| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Radiological Associates of Sacramento, Sutter Medical Center, Sacramento, Calif (D.J.S.); and Department of Neurology, New York University, New York, NY (O.D.). Received April 21, 2005; final version accepted April 22. Address correspondence to D.J.S., 1500 Expo Parkway, Sacramento, CA 95815 (e-mail: seidenwurmd{at}radiological.com).
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION HISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
|
HISTORICAL BACKGROUND |
|---|
|
TOPINTRODUCTIONHISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
The social and scientific evolution of the understanding of the mind in Western society provides the bases and biases for our current views (7,8). Physical theories date back nearly 2500 years, when Democritus posited that the soul or mind is composed of small round atoms (6). Descartes' dictum "I think, therefore I am" endorsed a unified noncorporeal consciousness (6). The mind was neither physical nor divisible. Hume's (9) argument that "reason is and ought to be the slave of passion" makes the limbic cortex the master of the neocortex and the peripheral the master of the central. It also links human intellect with cerebral function and the minds of animals. Similar debates have surrounded social and moral issues. For Rousseau, the mind was a blank slate (6). People were fundamentally just, fair, reasonable, and cooperative but were debased by civilization, which alienated us from nature. Hobbes, on the other hand, characterized life in the primitive state as "solitary, poor, nasty, brutish, and short" and civilization as, well, civilizing (6).
Theologic and philosophic topics are now within the purview of science (8). Consider morality: Plato and Aristotle emphasized that morality reflects human frailty and that we should avoid bad and cultivate the good (8). Kant focused on the intentions of the agent, not the results (8). Perplexing paradoxical situations occur in which values conflict, leading to more recent interest in the social and emotional aspects of judgment, in contrast to pure reason. Patients with orbitofrontal lesions can develop sociopathic behavior, showing a complete loss of morals and comportment in their behavior, yet their verbal responses reflect an intellectual understanding of proper behavior (6,10). How much of morality is organized in the cortex?
Even truth and beauty have fallen under the gaze of science. Do these concepts refer to the objects being considered or to the subjective state of the observer? Beauty may reside within the beholder or may be a characteristic of the person, artwork, or natural phenomenon in question. Universal concepts of human beauty such as symmetry, certain patterns of facial features, and other physical proportions have been observed and may serve as fitness indicators in mate selection (11). Also, certain artistic and architectural proportions are common to diverse cultures, suggesting that preference for these proportions has a biologic basis.
Truth is another critical element in cognition and behavior in which the relation of the observer to the observed is crucial. We need to know which statements conform to reality, whether or not others are telling the truth when they have reason to lie, and how to recognize the truth ourselves when we have a motivation, even if altruistic, to lie. We are coming to the point where we can ask whether the moral choices involved are determined by the function of our brains. We can test whether dissonant statements are processed differently.
Social science approaches to these questions are severely constrained. Darwin's The Expression of the Emotions in Man and Animals was an early tour de force in linking disparate cultures and species among whom evolutionary homology is strongly supported by the nearly identical expression of primary emotions such as anger and fear. The universal human capacity for language and syntax, mathematic concepts, and self-awareness are similar across cultures and are likely genetically encoded. Anthropologists, sociologists, and historians have sought functions and features that are present in all human cultures and hence would seem to have a biologic basis. They have also sought cultural exceptions, because a single exception can help refute a putatively universal human characteristic, while many positive examples are inconclusive. Travel to isolated locations, problems of cultural interpretation and contamination, and even questions about the veracity of the local subjects, who may exaggerate, have called these sorts of investigations into question (12).
|
RECENT TECHNOLOGIC DEVELOPMENTS |
|---|
|
TOPINTRODUCTIONHISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
Structural computed tomography (CT) and MR imaging greatly simplified the correlation of lesion location and symptom. One no longer needed to harvest and examine brains. This dramatically expanded the subject pool and the range of symptom severity and permitted data pooling from multiple subjects. The temporal relationship between symptom expression and the anatomic image could be more precisely determined, as sensitive techniques enabled identification of functional-structural relations for even transient symptoms. However, neuroanatomic imaging is limited because symptoms associated with multifocal or diffuse lesions defy straightforward anatomic correlation, abnormal areas may not be depicted as abnormal, and the axonal connections rather than the processing centers themselves may be damaged. The greatest limit of structural MR imaging, of course, is that brain activity is invisible.
Earlier functional imaging techniques, such as single photon emission CT and positron emission tomography, provided a wealth of information, but they are limited by the need to inject radioisotopes and by the capability to study only a single state change per subject. Their time course is also much longer than that of functional MR imaging, which can probably resolve phenomena that take place at the scale of seconds. Thus, more complex functions cannot be evaluated well with nuclear medicine techniques. Early functional MR imaging techniques that depended on gadolinium chelate injections were similarly limited.
During the past decade, the development of functional MR imaging technology has allowed the imaging investigation of problems in the humanities and social sciences. For the first time, it is ethically possible to study how neural function may relate to the moral, aesthetic, mathematic, and linguistic concepts that define us. Functional MR imaging technology is widely used in clinical practice to map functional areas of the cortex and to determine their proximity to disease before surgical resection is performed. Fiber tracts that connect eloquent brain structures can also be visualized with new techniques, reducing the risk that vital tracts will be disconnected during resection of a lesion outside the eloquent cortex (13).
Functional MR imaging techniques that enable localization of cerebral functions rely on the biologic principle that local neuronal activity is associated with local blood flow. The physical principle that enables depiction of this biologic phenomenon is blood oxygen level–dependent (BOLD) contrast, which exploits changes in magnetic susceptibility. This provides a powerful correlation of the MR imaging signal with local neurologic activity. Usually performed with imaging techniques available in clinical practice, BOLD imaging can now be performed with many clinically based imaging systems. Because changes in local blood flow are small, the MR signal changes are difficult to detect. For this reason, subtraction and coregistration techniques and statistical analysis of the data are performed to map the functional data onto anatomic images. It is not clear if the signal changes depicted in functional MR imaging experiments relate directly to the brain structures themselves or if they are produced by changes in blood oxygen levels in vascular structures. Almost certainly, the signal is vascular. This "brain versus vein" controversy is of interest, but the temporal and spatial resolution limits of the current techniques render the controversy secondary to the larger theoretic questions (14).
|
EXPERIMENTAL DESIGN AND METHODOLOGIC LIMITATIONS |
|---|
|
TOPINTRODUCTIONHISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
Experimental designs become more challenging and complex as the refinement of the cognitive function being studied becomes increasingly precise. Suppose, for example, that one proposes that a specific method of graphic representation is fundamentally encoded within the genetic or structural makeup of the brain. Examples of this sort might be alphabetic writing, numeric systems, graphic comparisons, and tree diagrams. To demonstrate that these might be fundamentally salient representational systems by using functional MR imaging techniques, one would need to construct an experimental paradigm in which the same information is represented several ways and then compare the cerebral activation of the same individuals subjected to the information presented different ways in random sequence. Presumably, in the positive result, the activation pattern would differ depending on the manner in which the cognitive task was presented. Related cognitive processes such as attention and working memory, which are tapped in most tasks, must also be considered and controlled because changes in novelty or complexity can either mask relevant structures or falsely identify others (15,16).
The investigator must also specify, in advance, the criteria by which cerebral processes are likely to result from, or fit directly into, a basic cerebral mechanism rather than result from simple learning or from a more complex cultural overlay. Such criteria necessarily rely on principles of parsimony. Thus, the processes that are more fundamentally encoded will be associated with the activation of fewer and more compact areas. These principles follow from fundamental neuroanatomic concepts in which myriad features of the environment such as visual, spatial, and tonal data are mapped at the microanatomic level (17). Additional complications ensue when one considers that inhibition is an active cerebral process that is relevant to functional units from small neuronal groups to wide cortical fields. Inhibition may be indistinguishable from activation during functional MR imaging experiments performed by using current methods.
Cognitive functions such as language, mathematics, and navigation are performed by networks that span multiple cortical lobes and subcortical structures. These functions comprise different elements and are encoded in different areas. For language, there are the classic neurologic levels of comprehension, repetition, and naming, but within naming, we now know that human names, names for tools, and action verbs are localized to different areas (18,19). During navigation, the use of spatial relations as opposed to landmarks activates different brain areas (20). Complex functions that are tested clinically or in a cognitive laboratory often employ structures that originally evolved for different tasks. Thus, in addition to the attentional, mnemonic, motivational, and other systems that are often tapped in such experiments, experimental paradigms may activate multiple cognitive functions. We must therefore carefully specify the criteria for basic brain mechanisms.
The technical and theoretic limitations of functional MR imaging methods increase dramatically as one moves from mapping finger movements to mapping morality, mathematics, or social function. Functional MR imaging data may be limited on temporal, spatial, statistical, and acoustic grounds. The time course of the functional MR imaging phenomenon differs from the time course of neuronal activity. The mapping of functional MR imaging signal onto anatomic images can be degraded by patient motion. As is the case with all anatomic structures, there is individual variation in neuroanatomic structures. The large number of imaging voxels and the small magnitude of the BOLD phenomenon requires large numbers of comparisons that must be accounted for in the statistical analysis. Appropriate sample sizes and trial numbers are needed to avoid mistaking chance phenomena for real relationships and to identify true relationships that are near the threshold for detection. The MR imaging instrument is artificially loud and physically constrained. This environment and the level of stress it evokes could alter the data for some subjects.
Although some experiments can be performed repeatedly in the same subjects, those involving memory, for example, can be performed only once, and then only in a particular order. This problem affects many other areas of investigation, particularly if the task in question is designed to produce a novel, counterintuitive, socially discordant, or otherwise embarrassing or dissonant result. Careful attention must be paid to the experimental design to avoid these problems.
Diffusion-tensor imaging and MR tractography illustrate network relations. By defining the fiber tract relationships between different areas that are activated during a motor or cognitive task, direct anatomic relationships can be identified. Hardwired functions are more likely to be those that involve directly interconnected structures. Diffusion-tensor imaging is based on the principle that some structures within the brain demonstrate directional restriction of brownian motion. By mapping directional diffusion, one can precisely localize nerve fiber tracts. An intrinsic brain function is probably performed by cortical and subcortical structures with direct connections. Further confirmation of such linkages could be found in neuroanatomic or neurochemical studies, but these are hampered by methodologic difficulties. With time, diffusion-tensor imaging may help us understand the networks of feedback fibers and inhibitory circuits, which currently vex our understanding. Further progress will be made if direct imaging of fiber connections in multiple subjects helps overcome the problem of individual anatomic variation (21).
Linguistic differences between subjects deserve consideration, and not only in the study of language itself. Native speakers process a language differently than those who learned the language later in life; care must be taken to avoid resulting pitfalls. Furthermore, because we know that distinct cerebral regions process discrete components of language, we must be careful to validate any conclusions drawn from functional MR imaging work in subjects whose primary languages are from different linguistic families. To eliminate the effect of cultural variables, which represent the environmental side of the nature-nurture dichotomy, subjects from different cultural groups will need to be studied. Given the nuances of language and environment, it will be a great challenge to produce the same task in different linguistic and cultural contexts. Multilingual functional MR imaging paradigms should help isolate these factors. Investigations in multilingual and monolingual populations would be expected to yield similar results when the functions are intrinsic to human biology and the paradigm truly measures only the desired question. Even in this case, however, cultural-environmental differences superimposed on innate biologic functions such as language could produce variable findings even for an innate function (22).
|
CONTRIBUTIONS OF FUNCTIONAL MR IMAGING |
|---|
|
TOPINTRODUCTIONHISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
Results of complex functional MR imaging experiments are helping us understand the cerebral mechanisms of truth telling and lying. When subjects deceive, frontal lobe areas are activated (24). Faro et al (25) have ingeniously demonstrated that specific frontal, temporal, and limbic regions are activated when subjects lie. These activation patterns correlate well with polygraph findings in these naive subjects. Even when the deliberate misstatement concerns a fictitious, although extremely engaging, scenario, deception and truth telling can be measured with functional MR imaging. More study is needed to determine if truth itself—for example, in mathematics—is neurologically encoded.
With respect to aesthetics, the picture is becoming clearer as well. There appear to be structural correlates of aesthetic preferences for art. Functional MR imaging experiments have revealed decreased activation of the right caudate nucleus in response to visual stimulation of decreased self-rated preference and activation of bilateral occipital and fusiform gyri and left cingulate sulci in response to increasing preference (26). Results of other studies have demonstrated activation of reward-associated structures in response to faces that are judged attractive, independent of facial expression. Neural correlates of compelling musical features have also been demonstrated. It appears more and more likely that aesthetic preferences may reflect the observer as much as the observed (26,27). The medial orbitofrontal cortex is activated while viewing attractive faces (28). Indeed, when beautiful paintings are viewed, the orbitofrontal cortex is activated in a way that is different from the way it is engaged when ugly paintings are viewed (29,30).
Those of us involved in the use of MR imaging technology on a daily basis have a unique opportunity (and responsibility) to engage this exciting frontier. As a discipline, radiology must not stand on the sidelines, with radiologists relegated to the status of mere technicians. We can become leaders in this exciting field. The most profound advances will likely come from simple, creatively designed, conceptually elegant, intellectual paradigms that are tested by using standard functional MR imaging techniques. We are experiencing a new enlightenment. It is again possible to advance fundamental questions of philosophy and epistemology by using the tools of clinical practice.
More immediately, perhaps, the radiology community should attend to these issues because the power of functional MR imaging has been recognized by those who specialize in neuromarketing and employ functional MR imaging to examine the activation of neuroanatomic sites related to positive mental states—for example, pleasure, reward, social acceptance, and aesthetic preference. Functional MR imaging can be a tool in product development or in marketing, and this may extend into political and cultural domains.
In summary, recent technologic advances have placed within our hands tools that promise to answer some of the most perplexing and long-standing questions we can ask about ourselves. Radiologists work every day with instruments that give us the capability to explore the nature of the human mind, provided we ask questions properly and have the audacity to ask them at all. Not since the Enlightenment have physicians had this potential to advance the fundamental knowledge of human nature through the use of routine clinical techniques.
|
References |
|---|
|
TOPINTRODUCTIONHISTORICAL BACKGROUNDRECENT TECHNOLOGIC DEVELOPMENTSEXPERIMENTAL DESIGN AND...CONTRIBUTIONS OF FUNCTIONAL MR...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
