Spatiotemporal brain activity associated with hearing and reading in patients with verbal hallucinations: An fMRI study.

Abstract

Auditory verbal hallucinations (AVH) refer to the perception of speech devoid of external speakers. We previously suggested that while AVH experiences are that of speech perception, the underlying brain dysfunction involves inner speech generation (thinking in words).1 Inner speech is associated with a perceptual experience – an inner voice – and abnormalities of inner speech generation could lead to abnormalities in inner voice experiences – AVH. Consistently, a recent meta-analysis of imaging studies has shown AVH association with activity in both Broca's and Wernicke's areas.2 The present study examines the above hypothesis; that is: speech generation, but not speech perception, is associated with altered brain activity in AVH patients. We used functional magnetic resonance imaging to investigate brain activity during hearing (speech perception) and reading (speech generation) in seven schizophrenia patients with AVH and eight healthy control subjects. The protocol conformed to ethical standards and was approved by the University of Minnesota institutional review board. Informed consent was obtained and anonymity was preserved. To address variability of the hemodynamic response across subjects and across brain areas, we processed subject-level data with finite impulse response. In final group analyses, mixed models were used (see Appendix S1 for detailed information about methods). As expected, the neural correlates for reading, but not for hearing, differentiated patients from controls (Fig. 1). In all subjects, reading was associated with sequential activation of midline structures (cingulate and paracingulate) and bilateral insular, motor, auditory, and language cortices. In patients, higher activity in brain midline structures and in minor hemisphere homologues of Broca's and Wernicke's areas was noted. To the extent that speech generation (reading) engages similar neural resources to that engaged by inner speech generation,3 this study suggests that AVH reflect inner speech generation disorder. Consistent with these findings, altered activity of the brain midline structures during the recall of the self–other source of speech4, 5 and altered connectivity of these structures with language areas6 were reported in AVH populations. Our results, echoing the agent–observer distinction literature, suggest that the speaker of the speaker-less ‘voices’ is the self; in other words, the self is the agent, not the observer of the ‘voices.’ Although not invariably, AVH are perceived as the voices of others, which reflects agency externalization of inner speech. Our data shed light on the mechanism of this phenomenon. From a Cartesian perspective, agency externalization violates the first-person immunity principle7; that is: ‘If I think an apple is green, I can be wrong about the color of the apple but I cannot be wrong about the fact that I think an apple is green.’ The case of an AVH patient appears to be: ‘I thought of something but it is not me who was thinking.’ This paradox could be related to the difference between the self and the brain – between ‘I’ and ‘my brain.’ Irrespective of the philosophical views, it is agreed upon that the self and the brain are co-related. Descartes postulated that the self–brain interface occurs in the pineal gland7; he was not always right. Current neuroscience research implicates brain midline structures in a wide range of experiences of the self.8 In AVH literature, agency externality has often been considered to reflect abnormal corollary discharge (connectivity) between Broca's and Wernicke's areas, which is presumed to result in an abnormal perceptual appraisal of one's own inner voice and subsequent nonrecognition of this speech as one's own. While there is evidence for corollary discharge deficit,9 it may not sufficiently explain agency externalization. In a positron emission tomography study of speech generation, we have previously shown that hallucinating patients had increased activity in Wernicke's area and abnormal laterality of the activity of the supplementary motor area (SMA).10 The former, but not the latter, is consistent with a corollary discharge deficit. The abnormal SMA laterality, a midline structure implicated in self-initiated actions, during speech generation could lead to inefficient appraisal of the self-initiated source of speech. We suggested that an abnormal corollary discharge results in altered perception of one's own speech while the abnormal activity of the SMA results in agency externalization of these perceptions. The present study provides further evidence that inner speech agency externalization in AVH reflects a more complex neural dysfunction than previously considered. We suggest that altered activity of brain midline structures during inner speech in patients could blur the self-agency of that speech. Furthermore, the increased activity of minor hemisphere homologues of language areas could reflect altered perceptual experience of the inner voice (homologue of Wernicke's area) as well as altered agency experience (homologue of Broca's area). This work was supported by the Indiana University Medical School (M.S.), the William and Martha Muska Foundation (M.S.), and a Career Development Award from the US Department of Veterans Affairs (G.Y.). All authors report no conflicts of interest to be disclosed. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.