New technologies shed light on schizophrenia

October 25, 2004

Diffusion tensor imaging, a type of MRI, relates the size of keybrain-fiber bundles to different neuropsychological signs

Washington -- Researchers at the Boston Veterans Affairs Health Care System - Brockton Division, Harvard Medical School, and the University of Massachusetts-Boston are using new imaging technology to gather valuable information about the brains of people with schizophrenia. This new variety of magnetic resonance imaging (MRI) is called diffusion tensor imaging (DTI). Using DTI on patients with schizophrenia, neuropsychologists have related smaller sizes in two distinct webs of brain fibers to two distinct types of cognitive malfunction.

The findings appear in the October issue of Neuropsychology, which is published by the American Psychological Association (APA).

Diffusion tensor imaging (DTI) uses a regular MRI machine to analyze the movement of water molecules in and around the fibers that connect different parts of the brain. Neuroscientists use DTI to track indicators of brain "connectivity" - factors such as the number, thickness, density and arrangement of axons (the hair-like extensions of neurons, which send messages to other neurons) and thickness of the insulating/conducting fatty myelin sheath in which they are embedded. If weaker structural integrity reduces connectivity, lead author Paul Nestor, PhD, says it may mean that, "different brain areas do not communicate as well - with less synchrony or harmony, akin to an orchestra or band playing out of synch."

The researchers conducted neuropsychological tests on 41 patients with schizophrenia and 46 healthy controls, and used DTI scans on a 14-person subset of people with schizophrenia and healthy controls, a sample size typical of seminal studies of the human brain and comparable to early studies using functional MRI.

Brain images from the schizophrenic patients showed abnormalities in two functionally and anatomically different neural pathways - the uncinate fasciculus (UF) and the cingulate bundle (CB). Compared with age-matched controls, patients had smaller UF and CB. These bunches of axons are wrapped in myelin sheaths and bundled like electrical wire. The UF connects different parts of the frontal and temporal lobes and the CB connects parts of the prefrontal-cingulate regions. Each of these fiber tracts may help to define distinct neural networks. "We presume that the health of these fibers reflects the degree to which different parts of the brain are able to communicate," says Nestor.

Patients with significantly smaller left-side UF bundles than those of age-matched controls had worse declarative-episodic verbal memory; they found it significantly harder than did healthy controls to describe prior events in words. Schizophrenic patients with smaller left-side CB tracts were not as good as aged-matched healthy controls in adjusting their performance based on monitoring as they went along -- a type of error that hurts "executive functioning." Disturbed information processing lies at the heart of schizophrenia and sometimes predates the onset of illness.

The high-tech scans support the concept of schizophrenia as a serious cognitive disorder whose diverse symptoms rest in diverse anatomical differences which, consistent with the current findings, are based in neural fiber networks. Symptoms may result not from faults in single "components" but rather from differences in webs of neural connections. For example, a smaller UF bundle might not do as well in orchestrating important signals related to verbal processes, such as using words to organize thoughts.

These neural networks are known to contribute to a wide range of cognitive functions. Write the authors, "Correlations [between fiber-tract size and cognitive problems] underscore the widely distributed nature of higher cognition in the brain and caution against drawing simple isomorphic relationships between function and anatomy." Clearly, healthy brain function depends on networking as a whole, not only discrete parts acting alone.

The authors write, "[Our] findings may [help us to] develop neuropsychological models of schizophrenia that point to a core disturbance in the synchronization and integration of local brain networks and wider scale neural systems."

American Psychological Association