This reading and course assumes a basic understanding of cell biology and action potentials.

If you are unsure

about any of these topics, please review the background material provided on the Neural Recordings
Supplemental Reading Material page.

Introducing Electrophysiology
Electrophysiology is the study of the electrical properties of cells, and electrophysiological techniques allow
researchers to record and manipulate the electrical activity of cells such as neurons.

Recording From Single Cells: Adapted from Handbook of Neural Activity Measurement Intracellular Recordings
Intracellular recordings allow us to study the electrical events in a single neuron that are smaller than an
action potential threshold and lead up to the generation and/or inhibition of action potentials. There are several
intracellular recording techniques, but we will discuss the most current and common way of intracellularly
recording from an entire single neuron, which is whole-cell patch clamp.

Whole-Cell Patch Clamp

Whole-cell patch clamp is typically performed on recently obtained brain slices (a few hours old maximum)
bathed in oxygenated artificial cerebrospinal fluid (ACSF) to keep the cells alive during the recording process.
The electrophysiologist will use a powerful microscope to guide the extremely thin end of an electrode-
containing glass pipette to the surface of a neuron, and will then use suction to break off a piece of the cell
membrane and form a tight gigaohm seal with the remaining ends.

Figure 1. Whole-cell patch clamp procedure: 1) The electrophysiologist first guides the recording pipette to the surface of the soma. 2) Then, mild suction
from the pipette creates tight contact between the pipette and the membrane. 3) A strong pulse of suction then tears the membrane under the pipette. The
remaining ends form a tight, gigaohm seal with the interior of the pipette, making the cytoplasm continuous with the pipette interior. Image from Leica
Microsystems.

This high resistance seal is important for two reasons: 1) it keeps the cell alive by preventing the cytosol
from leaking out of the cell when the cell membrane breaks 2) it isolates the recording from external currents
coming from surrounding cells. With a patch clamp, electrophysiologists can measure the intracellular voltage
over time to create the plots in the video you watched prior to this reading. Key characteristics of intracellular
recordings are the consistent nature of their resting potential (approximately -70 mV) and the consistent
waveform of action potentials. That is, intracellular action potentials have remarkably consistent durations,
baseline voltages, and peak voltages. This also means that intracellular action potentials will always have
positive deflection (i.e., going from a more negative value to a more positive value).

Single-Unit Extracellular Recordings:

A major advantage of extracellular recordings compared to intracellular recordings, is that extracellular
recordings can be done in vivo. Like patch-clamp recordings, single-unit extracellular recordings use an
extremely fine electrode that is placed within the neural tissue. Unlike patch-clamp recordings, the electrode is
placed near the soma just outside the cell, and voltages from the extracellular fluid are recorded. The term
“single-unit” comes from the fact that the electrode is small enough and close enough to record from only a
single nearby neuron.
Single-unit extracellular voltage measurements will often appear inverted compared to intracellular
recordings. This is because during a neuron depolarization, an extracellular recording electrode is measuring
cations leaving, whereas an intracellular recording electrode would be measuring cations approaching. The
extracellular voltage measurements will also be considerably smaller than intracellular voltage measurements.
Voltage changes due to an action potential recorded extracellularly are much smaller than those recorded
intracellularly because relative changes in ion concentration are much smaller in extracellular fluid.
Figure 2. Simultaneous recording of the intracellular and extracellular spike with a double electrode. The monophasic depolarization seen by the
intracellular electrode corresponds with a positive-negative-positive (triphasic) waveform recorded by the extracellular electrode. Note different scales for
extracellular and intracellular recordings. Adapted from Terzuolo & Araki, Ann N Y Acad Sci. 1961.

Spike Detection
Both intracellular recordings and extracellular recordings can be used to detect discrete action potentials.
Often this this done by setting a spike threshold on the recording amplitude. For example, when the value of
the intracellular recording exceeds 50 mV, we can reasonably assume that a single action potential has
occurred. By looking at the signal over time, and measuring each time the signal passes 50 mV (or some other
threshold value), we can identify discrete timepoints when action potentials occurred, commonly referred to as
spikes. The process of visualizing spikes in their discrete format (zero or one) is often referred to as a raster
plot. A raster plot of discrete spikes can also be generated from extracellular recordings – often using a
negative threshold at a smaller value. For example, when the extracellular recording drops below -5 mV, we
could assume a spike has occurred.

Figure 3. Example intracellular and extracellular recordings occurring at the same times. Arbitrary spike-detection thresholds are shown in red, green and
blue, and the resulting spikes, shown as discrete ticks in time known as a raster plot. Note how the number of spikes detected from the extracellular
waveform can vary greatly based on the choice of the spike detection threshold. Adapted from Ishimatsu & Williams, J Neurosci. 1996
Multi-Unit Extracellular Recordings: Adapted from Buzsáki et al., Nat Rev Neurosci. 2012 and Blache et al., J Neuro Methods, 2005.
In most cases, a single electrode placed in extracellular fluid will often record from multiple nearby neurons.
In cases where multiple neurons are close enough to the same recording electrode, extracellular action
potentials can be detected from both neurons. When multiple neurons are present in a single electrode
recording, it’s referred to as “Multi-Unit Extracellular Recordings”, or “multi-unit activity” for brief. When multiple
extracellular waveforms are detected from a single recording channel, spike sorting can be used to sort out
which waveforms belong to the different nearby neurons. Spike sorting is the process of separating recorded
extracellular action potentials into different clusters such that extracellular action potentials with similar
waveforms constitute individual clusters. These individual clusters are assumed to represent individual
neurons. However, spike sorting algorithms are imperfect and can incorrectly assign spikes to clusters. For
example, sometimes two different neurons may produce a very similar waveform, resulting in someone
mislabeling multi-unit activity as single-unit activity. On the other hand, sometimes a single neuron can produce
two different extracellular waveforms, resulting in someone mislabeling single-unit activity as multi-unit activity.

Amplitude and Signal-to-Noise Ratio of Spikes

The amplitude of a recorded spike can be measured by its peak-to-peak voltage. When recording
intracellularly, the peak-to-peak voltage is extremely consistent, going from approximately -70 mV to
approximately +30 mV. Thus, the amplitude of an intracellular spike is typically around 100 mV. In contrast, the
amplitude of an extracellular spike is much smaller, typically less than 500 µV. This is mainly because relative
changes in ion concentration are much smaller in extracellular fluid. But in addition, as you move further and
further away from a neuron, the relative changes in ion concentration become smaller and smaller.
The signal-to-noise ratio of a recorded spike can be measured by its peak-to-peak voltage (i.e., the
amplitude of the action potential signal) divided by the standard deviation of the voltage during rest (i.e., the
background electrical noise). The signal-to-noise ratio of intracellular spikes is very high because the resting
membrane potential is fairly consistent and because the amplitude of the action potential is very large. In
contrast, extracellular action potentials have much smaller amplitudes, as noted above. In addition,
extracellular recordings tend to have much more background noise. In active neural tissue, there are many
neurons, each firing their own action potentials. An extracellular electrode will record large changes in voltage
due to ion changes resulting from a nearby neuron’s action potential. But that same extracellular electrode will
also record small changes in voltage due to ion changes resulting from a distant neuron’s action potential.
Summing the relative contributions of distant neurons’ action potential over space and time, results in
background fluctuations in the recorded voltage. This contributes to a noisier background signal, and thus a
lower signal-to-noise ratio.
In the case where the signal-to-noise ratio is so poor, that you can’t readily detect an action potential from
an extracellular recording, we will often label this recording as a local field potential (LFP) instead of multi-unit
activity. The term “local field potential” is a regrettable malapropism, because the recording is actually from
many distant neurons that are not truly “local.” However, in neuroscience, we continue to use the term LFP
because it is familiar.

Local Field Potential (LFP) versus Multi-Unit Activity (MUA)

LFP broadly refers to the spatiotemporal summations of multiple neurons across a large volume of neural
tissue. In contrast to MUA, LFP does not contain detectable spikes. That is, no such spike detection threshold
could be established to reliably detect action potentials. As a reminder, this happens when the signal-to-noise
ratio is too low, meaning the action potential amplitude is too small (i.e., the nearest neuron is far away) and
the background noise is high (i.e., the contributions from multiple distant neurons are greater than that of the
single closest neuron). Because multiple distant neurons are the primary source of the LFP signal, LFP is both
the spatial and temporal summation of electrical neural activity.
Another key distinction between LFP and MUA is the sampling rate. With MUA, the intention is to detect
extracellular action potentials from a few nearby neurons, or even a single nearby neuron (i.e., single-unit
activity; SUA). Extracellular action potentials typically occur on the order of approximately 1 millisecond. If you
recorded data with only a 1 kHz sampling rate, the entire action potential could take place in between your
samples, and you would never know. The Nyquist Theorem states that in order to reconstruct the signal fully,
you have to sample the signal at least twice the highest frequency content of the signal (~5 kHz for neural
data). Thus, the minimum sampling rate to recreate an extracellular action potential would be ~10 kHz.
Although 10 kHz is likely the theoretical minimum, in practice, most electrophysiologists sample MUA and SUA
at 20 – 40 kHz. In contrast, LFP data is typically sampled at 1 – 5 kHz, because there is no need to capture the
fast spike events and because the LFP signal is already a spatiotemporal summation of many events.

Different LFP Waveforms: Adapted from Buzsáki, Anastassiou, & Koch, Nat Rev Neurosci. 2012
Electric current contributions from all active cellular processes within a volume of brain tissue superimpose
at a given location in the extracellular medium and generate a potential, Ve (a scalar measured in Volts), with
respect to a reference potential. The difference in Ve between two locations gives rise to an electric field (a
vector whose amplitude is measured in Volts per distance) that is defined as the negative spatial gradient of
Ve. Electric fields can be monitored by extracellularly placed electrodes with submillisecond time resolution and
can be used to interpret many facets of neuronal communication and computation. A major advantage of
extracellular field recording techniques is that, in contrast to several other methods used to investigate network
activity, the biophysics related to these measurements are well understood. This has enabled the development
of reliable and quantitative mathematical models to elucidate how transmembrane currents give rise to the
recorded electric potential.
Historically, Ve has been referred to as the electroencephalogram (EEG) when recorded from the scalp,
as the electrocorticogram (ECoG) when recorded by subdural grid electrodes on the cortical surface, and as
the local field potential (LFP; also known as micro-, depth or intracranial EEG) when recorded by a small
size electrode in the brain. Ve can also be calculated for other biological tissues that generate electricity such
as skeletal muscles (electromyogram, EMG) and cardiac muscles (electrocardiogram, EEG). These signals
can all broadly be classified as “local field potentials” because they all record spatiotemporal summations of
electric potentials generated by tissue, and because they are all sampled at a similar sampling frequency (~1 –
5 kHz).
Electroencephalography (EEG) is one of the oldest and most widely used methods for the investigation of
the electric activity of the brain. The scalp electroencephalogram, recorded by a single electrode, is a
spatiotemporally smoothed version of the local field potential (LFP), integrated over an area of 10 cm2 or more.
Under most conditions, it has little discernible relationship with the firing patterns of the contributing individual
neurons, and this is largely due to the distorting and attenuating effects of the soft and hard tissues between
the current source and the recording electrode. The recently introduced ‘high- density’ EEG recordings, in
combination with source-modelling that can account for the gyri and sulci (as inferred from structural MRI
imaging) of the subject, have substantially improved the spatial resolution of EEG.
Electrocorticography (ECoG) is becoming an increasingly popular tool for studying various cortical
phenomena in clinical settings. It uses subdural platinum–iridium or stainless-steel electrodes to record electric
activity directly from the surface of the cerebral cortex, thereby bypassing the signal-distorting skull and
intermediate tissue. The spatial resolution of the recorded electric field can be substantially improved (<5 mm2)
by using flexible, closely spaced subdural grid or strip electrodes.
EEG and ECoG mainly sample electrical activity in the cortex's superficial layers. Electrical events at
deeper locations can be explored by inserting metal or glass electrodes, or silicon probes into the brain to
record the LFP (also known as ‘micro-EEG’). Recording the wide-band signal (direct current to 40 kHz) —
which contains both action potentials and other membrane potential-derived fluctuations in a small neuronal
volume — using a microelectrode yields the most informative signal for studying cortical electrogenesis. This
wide-band signal sampled between 20 – 40 kHz gives rise to the MUA and SUA signals described earlier.
Many observation points, with short distances between the recording sites and with minimal impact on brain
tissue, are needed to achieve high spatial resolution. In principle, the spiking activity of nearly all or at least a
representative fraction of the neuron population in a small volume can be monitored with a sufficiently large
density of recording sites. Additional clues about the intracellular dynamics can be deduced from the waveform
changes of the extracellular action potentials. Progress in this field has been accelerated by the availability of
micro-machined silicon-based probes with ever-increasing numbers of recording sites.

Figure 4. A) Illustration of different extracellular recording technologies. EEG records from outside of the skull, on the scalp. ECoG records from the
surface of the brain. LFP is recorded from an electrode implanted inside of the brain tissue. Note the differences in the recordings (both in terms of
magnitude and temporal resolution) as distance from the neurons increases. B) Example of intracellular and extracellular recordings taken from layer V
of the brain. Intracellular action potentials are always the same magnitude, are always positive, and always monophasic. The raster plot shows the
discrete spike events. Multiunit activity is recorded from the extracellular electrode. LFP is recorded from the surface of the brain. Note the differences in
magnitude of the recordings (see bottom right legend) and temporal features. From Obien, Deligkaris, Bullmann, Bakkum, & Frey. Front Neurosci.

Origin of Extracellular Waveforms: adapted from Adapted from Buzsáki, Anastassiou, & Koch, Nat Rev Neurosci. 2012 and from
[Link]
Any excitable membrane — whether it is a spine, dendrite, soma, axon or axon terminal — and any type of
transmembrane current contributes to the extracellular field. The field is the superposition of all ionic
processes, from fast action potentials to the slowest fluctuations in glia. All currents in the brain superimpose at
any given point in space to yield Ve at that location. Thus, any transmembrane current, irrespective of its origin,
leads to an intracellular as well as an extracellular (that is, LFP) voltage deflection. The characteristics of the
LFP waveform, such as the amplitude and frequency, depend on the proportional contribution of the multiple
sources and various properties of the brain tissue. The larger the distance of the recording electrode from the
current source, the less informative the measured LFP becomes about the events occurring at the location(s)
of the source(s). This is mainly owing to the fact that the Ve amplitude scales with the inverse of the distance r
between the source and the recording site, and to the inclusion of other (interfering) signals (leading to ‘spatial
averaging’). In addition to the magnitude and sign of the individual current sources, and their spatial density,
the temporal coordination of the respective current sources (that is, their synchrony) shapes the extracellular
field. Thus, extracellular currents can emerge from multiple sources such as: synaptic activity, action potentials,
calcium spikes, gap junctions, neuron-glia interactions, etc.
Figure 5. Examples of extracellular recordings taken in two-dimensional space outside of a neuron. The same traces are shown normalized to the negative
peak (bottom right panel). Note the widening of the spike with distance from the soma, owing to greater contributions from dendritic currents and intrinsic
filtering of high-frequency currents by the cell membrane. Note the shift in polarity depending on location as well. From Buzsáki, Anastassiou, & Koch, Nat
Rev Neurosci. 2012

Figure 5 highlights one final important point about extracellular action potentials: they can be both positive
and negative, and both biphasic and triphasic. The action potential recorded with an extracellular electrode is
produced by currents that are induced to flow in the extracellular space around an active neuron. A
straightforward approach to these current flows is provided by volume conductor theory. In this approach, the
neuron is visualized as surrounded by an extracellular medium with low uniform resistance; that is, a “volume
conductor.” In the simplest case, one can imagine an isolated axon in a saline bath. When the axon is at rest,
the membrane potential is uniform along its entire length, and there is no current flowing inside or outside the
cell (Fig. 2– 1A).
 However, if the axon is depolarized at some point along the membrane, the potential difference between
the depolarized and resting regions will cause current to flow (Fig. 2–1B). Current flows inward at the active
region, and an electrode that is adjacent to the axonal membrane at this point will be negative with respect to a
distant indifferent electrode. The active region is referred to as a current “sink.” Inactive regions of the
membrane adjacent to the depolarized area are then said to act as a “source” of current for the active region.
Because current flows outward at the source, an electrode recording from a region of membrane that is acting
as a current source will be positive relative to a distant indifferent electrode.
At least in theory, one should be able to model the distribution of sources and sinks associated with action
potential discharge in an active neuron. This proves to be a straightforward task when considering the isolated
axon (Fig. 2–2). As an action potential approaches, reaches, and then moves away from an electrode adjacent
to a spot somewhere along the axon, the electrode will at first register a positive potential as the membrane
under the electrode serves as a current source for the depolarized membrane some distance away. Then,
when the action potential reaches the region underlying the electrode, the depolarized membrane will act as a
sink. The extracellular electrode will therefore record a negative potential. Finally, as the action potential
continues past the electrode, the underlying membrane is repolarized and once again serves as a current
source. The recording electrode will therefore again be positive relative to a distant reference electrode.

The action potential recorded by an electrode adjacent to an isolated axon should therefore be triphasic.
This theoretical prediction has been confirmed, as shown in the example from an early experiment by Terzuolo
and Araki in 1961 (see blue and red traces in Fig. 2 from earlier). As predicted by volume conductor theory, the
potential recorded with the extracellular electrode was triphasic, and the negative phase of the extracellular
spike coincided with the depolarization seen by the intracellular electrode. The late positive phase of the
extracellular potential corresponded to the repolarization of the membrane recorded intracellularly.
Similar considerations can be applied when modeling the action potential recorded close to the cell soma.
As shown in the simple neuron model in Figure 2–4, depolarization at the soma should be reflected in negative
potential at the electrode. The soma membrane would subsequently become a source as the action potential
moved away down the axon. As a result, the waveform recorded with an extracellular electrode near the soma
should in theory be biphasic, with an initial negative component followed by a later positivity. However, actual
neurons are somewhat more complex than the “lollipop” envisioned in Figure 2–4. Cell morphology and
distribution of active conductances on the somatic and dendritic membrane, the location of the electrode
relative to the cell body, and the state of cell excitability all complicate the simple picture derived from volume
conductor theory. As a result, somatic recordings vary in the relative amplitude of the negative and positive
phases of the waveform and often show distinct inflection points on either phase. See Figure 5 for a more
complex example of extracellular recordings around a pyramidal neuron, which includes both positive and
negative waveforms depending on the recording location.