C H A P T E R 42

Basic Principles
of Sensory Physiology
CHAPTER CONTENTS
!." Electrical Signals in Neurons !.! Sensory Coding: How Neurons METHOD: Brain Imaging
Recording Electrical Signals Represent Information Distributed Representation
in Neurons Specificity Coding Connections Between Brain Areas
METHOD: The Setup for Recording Sparse Coding METHOD: The Resting State Method
From a Single Neuron Population Coding of Measuring Functional Connectivity
Basic Properties of Action Potentials TEST YOURSELF 2.1 SOMETHING TO CONSIDER:
Chemical Basis of Action Potentials !.# Zooming Out: Representation The Mind–Body Problem
Transmitting Information Across in the Brain TEST YOURSELF 2.2
a Gap Mapping Function to Structure
THINK ABOUT IT

Some Questions We Will Consider: receptor had been stimulated, then the straight-through
method would work. But the purpose of electrical signals in
■ How do neurons work, and how does neural firing under- the nervous system goes beyond signaling that a receptor was
lie our perception? (p. 21) stimulated. The information that reaches the brain and then
■ How do perceptual functions map onto the structure of continues its journey within the brain is much richer than this.
the brain? (p. 30) As we will see in this and upcoming chapters, there are neurons
■ How is brain activity measured both within a brain area in the brain that respond to certain stimuli like slanted lines,
and between different brain areas? (p. 31) faces, movement across space in a specific direction, movement
across the skin in a specific direction, or salty tastes. These

T
neurons didn’t achieve these properties by receiving signals
wo cars start at the same place and drive to the same through a straight-line transmission system from receptors
destination. Car A takes an express highway, stopping to brain. They achieve these properties by neural processing—the
only briefly for gas. Car B takes the “scenic” route— interaction of the signals of many neurons (see page 8).
back roads that go through the countryside and small towns, Because the activity of individual neurons and neural pro-
stopping a number of times along the way to see some sights cessing carried out by large numbers of neurons create our
and meet some people. Each of Car B’s stops can influence its perceptual experiences, it is important to understand the basic
route, depending on the information its driver receives. Stop- mechanisms behind neural responding and neural processing.
ping at a small-town general store, the driver of Car B hears We begin by describing electrical signals in neurons.
about a detour up the road, so changes the route accordingly.
Meanwhile, Car A is speeding directly to its destination.
The way electrical signals travel through the nervous system
is more like Car B’s journey. The pathway from receptors to brain
2.1 Electrical Signals
is not a nonstop expressway. Every signal leaving a receptor trav-
els through a complex network of interconnected signals, often
in Neurons
meeting, and being affected by, other signals along the way. Electrical signals occur in structures called neurons, like the ones
What is gained by taking a complex, indirect route? If the shown in Figure 2.1. The key components of neurons, shown in
goal were just to send a signal to the brain that a particular the neuron on the right in Figure 2.1, are the cell body, which

21

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Touch receptor Dendrite
Stimulus from Nerve fiber
environment Axon or nerve fiber
Synapse

Cell body
Electrical
signal

Figure 2.1 The neuron on the right consists of a cell body, dendrites, and an axon, or nerve fiber. The
neuron on the left that receives stimuli from the environment has a receptor in place of the cell body.

contains mechanisms to keep the cell alive; dendrites, which Recording Electrical Signals in Neurons
branch out from the cell body to receive electrical signals from
other neurons; and the axon, or nerve fiber, which is filled with Electrical signals are recorded from the axons (or nerve fibers)
fluid that conducts electrical signals. There are variations on this of neurons using small electrodes to pick up the signals.
basic neuron structure: Some neurons have long axons; others
have short axons or none at all. Especially important for percep- METHOD The Setup for Recording From a Single
tion are sensory receptors (see Figure 1.5), which are neurons special- Neuron
ized to respond to environmental stimuli. The receptor on the left
Figure 2.2a shows a typical setup used for recording from a sin-
in Figure 2.1 responds to touch stimuli.
gle neuron. There are two electrodes: a recording electrode,
Individual neurons do not, of course, exist in isolation.
shown with its recording tip inside the neuron,1 and a reference
There are hundreds of millions of neurons in the nervous sys-
electrode, located some distance away so it is not affected by
tem, and each neuron is connected to many other neurons. As
the electrical signals. These two electrodes are connected to a
we will discuss later in this chapter, these connections are ex-
meter that records the difference in charge between the tips of
tremely important for perception. To begin our discussion of
the two electrodes. This difference is displayed on a computer
how neurons and their connections give rise to perception, we
screen, like the one shown in Figure 2.3, which shows electrical
focus on individual neurons.
signals being recorded from a neuron.
One of the most important ways of studying how electrical
signals underlie perception is to record signals from single neu-
rons. We can appreciate the importance of being able to record
When the axon, or nerve fiber, is at rest, the difference in
from single neurons by considering the following analogy: You
the electrical potential between the tips of the two electrodes
walk into a large room in which hundreds of people are talking
is –70 millivolts (mV, where a millivolt is 1/1,000 of a volt),
about a political speech they have just heard. There is a great deal
as shown on the right in Figure 2.2a. This means that the in-
of noise and commotion in the room as people react to the speech.
side of the axon is 70 mV more negative than the outside. This
Based on hearing this “crowd noise,” all you can say about what is
value, which stays roughly the same as long as there are no
going on is that the speech seems to have generated a great deal of
signals in the neuron, is called the resting potential.
excitement. To get more specific information about the speech,
Figure 2.2b shows what happens when the neuron’s recep-
you need to listen to what individual people are saying.
tor is stimulated so that a signal is transmitted down the axon.
Just as listening to individual people provides valuable in-
As the signal passes the recording electrode, the charge inside
formation about what is happening in a large crowd, record-
the axon rises to +40 mV compared to the outside. As the signal
ing from single neurons provides valuable information about
continues past the electrode, the charge inside the fiber reverses
what is happening in the nervous system. Recording from
course and starts becoming negative again (Figure 2.2c), until it
single neurons is like listening to individual voices. It is im-
returns to the resting level (Figure 2.2d). This signal, identified
portant to record from as many neurons as possible, of course,
by the predictable rise and fall of the charge inside the axon rela-
because just as individual people may have different opinions
tive to the outside, is called the action potential, and lasts about
about the speech, different neurons may respond differently to
1 millisecond (ms, 1/1,000 second). When we refer to neurons as
a particular stimulus or situation.
“firing,” we are referring to the neuron having action potentials.
The ability to record electrical signals from individual
neurons ushered in the modern era of brain research, and in
1
the 1950s and 1960s, development of sophisticated electronics In practice, most recordings are achieved with the tip of the electrode positioned
just outside the neuron because it is technically difficult to insert electrodes into the
and the availability of computers made possible more detailed neuron, especially if it is small. However, if the electrode tip is close enough to the
analysis of how neurons function. neuron, the electrode can pick up the signals generated by the neuron.

22 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Recording electrode Reference Figure 2.2 (a) When a nerve fiber is at rest,
(inside axon) Meter electrode there is a difference in charge of –70 mV between
Resting
Push (outside potential the inside and the outside of the fiber. This
axon) difference, which is measured by the meter
–70
indicated by the blue circle, is displayed on the
Time right. (b) As the nerve impulse, indicated by
Pressure-sensitive receptor the red band, passes the electrode, the inside
(a)

Charge inside fiber relative to outside (mV)

of the fiber near the electrode becomes more
Nerve positive. This positivity is the rising phase of the
+40
impulse action potential. (c) As the nerve impulse moves
past the electrode, the charge inside the fiber
becomes more negative. This is the falling phase
–70 of the action potential. (d) Eventually the neuron
returns to its resting state.

(b)

–70

(c)

Back at
resting
level

–70

(d)

Your experience of seeing the words on this page, hearing the

sounds around you, and tasting your food all start with electri-
cal signals in neurons. In this chapter we will first describe how
individual neurons work, and will then consider how the activity
of groups of neurons is related to perception. We begin by describ-
ing basic properties of the action potential and its chemical basis.

Basic Properties of Action Potentials

An important property of the action potential is that it is a
propagated response—once the response is triggered, it travels all
the way down the axon without decreasing in size. This means
that if we were to move our recording electrode in Figure 2.2 to a
position nearer the end of the axon, the electrical response would
take longer to reach the electrode, but it would still be the same
Bruce Goldstein

size (increasing from 270 to 140 mV) when it got there. This is
an extremely important property of the action potential because
it enables neurons to transmit signals over long distances.
Figure 2.3 Electrical signals being displayed on a computer
screen, in an experiment in which responses are being recorded
Another property is that the action potential remains the
from a single neuron. The signal on the screen shows the difference same size no matter how intense the stimulus is. The three records
in voltage between two electrodes as a function of time. In this in Figure 2.4 represent the axon’s response to three intensities
example, many signals are superimposed on one another, creating a of pushing on the skin. Each action potential appears as a sharp
thick white tracing. (Photographed in Tai Sing Lee’s laboratory at Carnegie Mellon University) spike in these records because we have compressed the time scale

2.1 Electrical Signals in Neurons 23

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 K+
Na+ Na+
Na+
Na+

(a) K+
K+
K+

Na+
K+
Na+
(b) Na+
K+
Na+ Na+

Figure 2.5 A nerve fiber showing the high concentration of sodium

(c) ions (Na+) outside the fiber and potassium ions (K+) inside the fiber.
Other ions, such as negatively charged chlorine, are not shown.
Time
Pressure on Pressure off

Figure 2.4 Response of a nerve fiber to (a) soft, (b) medium, and The key to understanding the “wet” electrical signals trans-
(c) strong stimulation. Increasing the stimulus strength increases mitted by neurons is understanding the components of the neu-
both the rate and the regularity of nerve firing in this fiber, but has ron’s liquid environment. Neurons are bathed in a liquid solution
no effect on the size of the action potentials. rich in ions, molecules that carry an electrical charge (Figure 2.5).
Ions are created when molecules gain or lose electrons, as hap-
pens when compounds are dissolved in water. For example, add-
to display a number of action potentials. Figure 2.4a shows how ing table salt (sodium chloride, NaCl) to water creates positively
the axon responds to gentle stimulation applied to the skin, and charged sodium ions (Na+) and negatively charged chlorine ions
Figures 2.4b and 2.4c show how the response changes as the (Cl–). The solution outside the axon of a neuron is rich in posi-
pressure is increased. Comparing these three records leads to an tively charged sodium (Na+) ions, whereas the solution inside the
important conclusion: Changing the stimulus intensity does axon is rich in positively charged potassium (K+) ions. This distri-
not affect the size of the action potentials but does affect the rate bution of ions across the neuron’s membrane at rest is important
of firing. to maintaining the –70 mV resting potential, as well as to the ini-
Although increasing the stimulus intensity can increase tiation of the action potential itself.
the rate of firing, there is an upper limit to the number of nerve You can understand how these ions result in the action po-
impulses per second that can be conducted down an axon. This tential by imagining yourself just outside an axon next to a re-
limit occurs because of another property of the axon called the cording electrode (Figure 2.6a). (You will have to shrink your-
refractory period—the interval between the time one nerve im- self down to a very small size to do this!) Everything is quiet
pulse occurs and the next one can be generated in the axon. until incoming signals from other neurons trigger an action
Because the refractory period for most neurons is about potential to begin traveling down the axon. As it approaches,
1 millisecond, the upper limit of a neuron’s firing rate is about you see positively charged sodium ions (Na+) rushing into the
500 to 800 impulses per second. axon (Figure 2.6b). This occurs because channels in the mem-
Another important property of action potentials is illus- brane that are selective to Na+ have opened, which allow Na+
trated by the beginning of each of the records in Figure 2.4. No- to flow across the membrane and into the neuron. This open-
tice that a few action potentials are occurring even before the ing of sodium channels represents an increase in the mem-
pressure stimulus is applied. Action potentials that occur in brane’s selective permeability to sodium, where permeability
the absence of stimuli from the environment are called sponta- refers to the ease with which a molecule can pass through the
neous activity. This spontaneous activity establishes a baseline membrane and selective means that the fiber is permeable to
level of firing for the neuron. The presence of stimulation usu- one specific type of molecule, Na+ in this case, but not to oth-
ally causes an increase in activity above this spontaneous level, ers. The inflow of positively charged sodium causes an increase
but under some conditions, which we will describe shortly, it in the positive charge inside the axon from the resting poten-
can cause firing to decrease below the spontaneous level. tial of –70 mV until it reaches the peak of the action potential
of +40 mV. An increase in positive charge inside the neuron
Chemical Basis of Action Potentials is called depolarization. This quick and steep depolarization
from –70 mV to +40 mV during an action potential is referred
What causes these rapid changes in charge that travel down the to as the rising phase of the action potential (Figure 2.6b).
axon? Because this is a traveling electrical charge, we might be Continuing your vigil, you notice that once the charge in-
tempted to equate it to the electrical signals that are conducted side the neuron reaches +40 mV, the sodium channels close (the
along electrical power lines or the wires used for household appli- membrane becomes impermeable to sodium) and potassium
ances. But action potentials create electricity not in the dry envi- channels open (the membrane becomes selectively permeable to
ronment of metal wires, but in the wet environment of the body. potassium). Because there were more potassium ions (K+) inside
24 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Figure 2.6 How the flow of sodium and potassium
creates the action potential. (a) When the fiber is at rest,
Resting
there is no flow of ions, and the record indicates the
potential
–70 mV resting potential. (b) Ion flow occurs when an
–70 action potential travels down the fiber. Initially, positively
charged sodium (Na+) flows into the axon, causing the
Time
inside of the neuron to become more positive (rising
(a)
phase of the action potential). (c) Later, positively charged
+40 potassium (K+) flows out of the axon, causing the inside
K+
of the axon to become more negative (falling phase of the
Sodium
flows action potential). (d) When the action potential has passed
into axon the electrode, the charge returns to the resting level.
Na+
Charge inside fiber relative to outside (mV)
–70

(b)

K+ Potassium
flows out
of axon
Na+
–70

(c )

Back at
K+ resting Receiving
level neuron
Receptor
Na+
–70 Axon
Stimulus

Nerve
(d) impulse
(a)

Neurotransmitter
than outside the neuron while at rest, positively charged potas- molecules Receiving
sium rushes out of the axon when the channels open, causing neuron
the charge inside the axon to become more negative. An increase Synaptic vesicle
in negative charge inside the neuron is called hyperpolariza-
tion. The hyperpolarization from +40 mV back to –70 mV is Axon of
the falling phase of the action potential (Figure 2.6c). Once sending neuron
the potential has returned to the –70 mV resting level, the K+
flow stops (Figure 2.6d), which means the action potential is
over and the neuron is again at rest. (b)
After reading this description of ion flow, students often
ask why the sodium-in, potassium-out flow that occurs during
the action potential doesn’t cause sodium to build up inside
the axon and potassium to build up outside. The answer is Receptor site
that a mechanism called the sodium-potassium pump keeps this Neurotransmitter
molecules
buildup from happening by continuously pumping sodium
out and potassium into the fiber.

Transmitting Information Across a Gap

(c)
We have seen that action potentials caused by sodium and po-
Figure 2.7 Synaptic transmission from one neuron to another.
tassium flow travel down the axon without decreasing in size.
(a) A signal traveling down the axon of a neuron reaches the
But what happens when the action potential reaches the end of synapse at the end of the axon. (b) The nerve impulse causes the
the axon? How is the action potential’s message transmitted to release of neurotransmitter molecules (red) from the synaptic
other neurons? The problem is that there is a very small space vesicles of the sending neuron. (c) The neurotransmitters fit into
between neurons, known as a synapse (Figure 2.7). The dis- receptor sites that are shaped like the transmitter and cause a
covery of the synapse raised the question of how the electrical voltage change in the receiving neuron.
2.1 Electrical Signals in Neurons 25

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 0 40
Level of
Charge inside fiber

depolarization
needed to

Charge inside fiber

trigger an
action potential 0

–70

Depolarization
(a) (Excitatory)
–70

Depolarization
0 (b) (Excitatory)
Level to
trigger an
Charge inside fiber

action potential

–70

Hyperpolarization
(c) (Inhibitory)

Figure 2.8 (a) Excitatory transmitters cause depolarization, an increased positive charge inside the neuron. (b) When the level
of depolarization reaches threshold, indicated by the dashed line, an action potential is triggered. (c) Inhibitory transmitters cause
hyperpolarization, an increased negative charge inside the axon.

signals generated by one neuron are transmitted across the Figure 2.8a shows this effect. Notice, however, that this re-
space separating the neurons. As we will see, the answer lies in sponse is much smaller than the depolarization that happens
a remarkable chemical process that involves molecules called during an action potential. To generate an action potential,
neurotransmitters. enough excitation must occur to increase depolarization to the
Early in the 1900s, it was discovered that when action level indicated by the dashed line.
potentials reach the end of a neuron, they trigger the re- How does the receiving neuron get enough excitation to
lease of chemicals called neurotransmitters that are stored reach this level? The answer is that it might take more than
in structures called synaptic vesicles at the end of the sending one excitatory response, such as what occurs when multiple
neuron (Figure 2.7b). The neurotransmitter molecules flow neurotransmitters from a number of incoming neurons all
into the synapse to small areas on the receiving neuron called reach the receptor sites of the receiving neuron at once. If the
receptor sites that are sensitive to specific neurotransmitters resulting depolarization is large enough, an action potential
(Figure 2.7c). These receptor sites exist in a variety of shapes is triggered (Figure 2.8b). Depolarization is an excitatory re-
that match the shapes of particular neurotransmitter mol- sponse because it causes the charge to change in the direction
ecules. When a neurotransmitter makes contact with a re- that triggers an action potential.
ceptor site matching its shape, it activates the receptor site An inhibitory response occurs when the inside of the neu-
and triggers a voltage change in the receiving neuron. A neu- ron becomes more negative, or hyperpolarized. Figure 2.8c
rotransmitter is like a key that fits a specific lock. It has an shows this effect. Hyperpolarization is an inhibitory response
effect on the receiving neuron only when its shape matches because it causes the charge inside the axon to move away
that of the receptor site. from the level of depolarization, indicated by the dashed line,
Thus, when an electrical signal reaches the synapse, it trig- needed to generate an action potential.
gers a chemical process that causes a new electrical signal in the We can summarize this description of the effects of ex-
receiving neuron. The nature of this signal depends on both citation and inhibition as follows: Excitation increases the
the type of transmitter that is released and the nature of the chances that a neuron will generate action potentials and is
receptor sites in the receiving neuron. Two types of responses associated with increasing rates of nerve firing. Inhibition
can occur at these receptor sites, excitatory and inhibitory. An decreases the chances that a neuron will generate action po-
excitatory response occurs when the neuron becomes depolar- tentials and is associated with lowering rates of nerve firing.
ized, and thus the inside of the neuron becomes more positive. Since a typical neuron receives both excitation and inhibition,

26 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 the response of the neuron is determined by the interplay
of excitation and inhibition, as illustrated in Figure 2.9. In 2.2 Sensory Coding:
Figure 2.9a, excitation (E) is much stronger than inhibition
(I), so the neuron’s firing rate is high. However, as inhibition How Neurons Represent
becomes stronger and excitation becomes weaker, the neu-
ron’s firing decreases, until in Figure 2.9e, inhibition has Information
eliminated the neuron’s spontaneous activity and has de-
Now that we have an understanding of the basics of neural
creased firing to zero.
functioning, we can go one step further and think about how
Why does inhibition exist? If one of the functions of a
these neural processes underlie perception. How is it that a
neuron is to transmit its information to other neurons, what
neuron “represents” information, like the taste of the salt in
would be the point of decreasing or eliminating firing in the
your stew? Is there a “salty” neuron in your brain that only fires
next neuron? The answer is that the function of neurons is
in response to salt, and so causes you to perceive “saltiness”?
not only to transmit information but also to process it, and,
Or does the pattern of firing of a group of neurons in one brain
as we will see in Chapter 3, both excitation and inhibition are
area, or perhaps many brain areas, result in our perception of
involved in this processing.
saltiness? The problem of neural representation for the senses
has been called the problem of sensory coding, where the sen-
sory code refers to how neurons represent various characteris-
Excitation stronger
tics of the environment.
E Electrode

(a)
Specificity Coding
I
One way in which neurons can represent sensory informa-
tion is demonstrated in the “salty neuron” example above—
E the idea that one neuron can represent one perceptual
experience, like the taste of salt. This notion of a specialized
(b)
neuron that responds only to one concept or stimulus is
I called specificity coding. An example of specificity coding
from the sense of vision is illustrated in Figure 2.10, which
shows how a number of neurons respond to three different
E
faces (the actual firing rates don’t matter; they’re made up for
(c) the sake of this example). Only neuron #4 responds to Bill’s
face, only #9 responds to Mary’s face, and only #6 responds
I
to Raphael’s face. Also note that the neuron specialized to
respond only to Bill, which we can call a “Bill neuron,” does
E not respond to Mary or Raphael. In addition, other faces or
types of objects would not affect this neuron. It fires only in
(d)
response to Bill’s face.
I This idea that one neuron can represent one stimulus or
concept, such as a face, dates back to the 1960s (Konorski,
1967; see also Barlow, 1972; Gross, 2002). At that time, Je-
E rome Lettvin proposed the somewhat tongue-in-cheek idea
(e) that neurons could be so specific that there could be one
neuron in your brain that fires only in response to, say, your
I grandmother. This highly specific type of neuron, dubbed
by Lettvin as a grandmother cell, would respond to your
Inhibition stronger Stimulus on Stimulus off
grandmother “… whether animate or stuffed, seen from be-
Figure 2.9 Effect of excitatory (E) and inhibitory (I) input on the fore or behind, upside down or on a diagonal or offered by
firing rate of a neuron. The amount of excitatory and inhibitory input caricature, photograph or abstraction” (Lettvin, as quoted in
to the neuron is indicated by the size of the arrows at the synapse. Gross, 2002).
The responses recorded by the electrode are indicated by the
According to Lettvin, even just thinking about the idea of
records on the right. The firing that occurs before the stimulus is
your grandmother, not just the visual input, could make your
presented is spontaneous activity. In (a), the neuron receives only
excitatory transmitter, which causes the neuron to fire. In (b) to
grandmother cell fire. Along this reasoning, you would also
(e), the amount of excitatory transmitter decreases while the have a “grandmother cell” for every face, stimulus, and concept
amount of inhibitory transmitter increases. As inhibition becomes that you’ve ever encountered—a specific neuron to represent
stronger relative to excitation, firing rate decreases, until eventually your professor, one for your best friend, one for your dog, and
the firing rate becomes zero. so on. Perhaps you even have grandmother cells that respond

2.2 Sensory Coding: How Neurons Represent Information 27

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Figure 2.10 Specificity coding, in which each face
SPECIFICITY CODING
causes a different neuron to fire. Firing of neuron
4 signals “Bill”; neuron 9 signals “Mary”; neuron
6 signals “Raphael.”
Firing rate

1 2 3 4 5 6 7 8 9 10
(a) Bill Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(b) Mary
Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(c) Raphael Neuron number

to specific information in other senses as well, like one neuron like the Sydney Opera House, and not any other buildings
per song that you know or food that you’ve eaten. Could it be or objects, indicating that these specific cells can be found
that we have such specific representations of stimuli and con- not just for people but for other objects as well.
cepts that we’ve encountered?
Evidence that provided some insight into this question
came from R. Quian Quiroga and colleagues (2005; 2008)
who recorded from the temporal lobe of patients undergo-
ing brain surgery for epilepsy. (Stimulating and recording
from neurons is a common procedure before and during
brain surgery, because it makes it possible to determine the
exact layout of a particular person’s brain.) These patients
were presented with pictures of famous people from different
viewpoints, as well as other things such as other faces, build-
ings, and animals, in order to see how the neurons responded.
Not surprisingly, a number of neurons responded to some of
these stimuli. What was surprising, however, was that some
neurons responded to a number of different views of just one
person or building, or to a number of ways of representing
that person or building.
For example, Figure 2.11 shows a particular neuron
that responded to pictures of the actor Steve Carell and not
to other people’s faces (Quiroga et al., 2008). Neurons were
also found that responded to just the actress Halle Berry, in-
cluding pictures of her from different films, a sketch draw-
ing of her, and even just the words “Halle Berry” (Quiroga Figure 2.11 Records from a neuron in the temporal lobe that
et al., 2005). Thus, these neurons were not just responding responded to different pictures of Steve Carell similar to the ones
shown here (top records) but which did not respond to pictures of
to the visual input of the famous person’s face, but also to
other well-known people (bottom records). (From Quiroga et al., 2008)
the concept of that particular person. Likewise, other neu- Photos: Frederick M. Brown/Getty Images; AP Images/Todd Williamson; Photos 12/Alamy Stock Photo; JStone/
rons were found that responded just to certain buildings, [Link]; Young Nova/[Link]; s_bukley/[Link]

28 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Figure 2.12 Sparse coding, in which each face’s
SPARSE CODING
identity is indicated by the pattern of firing of a
small number of neurons. Thus, the pattern created
by neurons 2, 3, 4, and 7 signals “Bill”; the pattern

Firing rate
created by 4, 6, and 7 signals “Mary”; the pattern
created by 1, 2, and 4 signals “Raphael.”

1 2 3 4 5 6 7 8 9 10
(a) Bill Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(b) Mary
Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(c) Raphael Neuron number

At this point, you might be thinking that Quiroga and silent. Going back to our example of using faces as stimuli, as
coworkers’ study provides evidence for grandmother cells. After shown in Figure 2.12a, sparse coding would represent Bill’s
all, these neurons were responding to highly specific stimuli! face by the pattern of firing of a few neurons (neurons 2, 3, 4,
While this finding does seem consistent with the idea of grand- and 7). Mary’s face would be signaled by the pattern of firing of
mother cells, it does not prove that they exist. The researchers a few different neurons (neurons 4, 6, and 7; Figure 2.12b), but
themselves even say that their study doesn’t necessarily sup- possibly with some overlap with the neurons representing Bill,
port the notion of grandmother cells. In fact, Quiroga and col- and Raphael’s face would have yet another pattern (neurons 1,
leagues point out that they had only 30 minutes to record from 2, and 4; Figure 2.12c). Notice that a particular neuron can
these neurons, and that if more time were available, it is likely respond to more than one stimulus. For example, neuron #4
that they would have found other faces, places, or objects that responds to all three faces, although most strongly to Mary’s.
would have caused these neurons to fire. In other words, the There is evidence that the code for representing objects in
“Steve Carell neuron” might actually have been responsive to the visual system, tones in the auditory system, and odors in
other faces or objects as well, had more options been tested. the olfactory system may involve a pattern of activity across a
In fact, the idea of grandmother cells is not typically relatively small number of neurons, as sparse coding suggests
accepted by neuroscientists today, given the lack of confirma- (Olshausen & Field, 2004).
tory evidence and its biological implausibility. Do we really have
one neuron to represent every single concept we’ve encoun-
tered? It’s unlikely, given how many neurons would be required. Population Coding
An alternative to the idea of specificity coding is that a number While sparse coding proposes that the pattern of firing across
of neurons—rather than just one—are involved in representing a small number of neurons underlies neural representation,
a perceptual experience. population coding proposes that our experiences are repre-
sented by the pattern of firing across a large number of neu-
rons. According to this idea, Bill’s face might be represented
Sparse Coding by the pattern of firing shown in Figure 2.13a, Mary’s face
In their 2008 article, Quiroga and coworkers proposed that by a different pattern (Figure 2.13b), and Raphael’s face by
sparse coding, rather than specificity coding, was more likely to another pattern (Figure 2.13c). An advantage of population
underlie their results. Sparse coding occurs when a particular coding is that a large number of stimuli can be represented,
stimulus is represented by a pattern of firing of only a small because large groups of neurons can create a huge number of
group of neurons, with the majority of neurons remaining different patterns. As we will see in upcoming chapters, there

2.2 Sensory Coding: How Neurons Represent Information 29

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Figure 2.13 Population coding, in which the face’s
POPULATION CODING
identity is indicated by the pattern of firing of a large
number of neurons.!

Firing rate

1 2 3 4 5 6 7 8 9 10
(a) Bill Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(b) Mary
Neuron number
Firing rate

1 2 3 4 5 6 7 8 9 10
(c) Raphael Neuron number

is good evidence for population coding in each of the senses,

and for other cognitive functions as well. 2.3 Zooming Out:
Returning to the question about how neural firing can
represent perception, we can state that part of the answer is Representation in the Brain
that perceptual experiences—such as the experience of the
So far in this chapter, we’ve been focusing mainly on how
aroma of cooking or the appearance of the objects on the table
neural firing “represents” information that’s out there in the
in front of you—are represented by the pattern of firing of
world, such as a face. But as we’ll see throughout this book as
groups of neurons. Sometimes the groups are small (sparse
we explore each of the senses, perception goes beyond individ-
coding), sometimes large (population coding).
ual neurons or even groups of neurons. To get a more complete
picture of the physiology of perception, we must zoom out
TEST YOURSELF 2.1 from neurons and consider representation in the brain more
broadly, including different brain areas and the connections
1. Describe the basic structure of a neuron. between those areas.
2. Describe how to record electrical signals from a neuron.
3. What are some of the basic properties of action potentials?
4. Describe what happens when an action potential travels Mapping Function to Structure
along an axon. In your description, indicate how the How do perceptual functions, like perceiving faces, map onto
charge inside the fiber changes, and how that is related the structure of the brain? The general question of how dif-
to the flow of chemicals across the cell membrane. ferent functions map onto different brain areas can be dated
5. How are electrical signals transmitted from one neuron all the way back to the 18th century with German physiolo-
to another? Be sure you understand the difference gist Franz Joseph Gall and his colleague Johann Spurzheim.
between excitatory and inhibitory responses. Using prison inmates and mental hospital patients as his par-
6. What is a grandmother cell? Describe Quiroga and col- ticipants, Gall claimed to observe a correlation between the
leagues’ experiments on recordings from neurons in shape of a person’s skull and their abilities and traits, which
patients undergoing surgery for epilepsy. he called “mental faculties.” Based on his observations, Gall
7. What is the sensory code? Describe specificity, sparse, concluded that there were about 35 different mental faculties
and population coding. Which types of coding are most that could be mapped onto different brain areas based on the
likely to operate in sensory systems? bumps and contours on the person’s skull, as in Figure 2.14—
an approach that Spurzheim called phrenology. For instance,

30 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Wernicke’s
area

Library of Congress, Prints & Photographs Division, Reproduction number LC-DIG-pga-07838 (digital file from original item) LC-USZC4-4556
(color film copy transparency) LC-USZCN4-195 (color film copy neg.) LC-USZ62-2550 (b&w film copy neg.)
Broca’s
area

Figure 2.15 Broca’s and Wernicke’s areas. Broca’s area is in the

frontal lobe, and Wernicke’s area is in the temporal lobe.

Broca’s and Wernicke’s areas provided early evidence

for modularity, and since Broca’s and Wernicke’s pioneering
research, there have been many other studies relating the loca-
tion of brain damage to specific effects on behavior—a field
now known as neuropsychology. We will describe more exam-
ples of neuropsychological case studies throughout this book,
including more on Broca and Wernicke in Chapter 14.
While neuropsychology has provided evidence for modu-
Figure 2.14 How different functions map onto the structure of the
larity, studying patients with brain damage is difficult for
head, according to phrenology. numerous reasons, including the fact that the extent of each
patient’s damage can differ greatly. A more controlled way that
modularity has been studied is by recording brain responses
in neurologically normal humans using brain imaging, which
a ridge on the back of your head might mean that you’re a lov- makes it possible to create pictures of the location of the
ing person, while a bump on the side means that you’re good brain’s activity.
at musical perception.
Although phrenology has now been debunked as a method,
it was the first proposal that different functions map onto dif-
ferent areas of the brain—a concept that is still discussed today. METHOD Brain Imaging
The idea that specific brain areas are specialized to respond to In the 1980s, a technique called magnetic resonance imaging
specific types of stimuli or functions is called modularity, with (MRI) made it possible to create images of structures within
each specific area called a module. the brain. Since then, MRI has become a standard technique
Early evidence supporting modularity of function came for detecting tumors and other brain abnormalities. While this
from case studies of humans with brain damage. One such technique is excellent for revealing brain structures, it doesn’t
historical case study was conducted by French physician indicate neural activity. Another technique, functional magnetic
Pierre Paul Broca (1824–1890), who saw a patient with a very resonance imaging (fMRI), has enabled researchers to deter-
specific behavioral deficit: the patient could only speak the mine how various types of cognitions, or functions (hence the
word “tan,” although his speech comprehension and other “functional” part of functional MRI), activate different areas of
cognitive abilities appeared to be intact. Examination of Tan’s the brain.
brain after his death showed that he had a lesion to his left Functional magnetic resonance imaging takes advantage of
frontal lobe (Figure 2.15). Soon thereafter, Broca saw other the fact that blood flow increases in areas of the brain that are
patients with similar defects in speech production who had activated. The measurement of blood flow is based on the fact that
damage to that same area of the brain. Broca therefore con- hemoglobin, which carries oxygen in the blood, contains a ferrous
cluded that this area was the speech production area, and it (iron) molecule and therefore has magnetic properties. If a mag-
came to be known as Broca’s area. Another early researcher, netic field is presented to the brain, the hemoglobin molecules
Carl Wernicke (1848–1905) identified an area in the temporal line up, like tiny magnets. Areas of the brain that are more active
lobe which was involved in understanding speech, and which consume more oxygen, so the hemoglobin molecules lose some
came to be known as Wernicke’s area (Figure 2.15).

2.3 Zooming Out: Representation in the Brain 31

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
of the oxygen they are transporting, which makes them more mag- Frontal
cortex Superior temporal
netic and increases their response to the magnetic field. The fMRI
(FC) sulcus
apparatus, also known as the “scanner,” determines the relative (STS)
activity of various areas of the brain by detecting changes in the
magnetic response of the hemoglobin.
The setup for an fMRI experiment is shown in Figure 2.16a, Occipital
with the person laying down such that their head is in the scanner. cortex
As a person engages in a task, such as listening to certain sounds, (OC)
the activity of the brain is recorded. Importantly, fMRI is limited
in that it can’t record activity from individual neurons. Instead,
Temporal Fusiform gyrus (FG)
what’s being recorded is activity in subdivisions of the brain called lobe (underside of
voxels, which are small cube-shaped areas of the brain about 2 the brain)
or 3 mm on a side. Due to their size, each voxel contains many
Figure 2.17 Location of the superior temporal sulcus (STS) in
neurons. Voxels are not brain structures but are simply small units
the temporal lobe, where the “voice area” of the brain resides
of analysis created by the fMRI scanner. One way to think about according to a modular view of speech perception.
voxels is that they are like the small square pixels that make up the
image on your computer screen, but because the brain is three-
dimensional, voxels are small cubes rather than small squares.
Figure 2.16b shows the results of an fMRI scan. Increases or by Belin and coworkers (2000) which asked whether there is
decreases in brain activity associated with cognitive activity are a brain area that responds specifically when you hear a voice,
indicated by colors, with specific colors indicating the amount of compared to when you hear other sounds. The participants
activation; usually “hotter” colors like red indicate higher activa- reclined in the fMRI scanner and passively listened to vocal
tion, while “cooler” colors like blue indicate lower activation. sounds on some trials and non-vocal sounds, like environmen-
Note how the brain appears pixelated in Figure 2.16b; each of tal sounds, on other trials. The results of the study revealed an
those small units is a voxel! area in the temporal lobe—the superior temporal sulcus (STS),
It is important to realize that these colored areas do not appear (Figure 2.17)—that was activated significantly more in res-
as the brain is being scanned. They are determined by a calculation ponse to vocal sounds than non-vocal sounds. This area was
in which brain activity that occurred during the cognitive task is therefore dubbed the “voice area” of the brain, given its highly
compared to baseline activity (like while the participant is at rest) specialized response.
or a different task. The results of this calculation, which indicate The fact that a specific function—in this case, voice
increases or decreases in activity in specific areas of the brain, are perception—was able to be mapped onto a specific area of the
then converted into colored displays like the one in Figure 2.16b. brain in this fMRI study supports a modular view of represen-
tation. Throughout this book, we will see many more exam-
ples of modularity in the brain in other senses. For instance,
in Chapter 5, we’ll discuss fMRI research suggesting that there
are specific brain areas for perceiving faces versus other objects.
Many researchers have used brain imaging techniques But we’ll also see how representation in the brain often goes
like fMRI in an attempt to map a certain function onto a spe- beyond individual modules. As we will see next, specific per-
cific area of the brain. One example of this from speech per- ceptions are often associated with networks of brain areas that
ception (continuing from our discussion of Broca) is a study are distributed across the cortex.

Figure 2.16 (a) A person in a brain scanner.

(b) fMRI record. Each small square represents
a voxel, and the colors indicate whether brain
activity increased or decreased in each voxel. Red
and yellow indicate increases in brain activity;
blue and green indicate decreases.
Source: From Ishai et al., 2000
Photodisc/Jupiter Images

Percent Activation

–1 0 +1 +2
(a) (b)

32 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Distributed Representation Sensory aspects,
Location
Thinking about the brain in terms of modularity is still dis-
cussed today (Kanwisher, 2010). But starting in the late 1900s,
researchers also began considering how multiple brain ar-
Unpleasantness
eas work together. One such researcher is computer scientist Motivation,
Geoffrey Hinton, who, along with his colleagues James McClel- Temperature
Attention,
land and David Rumelhart, proposed that the brain represents Memory Relevance
information in patterns distributed across the cortex, rather Emotion,
than in one single brain area—a concept known as distributed Avoidance
representation (Hinton et al., 1986). A distributed approach
to representation focuses on the activity in multiple brain
areas and the connections between those areas. Hinton’s life-
long work on distributed representation, as modeled by com-
puter programs, won him a Turing Award (also known as the
“Nobel Prize of Computing”) in 2018, which speaks to the
importance of this view of representation. Figure 2.18 Areas that are involved in the perception of pain. Each
area serves a different aspect of pain perception.
One example of distributed representation is how the
brain responds to pain. When you experience a painful stim-
ulus, like accidentally touching a hot stove, your perception wide area of the cortex (Ishai et al., 1999; 2000). We’ll discuss
involves multiple components. You might simultaneously modular and distributed representation of objects further in
experience the sensory component (“it feels burning hot”), Chapter 5.
an emotional component (“it’s unpleasant”), and a reflexive
motor component (pulling your hand away). These different
aspects of pain activate a number of structures distributed
Connections Between Brain Areas
across the brain (Figure 2.18). Thus, pain presents a good ex- We’ve seen how a given perceptual experience can involve mul-
ample of how a single stimulus can cause widespread activity. tiple brain areas. But what about the connections between
Another example of distributed representation is shown those areas? Recent research has shown that connections
in Figure 2.19. Figure 2.19a shows that the maximum activity between brain areas may be just as important for perception as
for houses, faces, and chairs occurs in separate areas in the cor- the activity in each of those areas alone (Sporns, 2014).
tex. This finding is consistent with the idea that there are areas There are two different approaches to exploring the con-
specialized for specific stimuli. If, however, we look at Figure nections between brain areas. Structural connectivity is the
2.19b, which shows all of the activity for each type of stimulus, “road map” of fibers connecting different areas of the brain.
we see that houses, faces, and chairs also cause activity over a Functional connectivity is the neural activity associated

Houses Faces Chairs

(a) Segregation by category (b) Response magnitude

Maximal Response to: Percent Activation

Bruce Goldstein

Houses Faces
–1 0 +1 +2
Chairs No difference

Figure 2.19 fMRI responses of the human brain to houses, faces, and chairs. (a) Areas activated most
strongly by each type of stimulus; (b) all areas activated by each type of stimulus, showing that each type of
stimulus activates multiple areas. (From Ishai et al., 2000)

2.3 Zooming Out: Representation in the Brain 33

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
with a particular function that is flowing through this
structural network. The distinction between structural and METHOD The Resting State Method of Measuring
functional connectivity is similar to the one we described in Functional Connectivity
“Method: Brain Imaging,” where the structure of the brain is Resting-state functional connectivity is measured as follows:
measured using magnetic resonance imaging (MRI) and the 1. Use task-related fMRI to determine a brain location associated
functioning of the brain is measured by functional magnetic with carrying out a specific task. For example, movement of
resonance imaging (fMRI). the finger causes an fMRI response at the location marked Mo-
Here’s another way to think about structural and func- tor (L) in Figure 2.20a. This location is called the seed location.
tional connectivity. Picture the road network of a large city. On
2. Measure the resting-state fMRI at the seed location. The rest-
one set of roads, cars are streaming toward the shopping area
ing-state fMRI of the seed location, shown in Figure 2.20b,
just outside the city, while on other roads cars are traveling
is called a time-series response because it indicates how the
toward the city’s business and financial district. One group of
response changes over time.
people is using roads to reach places to shop; another group
is using roads to get to work or conduct business. Thus, the 3. Measure the resting-state fMRI at another location, which
road map is analogous to the brain’s structural pathways and is called the test location. The response of the test location
connections, and the different traffic patterns are analogous Somatosensory, which is located in an area of the brain
to the brain’s functional connectivity. Just as different parts responsible for sensing touch, is shown in Figure 2.20c.
of the city’s road network are involved in achieving different 4. Calculate the correlation between the seed and test
goals, so different parts of the brain’s neural network are in- location responses. The correlation is calculated using a
volved in carrying out different cognitive or motor goals. complex mathematical procedure that compares the seed
One way of measuring functional connectivity involves us- and test responses at a large number of places along the
ing fMRI to measure resting state activity of the brain. To un- horizontal time axis. Figure 2.21a shows the response at
derstand what this means, let’s return to the “Brain Imaging” the Somatosensory test location superimposed on the
method on page 31. That method described the fMRI measured seed response. The correspondence between these re-
as a person is engaged in a specific task, such as listening to cer- sponses results in a high correlation, which indicates high
tain sounds. This type of fMRI is called task-related fMRI. It is functional connectivity. Figure 2.21b shows the seed re-
also possible to record fMRI when the brain is not involved in a sponse and the response at another test location. The poor
specific task. This fMRI is called the resting-state fMRI. Resting- match between these two responses results in a low corre-
state fMRI is used to measure functional connectivity, as follows: lation, which indicates poor or no functional connectivity.

Figure 2.20 How functional Motor (L) Motor (R) Somatosensory

connectivity is determined by the
resting-state fMRI method. (a) Left
hemisphere of the brain, showing the
seed location Motor (L) in the left motor
cortex, and a number of test locations,
each indicated by a dot. Test location
Motor (R) is in the right motor cortex on (b) Response at Motor (L) seed location
the other side of the brain from Motor
(L). Test location Somatosensory is in
the somatosensory cortex, which is
involved in perceiving touch. (b) Resting
level fMRI response of the Motor (L)
seed location. (c) Resting level fMRI
response of the Somatosensory test
location. The responses in (b) and (c) are
4 seconds long. (Responses courtesy of Ying-Hui Chou) (a) (c) Response at Somatosensory test location

Figure 2.21 Superimposed seed response

(black) and test-location response (red).
(a) Response of the Somatosensory test
location, which is highly correlated with
the seed response (correlation = 0.86).
(b) Response of another test location, which
is poorly correlated with the seed response
(correlation = 0.04). (Responses courtesy of Yin-Hui Chou) (a) Seed response (black) and test response at (b) Seed response (black) and test response at
Somatosensory test location (red) another test location (red)
Correlation = 0.86 Correlation = 0.04

34 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 Figure 2.22 shows the time-series for the seed location moment-to-moment functional connectivity across a net-
and a number of test locations, and the correlations between work of brain areas. The participants’ task was to detect
the seed and test locations. The test locations Somatosensory a very quiet sound that was only perceptible 50 percent of
and Motor R are highly correlated with the seed response and the time—in other words, it was at the participant’s detec-
so have high functional connectivity with the seed location. tion threshold (see Chapter 1, page 14). The researchers found
This is evidence that these structures are part of a functional that the strength of functional connectivity immediately
network. All of the other locations have low correlations so are before the detection task predicted how likely it was that
not part of the network. the person would hear the sound. So, the person was more
Resting-state functional connectivity is one of the main likely to report hearing the sound when their neural con-
methods for determining functional connectivity, but there nections were stronger. Other research has observed similar
are also other methods. For example, functional connectiv- effects in other senses. For example, a person’s resting-state
ity can be determined by measuring the task-related fMRI at functional connectivity can predict whether or not they
the seed and test locations and determining the correlations will perceive a hot stimulus on their foot as painful (Ploner
between the two responses. et al., 2010).
It is important to note that saying two areas are function- By examining the structural and functional connectivity
ally connected does not necessarily mean that they directly between brain areas in a network, in addition to the activation
communicate by neural pathways. For example, the response in each brain area alone, researchers can get a more compre-
from two areas can be highly correlated because they are both hensive picture of how the brain represents our perceptual
receiving inputs from another area. Functional connectivity experiences.
and structural connectivity are not, therefore, the same thing,
but they are related, so regions with high structural connectiv-
ity often show a high level of functional connectivity (van den SOMETHING TO CONSIDER:
Heuvel & Pol, 2010).
So why does it matter if certain brain areas are function- The Mind–Body Problem
ally connected? What does this really tell us about percep-
tion? One example of how functional connectivity can help The main goal of our discussion so far has been to explore the
us understand perception is that it can be used to predict electrical signals that are the link between the environment
behavior. A recent experiment by Sepideh Sadaghiani and and our perception of the environment. The idea that nerve
coworkers (2015) explored this by using fMRI to look at impulses can represent things in the environment is what is

Seed 0.74 Motor R 0.86 Somatosensory

0.09 –0.13

0.14 0.04

Figure 2.22 Resting-state fMRI responses for the Motor (L) seed, test locations Motor (R), Somatosensory,
and five test locations in other parts of the brain. The numbers indicate correlations between the seed response
and each test-location response. Responses Motor (R) and Somatosensory have been singled out because they
have high correlations, which indicates high functional connectivity with the seed. The other locations have low
correlations so are not functionally connected to the seed location. (Responses courtesy of Yin-Hui Chou)

Something to Consider: The Mind–Body Problem 35

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
behind the following statement, written by Bernita Rabinovitz, Experience
a student in my class.

A human perceives a stimulus (a sound, a taste, etc.).

This is explained by the electrical impulses sent to the
brain. This is so incomprehensible, so amazing. How Correlation
can one electrical impulse be perceived as the taste of
a sour lemon, another impulse as a jumble of brilliant
“Susan’s face” “red”
blues and greens and reds, and still another as bitter,
cold wind? Can our whole complex range of sensations (a) Typical physiological experiment
be explained by just the electrical impulses stimulating
the brain? How can all of these varied and very concrete
sensations—the ranges of perceptions of heat and cold, Experience
colors, sounds, fragrances and tastes—be merely and so
Na+
abstractly explained by differing electrical impulses?

When Bernita asks how hot and cold, colors, sounds, fra-
grances, and tastes can be explained by electrical impulses, she Cause
is asking about the mind–body problem: How do physical
processes like nerve impulses (the body part of the problem) “Susan’s face” “red”
become transformed into the richness of perceptual experience
(the mind part of the problem)? (b) Mind–body problem
As we continue on to discuss vision in the following chap- Figure 2.23 (a) This illustrates the situation for most of the
ters, and the other senses later in this book, we’ll see many exam- physiological experiments we will be describing in this book, which
ples of the connections between electrical signals in the nervous determine correlations between physiological responding such as
system and what we perceive. We will see that when we look out nerve firing and experiences such as perceiving “Susan’s face” or
at a scene, countless neurons are firing—some to the basic fea- “red.” (b) Solving the mind–body problem requires going beyond
tures of a stimulus (discussed in Chapter 4), and others to entire demonstrating correlations to determine how ion flow or nerve
objects, like faces or bodies (discussed in Chapter 5). firing causes the experiences of “Susan’s face” or the color “red.”
You may think that all of these connections between
electrical signals and perception provide a solution to the
mind–body problem. This is not, however, the case, because TEST YOURSELF 2.2
as impressive as these connections are, they are all just corre- 1. What is phrenology, and what insight did it provide into
lations—demonstrations of relationships between neural firing neural representation?
and perception (Figure 2.23a). But the mind–body problem 2. Explain how neuropsychological case studies can sup-
goes beyond asking how physiological responses correlate with port a modular view of neural representation, using
perception. It asks how physiological processes cause our expe- Broca’s research as an example.
rience. Think about what this means. The mind–body prob-
3. Describe the technique of brain imaging. How can fMRI
lem is asking how the flow of sodium and potassium ions
be used to study modularity?
across membranes that creates nerve impulses becomes trans-
4. What is distributed representation? Provide an example
formed into the experience we have when we see a friend’s face
from one of the senses.
or when we experience the color of a red rose (Figure 2.23b).
Just showing that a neuron fires to a face or the color red doesn’t 5. Discuss the difference between structural and functional
answer the question of how the firing creates the experience of connectivity. Which technique might be used if one was
seeing a face or perceiving the color red. interested in studying the neural connections associated
Thus, the physiological research we describe in this book, with a certain task, and why?
although extremely important for understanding the physi- 6. Describe how functional connectivity is determined.
ological mechanisms responsible for perception, does not What is the resting-state method?
provide a solution to the mind–body problem. Researchers 7. How can functional connectivity provide insight into
(Baars, 2001; Crick & Koch, 2003) and philosophers (Block, perception?
2009) may discuss the mind–body problem, but when re- 8. What is the mind–body problem? Why do we say that
searchers step into the laboratory, their efforts are devoted demonstrating connections between nerve firing and a
to doing experiments like the ones we have discussed so far, particular stimulus like a face or a color does not solve
which search for correlations between physiological responses the mind–body problem?
and experience.

36 CHAPTER 2 Basic Principles of Sensory Physiology

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 THINK ABOUT IT
1. Because the long axons of neurons look like electrical wires, 2. We described pain as consisting of multiple components.
and both neurons and electrical wires conduct electricity, it Can you think of ways that other objects or experiences
is tempting to equate the two. Compare the functioning of consist of multiple components? If you can, what does
axons and electrical wires in terms of their structure and the that say about the neural representation of these objects or
nature of the electrical signal they conduct. experiences?

KEY TERMS
Action potential (p. 22) Ions (p. 24) Resting potential (p. 22)
Axon (p. 22) Magnetic resonance imaging (MRI) Resting-state fMRI (p. 34)
Brain imaging (p. 31) (p. 31) Resting-state functional connectivity
Broca’s area (p. 31) Mind–body problem (p. 36) (p. 34)
Cell body (p. 21) Modularity (p. 31) Rising phase of the action potential
Dendrites (p. 22) Module (p. 31) (p. 24)
Depolarization (p. 24) Nerve fiber (p. 22) Seed location (p. 34)
Distributed representation (p. 33) Neurons (p. 21) Sensory coding (p. 27)
Excitatory response (p. 26) Neuropsychology (p. 31) Sparse coding (p. 29)
Falling phase of the action potential (p. 25) Neurotransmitters (p. 26) Specificity coding (p. 27)
Functional connectivity (p. 33) Permeability (p. 24) Spontaneous activity (p. 24)
Functional magnetic resonance imaging Phrenology (p. 30) Structural connectivity (p. 33)
(fMRI) (p. 31) Population coding (p. 29) Synapse (p. 25)
Grandmother cell (p. 27) Propagated response (p. 23) Task-related fMRI (p. 34)
Hyperpolarization (p. 25) Receptor sites (p. 26) Test location (p. 34)
Inhibitory response (p. 26) Refractory period (p. 24) Wernicke’s area (p. 31)

Key Terms 37

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
 One message of this book is that per-
ceptual experience is shaped by prop-
erties of the perceptual system. The
sharp, colorful scene represents per-
ception created by activation of cone
receptors in the retina. The less focused,
grey-scale scene represents perception
created by activation of rod receptors in
the retina.

Bruce Goldstein

Learning Objectives
After studying this chapter, you will be able to …
■ Identify the key structures of the eye and describe how they ■ Describe how lateral inhibition and convergence underlie
work together to focus light on the retina. center-surround antagonism in ganglion cell receptive fields.
■ Explain how light is transduced into an electrical signal. ■ Understand the development of visual acuity over the first
■ Distinguish between the influence of rods and cones on year of life.
perception in both dark and light environments.
■ Use your knowledge of neural processing to explain how
signals travel through the retina.

Copyright 2022 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.