Functional Magnetic Resonance Imaging During
Auditory Verbal Working Memory in Nonpsychotic
Relatives of Persons with Schizophrenia: A Pilot Study
Heidi W. Thermenos, Larry J. Seidman, Hans Breiter, Jill M. Goldstein, Julie M. Goodman,
Russell Poldrack, Stephen V. Faraone, and Ming T. Tsuang
Background: First-degree relatives of persons with schizophrenia carry elevated genetic risk for the illness and show deficits on
high-load information processing tasks. We used functional magnetic resonance imaging (fMRI) to test whether nonpsychotic relatives
show altered functional activation in the prefrontal cortex (PFC), thalamus, hippocampus, and anterior cingulate during a working
memory task requiring interference resolution.
Methods: Twelve nonpsychotic relatives of persons with schizophrenia and 12 healthy control subjects were administered an auditory,
verbal working memory version of the Continuous Performance Test during fMRI. An asymmetric, spin-echo, T2*-weighted sequence
(15 contiguous, 7-mm axial slices) was acquired on a full-body MR scanner. Data were analyzed by Statistical Parametric Mapping
(SPM).
Results: Compared with control subjects, relatives showed greater task-elicited activation in the PFC and the anterior and dorsomedial
thalamus. When task performance was controlled, relatives showed significantly greater activation in the anterior cingulate. When
effects of other potentially confounding variables were controlled, relatives generally showed significantly greater activation in the
dorsomedial thalamus and anterior cingulate.
Conclusions: This pilot study suggests that relatives of persons with schizophrenia have subtle differences in brain function in the
absence of psychosis. These differences add to the growing literature identifying neurobiological vulnerabilities to schizophrenia.
Key Words: Schizophrenia, genetics, functional magnetic reso-
nance imaging, working memory, interference, cingulate
S
chizophrenia is thought to have a complex mode of
inheritance because of its high prevalence and sporadic
familial transmission (Freedman et al 2001b; Tsuang et al
1999). A majority of genetic studies have relied on the diagnosis
of schizophrenia as the target phenotype, but robust susceptibil-
ity genes have yet to be identified. Alternatively, genetic studies
might benefit from the identification of neurobiological pheno-
types derived from information processing or brain imaging
abnormalities that characterize schizophrenia (Faraone et al
2001; Seidman and Wencel 2003). Nonpsychotic, first-degree
relatives of persons with schizophrenia, who on average share
50% of their genes with ill relatives, are advantageous to study
because they are free of psychosis and its many associated
confounds (e.g., antipsychotic medications).
Structural brain imaging studies of nonpsychotic relatives
demonstrate a number of significant volume differences similar
to those seen in schizophrenia (Seidman et al 2004). In contrast
to the widespread brain abnormalities observed in schizophrenia
(Harrison 1999; McCarley et al 1999), nonpsychotic relatives
appear to have differences in a smaller number of brain regions
(Seidman 1997; Seidman et al 2004). To date, relatives most
consistently demonstrate volumetric alterations in the medial
temporal lobe and the thalamus (Keshavan et al 1997; Lawrie et
al 1999; Seidman et al 1999, 2002a, 2004; Staal et al 1998). In
relatives, hippocampal volume appears to be meaningfully re-
lated to verbal declarative memory deficits (O’Driscoll et al 2001;
Seidman et al 2002a).
While the structural imaging studies highlight medial tempo-
ral lobe differences, neurocognitive and electrophysiologic mea-
sures also point to possible dysfunction of the prefrontal cortex
(PFC) and associated regions, including the thalamus and hip-
pocampus. Nonpsychotic relatives show impairment on tests of
sustained attention, working memory (WM), verbal declarative
memory, sensorimotor gating, and saccadic inhibition (Caden-
head et al 2000; Cornblatt and Keilp 1994; Freedman et al 2001a;
Kremen et al 1994; Nuechterlein et al 1994; Park et al 1995), tasks
that are associated with hippocampal, thalamic, and prefrontal
cortical functions (Aggleton et al 1995; Agster et al 2002; Cirillo
and Seidman 2003; Goto et al 2002). Performance of relatives on
these tasks appears to be especially sensitive to conditions of
high interference (Seidman et al, unpublished data), which is
important because of the presumed roles of both the hippocam-
pus and thalamus in sensorimotor gating and modulation of
interference.
Only a few functional neuroimaging studies have assessed
whether nonpsychotic relatives manifest functional differences in
the PFC, thalamus, or hippocampus. A recent functional mag-
netic resonance imaging (fMRI) study of relatives performing the
N-back WM task revealed exaggerated activity in the right
dorsolateral PFC (DLPFC; Callicottt et al 2003). A positron
emission tomography study of obligate-carrier relatives perform-
ing a verbal fluency task showed less functional connectivity
between the left PFC and the precuneus, and activation in the
From the Department of Psychiatry (HWT, LJS, JiMG, MTT), Harvard Medical
School, Massachusetts Mental Health Center, Boston; Brockton/West
Roxbury Veterans Affairs Medical Center (LJS, MTT), Brockton; and Mas-
sachusetts General Hospital (HWT, LJS, HB, JiMG, SVF, MTT), Boston;
Department of Radiology (HB, JiMG), Harvard Medical School, Massa-
chusetts General Hospital, Boston; Department of Epidemiology (MTT),
Harvard School of Public Health, Boston; Martinos Center for Biomedical
Imaging (HWT, HB, JuMG), Massachusetts Institute of Technology, Har-
vard Medical School and Massachusetts General Hospital, Charlestown;
Harvard Institute of Psychiatric Epidemiology and Genetics (LJS, JiMG,
SVF, MTT), Boston, Massachusetts; and Department of Psychology and
Brain Research Institute (RP), University of California, Los Angeles.
Address reprint requests to Heidi W. Thermenos, Ph.D., or Larry J. Seidman,
Ph.D., Massachusetts General Hospital-East, Building 149, Room 2602E,
13th Street, Charlestown, MA 02129.
Received May 12, 2003; revised November 18, 2003; accepted November 20,
2003.
BIOL PSYCHIATRY 2004;55:490 –500 0006-3223/04/$30.00
doi:10.1016/j.biopsych.2003.11.014 © 2004 Society of Biological Psychiatry