Functional imaging abnormality has been proposed as a trait marker or endophenotype for schizophrenia. Although studies examining dorsolateral prefrontal cortex activation in relatives of schizophrenic patients have not had consistent findings, failure of de-activation in the medial frontal cortex may be a promising alternative candidate.

Twenty-nine patients with schizophrenia, 29 of their siblings and 58 healthy controls underwent fMRI during performance of the n-back working memory task. An ANOVA was fitted to individual whole-brain maps from each set of patient-relative-matched pair of controls. Clusters of significant difference among the groups were used as regions-of-interest to compare mean activations/de-activations between them.

Four clusters of significant difference were found. The schizophrenic patients, but not the relatives, showed reduced activation compared to the controls in the left lateral frontal cortex, the left basal ganglia and the cerebellum. In contrast, both the patients and the relatives showed significantly reduced de-activation compared to the healthy controls in the medial frontal cortex, with evidence of an intermediate pattern in the relatives. Failure of de-activation was not associated with schizotypy scores or presence of psychotic experiences in the relatives.

Both schizophrenic patients and their relatives show altered task-related de-activation in the medial frontal cortex. This in turn suggests that default mode dysfunction may function as a trait marker or endophenotype for schizophrenia.

Table. 1

Demographic characteristics of the patinents (n = 29), healthy relatives (n = 29) and controls (n = 58)

Fig. 1

Brain regions showing a significant effect in the 2-back versus baseline contrast in controls (a), healthy first-degree relatives (b) and in schizophrenic patients (c) Yellow indicates a positive association (activation) with the task. Blue indicates areas where the task led to a decrease in the blood oxygenation level-dependent (BOLD) response (deactivation). Numbers refer to Montreal Neurological Institute (MNI) z coordinates of the slice shown. The right side of each image represents the right side of the brain. Results are thresholdedd at z = 3.5 and corrected for multiple comparisons across space using Gaussian random field methods.

Fig. 2

Clusters of difference between schizophrenic patients (SCZ, n = 29), their first-degree healthy relatives (REL, n = 29) and healthy controls (CTRL, n = 58) (a) Left caudate gryus, globus pallidus and putamen activation, (b) bilateral cerebellar activation, (c) left dorsolateral activation and (d) anterior cingulate deactivation. The right side of each image represents the right side of the brain. Whole brain results are corrected for multiple comparisons across space using Gaussian random field methods. Analyses within ROIs were FDR-corrected for multiple comparisons.