Brain Imaging Provides A Peek Into The Human Psyche

During the past 150 years, the study of brain function has enlisted an increasing number of scientific disciplines in its service, beginning with anatomy and continuing with physiology, psychology, pharmacology, and molecular biology. Within the past 15 years, brain imaging has emerged as a tool that can integrate findings from these diverse disciplines through its intimate access to the working brain. Positron-emission tomography (PET) and functional magnetic resonance imaging (fMRI) are the imaging technologies applied most often by neuroscience researchers. International Society for Magnetic Resonance in Medicine 2118 Molvia St. Berkeley, Calif. 94704 (510) 841-1899 Fax: (510) 841-2340 http://www.ismrm.org 5,300 members President: Gary Glover Journals: Journal of Magnetic Resonance Imaging, Magnetic Resonance in Medicine Organization for Human Brain Mapping Inc. P.O. Box 425464 Cambridge, Mass. 02142-0009 E-mail: hbm96@nmr.mgh.harvard.edu http://mbb.harvard.edu/HBM 700 members Chairman: Bernard Mazoyer Society of Nuclear Medicine 1850 Samuel Morse Dr. Reston, Va. 20190 (703) 708-9000 Fax: (703) 708-9015 http://www.snm.org 16,000 members President: William Strauss Journals: Journal of Nuclear Medicine, Journal of Nuclear Medicine Technology Though much is still unknown about the workings of the central nervous system, brain imaging has confirmed and extended earlier observations about major neurological and psychiatric conditions. It has also been applied to practically any cognitive task that could be structured into a controlled experiment. For example, fMRI was used to compare brain activation when subjects read kanji (ideogramic) vs. kana (phonogramic) Japanese characters (T. Makabe et al., Nippon Rinsho, 55:1699-705, 1997). And in a study of spatial memory, London taxi drivers underwent PET scans as they recalled complex routes around the city (E.A. Maguire et al., Journal of Neuroscience, 17:7103-10, 1997). Because PET and fMRI are new technologies, many brain-imaging studies have focused on methodological questions. Some of the most intriguing imaging experiments, however, have explored issues in psychology and psychiatry, disciplines that often resist a hard-science approach. A BOON FOR RESEARCH: With fMRI technology available, the number of learning and memory imaging experiments has ballooned, declares Susan Bookheimer, a UCLA psychiatry professor. The hippocampus has long been deemed crucial to learning and memory, but early PET studies failed to show activation of this brain region, according to Susan Bookheimer, a psychiatry professor at the University of California, Los Angeles, School of Medicine. "Designing a control task for memory" was a problem, she explains. "You can't tell the brain to stop remembering." PET researchers eventually had subjects perform increasingly difficult memory tasks and noted a corresponding increase in hippocampal activation (P.M. Grasby et al., Neuroscience Letters, 163:185-8, 1993). The laboratory of Marcus Raichle, a professor of radiology and neurology at Washington University School of Medicine in St. Louis, became especially adept at using PET to explore memory. But it is only since the advent of fMRI, with its high spatial resolution and wide availability, that the number of learning and memory imaging experiments has ballooned, Bookheimer says. In a study of motor-skill learning. Leslie Ungerleider, chief of the Laboratory of Brain and Cognition at the National Institute of Mental Health (NIMH), scanned six subjects with fMRI before and after they had practiced a finger-opposition task. By the fourth week of practice, the area of motor cortex activated by the task had expanded (A. Karni et al., Nature, 377:155-8, 1995). But all forms of prior experience don't have the same effect. Studies of "priming"--in which a subject must complete words after being presented with their initial letters--have found that less neural activity is needed to process words shown earlier (L.R. Squire et al., Proceedings of the National Academy of Sciences, 89:1837-41, 1992). According to Ungerleider, recent work in other laboratories suggests that the changes she observed in the primary motor cortex may occur earlier in other brain regions. In scans encompassing wider swaths of the brain, she is tracking training-induced changes in cerebellum, striatum, and premotor cortex. CRITICAL VIEW: Harvard's Daniel Schacter has used both PET and fMRI technology, but he suggests that conclusions about memory drawn from brain imaging should be viewed with caution. Two studies by Daniel Schacter, chairman of Harvard University's psychology department, suggest that conclusions about memory drawn from brain imaging should be viewed with caution. Schacter compared brain activation when subjects recognized printed words to be words they had heard several minutes earlier (veridical memory) with activation when subjects falsely recognized printed words as words they had not, in fact, heard earlier (illusory memory). The strongest finding of the earlier study, which involved PET, was that veridical, but not illusory, memories activated the left temporoparietal (TP) cortex, a region that other research has linked to retention of auditory information (D.L. Schacter et al., Neuron, 17:267-74, 1996). Several lawyers soon contacted Schacter about using PET in cases that relied on the accuracy of memories. "I didn't think that this would be useful in court," he recalls. (PET scans, however, have long been introduced in civil and criminal cases to show brain damage.) His doubts were confirmed by an fMRI study in which neither veridical nor illusory memory activated the TP cortex (D.L. Schacter et al., NeuroImage, 6:259-69, 1997). Since there were some procedural differences from the PET experiment, Schacter wrote that the effects of these types of memory on the brain "may vary across testing formats." He now says he is "much more impressed with the similarities than the differences" in activation and plans to explore these similarities further. PET, unlike fMRI, can monitor cerebral metabolism and track the binding of drugs to receptors in the brain. Researchers typically measure metabolism by injecting subjects with the glucose analog fluorodeoxyglucose labeled with fluorine-18. They monitor receptor binding by administering drugs labeled with carbon-11. Steven Paul, vice president of research and development at Eli Lilly and Co. in Indianapolis, says PET played a role in the development of such medications as Prozac and the anti-psychotic olanzapine (Zyprexa). PET has also been applied in studies of the effects of illegal drugs such as cocaine. In the last decade, Nora Volkow, director of nuclear medicine at Brookhaven National Laboratory in Upton, N.Y., has examined how cocaine abuse affects metabolism, dopamine transporters, and dopamine D2 receptors in the brain. Another receptor linked to cocaine abuse is the µ opioid receptor. Using PET and a synthetic µ ligand, a group headed by J. James Frost, a radiology professor at Johns Hopkins University School of Medicine, imaged addicts who had recently stopped taking the drug (J.-K. Zubieta et al., Nature Medicine, 2:1225-9, 1996). Addicts showed higher ligand binding than did control subjects, meaning that more µ receptors were available in their brains. Furthermore, right after withdrawal, addicts' greater craving for cocaine correlated with greater µ receptor availability. Frost says this linkage of craving and a chemical change in the brain was unprecedented. In a still-unpublished study, Frost tracked former addicts for 90 days and found that those whose available µ receptors had fallen to normal levels stayed off cocaine for at least several weeks. Those whose µ availability had remained stubbornly high, however, relapsed within days. "This [method] could be useful in the future for identifying subjects who are more likely to relapse," Frost observes. He plans similar experiments on the * opioid receptor, which is also associated with cocaine addiction, and is seeking funding to image opioid receptors in chronic alcoholics. TALKING THERAPY: UCLA's Jeffrey Schwartz, who studied the impact of a "talking" treatment during a PET study, says that effective cognitive-behavior therapy can alter brain metabolism without drug treatments. Like addiction, obsessive-compulsive disorder (OCD) has been linked to brain abnormalities. PET scans indicate that OCD patients exhibit increased glucose metabolism in the orbitofrontal cortex, caudate nucleus, and anterior cingulate gyrus, according to Jeffrey Schwartz, a psychiatry professor at UCLA. Serotonin reuptake inhibitors, such as Prozac, that reduce OCD symptoms also reduce this abnormally high metabolism. In what Schwartz considers a unique PET study on the physiological impact of a "talking" treatment, he and his colleagues at UCLA showed that cognitive-behavior therapy, when it was effective, altered brain metabolism, much as medication had (J.M. Schwartz et al., Archives of General Psychiatry, 53:109-13, 1996). "People can change how their own brain works without using drugs," enthuses Schwartz, who adds that panic disorder and depression are other psychiatric conditions ripe for this kind of experiment. David Rosenberg, a psychiatry professor at Wayne State University School of Medicine in Detroit, is planning a PET study on how behavior therapy might affect abnormal rates of synthesis of the neurotransmitter serotonin in OCD patients. Brain imaging has afforded the first intimate scientific access to two activities so central to human life and culture that they have long been the subject of hosts of untestable hypotheses: reading and sleeping. WORDS ON THE BRAIN: Bennett Shaywitz and colleagues at Yale have successfully used fMRI to study how different people use different parts of their brains during reading, comprehension, and speaking tasks. Four years ago, a group at Yale University decided to use fMRI to study how readers recognize words, sound them out, and extract meaning. Bennett Shaywitz, a pediatrics and neurology professor at Yale, recalls, "What we observed that really jumped out at us--and it jumps out from the cover of Nature--is that if you group the 19 men and 19 women and look at patterns during the phonology task, you see these dramatic differences." All 19 male subjects activated only the left sides of their brains, while 11 of the 19 females activated both the left and right sides (B.A. Shaywitz et al., Nature, 373:607-9, 1995). Shaywitz notes that it was at the urging of his wife, Sally Shaywitz, a member of the research team, that women were included in the experiments. In further studies, the Yale team has found that "the degree of [brain] lateralization predicts the reading style," he says. "If you're left-lateralized, you tend to read in a global way. You take larger chunks. If you're a bilateral reader, you read in a more bit-by-bit way." RESEARCHING DYSLEXIA: Studies using fMRI imaging have advanced the understanding of dyslexia, and Georgetown's Guinevere Eden is looking further into how abnormalities in parts of the brain contribute to reading problems. The complexity of the reading process has been highlighted by research on dyslexics, problem readers with well-established deficits in phonological awareness. It is not intuitively obvious that their brains would process visual motion differently, but that is what Guinevere Eden, then a postdoctoral fellow at NIMH, and her colleagues found, using fMRI (G.F. Eden et al., Nature, 382:66-9, 1996). When they presented moving dots to dyslexic subjects, an area of visual cortex known as V5/MT did not activate as much as it did in nondyslexics. Eden, now an assistant professor in the neurology department at Georgetown University Medical Center, says, "The visual hypothesis has been controversial for a long time, but I think people welcomed the paper because they find it comforting to know that there's a biological basis for dyslexia." Recent fMRI findings from other groups confirm that dyslexia involves visual as well as language areas of the cortex. Eden is now focusing on how activation abnormalities in the cerebellum and parietal cortex could contribute to reading problems. Sleep research has also generated its share of controversial and unexpected findings. A recent PET study of sleep presumed that the rapid-eye-movement (REM) phase associated with dreaming would activate the area of visual cortex that is active when awake subjects imagine a visual scene (A.R. Braun et al., Science, 279:91-5, 1998). REM not only failed to have that effect but also shut down the prefrontal cortex, a finding that punctured Freudian theory, according to Allen Braun, acting chief of the language section of the National Institute on Deafness and Other Communication Disorders (NIDCD) and lead author of the paper. UNDOING FREUD: A recent PET study of sleep failed to validate a Freudian theory about the REM phase and dreams, says Allen Braun of NIDCD. "Freud said that dream content has to be monitored by a sensor that screens wishes that are emerging from the unconscious, sees that they're unacceptable, and then critically disguises them in acceptable but very obscure dream symbols," explains Braun. "If you had to pick any single region of the brain that subserves functions of self-monitoring and censorship, that would be the pre-frontal cortex," which was inactive during REM sleep. When Braun expressed that view in the New York Times (N. Wade, Jan. 6, 1998, page F6), critical letters inevitably followed (H. Shevrin, P.W. Mosher, and A. Adams-Silva, New York Times, Jan. 12, 1998, page A20). Braun is now hoping to image sleeping subjects using fMRI. He is not concerned about the loud noise produced by MRI scanners. "People sleep in them all the time," he observes. "I think there's something soothing about them." He is more worried about fMRI-associated artifacts. "People [using fMRI] are still primarily reporting changes in the cortex," Braun says. "Reliable subcortical signal--thalamus, basal ganglia--is still a potential problem." Douglas Steinberg is a freelance science writer based in New York.

Brain Imaging Provides A Peek Into The Human Psyche

Interested in reading more?

Become a Member of