Chapter 1: Introduction to Social and Personality
Neuroscience Methods
What is Social and Personality Neuroscience?
Social psychology is often defined as the scientific study of how the thoughts, feelings, and behaviors
of an individual are influenced by the actual, imagined, or implied presence of others.
Personality psychology is the scientific study of how dispositional aspects of the individual influence
his or her thoughts, feelings, and behavior.
Social and personality neuroscience: interest in understanding the neurobiological substrates and
correlates of social and personality processes and behaviors. It emphasizes the relationships among
different levels of organization – form the molecule to the cell to the organ, system, person,
interpersonal, social group, and societal level. This multilevel approach, it is hoped,, will lead to an
increased understanding of the human mind and behavior.
Advantages of Social and Personality Neuroscience Methods
The appeal of neuroscience methods for social and personality psychology it that humans can be
noninvasively monitored in various situations.
Neuroscience methods are invaluable for social and personality psychology, because an appreciation
of the underlying biological and chemical processes will lead to a more complete understanding of
psychological and behavioral processes.
To fully understand behavior and psychology, researchers need to ‘get under the hood.’ But knowing
the outward behaviors and gauge reports will assist them in knowing where to look under the hood.
For a more complete understanding of human behavior, psychology needs neuroscience, and
neuroscience needs psychology.
Considerations for Using Neuroscience Methods
Most social and personality neuroscience studies involve intact humans and are thus noninvasive.
No single technique can measure all biological activities with excellent spatial and temporal
resolutions. That is, no one measure of brain function captures neuronal activity as it unfolds on the
order of milliseconds or nanoseconds (temporal resolution), or can specify exactly which neurons are
activated on the order of millimeters or nanometers (spatial resolution).
Event-related potential (ERPs) are often lauded for their excellent temporal resolution (milliseconds)
but damned for their poor spatial resolution (centimeters).
Functional magnetic resonance imaging (fMRI), is praised for its excellent spatial resolution
(millimeters) but condemned for its poor temporal resolution (seconds).
Brain activations obtained from ERPs or fMRI are essentially correlational, in that a psychological state
is manipulated and the brain activation is measured. Here it is impossible to determine causality.
Patient methods can assist in establishing causality, as the psychological task performance of
individuals who have suffered damage to one brain region can be compared to that of individuals
who have suffered damage to another brain region on psychological tasks. If the groups differ in their
performance, then the difference is probably due to processes supported by that specific brain
region. However, the lesions often involve a number of brain regions.
Transcranial magnetic stimulation (rTMS) can be used to increase or decrease neuronal activity
temporarily and non-invasively over particular areas. However, the ‘virtual lesions’ cannot penetrate
too deeply into the brain, and the spatial resolution of the method is not very precise.
For examination of hormones, many studies have manipulated or measured psychological variables
and then related these variables to hormone responses. Greater causal certainty can be obtained,
however, with manipulations of hormones.
,Chapter 9: Electroencephalographic Methods in Social
and Personality Psychology
Physiology Underlying Electroencephalography
EEG is the recording of electrical brain activity from the human scalp. The observed EEG at the human
scalp is the result of electrical voltages generated inside the brain.
Electrical activity associated with neurons comes from action potentials and postsynaptic potentials.
action potential: a discrete voltage spike that runs from the beginning of the axon at the cell body to
the axon terminals where neurotransmitters are released.
Postsynaptic potential: a voltage that occurs when the neurotransmitters bind to receptors on the
membrane of the postsynaptic cell. This causes ion channels to open or close, and it leads to a graded
change in the electrical potential across the cell membrane.
Recording of individual neurons or single-unit recordings measure action potentials and not
postsynaptic potentials.
Scalp-recorded EEG does not measure action potentials because of the timing of potentials and the
physical arrangement of axons. That is, unless the neuron fire within microseconds of each other,
action potentials in different axons will typically cancel each other out.
Postsynaptic potentials are longer in duration than action potentials. They are also mostly confined to
dendrites and cell bodies, and occur instantaneously rather than traveling down axons at a fixed rate.
These factors allow postsynaptic potentials to summate rather than cancel each other out.
Scalp-recorded electrical activity is the result of activity of populations of neurons.
For activity to be recorded at the scalp, the electric fields generated by each neuron must be oriented
in such a way that their effects accumulate. That is, the neurons must be arranged in an open as
opposed to a close field.
In an open field, neurons’ dendrites are all oriented on one side of the structure, while their axons all
depart from the other side. Open fields are present where neurons are organized in layers, as in most
of the cortex, parts of the thalamus, the cerebellum, and other structures.
Raw EEG signal is a complex waveform that can be analyzed in the temporal domain or frequency
domain. Processing of the temporal aspect is typically done with event-related potential designs and
analyses. In frequency analyses, the frequency is specified in hertz or cycles per second.
Recording
In contemporary social and personality research, EEG is recorded from electrodes that are mounted
in a stretch Lycra electrode cap. Electrodes are often made of tin or silver/silver chloride; the latter
are nonpolarizable, but are typically much more expensive.
Because most modern EEG amplifiers with high input impedance use very low electrode currents,
polarizable electrodes can often be used to record slow potentials without distortion. However, for
frequencies less than 0.1 Hz, nonpolarizable electrodes are recommended.
The first letter of the name of the electrode refers to the brain region over which the electrode sits;
thus Fp refers to frontal pole, F to frontal region, C to central region, P to parietal region, T to
temporal region, O to occipital region. Electrodes in between the regions are often designated by
using to letter, like FC for frontal-central. After the letter is a number or another letter. Odd numbers
are used to designate sites of the left side of the head, and even numbers on the right side. Numbers
increase as distance form the middle of the increases. The letter z is used to designate the midline,
which runs from front to back of the head.
Caps often contain a ground electrode, which is connected to the isoground of the amplifier and
assists in reducing electrical noise.
Recording of eye movements (EOG) is also carried out, to facilitate artifact scoring of the EEG. EOG
can be recorded from the supra- and suborbit of the eyes to assess vertical eye movements, and form
the left and right outer canthus to assess horizontal eye movements.
Additional electrodes are often placed on earlobes, so that offline digitally derived references can be
computed.
,Sites where electrodes will be placed must be abraded to reduce electrode impedances, typically
under 5000 ohms.
Conductive gel is used as a medium between the scalp and electrodes.
EEG, EOG, and other signals are amplified with bioamplifiers. For EEG frequency analyses, the raw
signals are often bandpass-filtered online, because the frequencies of interest fall between 1 and 40
Hz. Online notch filters are also often used to reduce electrical noise further.
From the amplifiers, the raw signals are then digitized onto a computer at a sampling rate greater
than twice the highest frequency of interest. This sampling rate is necessary because of the Nyquist
theorem, which states that exact reconstruction of a continuous signal from its samples is possible if
the signal is off limited bands and is sampled at least twice as great as the actual signal bandwidth. If
this sampling condition is not met, then frequencies will overlap; that is, frequencies above half the
sampling rate will be reconstructed as, and appear as, frequencies below half the sampling rate. This
distortion is called ‘aliasing’, because the reconstructed signal is said to be an alias of the original
signal.
Preparing the Participant
We avoid referring to electricity or anything that sound painful to participants. We also adopt the
mindset of a person who has done this very routine procedure many times. We also encourage
research assistants to avoid being too friendly or displaying too much empathy, as doing so could
drastically alter the mood of the participants.
With most EEG system, electrical impedance of the scalp will need to be brought under 5000 ohms.
To assist in reducing impedance, we ask participants to ‘brush your hair vigorously about 5 min. make
sure to press the brush hard against your scalp as you brush. It helps with he hookup process. I will
tell you when you can stop.’
After brushing their scalps, participants are told that we’re going to use an exfoliant to clean some
areas of their skin and use rubbing alcohol to remove the exfoliant. We clean their foreheads.
Earlobes, temples, and areas above/below the eyes with mildly abrasive cleaning solution and gauze
pad. We follow the cleaning by wiping the areas with alcohol.
The capping procedure: measure the length from the nasion to the inion with a metric tape. Then
10% of this distance is calculated and measured from the nasion and this spot will be marked with a
wax pencil. Cap size is determined by measuring the distance around the participant’s head, crossing
the marks on the forehead and the inion. Then two adhesive, cushioned collars are placed on cap
sites Fp1 and Fp2. These collars are then adhered to the forehead, in line with the wax pencil marks,
and centered over the nose. The cap is then stretched towards the back of the head and down. Then
me remeasure from the naison to the inion and ensure that Cz is hallway between these sites.
After attaching the cap’s connectors to the preamp, we place electrodes on each earlobe by placing
an adhesive collar on the flat side of and sticking it on the ear; additional adhesive collars can be
placed on top of the electrode, to ensure that the electrode remains attached. We then fill the sensor
with gel and abrade the ground site with the blunt tip of a wooden Q-tip or blunt tip of a needle, and
appy gel with a syringe. Then we attach the chin strap to the cap and position it under the
participant’s chin.
For measuring vertical eye movements, one electrode is placed 10% of the inion-nasion distance
above the pupil, whereas another is placed 10% of the inion-nasion distance below the pupil. These
electrodes are referenced to each other, rather than to the EEG reference electrode.
For measuring horizontal eye movements, one electrode is placed on the right temple, and another is
placed on the left temple. These electrodes are also referenced to each other.
We attempt to reduce EOG impedance to only 10,000 ohms, so as to avoid overabrading the face.
Artifacts
Artifacts are much easier to deal with if they are prevented from occurring. However, if they are
recorded in the EEG, procedures exist to handle them.
, Muscle Artifact
EMG is typically of higher frequencies than EEG. However, some EMG blends in with the EEG
frequencies, so it is advisable to limit muscle artifacts by training the participants to limit muscle
movements. If they appear in studies where they should not occur, the artifacts can be removed
during the data-processing stage (often by hand).
Particularly in experiments that evoke emotion, muscle artifacts cannot be avoided. One way to
handle the EMG that may contaminate the EEG is to measure facial EMG and then use the facial EMG
responses in covariance analyses, to assess whether statistically adjusting the EEG data for several
possible covariates has eliminated the effect on EEG interest. Similarly, one could obtain EMG
frequencies form the EEG sites and use these EMG frequencies in covariance analyses. Another way
to address this concern is to assess the EEG frequency of interest in the facial EMG sites, and test
whether this covariate would account for the EEG effects.
Eye Movement Artifacts
They are often best dealt with in advance of EEG recording. Training participants to limit eye
movements during EEG recording is recommended. However, researchers must not encourage
participants to control their blinking, because blinks and spontaneous eye movements are controlled
by several brain systems in a highly autonomic fashion, and suppressing these systems may act as a
secondary task.
Epochs containing blinks should be removed from the EEG or corrected via a computer algorithm.
These algorithms rely on regression techniques. In general, the actual EEG time series represents a
new EEG from which the influence of the ocular activity is statistically removed. Then EOG-corrected
EEG data may be processed as would EEG data without EOG artifact. The latter has the advantage of
not losing data, as the eye movement rejection procedure does.
Eye movement artifact correction does not cause a distortion of frontal EEG asymmetry – a measure
of interest in contemporary social and personality psychology.
Nonbiological Artifacts
These are those that typically involve external electrical noise. Most of these problems can be dealt
with through filtering of these signal. High electrode impedances or a faulty ground connection can
increase 60-Hz noise. Electrodes need to be carefully washed after each use, to prevent corrosion and
thus to assist in prevention or artifacts.
Offline Data Processing
Referencing
There are no ‘inactive sites’, and all sites near the scalp reflect some EEG because of volume
conduction. To address this issue, some researchers suggest using an everage reference consisting of
the average of activity at all recorded EEG sites. The average reference should approximate an
inactive reference if a sufficiently large array of electrodes is placed in a spherical arrangement
around the head. That is, activity generated from dipoles will be positive at one site and negative at a
site 180 degrees opposite to this side, and thus the sum across sites should approach zero with a
representative sample of the sphere. Because electrodes are not placed under the head, the
assumption is rarely met. Moreover, use of smaller montages of electrodes causes more residual
activity in the average reference.
The use of linked ears could be used as a reference, because of the ears’ relatively low activity and
because the linking of the ears should theoretically center the reference on the head, making the
determination of lateralized activity more accurate. Physically linking the ears does not alter the
observed EEG asymmetries. However, when ears are linked prior to being input to the amplifier,
variations of the impedances of the left and right electrodes will change the spatial location of the
reference, and thus will potentially alter the magnitude and direction of any observed differences in
left versus right EEG activity.
