Abstract
Background
Auditory verbal hallucinations (AVH) are conscious sensory experiences occurring in the absence of external stimulation. AVH are experienced by 75% of individuals diagnosed with schizophrenia and can manifest in other neuropsychiatric disorders. However, AVH are also reported amongst healthy individuals. This implies that hearing voices is not necessarily linked to psychopathology. Amongst voice hearers, the likelihood of AVH seems to reflect individual differences in hallucination proneness (HP). The HP construct allows placing individuals on a psychosis continuum ranging from non-clinical to clinical experiences. Clinical voice hearers tend to misattribute internal events to external sources (externalization bias). Specifically, they seem to experience altered sensory feedback in response to self-initiated stimuli: Although more predictable, clinical voice hearers show similar, neurophysiological responses in reaction to self-initiated vs. externally presented stimuli. EEG studies suggest that this aberrance of prediction is associated with diminished N1-suppression effects that are observed in healthy individuals in response to self-initiated stimuli. Accordingly, clinical voice hearers may have problems differentiating between self-initiated and externally generated speech, potentially leading to externalization of their own speech. In line with this proposal, the current study focusses on non-clinical aspects of the psychosis continuum in healthy voice hearers and controls. This approach avoids confounding factors (medication, disease onset/duration etc.) that typically impede comparisons of clinical and non-clinical voice hearers. By utilizing insights on prediction from the forward model concept within the auditory-sensory domain, we want to investigate how N1-amplitudes in reaction to one’s own or someone else’s voice are modulated as a function of HP. Next to ascertaining the mechanism behind AVH, this research could give direction to identifying risk factors that potentiate the emergence of first-incidence psychosis.
Methods
HP was assessed by means of the Launay-Slade Hallucination Scale. Each participant’s voice was recorded prior to EEG data acquisition (monosyllabic utterances, “ah” & “oh”, duration = 500 ms). Voice stimuli were morphed with an anchor voice, so that voice identity could be alternated from self- to other-voice (0%, 40%, 50%, 60%, 100%). To contrast neurophysiological responses between self- vs. externally generated voice stimuli, a well-established motor-to-auditory paradigm was used: In a motor-to-auditory condition (MAC) participants were prompted to press a button, thereby eliciting a voice stimulus (self-initiation). In an auditory-only condition (AOC), participants were prompted to passively listen to the voice stimulus (external generation). The motor-only condition (MOC), in which participants executed the button press only, served as a control condition to correct for motor activity in MAC.
Results
Data from 38 participants replicate the classical N1-suppression effects for self-initiated vs. externally generated self-voice stimuli. This pattern of suppression is also visible for other-voice stimuli. Furthermore, current findings seem to replicate reversed N1-suppression for self-voice in individuals with high HP.
Discussion
Preliminary findings suggest that HP modulates voice identity processing. More specifically, HP determines how voice stimuli are processed within the internal and external domain. Particularly, individuals with high HP show a reversal of N1-suppression for self-voice stimuli, which corroborates the external biasing hypothesis.
Auditory verbal hallucinations and childhood trauma subtypes across the psychosis continuum: a cluster analysis
