DEAN F. SALISBURY, PHD
- Harvard title(s): Associate Professor, Department of Psychiatry, Harvard Medical School
- McLean title(s): Director, Cognitive Neuroscience Laboratory, McLean Hospital
- Email: dean_salisbury@hms.harvard.edu, dsalisbury@mclean.harvard.edu
- Telephone: (617) 855-3786
- Fax: (617) 855-3795
- Office Address: Cognitive Neuroscience Laboratory, NBG21, 115 Mill St., Belmont, MA 02478
- Degree(s):
- 1990 Ph.D.State University of New York, Stony Brook, New York (Biological Psychology)
- 1989 M.A.State University of New York, Stony Brook, New York (Biological Psychology)
- 1985 B.A.Whittier College, Whittier, California
- Fellowship:
- 1990 - 1992 Fellow, Clinical Research Training Program, Department of Psychiatry, Harvard Medical School
- Bio:
My research utilizes electrophysiology and multimodal brain imaging. I specialize in event-related potential macropotential recordings in humans and have collaborated over the last decade with researchers using MR and fMR imaging methods. Models of normal brain function need to be constrained by an awareness of differences that arise when the brain is compromised. To that end, I have examined thought disturbance (cognitive-level) and basic auditory processing abnormalities (sensory-level) in psychosis and wellcontrols.
Thought disorder is a cardinal symptom of schizophrenia, and is inferred through abnormal language. Disordered thought has a high genetic penetrance, and is detected in relatives of schizophrenia patients ata high rate, even when these relatives do not exhibit overt psychosis. Thus, examination of thinking problemsmay be especially insightful for understanding schizophrenia. The broad aim of my work is to combine behavioral measures and brain activity measure to further understand the underlying interplay between semantic memory neural networks subserving concept storage and verbal working memory systems that allow adaptive and flexible human behavior in the face of unique current situations. More specifically, the aims are to understand automatic semantic activation versus controlled verbal working memory inhibition and to relate these dysfunctions to underlying brain systems, so as to understand the neurophysiology of psychosis. Automatic memory processes and related brain activity during language processing tasks using lexical ambiguities (homographs like toast, box, or panel) are contrasted with controlled contextual inhibitory processes and related brain activity. In addition to homograph processing and expectancy-based priming, we measure high frequency brain activity (gamma waves) to lexical access to examine semantic memory neural network structure across the brain. Memories are stored in distributed networks, and gamma activity allows the distributed brain areas to work in concert. We also examine the effects of recent memory traces and recent response outcome traces on current behavior, to assess how memory traces benefit or detract from current performance, especially in patients. The studies will help determine where in the information processing stream most blame can be laid for thought disorder, early in the automatic semantic memory system or later in the controlled working memory system. These results, in turn, when related to underlying brain systems will help to clarify the nature of thought disorder and cognitive dysfunction in schizophrenia with respect to actual brain structure and function.
Another program of research looks at basic auditory sensory processing in first-episode psychotic patients and chronically-ill psychotic patients to determine where abnormalities occur in patients in basic auditory processing relative to controls, and where patients with schizophrenia differ from patients with mania and affective psychosis. In addition, MR measures allow us to determine whether the neural substrates of these brain processes are also affected in the diseases, and to determine whether we can find brain-behavior relationships between the size of the electrical brain responses we record and the structural measures derived from MR. We also can determine the degree to which such brain-behavior relationships are present at first psychotic episode and how they develop over time with disease duration through our cross-sectional measures of first-episode patients and chronic patients, and through longitudinal assessments of first episode patients. Our data to date suggest that patients first hospitalized with schizophrenia show an abnormality that is lateralized to the auditory processing centers of the left temporal lobe in the P300 event-related brain potential, a brainwave associated with goal-directed stimulus detection. This P300 voltage abnormality is correlated with the size of abnormal cortical gray matter reductions in the left posterior superior temporal gyrus, a polymodal, complex tertiary area of temporal-parietal association cortex. Interestingly, patients first hospitalized with psychotic mania do not show this pattern of brain structural-functional abnormalities. Furthermore, patients with first-episode schizophrenia do not show abnormalities in an earlier brain potential termed mismatch negativity that reflects the early sensory memory automatic detection of stimulus deviance. However, the size of the mismatch negativity voltage is correlated with the size of Heschl's gyrus gray matter, which contains primary auditory cortex in patients at first schizophrenic episode, suggesting that mismatch negativity may be sensitive to the presence of an initial mild reduction of auditory cortical neuropil. Most importantly, our longitudinal data shows that mismatch negativity, which is in the normal range at first hospitalization, becomes progressively abnormal in patients along with the duration of their disease. This suggests that mismatch negativity may serve as a index of progressive disease pathological processes in the first few years of schizophrenia.
- Curriculum vitæ: (PDF format)
- Publications: Link
