Self-evaluation in schizophrenia: an fMRI study with implications for the understanding of insight

Abstract

Background: Lack of insight is a core feature of schizophrenia and is associated with structural brain abnormalities. The functional neuroanatomy of insight has only recently been investigated. When people evaluate their personality traits compared to those of another, activation is seen in central midline structures (CMS) of the brain. This study set out to compare cerebral activation in schizophrenia patients versus controls during a self-evaluation task which included positive and negative traits as well as mental and physical illness terms.

Methods: Eleven schizophrenia patients and 8 healthy controls, matched for age were studied. Insight was assessed using the Schedule for the Assessment of Insight-expanded version (SAI-E). FMRI data were obtained with a 1.5 Tesla GE system and interactions between participant group, self versus other, significant at the cluster level, were recorded.

Results: Significant hypoactivation in the medial superior frontal gyrus (dorsomedial prefrontal cortex) was observed in patients vs. controls during self-evaluation of all traits combined. A second cluster of hypoactivation in the posterior cingulate was also detected. When the response to individual traits was explored, underactivation in other frontal regions plus right inferior parietal lobule emerged and this tended to correlate, albeit weakly with lower insight scores. Further, there were areas of hyperactivation relative to controls in anterior cingulate, frontal and parietal regions (especially precuneus) which showed moderate inverse correlations with insight scores.

Conclusions: We have demonstrated that the CMS, identified as a key system underpinning self-evaluation, is dysfunctional in patients with schizophrenia, particularly dorso-medial PFC. This may have implications for lack of insight in schizophrenia. Hypofunction within the dorsomedial prefrontal region seems to be particularly important although other posterior and lateral cortical regions play a part and may modulate self-evaluative responses depending on the type of trait under consideration.

Figure 1

Blood oxygen level dependent activation maps. Upper Panel (cold colours): clusters of activation during person-evaluation where healthy controls (HC) were significantly greater than schizophrenia (Sz) patients. Lower Panel (warm colours): clusters of activation during person-evaluation where Sz were significantly greater than HC (see Table 4). Right side of axial image is right side of brain.

Figure 2

A. Regional cluster in left superior frontal gyrus (57 voxels) showing relative underactivity in schizophrenia patients (Sz) versus healthy controls (HC) during self-evaluation (vs other (Blair)) of traits. B. Cluster map (see Table 4). Right side of axial/coronal image is right side of brain.

Figure 3

(left). Regional cluster showing relative over-activity in schizophrenia patients (Sz) versus healthy controls (HC) during self-evaluation of mental illness traits (self vs. other (Blair)). B (right column). Pearson’s correlation with insight scores (Sz only) and scatter plot (see Table 4).

Figure 4

A (left). Regional cluster showing relative over-activity in schizophrenia patients (Sz) vs. healthy controls (HC) during self-evaluation of positive traits (self vs. other (Blair)). B (right column). Pearson’s correlation with insight scores (Sz only) and trait ownership (Sz and HC) and scatter plot (all participants) (see Table 5).

Figure 5

Regional clusters showing relative underactivity in schizophrenia patients (Sz) versus healthy controls (HC) during self-evaluation of traits.A. Positive traits (self vs. other (Blair)). B. Negative traits (self vs. other (Blair) see Table 5).