Professional Documents
Culture Documents
Raichle ME, Sci Amer 270, 1994
Raichle ME, Sci Amer 270, 1994
by Marcus E. Raichle
W
hat causes the pity we might velopments in imaging technology that posits that networks of neurons resid-
feel for the melancholy Dane the past few years have seen, most no- ing in strictly localized areas perform
in Hamlet or the chill during a tably positron-emission tomography thought processes. So just as speciÞc
perusal of the Raven? Our brains have and magnetic resonance imaging. Cou- members of a large orchestra perform
absorbed from our senses a printed se- pled with powerful computers, these together in a precise fashion to pro-
quence of letters and then converted techniques can now capture in real time duce a symphony, a group of localized
them into vivid mental experiences and images of the physiology associated brain areas performing elementary op-
potent emotions. The Òblack boxÓ de- with thought processes. They show how erations work together to exhibit an
scription of the brain, however, fails to speciÞc regions of the brain Òlight upÓ observable human behavior. The foun-
pinpoint the speciÞc neural processes when activities such as reading are per- dation for such analyses is that com-
responsible for such mental actions. formed and how neurons and their elab- plex behaviors can be broken down into
While philosophers have for centuries orate cast of supporting cells organize a set of constituent mental operations.
pondered this relation between mind and coordinate their tasks. The map- In order to read, for example, one must
and brain, investigators have only re- ping of thought can also act as a tool for recognize that a string of letters is a
cently been able to explore the connec- neurosurgery and elucidate the neural word; then recognize the meaning of
tion analyticallyÑto peer inside the diÝerences of people crippled by dev- words, phrases or sentences; and Þnal-
black box. The ability stems from de- astating mental illnesses, including de- ly create mental images.
pression and schizophrenia. The challenge, of course, is to deter-
I hasten to point out that mine those parts of the brain that are
the underlying assumptions active and those that are dormant dur-
6 5 4 3 2 1 of current brain mapping are ing the performance of tasks. In the
distinct from those held by past, cognitive neuroscientists have re-
early phrenologists. They pos- lied on studies of laboratory animals
ited that single areas of the and patients with localized brain in-
brain, often identiÞed by juries to gain insight into the brainÕs
bumps on the skull, uniquely functions. Imaging techniques, howev-
represented speciÞc thought er, permit us to visualize safely the
processes and emotions. In anatomy and the function of the nor-
contrast, modern thinking mal human brain.
1 RIGHT LEFT 2 3
T
ages themselves. Without the availabil- is determined by the radioactive com- hroughout the late 1970s and
ity of sophisticated computers, the pound injected. A radioactively labeled early 1980s, researchers rapidly
task would have been impossible to form of glucose, for example, measures developed PET to measure vari-
accomplish. brain metabolism because glucose is the ous activities in the brain, such as glu-
X-ray CT had two consequences. primary source of energy for neurons. cose metabolism, oxygen consumption,
First, it changed forever the practice of Louis SokoloÝ of the National Institute blood ßow and interactions with drugs.
medicine because it was much superior of Mental Health introduced this now Of these variables, blood ßow has
to standard x-rays. For the Þrst time, widely used autoradiographic method proved the most reliable indicator of
investigators could safely and eÝective- in 1977. moment-to-moment brain function.
ly view living human tissue such as the Investigators adept with tissue au- The idea that local blood ßow is inti-
brain with no discomfort to the patient. toradiography became fascinated when mately related to brain function is a sur-
Standard x-rays revealed only bone and CT was introduced. They suddenly real- prisingly old one. English physiologists
some surrounding soft tissue. Second, ized that if they could reconstruct the Charles S. Roy and Charles S. Sherring-
it immediately stimulated scientists and anatomy of an organ by passing an x- ton formally presented the idea in a
engineers to consider alternative ways ray beam through it, they could also re- publication in 1890. They suggested
of creating images of the bodyÕs interior construct the distribution of a previous- that an Òautomatic mechanismÓ regu-
using similar mathematical and comput- ly administered radioisotope. One had lated the blood supply to the brain. The
er strategies for image reconstruction. simply to measure the emission of ra- amount of blood depended on local var-
One of the Þrst such groups to be in- dioactivity from the body section. With iations in activity. Although subsequent
trigued by the possibilities opened by this realization was born the idea of au- experiments have amply conÞrmed the
computed tomography consisted of ex- toradiography of living human subjects. existence of such an automatic mecha-
perts in tissue autoradiography, a meth- A crucial element in the evolution of nism, no one as yet is entirely certain
od used for many years in animal stud- human autoradiography was the choice about its exact nature. It obviously re-
ies to investigate organ metabolism and of radioisotope. Workers in the Þeld se- mains a challenging area for research.
4 5 6
PET SCANS show active neural areas. In the far left column nouns (a ), the primary visual cortex lights up. When nouns
the left side of the brain is presented; the next columns show are heard (b ), the temporal lobes take command. Spoken
Þve horizontal layers (the right side faces to the right, with nouns minus viewed or heard nouns (c ) reveal motor areas
the front to the top). Each row corresponds to the diÝerence used for speech. Generating verbs (d ) requires additional
between a speciÞc task and the control state of gazing at a neural zones, including those in the left frontal and temporal
dot on a television monitor. When subjects passively view lobes corresponding roughly to BrocaÕs and WernickeÕs areas.
O
necessary. ne of the Þrst assignments in were asked to Þx their gaze on a pair
The fast decay and small amounts which PET blood-ßow mapping of small crosshairsÑthe arrangement
permit many measurements of blood has proved useful is in the study looks like a small plus signÑin the
ßow to be made in a single experiment. of language. The manner in which lan- middle of a television monitor. At the
In this way, PET can take multiple pic- guage skills are acquired and organized same time, a PET scan measured blood
tures of the brain at work. Each picture in the human brain has been the subject ßow in the brain, providing a snapshot
serves as a snapshot capturing the mo- of intense investigation for more than of mental activity.
mentary activity within the brain. Typi- a century. Work began in earnest in In the second level, subjects contin-
cal PET systems can locate changes 1861, when French physician Pierre Paul ued to maintain their gaze on the cross-
in activity with an accuracy of a few Broca described a patient whose dam- hairs as blood ßow was measured, but
millimeters. aged left frontal lobe destroyed the abil- during this scan they were exposed to
A distinct strategy for the functional ity to speak. ( To this day, patients who common English nouns. The nouns ei-
mapping of neuronal activity by PET has have frontal lobe damage and have trou- ther appeared below the crosshairs on
emerged during the past 10 years. This ble speaking are often referred to as the television monitor or were spoken
approach extends an idea Þrst intro- having BrocaÕs aphasia.) BrocaÕs studies through earphones (separate scans were
duced to psychology in 1868 by Dutch of language localization were comple- performed for visual and auditory pre-
physiologist Franciscus C. Donders. mented by Carl Wernicke, a German sentations). In the third level, subjects
Donders proposed a general method to neurologist. In 1874 Wernicke told of were asked to recite the word they
measure thought processes based on a people who had diÛculty comprehend- viewed or heard. Finally, in the fourth
simple logic. He subtracted the time ing language. They harbored damage to level, the subjects said out loud a verb
needed to respond to a light (with, say, the left temporal lobe, a region now usu- appropriate for the noun.
the press of a key) from the time need- ally referred to as WernickeÕs area. From Subtracting the Þrst level from the
ed to respond to a particular color of these beginnings has emerged a con- second isolated those brain areas con-
light. He found that discriminating color cept of language organization in the cerned with visual and auditory word
required about 50 milliseconds. In this human brain: information ßows from perception. Deducting the second level
way, Donders isolated and measured a visual and auditory reception to areas from the third pinpointed those parts
mental process for the Þrst time. in the left temporal lobe for compre- of the brain concerned with speech pro-
The current PET strategy is designed hension and then on to frontal areas duction. Subtracting level three from
to accomplish a similar subtraction but for speech production [see ÒSpecializa- level four located those regions con-
in terms of the brain areas implement- tions of the Human Brain,Ó by Norman cerned with selecting the appropriate
ing the mental process. In particular, im- Geschwind; SCIENTIFIC AMERICAN, Sep- verb to a presented noun.
ages of blood ßow taken before a task tember 1979 ]. The Þnal subtraction (speaking nouns
is begun are compared with those ob- All this information was gleaned from minus generating verbs) was of partic-
tained when the brain is engaged in that brain-damaged patients. Can investiga- ular interest, because it provided a por-
task. Investigators refer to these two pe- tors derive insight about language orga- trait of pure mental activity (perception
riods as the control state and the task nization from a healthy brain? In 1988 and speechÑor input and outputÑhav-
state. Workers carefully choose each my colleagues Steven E. Petersen, Mi- ing been subtracted away). This image
state so as to isolate as best as possible chael I. Posner, Peter T. Fox and Mark permitted us to view what occurs in our
a limited number of mental operations. A. Mintun and I at the Washington Uni- brains as we interpret the meaning of
Subtracting blood-ßow measurements versity Medical Center began a series of words and, in turn, express meaning
made in the control state from each studies to answer just this question. The through their use. It renders visible con-
task state indicates those parts of the initial study was based on a PET analy- scious function because much of our
brain active during a particular task. sis of a seemingly simple job: speaking thinking is carried out by concepts and
To achieve reliable data, workers take an appropriate verb when presented ideas represented by words.
the average of responses across many with a common English noun. For ex- The results of this study clearly dem-
individual subjects or of many experi- ample, a subject might see or hear the onstrate how brain imaging can relate
A
s cognitive neuroscientists dem-
onstrated the utility of PET tech-
nology, a newer method swiftly
emerged that could compete with PETÕs
INDIVIDUAL DIFFERENCE IMAGES
abilities. Magnetic resonance imaging,
or MRI, has now become a fairly com-
mon tool for diagnosing tissue damage.
Recent developments have vastly in-
creased the speed with which MRI can
form images, thus making it suitable
for research in cognitive neuroscience.
MRI derives from a potent laboratory
MEAN DIFFERENCE IMAGE technique known as nuclear magnetic
resonance ( NMR ), which was designed
IMAGE SUBTRACTION AND AVERAGING serve as the foundation of functional to explore detailed chemical features of
brain imaging. Researchers subtract the PET blood-ßow pattern of a control state molecules. It garnered a Nobel Prize for
from that of a task state to produce a diÝerence image (top row ). Data from diÝer- its developers, Felix Bloch of Stanford
ent subjects are averaged (bottom two rows ) to eliminate statistical ßuctuations. University and Edward M. Purcell of
T
he ability of MRI systems to mon- lustration above ], accurate localization control and emotion. Clearly, we are
itor the oxygen signal in real time of the source of brain activity remains headed toward a much richer grasp of
has suggested to some the possi- diÛcult with electrical recording de- the relation between the human mind
bility of measuring the time it takes for vices. Furthermore, the resolution be- and the brain.
diÝerent brain areas to exchange infor- comes poorer the deeper into the brain
mation. Conceptually, one might think we attempt to image.
of a network of brain areas as a group Neither MRI nor PET suÝers from
of individuals in the midst of a confer- this diÛculty. They both can sample all FURTHER READING
ence call. The temporal information parts of the brain with equal spatial INTRINSIC SIGNAL CHANGES ACCOMPANY-
sought would be equivalent to knowing resolution and sensitivity. As a result, a ING SENSORY STIMULATION: FUNCTION-
who was speaking when and, possibly, collaboration seems to be in the mak- AL BRAIN MAPPING WITH MAGNETIC RES-
ONANCE IMAGING. S. Ogawa, D. W. Tank,
who was in charge. Such information ing between PET and MRI and electrical
R. Menon, J. M. Ellermann, S.-G. Kim,
would be critical in understanding how recording. PET and MRI, working in a H. Merkle and K. Ugurbil in Proceedings
speciÞc brain areas coordinate as a net- combination yet to be determined, can of the National Academy of Sciences,
work to produce behavior. deÞne the anatomy of the circuits un- Vol. 89, No. 13, pages 5951Ð5955; July
The stumbling block, however, is the derlying a behavior of interest; electri- 1, 1992.
speed of neuronal activity compared cal recording techniques can reveal the SOMATOSENSORY CORTICAL PLASTICITY
with the rate of change of oxygenation course of temporal events in these spa- IN ADULT HUMANS REVEALED BY MAG-
NETOENCEPHALOGRAPHY. A. Mogilner et
levels. Signals from one part of the brain tially deÞned circuits.
al. in Proceedings of the National
can travel to another in 0.01 second or Regardless of the particular mix of Academy of Sciences, Vol. 90, No. 8,
less. Unfortunately, changes in blood technologies that will ultimately be used pages 3593Ð3597; April 15, 1993.
ßow and blood oxygenation are much to image human brain function, the IMAGES OF MIND. M. I. Posner and M. E.
slower, occurring hundreds of millisec- Þeld demands extraordinary resources. Raichle. W. H. Freeman and Company,
onds to several seconds later. MRI Expensive equipment dominates this 1994.
would not be able to keep up with the work. MRI, PET and MEG equipment PRACTICE-RELATED CHANGES IN HUMAN
ÒconversationsÓ between brain areas. costs from $2 million to $4 million and FUNCTIONAL ANATOMY DURING NON-
MOTOR LEARNING. M. E. Raichle, J. A.
The only methods that respond quickly is expensive to maintain. Furthermore,
Fiez, T. O. Videen, A.-M. K. MacLeod,
enough are electrical recording tech- success requires close collaboration J. V. Pardo, P. T. Fox and S. E. Petersen
niques. Such approaches include elec- within multidisciplinary teams of sci- in Cerebral Cortex, Vol. 4, No. 1, pages
troencephalography ( EEG ), which de- entists and engineers working daily 8Ð26; January/February 1994.
tects brain electrical activity from the with these tools. Institutions fortunate