More than a century has passed since the field of neuroscience was born, but its central questions remain the same: How does the brain control human behavior, and to what extent do different people’s brains function differently?

In 1991, a new technology was introduced that has since begun to demystify the brain: fMRI, or functional magnetic resonance imaging, which allows neuroscience researchers to illustrate the brain’s instant reactions and leaps of logic. You may be familiar with the images created by fMRI technology—depictions of the brain featuring gorgeous neon blobs on an otherwise grayish backdrop. These blobs (and yes, scientists actually call them that) represent areas in the brain that are activated by a given condition, such as seeing a snake, flames, or an angry human face. Today, the results of such fMRI experiments are giving researchers new insight into disorders as varied as overeating, autism, and anxiety.

“These techniques are extraordinarily revolutionary in terms of using functional imaging to give the brain a voice,” says Dr. Joy Hirsch.

During an fMRI study, a subject lies down in the middle of a giant magnetic tube. Once in the tube, they are presented with whatever task or stimulus the researcher has come up with, usually on a screen of some kind. While the subject is occupied, the fMRI machine makes a lot of noise and grabs hundreds of split-second portraits of blood flowing through the brain. This data is then subjected to layers of statistical analyses, the results of which are turned into those spectacular blob pictures.

When it was first introduced 17 years ago, fMRI was greeted with skepticism, but it is now the “technique of choice” according to Dr. Joy Hirsch, director of the fMRI lab at Columbia University’s Neurological Institute. In Hirsch’s lab alone, fMRI is being used to come up with preliminary answers to such questions as: How does the brain influence our ability to lose weight? How can autism be more effectively diagnosed and treated? Can fMRI predict when and if someone will regain consciousness after a traumatic brain injury?

In Hirsch’s fMRI obesity study, she discovered that when a person is in a “depleted” state—on a diet—the accompanying hormonal changes actually cause a shift in the brain regions that respond most strongly to the presence of food. The change is not in the dieter’s favor; the mobilized brain regions control emotional reactivity, which helps to explain how carefully planned, rational goals can disappear at the sight of a donut.

The work Hirsch’s lab is doing with minimally conscious patients serves an entirely different purpose. In many cases, when a patient has fallen into a coma after a brain injury, it is totally unclear if and when he or she will regain consciousness. Hirsch and her colleagues have begun to use fMRI with comatose patients, prompting them with various stimuli and looking at their brain function in order to determine their level of awareness. It’s no replacement for direct communication, but the scans can provide a precious point of contact with an individual otherwise far removed from reality. “These techniques are extraordinarily revolutionary in terms of using functional imaging to give the brain a voice,” Hirsch says.

FMRI has also revolutionized the study of psychiatric conditions, including post-traumatic stress disorder and generalized anxiety disorder. By scanning the brains of people with these clinical diagnoses, scientists have observed that an anxious individual’s brain responds to triggers differently than that of a non-anxious counterpart. In particular, scientists have documented significantly increased activity in these patients’ amygdalas, a brain structure involved in threat detection. The amygdala operates subconsciously, monitoring the world for anything that could be dangerous and drawing up instantaneous plans of defense upon registering the presence of a threat. In the brains of those with post-traumatic stress disorder, the fMRI reveals that the amygdala is more likely to be activated by stimuli that are neutral or merely ambiguous, and to stay active for longer, pumping excessive amounts of stress hormones through the blood stream, which can ultimately wreak havoc on the brain.

Until the advent of fMRI, the options for studying living human brains in such ways were severely constrained. One of the most common methods was EEG, or electroencephalography. EEG involves pasting a hairnet of electrodes onto a subject’s skull; the sensors detect electrical activity caused by neurons firing in the brain. Because they are outside the skull, however, the electrodes can’t tell what’s happening deep inside the brain, where so much of the action unfolds. Another highly regarded method was PET, or positron emission tomography. PET can pinpoint specific areas of the brain that are activated by a given task. The downside: Anyone undergoing a PET scan has radioactive chemicals injected into their bloodstream, which are both expensive and potentially unsafe.

An fMRI, on the other hand, requires no radioactive compounds. And though fMRI research is on the cutting edge, it is actually based on an old-fashioned observation. In 1880, an Italian peasant named Bertino cracked his skull in an accident. The injury was so deep that the frontal lobes of Bertino’s brain were visible—so exposed, in fact, that one could clearly see blood pulsing through the brain matter. Bertino’s physician Mosso saw something else, too. When the town’s church bells rang, the blood flow in Bertino’s frontal lobes suddenly swelled.

In true Italian style, Mosso asked: “Do the bells make you think of a prayer?”

“Yes,” Bertino affirmed.

Mosso had made the first connection between increased brain activity—in this case, a ritualized thought process—and increased blood flow.

The fMRI has done away with the need for cracked-open skulls. Instead, it harnesses the brain’s natural functions. Specificically, when any part of the brain is spurred into a state of increased activity, it needs oxygen, which is promptly delivered via the bloodstream. During an fMRI experiment, as the active brain neurons grab up that oxygen, they leave behind deoxygenated blood, which has a magnetic force. What the fMRI machine detects is the shifting proportions of oxygenated blood in any given region of the brain. This shift is considered the direct result of a shift in brain activity.

At every stage of this process, there are shortcomings, sources of distortions, and errors that no one has yet figured out how to correct. This year, a controversial paper originally titled “Voodoo Correlations in Social Neuroscience” claimed that the statistical methods used in many fMRI studies artificially inflated results. The questions the paper raises have yet to be resolved.

Still, many, if not most neuroscientists, are cautiously optimistic about the fMRI's potential to dramatically elevate our understanding of the brain. They emphasize, though, that we are not there yet.

Peter Freed, an assistant professor of psychiatry at Columbia University, compares the current state of fMRI research to the invention of the microscope 400 years ago. The possibilities of the microscope were amazing, but only when it had been tailored and tweaked could the scope see everything there was to be seen.

As it is improved and expanded, fMRI research may one day reveal an equally astonishing view. There are no limits on how much we can learn about ourselves by studying the way our brain behaves. Or as Freed puts it, “It is, in the end, brain function that determines what we experience—both joy and suffering.”

Casey Schwartz is a graduate of Brown University and has a master's degree in psychodynamic neuroscience from University College London. She has previously written for The New York Sun and ABC News. She's working on a book about the brain world.