Imaging Provides Insight, Leaves Room for More


by Guy McKhann, M.D.

May 1, 2007

“Build a better mousetrap and the world will beat a path to your door.”

—Ralph Waldo Emerson

Two articles in the April edition of Brain in the News show how far we have come—and where we may go next.

Some background is helpful. Early advances in imaging allowed us to analyze the structure of the brain. Not too many years ago, we all oohed and aahed over the structural outlines of the normal and abnormal brain we saw first with computerized tomography imaging, then with magnetic resonance imaging (MRI).

In more recent times, the outlining of fiber pathways by diffusion tensor imaging and detection of new areas of vascular damage in the brain by diffusion weighted imaging have been added to our toolkit for looking at brain structure. Currently we are evaluating, by both MRI and positron emission tomography (PET), whether we can diagnose and follow diseases in which abnormal products accumulate in the brain, such as the accumulation of amyloid in Alzheimer’s disease.

Looking only at the structure of the brain is like looking at a computer that hasn’t been turned on. You may get a general picture, but you don’t know what is happening. Functional imaging of the brain has changed that—we can now ask what parts of the brain we are using as we carry out various acts.

This type of imaging, whether by MRI or PET, is based on the fact that as groups of nerve cells are being used, there is a change in the local blood flow to these neurons. It is detection of these blood flow changes that is the basis for the signal.

Why would I want this information? It has several uses. If a patient is going to have brain surgery, I want to know where various functions such as language or body movement reside in that person’s brain. The surgeon can then stay away from those areas, if possible. If a patient has epilepsy, this type of imaging may tell me where the seizures start and how they spread. Finally, if the person’s brain is processing information differently, as occurs in dyslexia or autism, functional imaging may tell me what the basis is for these differences.

These are all clinical examples, but the most remarkable effect of functional imaging has been to bridge the gaps between neuroscience and psychology, particularly cognitive psychology. The article “Hearts and Minds” by Jonah Lehrer in the Boston Globe is representative of this new, symbiotic relationship.

Before the advances in imaging, clinical neuroscientists had little to offer those studying human cognitive psychology. To be sure, correlations were made between human studies of the living and ultimate findings during autopsy, or sophisticated physiological studies in nonhuman primates indicated how the brain was organized and how information was processed. But psychologists were asking questions at one level, neuroscientists at another.

Many people went into the neurosciences and psychology because they were interested in global questions about how the brain worked, such as the interactions of emotion and behavior, the basis of creativity, and consciousness. They soon backed away from these pursuits because they were too hard to study. We just didn’t have the tools to study the interaction of higher aspects of human brain functions. Now, with functional imaging, we may.

However, before we succumb to this “irrational exuberance” (to borrow an expression from another field), we should ask if current imaging techniques are telling us what we really want to know about how the brain processes information. My answer to that is “no”—current techniques are too slow.

Because we are imaging changes in blood flow, we are limited by the dynamics of this change in relation to the speed of neuronal processing: blood flow responses are 100 times slower than actual neuronal speeds. A second problem is that the brain works as a series of circuits. There are circuits for the control of movement, the production of language, or the recognition of a word, for example. Our current imaging techniques are limited in their ability to analyze these circuits.

The challenge is to develop new techniques that are actually linked, in real time, to neuronal activities in different, but related, parts of the brain. Help may be at hand. Alison Abbott of News@nature.com (“Proteins Make Light Work of Nerve Control,” April 5, payment required) describes the work of a psychiatrist at Stanford, Karl Deisseroth, who has engineered light-sensitive proteins capable of turning nerve cells “on” or “off” in real time.

These light-activated proteins, as tools to manipulate neuronal functions, may be the next “better mousetrap” in brain sciences.