis
of EEG and MEG: review of an emerging ﬁeld.
Clin Neurophysiol 116(10):2266–2301.
https://doi.org/10.1016/j.clinph.2005.
06.011
21.Wong CH, Siah KW, Lo AW (2018) Estima-
tion of clinical trial success rates and related
parameters. Biostatistics.https://doi.org/10.
1093/biostatistics/kxx069
22.Galambos R, Makeig S, Talmachoff PJ (1981)
A 40-Hz auditory potential recorded from the
human scalp. Proc Natl Acad Sci U S A 78(4):
2643–2647
23.Conti G, Santarelli R, Grassi C, Ottaviani F,
Azzena GB (1999) Auditory steady-state
responses to click trains from the rat temporal
cortex. Clin Neurophysiol 110(1):62–70
24.Kwon JS, O’Donnell BF, Wallenstein GV,
Greene RW, Hirayasu Y, Nestor PG, Hasselmo
ME, Potts GF, Shenton ME, McCarley RW
(1999) Gamma frequency-range abnormalities
to auditory stimulation in schizophrenia. Arch
Gen Psychiatry 56(11):1001–1005
25.Spencer K
alisbury DF, Shenton ME,
McCarley RW (2008) Gamma-band auditory
steady-state responses are impaired in ﬁrst epi-
sode psychosis. Biol Psychiatry 64(5):

High-Density Electroencephalography in Freely Moving Mice 13
369–375.https://doi.org/10.1016/j.
biopsych.2008.02.021
26.Thune H, Recasens M, Uhlhaas PJ (2016)
The
40-Hz auditory steady-state response in
patients with schizophrenia: a meta-analysis.
JAMA Psychiat 73(11):1145–1153.https://
doi.org/10.1001/jamapsychiatry.2016.2619
27.Schizophrenia Working Group of the Psychiat-
ric Genomics C (2014) Biological insights
from 108 schizophrenia-associated genetic
loci. Nature 511(7510):421–427.https://
doi.org/10.1038/nature13595
28.van Os J, Rutten BP, Poulton R (2008) Gene-
environment interactions in schizophrenia:
review of epidemiological ﬁndings and future
directions. Schizophr Bull 34(6):1066–1082.
https://doi.org/10.1093/schbul/sbn117
29.Geyer MA, McIlwain KL, Paylor R (2002)
Mouse genetic models for prepulse inhibition:
an early review. Mol Psychiatry 7(10):
1039–1053.https://doi.org/10.1038/sj.mp.
4001159
30.Powell CM, Miyakawa T (2006)
Schizophrenia-relevant behavioral testing in
rodent models: a uniquely human disorder?
Biol Psychiatry 59(12):1198–1207.https://
doi.org/10.1016/j.biopsych.2006.05.008
31.Wilson CA, Koenig JI (2014) Social interaction
and social withdrawal in rodents as readouts for
investigating the negative symptoms of schizo-
phrenia. Eur Neuropsychopharmacol 24(5):
759–773.https://doi.org/10.1016/j.
euroneuro.2013.11.008
32.Brenner CA, Krishnan GP, Vohs JL, Ahn WY,
Hetrick WP, Morzorati SL, O’Donnell BF
(2009) Steady state responses: electrophysio-
logical assessment of sensory function in
schizophrenia. Schizophr Bull 35(6):
1065–1077.https://doi.org/10.1093/
schbul/sbp091
33.Sohal VS, Zhang F, Yizhar O, Deisseroth K
(2009) Parvalbumin neurons and gamma
rhythms enhance cortical circuit performance.
Nature 459(7247):698–702.https://doi.org/
10.1038/nature07991
34.Cardin JA, Carlen M, Meletis K, Knoblich U,
Zhang F, Deisseroth K, Tsai LH, Moore CI
(2009) Driving fast-spiking cells induces
gamma rhythm and controls sensory responses.
Nature 459(7247):663–667.https://doi.org/
10.1038/nature08002
35.Shahriari Y, Krusienski D, Dadi YS, Seo M,
Shin HS, Choi JH (2016) Impaired auditory
evoked potentials and oscillations in frontal and
auditory cortex of a schizophrenia mouse
model. World J Biol Psychiatry 17(6):
439–448.https://doi.org/10.3109/
15622975.2015.1112036

Chapter2
Multisite Recording for the Analysis of Information Flow
Between Thalamocortical Regions
Hiroshi T. Ito
Abstract
High-density multichannel recordings are a powerful technique to investigate the information represented
in the activity of neuronal ensembles. Recent technological advancement has achieved the development of
small and low-weight recording devices, allowing for the use of hundreds of channels to monitor the neural
activity from freely behaving animals. Because of the substantial increase in the channel number of
recording devices, it is more common to record the neural activity from multiple brain regions simulta-
neously, allowing for researchers to investigate functional dependency and interactions5 between regions.
However, multiregional recordings often require different techniques from the ones used for single-region
studies. This chapter will review the methodological details of simultaneous recordings from multiple brain
regions as well as analysis techniques used for such datasets.
Key wordsSpatial navigation, Multiregional interactions, Ensemble analysis, Freely behaving record-
ings, Tetrodes
1 Introduction
Animals need to adapt their behaviors to dynamically changing
environments, and thus functional demands of the brain are differ-
ent from moment to moment. A large body of experimental evi-
dence suggests that the brain can cope with various behavioral
demands by changing functional interactions between regions.
Synchrony of oscillatory activity in the brain has been considered
a key mechanism to modulate multiregional interactions [1–3].
While many studies have demonstrated the difference in the degree
of synchrony depending on behavioral demands, direct demonstra-
tion of modulation of information ﬂow by this mechanism is still
experimentally challenging. Here, large-scale recordings of spike
activity together with local ﬁeld potentials from multiple brain
regions are a key technique to investigate this question.
Goal-directed spatial navigation is considered a behavior that
requires multiregional interactions in the brain. When animals
Robert P. Vertes and Timothy A. Allen (eds.),Electrophysiological Recording Techniques, Neuromethods, vol. 192,
https://doi.org/10.1007/978-1-0716-2631-3_2,©Springer Science+Business Media, LLC, part of Springer Nature 2022
15

intend to visit a particular location in space, they ﬁrst need to
identify their current position in the environment. Then, based
on previous experiences or memories about the environment,
they are required to estimate the direction and route to the desti-
nation. This entire process requires multistep computations,
including the process of incoming sensory signals, retrieval of
memory about the environment, and decision and planning of the
destination and next actions. Each of these processes requires
cooperation of multiplebrain
regions, and in support of this idea,
previous studies have reported temporal coordination of the activ-
ity between distant brain regions during navigation [4–8]. For
example, when ratsmake
decisions about the next route, spike
times of neurons in the prefrontal cortex, a brain area crucial for
action planning or decision making [9,10], are phase-locked to the
theta rhythm in local ﬁeld potentials in the hippocampus, a part of
the brain’s spatial representation system [11–13]. This temporal
coordination or synchrony is considered a mechanism of transfer-
ring information necessary for route decisions between the prefron-
tal cortex and the hippocampus [4,14,15]. Multiregional
recording further
helped identify the exact information content
that is transferred between regions [16–18]. These studies revealed
that the midline thalamic nucleus reuniens (RE) [18–20] (Fig.1a)
is an anatomical hub to transfer the animal’s next movement direc-
tion from the prefrontal cortex to the hippocampus during route
decisions. In the following sections, I will review the techniques
used in these studies as an example, but the same techniques should
also be applicable to other brain regions.
16 Hiroshi T. Ito
2 Construction of Microdrive for Multisite Recordings
2.1 Tetrode and
Microdrive Each channel of recording electrodes picks up voltage signals
derived from hundreds of neurons nearby. Voltage signals gener-
ated by neurons are usually associated with either spikes or synaptic
potentials. A high-pass ﬁlter is used to isolate signals derived from
spikes, because spike waveforms are mostly conﬁned in a high-
frequency spectral range compared to those of synaptic potentials.
A typical range of passbands used for spike detection is between
300 and 6000 Hz. To investigate the activity of individual neurons,
it also requires to classify and assign detected spikes to individual
neurons. While spike waveforms are often used as a criterion for
spike classiﬁcation, the waveforms of the same neurons are not
necessarily always the same, for example, during spike bursting,
leading to a possibility of misclassiﬁcation. Here, a high-density
channel conﬁguration of electrodes provides another criterion for
spike classiﬁcation [21,2
]. W hen the electrode has multiple chan-
nels in small space, signals from the same neuron can be detected
across multiple channels that are placed at different positions

relative to the neuron. Due to the decay of electrical signals along
the distance from the source, the ratio of spike amplitudes between
channels gives a distinct signature of the relative position of indi-
vidual neurons, improving the accuracy of spike classiﬁcation.
Multisite Recording for the Analysis of Information Flow Between... 17
Fig. 1Microdrive construction: (a ) A scheme showing anatomical locations of the
nucleus reuniens (RE) and the hippocampus CA1. (b ) Magnetic stirrer to make
tetrodes. The bottom white bar rotates to make twists of wires. (c ) Left. A Harlan
28 microdrive with split bundles. Right. The bottom side view of tetrode bundles
showing 28 holes for tetrodes and the inset showing an extended tetrode from
the bundle. d. Tetrode bundle is constructed by binding 14 metal tubings
together by heat-shrink tubings and applying solder in the middle. The bundle
is then cut in the middle, resulting in two bundles, each of which can hold
14 tetrodes
Tetrodes are a type of electrode with four independent channels
made by twisted wires. A tetrode is constructed manually from a
single 17μm wire made of 90% platinum and 10% iridium (Cali-
fornia Fine Wire). First, a wire loop is formed by connecting both
ends with heat-resistant copper tape. The wire loop is then twisted

and hanged over a magnetic stirrer using a magnetic bar to hold the
bottom twist of the wire (Fig.1b). After rotating the magnetic bar
for a few minutes, multiple twists of four wires will be formed. This
tetrode is then heated up to 210

C with a heat gun for 2 min for
sealing.
18 Hiroshi T. Ito
A microdrive of tetrodes can be obtained commercially, for
example, from Neuralynx or Axona, but it is also possible to con-
struct with your own design using a 3D printer. A Harlan 28 drive
from Neuralynx, for example, can hold 28 independently movable
tetrodes (Fig.1c), which is a good size for the recordings from the
hippocampus and the nucleus reuniens simultaneously in behaving
rats. The microdrive will be ﬁxed on the animal’s skull using several
anchor screws (e.g., M1.6, 3 mm) during the surgery. The ground
screws of the drive are typically placed on the skull above the
cerebellum for recordings from the hippocampus, the thalamus,
or other neocortical regions, as the cerebellar signals are presum-
ably largely uncorrelated with ones in these recording areas. The
use of reference tetrode is useful to reduce noise from muscle
activity (e.g., chewing noise), which helps obtain better spike wave-
forms for classiﬁcation.
2.2 Design of Split
Bundle For simultaneous multisite recordings, it is important to conﬁgure
the insertion angles and positions of individual tetrodes desirably so
that they can precisely target different brain structures. For this
purpose, we construct cannula bundles using 30 gauge hypodermic
tubings to hold individual tetrodes at precise positions. First, a
desired number of tubings are held together by using a pair of
heat-shrink tubings (Fig.1d). Then a thin layer of solder is applied
in the middle of the tubing to ﬁx them ﬁrmly. Finally, the soldered
region in the middle should be cut into two pieces, resulting in two
tetrode bundles. The separation distance and angle of each bundle
should be carefully adjusted for targeted brain regions and ﬁxed
with dental cement (Fig.1c).
Usually, each hypodermic tubing of the bundle is dedicated for
one tetrode. However, for targeting deep brain structures (>5mm
depth), the strength of tetrodes is sometimes not sufﬁcient to
penetrate the brain tissue. We thus sometimes make a bundle of
two tetrodes that are glued together and come out from one
hypodermic tubing. This strategy works well for recordings from
the thalamic nucleus reuniens that is located approximately 7 mm
below from the cortical surface [16].
2.3 Recording of the
Neural Activity
Together with the
Animal’s BehaviorsSeveral companies offer a complete set of recording systems with
software integrated with video tracking of the animal’s positions
and head directions, which is necessary for the investigation of
neural correlates to behaviors. If one wants to use an open-source
system (e.g., Open Ephys), it is necessary to combine the recording
system with a position tracking software to monitor the animal’s

Þ
Þ
behaviors. In such a case, it is important to make sure to use the
same time clock for position tracking and recording systems. Usu-
ally, all timestamps should be aligned to the clock signals of the
recording system. For example, in our system, whenever a new
image frame from CCD camera is acquired by the image grabber
(PC2-Comp Express, DALSA), a TTL pulse is sent to the recording
system so that the times of individual frame acquisitions are
registered as the timestamps of therecording data. Positions and
head directions of the animal can be extracted by monitoringtwo
LEDs
with different colors (e.g., red and green) mounted on the
headstage.
Multisite Recording for the Analysis of Information Flow Between... 19
3 Analysis of Multisite Recording Data
Simultaneous recordings from multiple brain regions expand the
possibility for analyses, because one can use not only typical meth-
ods used for single brain regions but also additional methods to
investigate interactions between the regions. This section describes
several useful techniques, using the analysis of thalamo-
hippocampal circuit as an example.
3.1 Spectral
Coherence AnalysisSpectral coherence analysis is a widely used technique to assess the
degree of synchrony of oscillatory activity between regions. This
technique is mathematically related to spectral power analysis; while
the spectral power analysis is based on an autocovariance function
of local ﬁeld potentials (LFPs) from a single brain region, the
spectral coherence analysis is on a cross-covariance function of
LFPs between regions.
Spectral power density:
S
xxωðÞ¼
X
τ¼1
τ?τ1
γ
xxτðÞe
τ2πiωτ
γ
xxτðÞ¼Ex tτμ
xðÞ x tþττμ
xðÞ?γ :Autocovarianceð
x
t:signal at timet,μ:signal average
Cross-spectrum:
S
xyωðÞ¼
X
τ¼1
τ?τ1
γ
xyτðÞe
τ2πiωτ
γ
xyτðÞ¼Ex tτμ
xðÞ y
tþτ
τμ
y
hi
:Cross‐covarianceð
y:s
he other brain region

20 Hiroshi T. Ito
Spectral coherence:
C
xyωðÞ¼
S
xyωðÞ
γ
γ
γ
γ
2
SxxωðÞSyyωðÞ
Therefore, spectral coherence is considered “normalized”
cross-spectrum. The calculation of spectral coherence does not
require spike data, and thus it can be performed if at least one
channel is located in individual brain regions.
While spectral coherence analysis provides the degree of coher-
ence across different frequencies, it does not give time-dependent
changes. When one wants to investigate behavior-dependent
changes of coherence, for example, during a goal-directed naviga-
tion task, a time-resolved strategy should be applied by using a
sliding time window. There are several choices of windows (e.g.,
Hanning or Hamming), which reduces the edge effects due to the
handling of short-duration datasets. It is also known that the multi-
taper method helps reduce estimation bias [23–25],
a useful
MATLAB toolbox, Chronux, is available for its
implementation [25].
In F
2a, t he spectral coherence analysis was applied when the
animal performed a continuous alternation task, in which the ani-
mal was required to choose either right or left turn at the
T-junction of the maze alternately to obtain a reward. The analysis
reveals a peak of coherence at the theta range of frequency
(6–12 Hz) between LFPs in CA1 and the nucleus reuniens (RE).
The RE has mutual anatomical connections with the prefrontal
cortex and also gives rise to inputs in the area CA1 and the sub-
iculum regions of the hippocampus [19,2
–28]. T he RE has been
considered a key anatomical hub for the communication between
the prefrontal cortex and the hippocampus because virtually no
direct projection exists from the prefrontal cortex to the hippocam-
pus [27]. Therefore, synchrony at the theta rhythm implies func-
tional coupling between RE and CA1, mediating prefrontal-
hippocampal interactions. To investigate if coherence is modulated
during the task phase, time-resolved spectral coherence analysis was
implemented between RE and CA1 using a sliding time window of
512 ms. The analysis shows a prominent coherence in the theta
frequency range that changes from time to time during the behav-
ioral task (Fig.2b)
he degrees of theta coherence are further
mapped on spatial positions of the maze using a color code,
which revealed a clear spatial bias of theta coherence (Fig.2c).
The theta coherence between CA1 and RE is higher at the stem
part of the maze in which the animals are required to make a
decision of the next trajectory. The enhanced synchrony between
CA1 and RE implies a change of information ﬂow, suggesting a key
role of their functional coupling during trajectory decisions.

Multisite Recording for the Analysis of Information Flow Between... 21
Fig. 2Spectral coherence analysis between CA1 and RE LFPs. (a ) Top: LFPs from the hippocampus CA1 and
the nucleus reuniens (RE). Bottom: the left two plots showing spectral power of LFPs from CA1 and RE. The
right plot shows spectral coherence. Note a peak of coherence at the theta range of frequency (6–12 Hz). (b )
Time-resolved spectral coherence of LFPs from CA1 and RE. The degree of coherence is color-coded. A
prominent coherence peak at the theta rhythm is observed, but it varies over time. (c ) The degree of spectral
coherence is plotted on spatial positions of the maze. Left. A rat performed a continuous alternation task in an
8-shaped maze. At the stem part of the maze, the animal was required to choose either right or left trajectory
based on the alternation rule (i.e., choosing the different direction from the previous trial). The right top panel
shows the spectral coherence between CA1 and RE at the theta rhythm on the maze. The right bottom panel
shows the phase coherence estimated by ISPC. In both plots, a prominent increase of theta coherence is
observed at the stem part of the maze
Caution for the interpretation of spectral coherence analysis is
that the degree of coherence reﬂects not only phase synchrony but
also the amplitude coordination of oscillatory activity. When one
wants to restrict the focus on phase synchrony, an alternative met-
ric, intersite phase clustering (ISPC) [29], can be used:
ISPC¼

1
N
X
N
t¼1
e
iθxt??θytðÞðÞ
j

22 Hiroshi T. Ito
whereθxandθyare instantaneous phases at a speciﬁc frequency
band, which can be estimated by a Hilbert transform described in
the following section. The data is sampled att¼1, 2,...,N, and the
phase consistency over time is calculated. A time-resolved version of
this method was used to assess the phase coordination between the
theta rhythms in CA1 and RE by using a sliding time window of
512 ms, which conﬁrmed the enhanced ISPC at the stem part of
the maze (Fig.2c). The results together suggest the enhanced
phase synchrony of theta rhythm between CA1 and RE at the
stem part of the maze.
3.2 Spike-Phase
Analysis Across Brain
Regions While coherence analysis based on local ﬁeld potentials (LFPs) is a
useful method to assess the degree of synchrony between brain
regions, it is important to note that LFPs are not necessarily local
signals. A study found that LFPs can spread more than a centimeter
range in the brain of nonhuman primates [30]. It is thus difﬁcult to
exclude a possibility that the measured spectral coherence is inﬂu-
enced by other brain regions near the recording electrodes. This
problem can be solved by using the correlation analysis based on
spike activity, because spikes can be detected only within a small
distance range from the electrode (~50μm) [31]. Therefore, spike-
based analysis is an important addition to conﬁrm synchrony
between regions.
3.3 Extraction of
Instantaneous Phase
by the Hilbert
Transform One strategy to investigate temporal coordination based on spike
activity is to measure the correlation of spike times of neurons in
one brain region to the phase of oscillatory activity in another brain
region. This analysis requires the estimates of instantaneous phases
of the oscillatory activity in LFPs, and they can typically be esti-
mated by using a Hilbert transform. The Hilbert transform is a
method for signal processing to obtain an analytic form of signals.
To implement this approach, the LFP signal should ﬁrst be ﬁltered
by a band-pass ﬁlter for the given frequency range. For example, if
one wants to estimate instantaneous phases of theta oscillations, the
passband of the ﬁlter should typically be in the range of 6–12 Hz. A
commonly used ﬁlter type for this purpose is either a Butterworth
inﬁnite-impulse-response (IIR) ﬁlter or a ﬁnite-impulse-response
(FIR) equiripple ﬁlter. The Butterworth ﬁlter is known to have a
ﬂat magnitude in the passband, but have a slow roll-off toward the
stopband. In contrast, the equiripple ﬁlter has small magnitude
ﬂuctuations (ripples) in the passband, but it is a linear phase ﬁlter
which guarantees the maintenance of waveform shapes, and fur-
thermore, its design algorithm, Parks-McClellan method, allows
for the ﬂexible speciﬁcation of the ripple size and roll-off speed of
the stopbands. After ﬁltering of the signal, the application of a
Hilbert transform will help separate amplitude and phase compo-
nents of the oscillatory activity as in the following procedure.

γ
Multisite Recording for the Analysis of Information Flow Between... 23
The LFP signal after the band-pass ﬁltering can roughly be
expressed with an amplitude-modulated trigonometric function:
utðÞﬃAtðÞcosωtþθðÞ
Here, our aim is to identify the functions
of instantaneous
amplitudesA(t)
and phaseswt+θ. The application of the Hilbert
transform will give the following signal, exchanging cosign with
sign function:
HilbertutðÞðÞ¼ AtðÞsinωtþθðÞ
This result is then used
to construct
the analytic form of the
signal:
u
atðÞ¼utðÞþiμHilbertutðÞðÞ
¼AtðÞcos ωtþθðÞþ iμsinωtþθðÞ½
¼AtðÞe
iωtþθðÞ
This analytic form enablesus to extract its amplitudeA(t)or
phasewt+θ, by simply calculatingthe
absolute value |u
a(t)| or
angular component arg[u
a(t)] of the signal, respectively.
The extracted phases by this method are then used to measure
the instantaneous phases at spike times, revealing the relationships
between spike times and oscillatory phases. For a quantitative anal-
ysis of spike phases, a circular histogram of spike phases can be
constructed to assess the degree of spike-phase modulation. The
phase bias of circular distribution is summarized by a mean vector
length that ranges from 0 to 1 (low to high phase bias). A useful
toolbox for circular statistics is available in MATLAB [32].
3.4 Spike-Field
Coherence Analysis
Revealed Phase-
Locking of RE Spikes
to CA1 Theta As describedin the previous section, spike-phase analysis based on a
Hilbert transfor
m requires the band-pass ﬁltering of the data at a
speciﬁc frequency range. A possible ambiguity here is the choice of
frequency range of the ﬁlter. For example, the passbands for gamma
oscillations used in previous studies are 30–80 Hz, 40–100 Hz, or
25–140 Hz, and it is not clear which range should be used for a
particular dataset. Furthermore, when one wants to have an over-
view of spike-phase relationships across a wide range of frequency,
the repetitive application of spike-phase analysis using different sets
of band-pass ﬁlters may result in a risk of multiple comparisons of
statistical tests.
Here, s
c oherence is another spike-phase estimation
method that does not require ﬁltering of the datasets. This method
is based on the analysis of a spike-triggered local ﬁeld potential
(LFP) average, and ﬁrst, short LFP segments around spike times
(e.g., 300 ms) are aligned so that spike times of a particular neuron
are at the center of the segments (Fig.3a). These LFP segments
around spike times are then averaged, giving a spike-triggered LFP
average. The key idea of this analysis is that while this averaging

24 Hiroshi T. Ito
Fig. 3Spike-ﬁeld coherence analysis between RE spikes and CA1 LFP. (a ) Procedure of spike-ﬁeld coherence
method. For each spike of a given RE cell, segments are extracted from CA1 LFP, so that the spike times are
aligned at the center of the segments. After averaging, a clear theta rhythm is observed in the real data, which
results in a prominent peak at the theta rhythm in the spectral power (left). By contrast, with simulated random
spikes, the LFP average as well as its spectral power are almost ﬂat. The spectral power of the averaged LFP

procedure will reduce spike-uncorrelated signals, the components
of LFP that are temporally consistent relative to the spike times will
become prominent. While the spike-triggered LFP average itself
gives an idea of spike-phase relationship, the shape of averaged
waveform also depends on the baseline spectral power of LFP
segments. This dependency can be canceled out by taking a nor-
malization procedure, calculating the ratio of spectral power
between the spike-triggered LFP average and the baseline LFP
segments across frequencies. The ﬁnalresult
is called spike-ﬁeld
coherence, a metric of spike-phase correlation in a wide range of
frequency [33]. Spike-ﬁeld coherence takes values from 0 to
1 (or 0–100%).
Multisite Recording for the Analysis of Information Flow Between... 25
As described in the procedure, the spike-ﬁeld coherence analy-
sis requires calculations of spectral power of short LFP segments.
One can use a conventional spectral power method based on Four-
ier transform. However, as this Fourier-based method is designed
for stationary signals, it is not necessarily ideal for detecting tran-
sient oscillatory events, such as bursts of gamma oscillations. It has
been suggested that such transient events can better be detected by
using a wavelet-based method. The wavelet-based method can be
used for spectral power estimation, by measuring the amplitudes of
the LFP signals that were convolved with complex Morlet wavelets
ranging from 4 to 140 Hz at 2 Hz intervals [7 ,18].
3.5 Comparison of
Spike-Phase Locking
Between Stem and
Nonstem Regions of
the Maze The analysis of spike-phase modulation is a useful technique to
understand the process of the brain’s signal between regions, and
it is often necessary to compare these values between groups, for
example, between different animals (e.g., normal vs. knockout),
experimental conditions (e.g., with or without optogenetic excita-
tion/inhibition), or behavioral states (e.g., task phases). Here, it is
important to note that spike-phase estimation methods described
in the previous sections are dependent not only on the degree of
spike-phase coupling but also on the number of spikes used to
calculate them [34–36]. This is because the estimation of these
metrics is biased; even if spikes are not correlated to LFPs at all,
the calculated coherence value will never become zero with a ﬁnite
number of spikes. As the table of Fig.3cshows, the degree of spike-
ﬁeld coherence is largely affected by the number of spikes used for
Fig. 3(continued) segment is then normalized, giving a spike-ﬁeld coherence. A prominent peak of coherence
is observed at the theta range, indicating that spikes of this particular RE cell are phase-locked to the theta
rhythm in CA1 LFP, but not in other frequency ranges (e.g., beta, gamma). (b ) Comparison of spike-ﬁeld
coherence between the stem and the nonstem parts of the maze. The coherence in the theta range is largely
increased at the stem part of the maze, indicating enhanced spike-time modulation by the CA1 theta. (c )
Inﬂuence of the spike number used to compute the spike-ﬁeld coherence. A small number of spikes generally
give a larger coherence value, and therefore, it is important to use the same number of spikes for the group
comparison

calculation. Therefore, for the comparison of spike-phase modula-
tion between groups, one should be careful to equalize the number
of spikes for each group. A bootstrap resampling procedure can be
used for this purpose. For example, we set the sampling number of
spikes to be 4/5 of the minimum number of spikes across the
groups. Then, a speciﬁed number of spikes are randomly chosen
from each group to estimate spike-ﬁeld coherence. This procedure
is repeated 100 times, andtheir averages are considered as a repre-
sentative value of thegroup
(e.g., [18]).
26 Hiroshi T. Ito
By applying this procedure, we compared the spike-ﬁeld coher-
ence between RE spikes and CA1 LFP between two task states,
either when animals are on the stem or nonstem parts of the maze
in the alternation task. On the stem part, animals are required to
make a trajectory decision at the upcoming T-junction, but on the
nonstem part, they can simply run along the track. First, we
observed a prominent peak at the theta range of frequency in the
spike-ﬁeld coherence plot, indicating that RE spikes are phase-
locked to the CA1 theta. This coherence was further enhanced
when the animals were on the stem part of the maze, suggesting
that spike times of RE neurons are temporally coordinated at the
phases of CA1 theta (Fig.3b). Together with the spectral coher-
ence analysis of LFPs, the results together conﬁrmed the idea that
CA1 and RE are functionally coupled at the theta rhythm, particu-
larly during trajectory decisions.
3.6 Signal
Directionality AnalysisAlthough spike-ﬁeld coherence analysis reveals temporal coordina-
tion between regions, the analysis does not provide directionality of
the impact. Several analyses are available to estimate the direction-
ality of signal ﬂow, such as Granger causality or transfer entropy
[37–40]. One can also examine the information ﬂow at a speciﬁc
frequency range, for example, by examining the microstructure of
temporal coupling between the CA1 theta and the spike activity in
another brain region [14 ,18,41].
This strategy is based on the
assumption of the signal time delay between regions due to the
ﬁnite traveling speed of spike activity along the axons.
To implement the signal directionality analysis between RE and
CA1, the mean vector lengths of spike phases of RE neurons are
calculated relative to a series of CA1 LFP traces that were band-pass
ﬁltered at 6–12 Hz and temporally shifted from200 ms to 200 ms
at 10 ms intervals. We can then identify the temporal shift that gives
the maximum mean vector length [14]. If the maximum is achieved
by shifting CA1 LFP backward, there must be a delay of the impact
of RE spikes on CA1 LFPs, indicating the directionality from RE to
CA1. By contrast, if the maximum can be achieved by shifting CA1
LFP forward, the results indicate the directionality from CA1 to RE
(Fig.4a)
he averaged mean vector length across all RE cells is
plotted in Fig.4b. The plot indicates that the peak of the mean
vector length is achieved when CA1 LFP is shifted ~50 ms

backward relative to RE spikes, indicating that RE spikes give the
maximum impact on CA1 theta after a small delay of axonal
conduction.
Multisite Recording for the Analysis of Information Flow Between... 27
Fig. 4Signal directionality analysis between RE spikes and CA1 LFP. (a) Illustrations showing an idea behind
the analysis. If CA1 inﬂuences RE spikes, the correlation will become stronger by temporally shifting CA1 LFP
forward, because it will cancel out the axonal conduction delay. The opposite is the case if RE inﬂuences CA1.
(b) The plot shows the mean vector lengths for all recorded RE cells relative to CA1 theta at different amounts
of time shifts (thick line, means; shaded, standard errors). The peak was observed by shifting CA1 LFP
backward, indicating the signal directionality from RE to CA1
4 Decoding Analysis to Investigate Populational Representations of Trajectories
Place cells in the hippocampus change their ﬁring rates depending
on the animal’s choice of trajectory on the T-junction of the maze
[42,43]. While this rate modulation is thought to reﬂect either
previous or next behavioral states of the animal, place cells also
change their ﬁring rates depending on the animal’s running
speed, head direction, or position [11,44,45], and therefore, the
analysis should account for the contributions of each of these
compounds separately. Here, ANOVA and ANCOVA can be used
to assess the signiﬁcance of trajectory dependency by considering
the contribution of each of these behavioral variables (e.g., [43]).
A
s A
NOVA or ANCOVA requires the segmentation of the data, the
stem position is ﬁrst divided into six equally sized bins (15 cm), and
the following behavioral parameters are measured for each bin and
each trial: (1) ﬁring rate, the number of spikes divided by the
amount of time spent in the bin; (2) running speed, the averaged
position shift per time in the bin; (3) head direction, the averaged
angle of two colored LEDs on the headstage; and (4) lateral posi-
tion, the averaged position perpendicular to the long axis of the
central stem. Then a two-way ANOVA is performed with trial type
(either left- or right-turn trials) and six spatial bins as independent
variables and ﬁring rate as the dependent measure. The cells with a
signiﬁcant main effect of the trial type are identiﬁed as potential

trajectory-dependent cells. For these cells, additional analysis is
performed to examine whether variations in speed, heading, or
lateral position might account for the differences in ﬁring rate
between trial types. This can be examined with a two-way
ANCOVA with trial type and bins as the independent variables,
ﬁring rate as the dependent measure, and speed, head direction,
and lateral position as covariates. Cells that continued to show a
signiﬁcant difference in ﬁring rate between left- and right-turn trials
(i.e., the change of ﬁring ratecannot
be explained by the animal’s
running speed, head direction, or lateral position) are classiﬁed as
trajectory dependent. This analysis revealed that 55.1% of CA1 cells
and 42.2% of RE cells are classiﬁed as trajectory dependent [16].
28 Hiroshi T. Ito
4.1 Population
Vector-Based
Decoding While the analysis based on ANOVA and ANCOVA can classify
each cell as either trajectory dependent or independent, this analysis
does not necessarily give the idea of information represented in a
brain area as a neural ensemble. A key advantage of decoding
analysis is the ability to evaluate the predictability of a particular
behavioral parameter from ensemble neural activity, and therefore,
it serves as a powerful strategy to assess its behavioral relevance.
Decoding analysis is usually performed in two steps, training
and testing phases. The dataset should be divided into two parts,
training and test datasets, where the training dataset is used to
optimize the decoding parameters and the test dataset is used
only for the evaluation of the decoder’s performance (Fig.5b).
This is because the decoding performance should always be evalu-
ated with samples that are not used for parameter optimization, and
especially in a machine-learning algorithm, one needs to be cau-
tious about the decoder’s overﬁtting to the training dataset, which
will result in poor performance in new datasets even if it has almost
perfect performance on the training dataset.
A relatively simple form of decoding can be performed by
constructing vectors of ﬁring rates of a neuronal ensemble (i.e.,
population vector). First, ﬁring rates of each neuron were normal-
ized to z scores, so that a neuronal bias due to the difference of
baseline ﬁring rates should be reduced. Then the averaged rate
vectors on the stem were calculated separately for either right- or
left-trajectory trials from the training dataset that is constructed by
removing one trial as a test dataset (Fig.5a). Finally, the difference
between dot products for a test rate vector and the individual
trajectory vectors gives the decoded trajectory as in the following
equation:
T¼signvc
rightvc left

in w
c
rightandc
leftare the population vectors from the training
dataset for right- or left-trajectory trials andvis a vector of ﬁring
rates of the same population of neurons on the test dataset.Tis the
output value of the classiﬁer; 1 or1 represents right or left

Fig. 5Decoding of the animal’s next trajectory from a population of neurons. (a ) Illustration showing a
construction of a population vector of ﬁring rates of neurons on the stem part of the maze. Each neuron shows
different levels of modulation in ﬁring rates between right- and left-oriented trajectories, which is summarized
as two population rate vectors for each of the two trajectories. The decoder works based on the similarity of a
given vector to these two population vectors. (b ) Cross-validation to test the decoder’s performance. The
dataset is divided into training and test datasets, and only the training dataset is used to construct the average
PVs for the two trajectories. The decoder’s performance is then tested with a test dataset by taking dot
products of a given vector with the PVs for the two trajectories. (c ) Linear classiﬁer in the machine-learning-
based approach. The ﬁring rates of individual neurons are weighted and summated to give an output. The
weight adjustment allows for increasing or decreasing the contribution of particular cells for the best
classiﬁcation performance. (d ) Performance of the decoders predicting the animal’s next trajectory based
on neurons in RE and CA1. The chance level is 50%. The performance was in general better for the SVM
decoder compared with the PV decoder. The high decoding performance of CA1 cells was signiﬁcantly
impaired by the lesions of RE, indicating the importance of RE for trajectory representation in CA1

γ
trajectory, respectively. This procedure was repeated for all trials to
be tested (i.e., leave-one-out cross-validation), and the classiﬁcation
accuracy on the test datasets was considered an estimate of the
decoding performance.
30 Hiroshi T. Ito
4.2 Support Vector
Machine for Decoding
of the Next Trajectory
Choice While a population vector decoder can achieve a good performance
when an ensemble of neurons represents one common feature such
as head direction or spatial position, neurons often exhibit mixed
representations of multiple modalities [46–50], and it is thus pref-
erable for a decoder to have a mechanism to adjust the contribution
of individual neurons. This can be achieved by a machine-learning-
based decoding strategy, such as a support vector machine. Unlike a
population vector method in which the decoding efﬁcacy can be
deteriorated by the inclusion of neurons that does not contribute to
the decoding feature, a machine-learning algorithm is able to dis-
regard such irrelevant information.
A support vector machine is one of the supervised learning
algorithms, in which new data are classiﬁed based on the class labels
of the training datasets. The performance of decoder should always
be tested with a dataset different from the one used in training (i.e.,
cross-validation), because even a decoder with perfect classiﬁcation
performance in the training dataset does not necessarily give good
performance in another dataset, due to overﬁtting of the decoder
that impairs the generalization ability. The support vector machine
is considered an algorithm with excellent generalization ability [51]
and is widely used for machine-learning-based classiﬁcation
problems.
While the support vector machine can be implemented with
nonlinear function, for simplicity, we use a linear decoder,
expressed as in the following equation, to predict the next trajec-
tory based on the spike rates on the stem (Fig.5c):
T¼bþw
1θF1þw2θF2þw3θF3...:¼bþw
T
F
F¼F
1,F2,F3,...?γ
T
,w¼w 1,w2,w3,...½
T
whereFis a vector ofﬁring rates for each cell,wis a vector of
respective weights,bis a scalar offset, andTis an output value of the
classiﬁer, either 1 for right turn orτ1 for left turn. The optimal
weights of the decoder were determined by a support vector
machine algorithm to maximize the separation margin for better
generalization performance. Brieﬂy, for a given numberNof trials
of rate-trajectory pairs, (F
i,T
i),i¼1,2,3,...N, we searched for
wthat satisﬁes the following condition:
min
w,b,ξ
1
2
w
T
wþC
X
N
i¼1
ξ
i
Subject toy
i
bþw
T
Fi
πμ
ξ1τξ
i,ξ
i,ξ0:
whereCi
enalty parameter for misclassiﬁcation, which we
usually set at 1, but changing the C value may improve the

performance depending on the dataset. A useful toolbox to imple-
ment support vector machine classiﬁcation with MATLAB is
available [52].
Multisite Recording for the Analysis of Information Flow Between... 31
4.3 Neurons in CA1
and RE Represent the
Next Trajectory in the
Alternation Task To assess the trajectory information represented in a neural ensem-
ble in CA1 and RE, we constructed a decoder to understand the
information about the animal’s next trajectory represented in these
neurons. The mean ﬁring rates on each spatial bin on the stem were
used as the inputs of the classiﬁer that predicts the animal’s next
trajectory. We used a one-leave-out cross-validation approach to
assess the performance of two decoders based on either population
vector or support vector machine. Both classiﬁers are able to predict
the animal’s next trajectory choice signiﬁcantly better than a chance
level of 50%, suggesting that neurons in both CA1 and RE repre-
sent the information about the next trajectory as a population
(Fig.5d). We further conﬁrmed that the support vector machine
classiﬁer performed better than a population vector decoder, indi-
cating excellent performance of the machine-learning-based
approach. To further conﬁrm the signal directionality of the trajec-
tory information, we recorded CA1 neurons from the animals with
lesions in RE. We then implemented the same decoding approach
based on a support vector machine and found that the decoding
performance of the animal’s next trajectory was signiﬁcantly
impaired in these animals (Fig.5d), suggesting that RE is necessary
for CA1 neurons to express the trajectory information.
5 Optogenetic Approach to Conﬁrm the Causality of Information Flow
A frequently asked question in multiregional studies is whether we
can tell the causality of impact between regions. While several
methods, including signal directionality analysis in the previous
section, are available to infer the causality, the direct method to
answer this question would be to manipulate the neural activity in
an anatomically and temporally speciﬁc manner. Historically, an
excitotoxic reagent, such as ibotenic acid or NMDA, has been
used to make permanent lesions in a particular brain region. For
example, as shown in the previous section, the trajectory-
dependent ﬁring of CA1 cells was largely abolished by lesions of
RE, indicating that RE plays a key role for the hippocampus to
express the trajectory information. However, the major caveat of
such lesion studies is that it is often difﬁcult to assess the role of the
brain region due to functional compensation by other brain areas
over time after damage. It has been demonstrated that different
behavioral deﬁcits are observed between acute and chronic inacti-
vation of a particular brain region [53 ]. Furthermore, in the case of
the trajectory-dependent ﬁring of place cells, the result of the lesion
experiment does not provide a clear answer to the question of

whether RE sends the trajectory information of ongoing trials
dynamically or is necessary for long-term circuit organization for
the hippocampus to express the trajectory information. In order to
make this distinction, temporally selective manipulation is neces-
sary. Here, the optogenetic manipulation technique [54,55]isa
great tool for spatiotemporal and cell-type-speciﬁc manipulations.
32 Hiroshi T. Ito
A light-application system for optogenetic experiments can
easily be coupled with the recording system, as several companies,
such as Neuralynx or Plexon, offer an integrated system for both
recordings and optogenetics. It is also possible to build a separate
laser system that can be synchronized with the recording system
through TTL-mediated timestamps. In our system, a diode-
pumped solid-state laser or diode laser (Shanghai Laser & Optics
Century) is used as a coherent light source that is coupled with an
optic-ﬁber patch cable through a collimator (Thorlabs). The other
end of the patch cable is then connected through ﬁber ferrules with
the optic ﬁber implanted in the animal’s brain. To achieve the
widespread impact in the region, a high-NA large-diameter ﬁber
is often preferable (400μm, 0.50 NA; FP400URT, Thorlabs).
We took an approach of virus-mediated expression of opsins in
the RE. The adeno-associated virus encoding opsins was injected
into RE, using a 10μl NanoFil syringe with a UMP3 pump con-
troller (World Precision Instruments). The injection rate was
100 nl/min and ~0.5μl in total for each injection site. The expres-
sion of opsins usually takes 3–4 weeks. Although the impact of
optogenetics can be validated by the ﬁber location as well as virus
expression pattern by histological examination after experiments, it
is ideal to assess the modulation of neuronal ﬁring in behaving
animals directly. For this aim, two tetrode wires were glued to the
optic ﬁber such that the tips of the tetrodes are extended ~0.75 mm
from the end of the optic ﬁber (Fig.6a). Individual wires of
tetrodes are then connected with a Mill-Max connector
(ED85100-ND, DigiKey) for the attachment of a recording device.
This conﬁguration allows for the measurement of the activity of
neurons near the optic ﬁber to assess the impact of the light
application (Fig.6b).
The laser light can be delivered whenever the silencing is nec-
essary by using a TTL-mediated switch of the laser source or by
using a mechanical shutter. Depending on experiments, the silenc-
ing may be necessary for a long duration, for example, in the range
of several minutes. However, such a long-time laser application is
known to generate heat that may inﬂuence neural activity nearby
[56]. Therefore, it is desirable to design experiments so that the
duration of continuous laser application should be as minimal as
possible.
As s
n Fig.6b, when a 635 nm laser (RLM635TA;
Shanghai Laser & Optics Century) was applied to RE neurons
expressing Jaws [57], ﬁring of a representative RE neuron was

strongly suppressed. The activity of a CA1 neuron recorded from
the same animal showed trajectory-dependent ﬁring when the laser
was off (Fig.). In the next behavioral session, the laser was
applied during the behaviors, resulting in silencing of RE neurons.
In order to minimize the duration of continuous laser application,
the laser was stopped during food consumption and restarted when
6c
Multisite Recording for the Analysis of Information Flow Between... 33
Fig. 6Optogenetic manipulation of the activity of RE neurons. (a ) Top: Schema showing the design of optrode.
Two tetrodes are glued to the optic ﬁber so that the tetrodes should extend ~0.75 mm from the tip of the optic
ﬁber. Bottom. After the injection of AAV-Jaws, the optrode was inserted so that the tip of the optic ﬁbers is
0.25–0.5 mm above the RE. (b ) Silencing of a representative RE cell during laser application. The top plot
shows the spike raster and the bottom shows the spike-count histograms. The 635 nm laser was applied
between 0 and 5 sec in the time axis. The spike activity was almost completely silenced during the laser
application. (c ) Trajectory-dependent ﬁring of a CA1 cell during the silencing of RE. A representative CA1 place
cell showing the trajectory-dependent rate change between right- and left-oriented trajectories. When RE
neurons were silenced by the laser application, the same cell still exhibited position-selective ﬁring at the
stem, but its trajectory-dependent rate change was largely diminished, indicating that RE cells dynamically
modulate the ﬁring rates of CA1 cells depending on the animal’s trajectory plan of ongoing trials

the animal initiated the next trials. Under the silencing of RE
neurons, while the same CA1 neuron still exhibited position-
selective ﬁring at the stem part of the maze, the neuron lost its
trajectory dependency and the ﬁring rates between right- and left-
oriented trials became almost the same. These data thus suggest
that RE neurons dynamically modulate ﬁring rates of CA1 cells
during behaviors, sending the information about the next trajec-
tories to the hippocampus.
34 Hiroshi T. Ito
References
1.Singer W (1993) Synchronization of cortical
activity and its putative role in information
processing and learning. Annu Rev Physiol
55:349–374
2.Gregoriou GG, Gotts SJ, Zhou H, Desimone
R (2009) Long-range neural coupling through
synchronization with attention. Prog Brain Res
176:35–45
3.Fries P (2015) Rhythms for cognition: com-
munication through coherence. Neuron 88:
220–235
4. Jones
MW,
Wilson MA (2005) Theta rhythms
coordinate hippocampal-prefrontal interac-
tions in a spatial memory task. PLoS Biol 3:
e402
5. Ji D, Wilson MA (2007) Coordinated memory
replay in the visual cortex and hippocampus
during sleep. Nat Neurosci 10:100–107
6.Tort AB, Kramer MA, Thorn C, Gibson DJ,
Kubota Y, Graybiel AM, Kopell NJ (2008)
Dynamic cross-frequency couplings of local
ﬁeld potential oscillations in rat striatum and
hippocampus during performance of a T-maze
task. Proc Natl Acad Sci U S A 105:
20517–20522
7.Colgin LL, Denninger T, Fyhn M, Hafting T,
Bonnevie T, Jensen O, Moser MB, Moser EI
(2009) Frequency of gamma oscillations routes
ﬂow of information in the hippocampus.
Nature 462:353–357
8. Ito HT (2017) Prefrontal-hippocampal inter-
actions for spatial navigation. Neurosci Res
129:2–7
9.Rainer G, Rao SC, Miller EK (1999) Prospec-
tive coding for objects in primate prefrontal
cortex. J Neurosci 19:5493–5505
10.Gold JI, Shadlen MN (2007) The neural basis
of decision making. Annu Rev Neurosci 30:
535–574
11.O’Keefe J, Dostrovsky J (1971) The hippo-
campus as a spatial map. Preliminary evidence
from unit activity in the freely-moving rat.
Brain Res 34:171–175
12.O’Keefe J, Nadel L (1978) The hippocampus
as a cognitive map. Oxford University Press,
Oxford, UK
13.Moser EI, Kropff E, Moser MB (2008) Place
cells, grid cells, and the brain’s spatial represen-
tation system. Annu Rev Neurosci 31:69–89
14.Siapas AG, Lubenov EV, Wilson MA (2005)
Prefrontal phase locking to hippocampal theta
oscillations. Neuron 46:141–151
15.Benchenane K, Peyrache A, Khamassi M, Tier-
ney PL, Gioanni Y, Battaglia FP, Wiener SI
(2010) Coherent theta oscillations and reorga-
nization of spike timing in the hippocampal-
prefrontal network upon learning. Neuron 66:
921–936
16. Ito HT, Zhang SJ, Witter MP, Moser EI,
Moser MB (2015) A prefrontal-thalamo-hip-
pocampal circuit for goal-directed spatial navi-
gation. Nature 522:50–55
17.Hallock HL, Wang A, Grifﬁn AL (2016) Ven-
tral midline thalamus is critical for
hippocampal-prefrontal synchrony and spatial
working memory. J Neurosci 36:8372–8389
18. Ito HT, Moser EI, Moser MB (2018) Supra-
mammillary nucleus modulates spike-time
coordination in the prefrontal-Thalamo-hippo-
campal circuit during navigation. Neuron 99:
576–587.e5
19.Herkenham M (1978) The connections of the
nucleus reuniens thalami: evidence for a direct
thalamo-hippocampal pathway in the rat. J
Comp Neurol 177:589–610
20.Vertes RP, Linley SB, Groenewegen HJ, Witter
MP (2014) Thalamus. In: Paxinos G (ed) The
rat nervous system. Elsevier Academic Press,
San Diego
21.McNaughton BL, O’Keefe J, Barnes CA
(1983) The stereotrode: a new technique for
simultaneous isolation of several single units in
the central nervous system from multiple unit
records. J Neurosci Methods 8:391–397
22.Gray C
aldonado PE, Wilson M,
McNaughton B (1995) Tetrodes markedly
improve the reliability and yield of multiple

Multisite Recording for the Analysis of Information Flow Between... 35
single-unit isolation from multi-unit record-
ings in cat striate cortex. J Neurosci Methods
63:43–54
23.Thomson DJ (1982) Spectrum estimation and
harmonic analysis. Proc IEEE 70(9):
1055–1096.https://doi.org/10.1109/
PROC.1982.12433
24.Mitra PP, Pesaran B (1999) Analysis of
dynamic brain imaging data. Biophys J 76(2):
691–708.https://doi.org/10.1016/s0006-
3495(99)77236-x
25.Bokil H, Andrews P
, Kulkarni JE, Mehta S,
Mitra PP (2010) Chronux: a platform for ana-
lyzing neural signals. J Neurosci Methods 192:
146–151
26.Wouterlood FG, Saldana E, Witter MP (1990)
Projection from the nucleus reuniens thalami
to the hippocampal region: light and electron
microscopic tracing study in the rat with the
anterograde tracer Phaseolus vulgaris-
leucoagglutinin. J Comp Neurol 296:179–203
27.Vertes RP, Hoover WB, Szigeti-Buck K, Ler-
anth C (2007) Nucleus reuniens of the midline
thalamus: link between the medial prefrontal
cortex and the hippocampus. Brain Res Bull
71:601–609
28.Dolleman-van der Weel MJ, Grifﬁn AL, Ito
HT, Shapiro ML, Witter MP, Vertes RP, Allen
TA (2019) The nucleus reuniens of the thala-
mus sits at the nexus of a hippocampus and
medial prefrontal cortex circuit enabling mem-
ory
and behavior
. Learn Mem 26:191–205
29.Cohen MX (2014) Analyzing neural time series
data. The MIT Press, Cambridge, MA
30.Kajikawa Y, Schroeder CE (2011) How local is
the local ﬁeld potential? Neuron 72:847–858
31.Buzsaki G (2004) Large-scale recording of
neuronal ensembles. Nat Neurosci 7:446–451
32.Berens P (2009) CircStat: a MATLAB toolbox
for circular statistics. J Stat Softw 31:1–21
33.Fries P, Reynolds JH, Rorie AE, Desimone R
(2001) Modulation of oscillatory neuronal syn-
chronization by selective visual attention. Sci-
ence 291:1560–1563
34.Sirota A, Montgomery S, Fujisawa S,
Isomura Y, Zugaro M, Buzsaki G (2008)
Entrainment of neocortical neurons and
gamma oscillations by the hippocampal theta
rhythm. Neuron 60:683–697
35.Rutishauser U, Ross IB, Mamelak AN, Schu-
man EM (2010) Human memory strength is
predicted by theta-frequency phase-locking of
single neurons. Nature 464:903–907
36.Vinck M, van Wingerden M, Womelsdorf T,
Fries P, Pennartz CM (2010) The pairwise
phase consistency: a bias-free measure of
rhythmic neuronal synchronization. Neuro-
Image 51:112–122
37.Granger CWJ (1969) Investigating causal rela-
tions by econometric models and cross-spectral
methods. Econometrica 37:424–438
38.Seth AK, Barrett AB, Barnett L (2015)
Granger causality analysis in neuroscience and
neuroimaging. J Neurosci 35:3293–3297
39.Walker BL, Newhall KA (2018) Inferring infor-
mation ﬂow in spike-train data sets using a trial-
shufﬂe method. PLoS One 13:e0206977
40.Bielczyk NZ, Uithol S, van Mourik T,
Anderson P, Glennon JC, Buitelaar JK (2019)
Disentangling causal webs in the brain using
functional magnetic resonance imaging: a
review of current approaches. Netw Neurosci
3:237–273
41.Hartung H, Brockmann MD, Poschel B, De
Feo V, Hanganu-Opatz IL (2016) Thalamic
and entorhinal network activity differently
modulates the functional development of
prefrontal-hippocampal interactions. J Neu-
rosci 36:3676–3690
42.Frank LM, Brown EN, Wilson M (2000) Tra-
jectory encoding in the hippocampus and ento-
rhinal cortex. Neuron 27:169–178
43.Wood ER, Dudchenko PA, Robitsek RJ,
Eichenbaum H (2000) Hippocampal neurons
encode information about different types of
memory episodes occurring in the same loca-
tion. Neuron 27:623–633
44.Huxter J, Burgess N, O’Keefe J (2003) Inde-
pendent rate and temporal coding in hippo-
campal pyramidal cells. Nature 425:828–832
45.Acharya L, Aghajan ZM, Vuong C, Moore JJ,
Mehta MR (2016) Causal inﬂuence of visual
cues on hippocampal directional selectivity.
Cell 164:197–207
46.Salinas E, Sejnowski TJ (2001) Gain modula-
tion in the central nervous system: where
behavior, neurophysiology, and computation
meet. Neuroscientist 7:430–440
47.Sargolini F, Fyhn M, Hafting T, McNaughton
BL, Witter MP, Moser MB, Moser EI (2006)
Conjunctive representation of position, direc-
tion, and velocity in entorhinal cortex. Science
312:758–762
48.Komorowski RW,
Manns JR, Eichenbaum H
(2009) Robust conjunctive item-place coding
by hippocampal neurons parallels learning what
happens where. J Neurosci 29:9918–9929
49.Rigotti M
arak O, Warden MR, Wang XJ,
Daw ND, Miller EK, Fusi S (2013) The impor-
tance of mixed selectivity in complex cognitive
tasks. Nature 497:585–590

36 Hiroshi T. Ito
50.Fusi S, Miller EK, Rigotti M (2016) Why
neurons mix: high dimensionality for higher
cognition. Curr Opin Neurobiol 37:66–74
51.Boser BE,
Guyon IM,
Vapnik
VN (1992) A
training algorithm for optimal margin classi-
ﬁers. Proceedings of the ﬁfth annual workshop
on computational learning theory, pp 144–152
52.Chang CC, Lin CJ (2011) LIBSVM: a library
for support vector machines. ACM Trans Intel
Syst Tec 2:1–27
53.Otchy TM, Wolff SB, Rhee JY, Pehlevan C,
Kawai R, Kempf A, Gobes SM, Olveczky BP
(2015) Acute off-target effects of neural circuit
manipulations. Nature 528:358–363
54.Bernstein JG, Boyden ES (2011) Optogenetic
tools for analyzing the neural circuits of behav-
ior. Trends Cogn Sci 15:592–600
55.Fenno L, Yizhar O, Deisseroth K (2011) The
development and application of optogenetics.
Annu Rev Neurosci 34:389–412
56.Owen SF, Liu MH, Kreitzer AC (2019) Ther-
mal constraints on in vivo optogenetic manip-
ulations. Nat Neurosci 22:1061–1065
57.Chuong A
iri ML, Busskamp V, Matthews
GA, Acker LC, Sorensen AT, Young A, Kla-
poetke NC, Henninger MA, Kodandaramaiah
SB et al (2014) Noninvasive optical inhibition
with a red-shifted microbial rhodopsin. Nat
Neurosci 17:1123–1129

n
Chapter3
Rabies Virus Tracing of Monosynaptic Inputs to Adult-Bor
Granule Cells
Carmen Vivar and Henriette van Praag
Abstract
Over the past two decades, adult hippocampal neurogenesis has become a well-established phenomenon.
However, our understanding of how adult-born hippocampal neurons process information and contribute
to memory formation has been limited by a strong focus on new neuron number, rather than on the
structure and function of the underlying circuitry. With the advent of new viral neuroanatomical tracers, we
are now able to expand our knowledge to the analysis of the physiological role of new neurons in the adult
brain network. In particular, combining the use of retrovirus to label dividing dentate gyrus progenitor cells
with recombinant rabies virus has enabled the investigation of speciﬁc monosynaptic connections between
new neurons and select brain areas and cell types. This technology, in combination with electrophysiological
approaches, has now allowed us to visualize and analyze the network of new neurons in the adult brain.
Moreover, changes in new neuron circuitry as a result of behavioral changes, genetic modiﬁcation, or
pathological conditions can be reliably monitored and characterized. Altogether, we provide the back-
ground and guidelines for the optimal use and application of this novel methodology.
Key wordsAdult Neurogenesis, Dentate gyrus, Rabies virus, Transsynaptic tracing, Circuitry
1 Introduction: Adult Hippocampal Neurogenesis
In adult mammals new neurons are generated in the dentate gyrus
of the hippocampus [1–4]. Adult hippocampal neurogenesis has
been implicated in cognition and mood regulation under both
physiological and pathological conditions [5 –7]. The development
and network integration of new neurons in the dentate gyrus
extends over weeks [8]. New hippocampal neurons are transiently
highly excitable with a lower threshold for long-term plasticity [9],
thereby providing enhanced structural and functional plasticity to
the hippocampal circuitry.
1.1 Adult-Born Neuron
Integration into the
Hippocampal NetworkElectrophysiological recordings from newly born neurons labeled
with retroviral vectors, selective for dividing hippocampal progeni-
tor cells, provide strong evidence that adult-born neurons are
Robert P. Vertes and Timothy A. Allen (eds.),Electrophysiological Recording Techniques, Neuromethods, vol. 192,
https://doi.org/10.1007/978-1-0716-2631-3_3,©Springer Science+Business Media, LLC, part of Springer Nature 2022
37

functionally and synaptically integrated into the adult hippocampus
[6,8,10,11]. Utilizing this approach to birth-date new neurons, as
well as transgenic mouse models, has shown that it takes about
1 month for neural progenitor cells to develop into mature granule
cells and several more months to mature fully [8,10]. However, the
characteristics of cells innervating adult-born neurons, such as
speciﬁc cell type, morphology, physiology, and brain area of origin,
have only recently begun to be explored [12–14]. Identiﬁcation of
thepatter
n of new neuron connectivity has allowed for a better
understanding of the functional signiﬁcance of adult-born neurons
in the adult mammalian brain [12]. In addition, studying how the
circuitry changes under physiological (e.g., exercise [13], enrich-
ment [15], learning, or pathological (e.g., depression, anxiety,
developmental disorders [16,17], or epilepsy [18,19] conditions)
will lead to uncovering the role of the adult-born neurons in these
processes, as well as understanding how they take place.
38 Carmen Vivar and Henriette van Praag
Over the past century, methods of mapping neural circuits have
improved. The use of neurotropic viruses that have the natural
ability to spread transneurally, and that have been genetically mod-
iﬁed to reduce their pathogenicity, as well as the addition of
reporter genes, control of synaptic spread, and pseudotyping for
select cell infection, have led to the uncovering of the relationship
between the organization and function of neural circuits through-
out the nervous system [20 ,21].
, the introduction of
rabies virus technology [22 ] and the further development of genet-
ically engineered rabies virus systems [23–26] have enabled the
mapping of speciﬁc neural circuits in living animals. Rabies virus
travels exclusively between connected neurons in retrograde direc-
tion (unidirectional) through chemical synapses regardless of their
phenotype, synaptic strength, or distance, with the exception of gap
junctions or volume transmission. The original vector allows for
stepwise identiﬁcation of neuronal connections of a progressively
higher-order [27 ]. Recently, the rabies virus system has been mod-
iﬁed to identify monosynaptic connections [20].
The r
irus envelope glycoprotein (Rgp) allows for retro-
grade infection through axon terminals and is essential for virus
packaging and transsynaptic spread, but not for transcription of the
gene or virus replication. Thus, the deletion of Rgp from the rabies
virus genome can restrict the transsynaptic spread after initial infec-
tion, unless there is an alternative source of Rgp, which will then
allow for tracing of ﬁrst-order synaptic connections (presynaptic
cells) [23,2
,26,28]. T he inclusion of reporter proteins into the
rabies virus genome results in the complete ﬁlling of axons and
dendrites of the presynaptic cells with ﬂuorescent proteins [green
ﬂuorescent protein (GFP), mCherry, etc.], providing detailed ana-
tomical and morphological information. In addition, direct visuali-
zation of the afferent cells in acute brain slices
makes electrophysiological recordings feasible. To introduce and

Rabies Virus Tracing of Monosynaptic Inputs to Adult-Born Granule Cells 39
complement Rgp in the starter cells (postsynaptic cells) different
viral vectors such as adeno-associated virus (AAV), lentivirus, retro-
virus, and genetically engineered mice, have been used to enable
transsynaptic tracing in vivo.This system has been instrumental in
deﬁning and analyzing patterns of connectivity in multiple brain
areas, such as visual cortex, cerebellum, olfactory bulb, and ventral
tegmental area [29,30]. This approach has allowed us and others to
identify and characterize the circuitry of adult-born neurons and
has contributed to our understanding of the functional signiﬁcance
of new neurons in the adult brain under physiological [12–15,31–
33]
and
pathological [16–19,34] conditions.
To trace the network of adult-born neurons, we developed a
dual-virus approach using retrovirus (expressing GFP, avian TVA
receptor, and Rgp (RV-SYN-GTRgp)) to label dividing adult pro-
genitor cells, in combination with Rgp deleted rabies virus expres-
sing MCherry (EnvA-ΔG-MCh). This method allows for
monosynaptic retrograde tracing to selectively map the circuitry
of adult-born hippocampal neurons and characterize electrophysi-
ological properties of afferent inputs to new neurons [8,1
,
13]. S
peciﬁcally, we used a retrovirus (RV-SYN-GTRgp) expressing
nucleus-localized GFP, the TVA receptor and Rgp under control of
the neuron-speciﬁc synapsin promoter to label only dividing neural
progenitor cells that differentiate into adult-born neurons over
time in the dentate gyrus of the hippocampus. RV-SYN-GTRgp
was injected into the dentate gyrus (ﬁrst injection) to induce the
expression of Rgp, TVA and GFP in the adult-born neurons. Sec-
ond, the retrovirus was complemented with the EnvA-pseudotyped
rabies virus (second injection), in which Rgp was replaced with the
ﬂuorescent reporter protein mCherry (EnvA-ΔG-MCh). Through
the interaction between EnvA, an avian leukosis sarcoma virus
glycoprotein and its cognate receptor, TVA, the pseudotyped rabies
virus selectively infected the adult-born neurons. EnvA-ΔG-MCh
rabies virus was then complemented with Rgp provided by RV-
SYN-GTRgp in the adult-born neurons, which allowed for retro-
grade spread through one synapse and labeling of the ﬁrst-order
presynaptic neurons (“traced cells”) with mCherry, a process that
typically takes about 3–10 days. Because of the absence of Rgp in
traced cells, the rabies virus will not spread any further (Fig.1a)

Thus, “
starter cells” are adult-born neurons infected by RV-SYN-
GTRgp and EnvA-ΔG-MCh rabies virus expressing GFP and MCh,
while traced cells are presynaptic neurons distinguished by the
expression of MCh only. Here we provide detailed guidelines for
the application of this method.

Random documents with unrelated
content Scribd suggests to you:

containing a part of this work or any other work associated with
Project Gutenberg™.
1.E.5. Do not copy, display, perform, distribute or redistribute
this electronic work, or any part of this electronic work, without
prominently displaying the sentence set forth in paragraph 1.E.1
with active links or immediate access to the full terms of the
Project Gutenberg™ License.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if
you provide access to or distribute copies of a Project
Gutenberg™ work in a format other than “Plain Vanilla ASCII” or
other format used in the official version posted on the official
Project Gutenberg™ website (www.gutenberg.org), you must,
at no additional cost, fee or expense to the user, provide a copy,
a means of exporting a copy, or a means of obtaining a copy
upon request, of the work in its original “Plain Vanilla ASCII” or
other form. Any alternate format must include the full Project
Gutenberg™ License as specified in paragraph 1.E.1.
1.E.7. Do not charge a fee for access to, viewing, displaying,
performing, copying or distributing any Project Gutenberg™
works unless you comply with paragraph 1.E.8 or 1.E.9.
1.E.8. You may charge a reasonable fee for copies of or
providing access to or distributing Project Gutenberg™
electronic works provided that:
• You pay a royalty fee of 20% of the gross profits you derive
from the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty

payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You provide a full refund of any money paid by a user who
notifies you in writing (or by e-mail) within 30 days of receipt
that s/he does not agree to the terms of the full Project
Gutenberg™ License. You must require such a user to return or
destroy all copies of the works possessed in a physical medium
and discontinue all use of and all access to other copies of
Project Gutenberg™ works.
• You provide, in accordance with paragraph 1.F.3, a full refund of
any money paid for a work or a replacement copy, if a defect in
the electronic work is discovered and reported to you within 90
days of receipt of the work.
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.E.9. If you wish to charge a fee or distribute a Project
Gutenberg™ electronic work or group of works on different
terms than are set forth in this agreement, you must obtain
permission in writing from the Project Gutenberg Literary
Archive Foundation, the manager of the Project Gutenberg™
trademark. Contact the Foundation as set forth in Section 3
below.
1.F.
1.F.1. Project Gutenberg volunteers and employees expend
considerable effort to identify, do copyright research on,
transcribe and proofread works not protected by U.S. copyright

law in creating the Project Gutenberg™ collection. Despite these
efforts, Project Gutenberg™ electronic works, and the medium
on which they may be stored, may contain “Defects,” such as,
but not limited to, incomplete, inaccurate or corrupt data,
transcription errors, a copyright or other intellectual property
infringement, a defective or damaged disk or other medium, a
computer virus, or computer codes that damage or cannot be
read by your equipment.
1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except
for the “Right of Replacement or Refund” described in
paragraph 1.F.3, the Project Gutenberg Literary Archive
Foundation, the owner of the Project Gutenberg™ trademark,
and any other party distributing a Project Gutenberg™ electronic
work under this agreement, disclaim all liability to you for
damages, costs and expenses, including legal fees. YOU AGREE
THAT YOU HAVE NO REMEDIES FOR NEGLIGENCE, STRICT
LIABILITY, BREACH OF WARRANTY OR BREACH OF CONTRACT
EXCEPT THOSE PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE
THAT THE FOUNDATION, THE TRADEMARK OWNER, AND ANY
DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE LIABLE
TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL,
PUNITIVE OR INCIDENTAL DAMAGES EVEN IF YOU GIVE
NOTICE OF THE POSSIBILITY OF SUCH DAMAGE.
1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you
discover a defect in this electronic work within 90 days of
receiving it, you can receive a refund of the money (if any) you
paid for it by sending a written explanation to the person you
received the work from. If you received the work on a physical
medium, you must return the medium with your written
explanation. The person or entity that provided you with the
defective work may elect to provide a replacement copy in lieu
of a refund. If you received the work electronically, the person
or entity providing it to you may choose to give you a second
opportunity to receive the work electronically in lieu of a refund.

If the second copy is also defective, you may demand a refund
in writing without further opportunities to fix the problem.
1.F.4. Except for the limited right of replacement or refund set
forth in paragraph 1.F.3, this work is provided to you ‘AS-IS’,
WITH NO OTHER WARRANTIES OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
1.F.5. Some states do not allow disclaimers of certain implied
warranties or the exclusion or limitation of certain types of
damages. If any disclaimer or limitation set forth in this
agreement violates the law of the state applicable to this
agreement, the agreement shall be interpreted to make the
maximum disclaimer or limitation permitted by the applicable
state law. The invalidity or unenforceability of any provision of
this agreement shall not void the remaining provisions.
1.F.6. INDEMNITY - You agree to indemnify and hold the
Foundation, the trademark owner, any agent or employee of the
Foundation, anyone providing copies of Project Gutenberg™
electronic works in accordance with this agreement, and any
volunteers associated with the production, promotion and
distribution of Project Gutenberg™ electronic works, harmless
from all liability, costs and expenses, including legal fees, that
arise directly or indirectly from any of the following which you
do or cause to occur: (a) distribution of this or any Project
Gutenberg™ work, (b) alteration, modification, or additions or
deletions to any Project Gutenberg™ work, and (c) any Defect
you cause.
Section 2. Information about the Mission
of Project Gutenberg™

Project Gutenberg™ is synonymous with the free distribution of
electronic works in formats readable by the widest variety of
computers including obsolete, old, middle-aged and new
computers. It exists because of the efforts of hundreds of
volunteers and donations from people in all walks of life.
Volunteers and financial support to provide volunteers with the
assistance they need are critical to reaching Project
Gutenberg™’s goals and ensuring that the Project Gutenberg™
collection will remain freely available for generations to come. In
2001, the Project Gutenberg Literary Archive Foundation was
created to provide a secure and permanent future for Project
Gutenberg™ and future generations. To learn more about the
Project Gutenberg Literary Archive Foundation and how your
efforts and donations can help, see Sections 3 and 4 and the
Foundation information page at www.gutenberg.org.
Section 3. Information about the Project
Gutenberg Literary Archive Foundation
The Project Gutenberg Literary Archive Foundation is a non-
profit 501(c)(3) educational corporation organized under the
laws of the state of Mississippi and granted tax exempt status
by the Internal Revenue Service. The Foundation’s EIN or
federal tax identification number is 64-6221541. Contributions
to the Project Gutenberg Literary Archive Foundation are tax
deductible to the full extent permitted by U.S. federal laws and
your state’s laws.
The Foundation’s business office is located at 809 North 1500
West, Salt Lake City, UT 84116, (801) 596-1887. Email contact
links and up to date contact information can be found at the
Foundation’s website and official page at
www.gutenberg.org/contact

Section 4. Information about Donations to
the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission
of increasing the number of public domain and licensed works
that can be freely distributed in machine-readable form
accessible by the widest array of equipment including outdated
equipment. Many small donations ($1 to $5,000) are particularly
important to maintaining tax exempt status with the IRS.
The Foundation is committed to complying with the laws
regulating charities and charitable donations in all 50 states of
the United States. Compliance requirements are not uniform
and it takes a considerable effort, much paperwork and many
fees to meet and keep up with these requirements. We do not
solicit donations in locations where we have not received written
confirmation of compliance. To SEND DONATIONS or determine
the status of compliance for any particular state visit
www.gutenberg.org/donate.
While we cannot and do not solicit contributions from states
where we have not met the solicitation requirements, we know
of no prohibition against accepting unsolicited donations from
donors in such states who approach us with offers to donate.
International donations are gratefully accepted, but we cannot
make any statements concerning tax treatment of donations
received from outside the United States. U.S. laws alone swamp
our small staff.
Please check the Project Gutenberg web pages for current
donation methods and addresses. Donations are accepted in a
number of other ways including checks, online payments and

credit card donations. To donate, please visit:
www.gutenberg.org/donate.
Section 5. General Information About
Project Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could
be freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose
network of volunteer support.
Project Gutenberg™ eBooks are often created from several
printed editions, all of which are confirmed as not protected by
copyright in the U.S. unless a copyright notice is included. Thus,
we do not necessarily keep eBooks in compliance with any
particular paper edition.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.
This website includes information about Project Gutenberg™,
including how to make donations to the Project Gutenberg
Literary Archive Foundation, how to help produce our new
eBooks, and how to subscribe to our email newsletter to hear
about new eBooks.

Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.
More than just a book-buying platform, we strive to be a bridge
connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.
Join us on a journey of knowledge exploration, passion nurturing, and
personal growth every day!
ebookbell.com

Electrophysiological Recording Techniques 2nd Edition Robert P Vertes Editor

About This Presentation

Slide Content

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

Electrophysiological Recording Techniques 2nd Edition Robert P Vertes Editor

About This Presentation

Slide Content

Slide 1

Slide 2

Slide 3

Slide 5

Slide 6

Slide 7

Slide 8

Slide 9

Slide 10

Slide 11

Slide 12

Slide 13

Slide 14

Slide 15

Slide 16

Slide 17

Slide 18

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Slide 24

Slide 25

Slide 26

Slide 27

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Slide 40

Slide 41

Slide 42

Slide 43

Slide 44

Slide 45

Slide 46

Slide 47

Slide 48

Slide 49

Slide 50

Slide 51

Slide 52

Slide 53

Slide 54

Slide 55

Slide 56

Slide 57

Slide 58

Slide 59

Slide 60

Slide 61

Slide 62

Slide 63

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

TLE-9-Prepare-Salad-and-Dressing.pptxkkk

LESSON 1 ABOUT MEDIA AND INFORMATION.pptx

GRADE-8-AQUACULTURE-WEEKQ1.pdfdfawgwyrsewru

Feelings PP Game FOR CHILDREN IN ELEMENTARY SCHOOL.pptx

Jeopardy_Figures_of_Speech_Template.pptx [Autosaved].pptx

Jeopardy_Figures_of_Speech.pptxvdsvdsvsdvsd