Lecture 3: Research

Methods

Paul Whissell, Ph.D.

Introduction to Neuroscience (HMB200H1)

1
Introduction
• Neuroscience is the study of the nervous system

• Difficult pursuit; the nervous system is enormously

complex and has many features

• A single technique can provide information on some –

but not all – features of the nervous system
• No technique is perfect

• Understanding the nervous system therefore requires

the use of a wide range of specialized techniques

2
Levels of Analysis

Macro
level

Micro
level

3
Overview
• Part 1: Studying Brain Structure/Function in Humans

• Part 2: Manipulating Brain Activity in Humans

• Part 3: Studying Brain Structure/Function in Animals

• Part 4: Manipulating Brain Activity in Animals

4
Part 1: Studying Brain
Structure/Activity in
Humans

5
Starting near the top…

Macro
level

6
General approach
Search for relationships (e.g. correlations) between brain
structure/activity and a particular state (e.g. a behavior,
disorder or environmental circumstance).

Control Altered
Case State

7
 REASONING:
If the structure/function of
Brain Region Y changes
with Behavior X, then
Brain Region Y might
contribute to Behavior X.
8
Studying the Brain in Humans
1. EEG/ERP

2. MEG

3. PET

4. MRI, DTI + fMRI

5. Lesion studies

We can use these techniques to observe correlations

between brain structure/activity, behavior and disorders. 9
For all techniques, 2 main features
Spatial resolution Temporal resolution
How effectively can we How effectively can we
distinguish between small measure changes in brain
brain regions? activity over time?

10
Two Kinds of Resolution
Spatial resolution Temporal resolution
Low High Low High

Seconds

Millimeters or less Seconds Milliseconds

to minutes

As spatial resolution As temporal resolution

increases, it is easier to increases, it is easier to tell
identify specific brain how neural activity changes
structures (even if they over time (even if it is
are tiny). changing very fast). 11
1) Electroencephalography (EEG)
Measures electrical activity in specific brain regions.
Useful in studies of arousal, consciousness + epilepsy.

12
What is an EEG measuring?

Electrodes are placed above a small amount of neurons.

13
Basis of the EEG signal
Axon Dendritic
Terminals terminals
Dendrites (output) from receptive
(input)
Transmitter neuron

Synapse

Cell body Axon

The activation of the synapse (L02)
is the basis of the EEG signal. 14
Basis of the EEG signal
SYNAPSE

Axon
Dendritic Terminal
terminal

Charged ions (+)

1. Transmitters stimulate receptors.

2. Receptor activation causes channels to open.
3. Channels allow charged ions (+) flow.
4. The movement of charge creates a potential
difference (volts/V) that is detectable by the EEG. 15
The EEG record
• Potential difference (volts or V, y axis) over time
(seconds or sec, x axis)

• Notice how the voltage varies in a ‘wave-like’ manner

over time, with peaks and valleys

• The variation in voltage over time gives us frequency

16
EEG in different states

Frequency
Ranges:
Beta = 13 - 30 Hz
Alpha = 7 – 13 Hz
Theta = 4 – 7 Hz
Delta = 1 – 4 Hz

What happens to
activity as you ‘fall
asleep’?

17
The Event-Related Potential (ERP)
• Neural activity related to a brief mental process
• ERP is measured using EEG equipment

18
ERP w/Language Processing

19
EEG + ERP – Summary
• Great temporal resolution (biggest advantage)
• Millisecond scale
• Great for measuring rapid changes in arousal/consciousness
(regular EEG) and rapid cognitive processes (ERP)

• Poor spatial resolution (biggest disadvantage)

• Difficult to determine which specific areas are active
• Deeper brain areas cannot be measured

20
2) Magnetoencephalography (MEG)
• Brain activity is associated with electrical currents,
which in turn generate magnetic fields

• If we measure magnetic fields, we can approximate

neural activity

21
MEG for specific events
• This study shows field strength (in microT, Y axis) over
time (ms, X axis) during processing of a face

22
MEG – Summary
• Great temporal resolution (comparable to EEG)

• Improved spatial resolution (better than EEG but not

as good as MRI)
• Magnetic signals attenuate less w/distance
• Overcomes one of the main problems w/the EEG

• Main disadvantage is expense and inconvenience

• Tough to measure certain behaviors in the device

• Currently, MEG studies are rare

• This could change; interest in diagnostic potential1
Aoe et al. 2019. Sci Rep. 23
3) Positron Emission Tomography

24
PET
• A synthetic radiotracer is injected into the subject

• This radiotracer reacts with tissue in the brain; this

reaction produces a signal that can be measured with
specialized equipment

• Wherever the radiotracer goes in the brain, there will

be a signal generated

• Two general purposes for neuroscientists:

• Measuring metabolic activity
• Characterizing distribution of specific substances
25
Measuring activity w/PET
• If the radiotracer is similar to glucose (such as
Fluorodeoxyglucose (18F)), it will be distributed in the
same way (i.e. to active neurons)
• Signal differences between brain regions reflect
differences in glucose demands (+ neuronal activity)

26
Measuring substances w/PET
• If the radiotracer is similar to opiate transmitters, it will
be distributed in the same way (i.e. it will bind to opiate
receptors)
• Signal differences between brain regions reflect
differences in opiate receptor occupancy

27
Diagnosing disorders w/PET?
• Proteins that are characteristic of pathology (e.g. B-
amyloid in Alzheimer’s Disease) could one day be
tracked with specialized radiotracers

• Though this application is intriguing it is not yet reliable

enough for diagnosis 28
PET – Summary
• Decent spatial resolution
• Better than EEG but worse than MRI

• Poor temporal resolution

• Difficult to resolve rapid changes in neural activity

• Due to its resolution issues, PET is no longer preferred

as a structural/functional measure

• Main current use in neuroscience is characterizing

substances (e.g. receptors, proteins) though not for
diagnostic purposes
29
4) MRI, DTI + Functional MRI

MRI = Magnetic Resonance Imaging

30
MRI method in brief
• Brain tissues are first magnetized in a strong magnetic
field; strength of the field matters

• When hit with a radiofrequency pulse, the magnetized

tissues emit a signal which depends upon their
structure

• If we capture these signals, we can use them to

artificially reconstruct the tissues of the brain

31
MRI reveals brain differences

Zhuo et al. 2018. Trans Psy. 32

What about white matter?
• Traditional MRI is great for imaging grey matter

• An adapted form of MRI, diffusion tensor imaging

(DTI), is often used for studying white matter

33
DTI and white matter pathology
• Cannabis use during adolescence is associated with
reduced growth of white matter tracts and poorer
verbal performance scores

Becker et al. 2015. Dev Cog Neuro. 34

The MRI (and DTI) are
used to assess structure.

Another adapted version

of MRI, fMRI, is used to
assess function.
35
Basis of the fMRI signal
• Active neurons use glucose + oxygen

• After blood delivers oxygen, it becomes deoxygenated

• If you knew the brain areas where there was a change

in blood oxygenation, you would know the brain areas
that are changing in activity
36
The fMRI signal
• Oxygenated and deoxygenated blood have different
magnetic properties which can be measured

• If you measure blood oxy/deoxy ratio in a given area,

you’ll have a correlate of neuronal activity in that area

37
fMRI for Behavior + Disorders

Brain activity Brain activity may differ with

differs by task. psychological disorders.
Sescousse et al. 2013. Brain. 38
One brain has many functions!
The amygdala is active during many different states!

Be mindful of reverse inference errors in fMRI studies!

39
The Problem of ‘Reverse Inference’
Psychological Forward inference Brain activity
Function X in Region Y

“Manipulating X leads to changes in Region Y,

so Y must be involved in X”

Reverse inference

“Psychological function X must be engaged

because Region Y is active”
Because Region Y can do many, many, many, many other things!
Poldrack, 2006. TICS. Credit Dr. Spaniol.
Zhuo et al. 2019. Transl Psychiatry.
https://www.youtube.com/watch?v=b64qvG2Jgro 40
fMRI – Summary
• Great spatial resolution (great for studying brain
structure)
• Best out of techniques we have covered today
• Can be ‘paired’ w/other techniques (e.g. PET)
• Can be used to study connectivity in the brain (as in resting-
state fMRI)

• Decent temporal resolution

• Better than PET (arguably), not as good as EEG or MEG
• Lag of seconds between activity and signal is still evident

• Popular technique for cognitive neuroscience, but

should be interpreted with care
41
No tool is perfect
• Test-rest reliability is poor to fair

• Limits strength of conclusions

• Limits use in diagnosis

• Cannot be used to measure certain behaviors because

it actively disrupts them

Elliot et al. 2020. Psychol Sci.

Travis et al. 2020. Front Psychol. 42
5) Lesion Studies
• While brain damage is an unfortunate tragedy,
important insights can be gained from its study

• Some brain lesions are associated with striking

behavioral deficits

• If damage to a specific brain region impairs a particular

behavior, it is plausible that the brain region controls
the behavior impaired

• Lesions studies are a major part of neuropsychology

and were critical to the history of neuroscience
43
Influential Lesion Studies
Patient HM – Removal of
Hippocampus + adjoining
areas; Impaired memory

Phineas Gage – Lesion of

Frontal Lobe; Impulsivity +
Impaired Social Behavior

Patient SM406 – Lesion of

Amygdala; Reduced fear

44
On case studies…
• Lesions rarely specific (many brain areas involved)
• Behavioral impairments could be due to damage in any one
of these areas

• Other behaviors could be impaired but not measured

• Effects of brain lesions could be more extensive than thought

• Cases are very rare (often N = 1)

• Is it repeatable? Or was it chance?
• Difficult to make a conclusive argument based on them

• Not experimental, no control for other variables (e.g.

individual differences, life history)
45
All these methods have
focused on overall brain
structure + function (macro).

Can we get more specific?

Can we get down to the level

of the cell (micro)?
46
Levels of Analysis

Micro
level

47
Single-cell Recording
• Recording activity of individual neurons

• Highly invasive; we rarely get the opportunity (unless

we are already performing surgery on someone)

Kreiman et al., 2000. 48

The ‘Jennifer Aniston’ Neurons
• In this experiment, they found neurons that fire
specifically to different pictures of the same person

What can you infer from this study?

Quiroga et al. 2005. Nature. 49
To the level of genes (T01)
• Genes encode for proteins, which are an essential
structural component of every organism

• Because genes determine the structure of the nervous

system, they also determine its functional output

DNA doctrine

50
Genetic testing
• Many methods
• Polymerase chain reaction/PCR
• Microarrays

• Identify genetic variants (T01) associated with specific

conditions (e.g. in schizophrenia)1

• Identify changes in gene expression with life

experiences (e.g. with meditation)2

• Use genetic variations to predict behavior (Behavioral

Genetics/PSY390)
1. Fromer et al. 2016. Nat Neurosci.
2. Vendetti et al. 2020. Front Psychol. 51
mRNA, Protein testing + more
• Expression of mRNA and proteins may vary in
disorders and diseases
• B-amyloid is higher in Alzheimer’s Disease (biomarker)
• Numerous other proteins in mental health disorders1
• Most often done post-mortem

• Neurotransmitter levels may also change

• In live subjects, we can measure precursors or related
compounds (e.g. low levels of 5-hydroxyindoleacetic acid in
urine suggest high levels of serotonin, seen in autism)1

1. Johnston-Wilson et al. 2000. Mol Psychiatry.

2. Muller et al. 2017. Neuroscience. 52
In animals…
• Social isolation is a form of
stress and a risk factor for
depression

• With social isolation of mice,

we see changes in gene
activity

• The mRNA and protein levels

(here) of some genes
change – sometimes by a lot

Serra et al. 2008. BRR. 53

Immunohistochemistry (IHC)
• Localize proteins + other substances in the brain with
specialized stains

• Used in humans + animals

• Post-mortem brain tissue is required

• Requires specialized antibodies

54
Mechanism of IHC
• Antibodies are a part of our immune system and can
tag + neutralize specific substances (antigens)

• We can engineer specific antibodies that tag proteins

present in the brain

55
Example
• Characterize different proteins (and thus cell types)

56
Part 2: Manipulating Brain
Activity in Humans

57
Moving forward -
• In science, we are most interested in compelling
evidence for causal relationships

• We want to be sure that Brain X causes Behavior Y

• To get evidence for this relationship, we need to do

more than measure brain activity/structure

• We need to change brain activity in a controlled

experimental setting and observe the consequences

• Three main strategies: TMS, EBS and Drugs

58
Transcranial Magnetic Stimulation
• Neuronal activity in specific brain areas is manipulated
by magnetic fields

• If TMS changes that behavior, we can infer that the

brain area targeted is involved in that behavior1

Promising treatment
for depression (see
Dr. Downar’s
amazing work)2

1. https://www.youtube.com/watch?v=mD34o-sW22A
2. https://rtmslab.com/ 59
TMS and cognitive processes

Balconi and Crivelli. 2020. 60

TMS – Therapeutic Applications
• TMS applied to the frontal lobe may help treat
depression1 and perhaps other conditions2

Gaynes et al. J Clin Psy. 2014.

Chou et al. 2020. Neurbiol Aging. 61
TMS – Summary
• Many great advantages
• Non-invasive
• Can be used multiple times in one subject
• Relatively fast (takes minutes)

• However, there are some notable disadvantages

• Unclear what stimulation parameters to use
• Difficulty targeting deep brain areas
• Relatively new; still underused

62
Electrical Brain Stimulation
• Pioneering work by Penfield with this technique lead to
the development of influential cortical maps (i.e.
sensory and motor homunculi, see L04)

• Can be used to locate epileptic foci (source of

seizures) 63
EBS – Summary
• Informative of causal relationships, but comes with
many drawbacks

• Highly invasive

• Only used when there is already another pathology

present (e.g. epilepsy) (potential confound)

• Few opportunities for use (small sample size)

• Concerns about generalization of findings

64
Comparing Techniques

Sejnowski et al. 2014. Nat Neurosci. 65

Other methods
• We can manipulate brain activity via drugs
• Giving haloperidol affects learning1

• We can alter neurotransmitter levels via diet

• Tryptophan loading to change serotonin levels2

Zernheld et al. 2004. JCN.

Silber and Schmitt. 2010. NBB. 66
Part 2: Using Animal
Models

67
Methods in Animals
1. Electrophysiological recordings
• Single or multiple cells

2. Targeted manipulation studies

• Lesions, Cannulation

3. Genetic manipulation studies

• Knock-out, knock-in and more

4. Optogenetics and Chemogenetics

68
1) Electrophysiological Recordings
• Study of electrical + physiological properties of neurons

• We can either study individual cells or large groups of

cells all at once

69
Recordings in the Hippocampus
• Shown here are recordings
from hippocampal cells
during are exploration

• Cell activity depends upon

position in the area

• Many cells in the

hippocampus have this
property (place cells)

Spiers and Barry. 2015. Behav Sci. 70

2) Targeting Manipulation Studies
• The rodent brain (mouse + rat) has been mapped
using a stereotaxic guide

• Using this guide, we can locate specific brain regions

we can target (e.g. via stereotaxic surgery)

71
Techniques available
• Lesion the region
• Irreversible, even controlled
lesions are not perfect

• Pharmacological treatment
• Inactivation (e.g. via GABA
agonist) is common
• Reversible

• Electrical stimulation

72
More targeted strategies
• We can implant electrodes
(to deliver stimulation)
cannulae (to deliver drugs)
to specific brain regions

• We can inhibit or excite

specific brain areas using
these methods

• Analogy: we are basically

adding a ‘switch’ to turn
brain regions on and off
73
Cannulation – Example
• Cannula implanted into hippocampus
• GABA agonist (muscimol) used to inhibit hippocampus
• If you pharmacologically inhibit the hippocampus
during sleep, memories of objects are not stored

Swangjanit et al. 2018. Nature. 74

3) Genetic studies
DNA

RNA

Proteins/Peptides (from amino acids)

Environmental
influences
Neuronal structure

Brain structure

Behavior

Variations in genes can lead to variations in

behavior. 75
LOGIC: If a gene is
‘causing’ a behavior, then
changing that gene should
change that behavior.
How do we change genes?

We use transgenic animals!

76
Transgenic animals
• A transgenic animal has a genome that has been
modified (e.g. to include genes from another source)

These transgenic animals express different fluorescent

proteins. 77
Mice are preferred transgenics
• Genome is well-understood;
similar structure to humans
• Most human genes have
mouse counterparts
• Disease genes in humans
often cause dysfunction in
mice

• Mouse populations are

cheap/easy to maintain
• breed quickly (reproductive
cycle ~ 2 months), large
litters (5-15)

78
Applications of Transgenic Models
• Remove a gene (knock-out mouse)
• Observe changes in behavior

• Add a gene (knock-in mouse)

• Useful for studying human genes not in mice
• Add reporter proteins (fluorescent proteins)
• Add drug-responsive and light-responsive receptors
(chemogenetics and optogenetics, T03)

• Many ways to modify genes

• Global, Selective and Inducible Transgenics

79
Global Transgenic Model
• If the gene is altered in all cells, we call this a global or
constitutive knockout mouse model

• Unfortunately, this does not allow us to determine how

the gene functions in different cell types

• To understand how a gene functions in a specific cell

type, we can use a conditional transgenic model 80
Conditional transgenic knock-out
• Genetic changes are
restricted to certain cell
types only

• Here, the NR1 gene is

removed from CA1
hippocampal cells only

• All other cells still have

the gene

81
Conditional transgenic knock-out
• In this study, one gene (the α5 receptor gene) was
specifically removed from dentate gyrus cells

• As a result, mice showed stronger + more resilient fear

conditioning over time

Engin et al. 2015. J Neurosci. 82

Conditional transgenic knock-in
• Fluorescent protein reporter genes (GFP and
mCherry) inserted into mouse genome + expressed in
different cells

• Allows for the study of specific cell types

83
Problems w/Transgenic Models
• In the models reviewed so far, the modification is
immediate (beginning at birth)

• Models of this type may trigger a compensatory

response that radically affects development
• This is a problem as we are our focus is the role of a gene in
behavior, not in development

• Additionally, some genetic modifications are not

possible using this approach (i.e. they are lethal to the
developing organism)

• For this reason, we find inducible models 84

Inducible Knockouts
• Knockout that is activated by a particular agent (e.g.
doxycycline) – without the agent, there is no knockout

• As we can control the presence of a drug, we can

control when the gene knockout occurs

85
Term Test and Practice
Questions
Term Test Format
• Administered July 13, 10 am – 12 pm

• Covers Lectures 1 – 3 and associated tutorials

• 15 multiple choice questions (1 mark each)

• 25 fill in the blank questions (1 mark each)

• 20 marks of written answer questions

• Out of 60; averages of 70 – 77% are typical 87

Describe the key charged
substances that contribute to the
resting membrane potential. In your
answer, highlight important
differences in the distribution of
each substance. (6)
Differentiate between MRI, fMRI
and DTI. (3)
What is a lesion study? Describe
the strengths and weaknesses of
lesion studies, highlighting their
significance to early neuroscience.
(5)
This question is on genes, disorders and the nervous system.

A. You think that Disorder X may be due to an abnormal gene variation (Gene
Y). How would you test this hypothesis in humans? (1)

B. Suppose you have a theory that Disorder X is indeed caused by reduced

expression of the product of gene Y, protein Y. How would you accumulate
evidence to support your hypothesis in humans? (1)

C. You think that reduced expression of protein Y is associated with reduced

activity of hippocampal neurons specifically. How would you test this theory in
animals? Please give the best model possible to answer this question. (3)