In Search of Memory: The Emergence of a New Science of Mind (11 page)

Freud called the period he spent studying nerve cells in simple organisms like crayfish, eels, and primitive fish “the happiest hours of my student life.” He left those basic research studies after he met and fell in love with Martha Bernays, whom he later married. In the nineteenth century, one needed an independent income in order to take on a career in research. In view of his poor financial position, Freud turned instead to the establishment of a medical practice that would earn him sufficient income to support a wife and family. Perhaps if a scientific career could have ensured a living wage then, as it does today, Freud would be known as a neuroanatomist and a co-founder of the neuron doctrine, instead of as the father of psychoanalysis.

H
ad I become a practicing psychoanalyst, I would have spent much of my life listening to patients talk about themselves—about their dreams and waking memories, their conflicts and their desires. This is the introspective method of “talk therapy” that Freud pioneered to arrive at deeper levels of self-understanding. By encouraging the free association of thoughts and memories, the psychoanalyst helps patients unpack the unconscious memories, traumas, and impulses that underlie their conscious thoughts and behavior.

In Grundfest’s laboratory I soon appreciated that to understand how the brain functions, I would have to learn how to listen to neurons, to interpret the electrical signals that underlie all mental life. Electrical signaling represents the language of mind, the means whereby nerve cells, the building blocks of the brain, communicate with one another over great distances. Listening in on those conversations and recording neuronal activity was, so to speak, objective introspection.

GRUNDFEST WAS A LEADER IN THE BIOLOGY OF SIGNALING
. From him I learned that thinking about the signaling function of nerve cells has proceeded in four distinct phases, reaching from the eighteenth century to a particularly clear and satisfying resolution in the work of Alan Hodgkin and Andrew Huxley two hundred years later. Throughout, the question of how nerve cells communicate has attracted some of the best brains in science.

The first phase dates to 1791, when Luigi Galvani, a biologist from Bologna, Italy, discovered electrical activity in animals. Galvani left a frog’s leg hanging on a copper hook from his iron balcony and found that the interaction of the two dissimilar metals, copper and iron, would occasionally cause the leg to twitch, as if it were animated. Galvani could also cause a frog’s leg to twitch by stimulating it with a pulse of electricity. After further study, he proposed that nerve cells and muscle cells are themselves capable of generating a flow of electrical current and that the twitch of muscles is caused by the electricity generated by muscle cells—not by spirits or “vital forces,” as was commonly believed at the time.

Galvani’s insight and his achievement in bringing nervous activity out of the realm of vital forces and into natural science was elaborated in the nineteenth century by Hermann von Helmholtz, one of the first scientists to bring the rigorous methods of physics to bear on a range of problems in brain science. Helmholtz found that the axons of nerve cells generate electricity not as a by-product of their activity, but as a means of producing messages that are carried along their whole length. These messages are then used to carry sensory information about the outside world into the spinal cord and the brain and to transmit commands for action from the brain and spinal cord to the muscles.

In the course of this work, Helmholtz made an extraordinary experimental measurement that changed thinking about electrical activity in animals. In 1859 he succeeded in capturing the speed at which these electrical messages are conducted and found to his amazement that electricity conducted along a living axon is fundamentally different from the flow of electricity in a copper wire. In a metal wire, an electrical signal is conducted at close to the speed of light (approximately 186,000 miles per second). Despite its speed, however, the strength of the signal deteriorates badly over long distances because it is propagated passively. If an axon relied on passive propagation, a signal from a nerve ending in the skin of your big toe would die out before it reached your brain. Helmholtz found that the axons of nerve cells conduct electricity much more slowly than wires do, and they do so by means of a novel, wavelike action that propagates actively at various speeds up to approximately 90 feet per second! Later studies showed that the electrical signals in nerves, unlike signals in wires, do not decrease in strength as they propagate. Thus, nerves sacrifice speed of conduction for active propagation, which ensures that a signal that arises in your big toe arrives at your spinal cord undiminished in size.

5–1
Edgar, Lord Adrian (1889–1977) developed methods of recording action potentials, the electrical signals nerve cells use for communication. (Reprinted from
Essentials of Neural Science and Behavior
, Kandel, Schwartz, and Jessell, McGraw-Hill, 1995.)

Helmholtz’s findings raised a set of questions that would occupy physiology for the next hundred years: What do these propagated signals, later called action potentials, look like and how do they encode information? How can biological tissue generate electrical signals? Specifically, what carries the current for the signals?

THE FORM OF THE SIGNAL AND ITS ROLE IN ENCODING
INFORMATION were addressed in the second phase, which began in the 1920s with Edgar Douglas Adrian’s work. Adrian (figure 5–1) developed methods of recording and amplifying the action potentials propagated along the axons of individual sensory neurons on the skin, thereby making the elementary utterances of nerve cells intelligible for the first time. In the process, he made several remarkable discoveries about the action potential and how it leads to what we perceive as a sensation.

To record action potentials, Adrian used a thin piece of metal wire. He placed one end of the wire on the outside surface of the axon of a sensory neuron on the skin and then ran the wire to both an ink oscillograph (so he could look at the shape and pattern made by the action potentials) and a loudspeaker (so he could hear them). Every time Adrian touched the skin, one or more action potentials were generated. Each time an action potential was generated, he heard a brief bang! bang! bang! over the loudspeaker and saw a brief electrical pulse on the ink writer. The action potential in the sensory neuron lasted only about 1/1000 of a second and had two components: a swift upstroke to a peak, followed by an almost equally rapid downstroke that returned it to the starting point (figure 5–2).

The oscillograph training and the loudspeaker both told Adrian the same remarkable story: all of the action potentials generated by a single nerve cell are pretty much the same. They are about the same shape and amplitude, regardless of the strength, duration, or location of the stimulus that elicits them. The action potential is thus a constant, all-or-none signal: once the threshold for generating the signal is reached, it is almost always the same, never smaller or larger. The current produced by the action potential is sufficient to excite adjacent regions of the axon, thus causing the action potential to be propagated without failure or flagging along the whole length of the axon at speeds of up to 100 feet per second, pretty much as Helmholtz had earlier found!

The discovery of the all-or-none characteristic of the action potential raised more questions in Adrian’s mind: How does a sensory neuron report the intensity of a stimulus—whether a touch is light or heavy, whether a light is bright or dim? How does it signal the duration of the stimulus? More broadly, how do neurons differentiate one type of sensory information from another, such as touch from pain, light, smell, or sound? How do they differentiate sensory information for perception from motor information for action?

5–2
Edgar Adrian’s recordings revealed the characteristics of the action potential. Recordings in single nerve cells showed that action potentials are all-or-none: once the threshold for generating an action potential is reached, the signal is always the same, both in amplitude and shape.

Adrian first addressed the question of intensity. In a landmark finding, he discovered that intensity results from the frequency with which action potentials are emitted. A mild stimulus, such as a gentle touch on the arm, will elicit just two or three action potentials per second, whereas a strong one, such as a pinch or bumping one’s elbow, could fire a hundred action potentials per second. Similarly, the duration of a sensation is determined by the length of time over which the action potentials are generated.

Next, he explored how information is conveyed. Do neurons use different electrical codes to tell the brain that they are carrying information about different stimuli, such as pain or light or sound? Adrian found that they did not. There was very little difference among the action potentials produced by neurons in the various sensory systems. Thus the nature and quality of a sensation—whether visual or tactile, for instance—does not depend upon differences in action potentials.

What, then, accounts for the differences in information carried by neurons? In a word, anatomy. In a clear confirmation of Cajal’s principle of connection specificity, Adrian found that the nature of the information conveyed depends on the type of nerve fibers that are activated and the specific brain systems to which those nerve fibers are connected. Each class of sensation is transmitted along specific neural pathways, and the particular kind of information relayed by a neuron depends on the pathway of which it is a part. In a sensory pathway, information is transmitted from the first sensory neuron—a receptor that responds to an environmental stimulus such as touch, pain, or light—to specific and specialized neurons in the spinal cord or in the brain. Thus visual information is different from auditory information because it activates different pathways.

In 1928 Adrian summarized his work in his characteristically vivid style: “all impulses are very much alike, whether the message is destined to arouse the sensation of light, of touch, or of pain; if they are crowded together, the sensation is intense; if they are separated by any interval, the sensation is correspondingly feeble.”

Finally, Adrian found that signals sent from motor neurons in the brain to the muscles are virtually identical to signals conveyed by sensory neurons from the skin to the brain: “the motor fibers transmit discharges which are almost an exact counterpart of those in the sensory fibers. The impulses…obey the same all-or-nothing principle.” Thus, a rapid train of action potentials down a particular neural pathway causes a movement of our hands rather than a perception of colored lights because that pathway is connected to our fingertips, not to our retinas.

Adrian, like Sherrington, extended Cajal’s neuron doctrine, which was based on anatomical observations, into the realm of function. But unlike Golgi and Cajal, who were locked in a bitter rivalry, Sherrington and Adrian were friends who lent each other support. For their discoveries regarding the function of neurons they shared the Nobel Prize in Physiology or Medicine in 1932. On hearing that he would share the prize with Sherrington, Adrian, who was a generation younger, wrote him:

Adrian had listened in on the bang! bang! bang! of neuronal signaling and discovered that the frequency of these electrical impulses represents the intensity of a sensory stimulus, but several questions remained. What lies beneath the nervous system’s remarkable ability to conduct electricity in this all-or-none manner? How are electrical signals turned on and off, and what mechanism is responsible for their rapid propagation along an axon?

THE THIRD PHASE IN THE HISTORY OF SIGNALING CONCERNS THE
mechanisms underlying the action potential and begins with the membrane hypothesis, first proposed in 1902 by Julius Bernstein, a student of Helmholtz and one of the most creative and accomplished electrophysiologists of the nineteenth century. Bernstein wanted to know: What mechanisms give rise to these all-or-none impulses? What carries the charge for the action potential?

Bernstein understood that the axon is surrounded by the cell surface membrane and that even in the resting state, in the absence of any neural activity, there exists a steady potential, or difference in voltage, across this membrane. He knew that the difference, now called the resting membrane potential, was of great importance for nerve cells because all signaling is based on changes in this, resting potential. He determined that the difference across the membrane is about 70 millivolts, with the inside of the cell having a greater negative charge than the outside.