39. Heightened Phonological Priming During Ambiguity: Modeling Auditory Hallucinations in Schizophrenia

Abstract

Background: The mechanism of auditory hallucinations has been avidly pursued for decades although its formation remains largely cryptic. Another domain of clinical symptoms in schizophrenia is reduced working memory, for which the “phonological loop” is the elementary component. The phonological loop is a subvocal rehearsal of verbal information to support working memory for holding information while other information is manipulated. We hypothesize that auditory hallucinations and working memory deficit in schizophrenia share a common deficit in the phonological loop operation, especially when the phonological loop is heavily taxed. Methods: We created a new active phonological priming paradigm to examine whether an abnormal phonological association to ambiguity may be an underlying mechanism of auditory hallucinations in schizophrenia, and tested it in 31 schizophrenia patient (SZ) and 20 healthy control (HC) participants. Using a set of pseudo-words to introduce semantically ambiguous stimuli yet requiring verbally generated semantic words to be linked to the otherwise ambiguous input, the paradigm obligates a phonological association and forces activation of the phonological loop. Once verbally primed, participants then rated their familiarity to each pseudo-word auditory stimulus, by comparing to real-word auditory stimulus. “Errors” are measured by rating of the familiarity to the pseudo- versus real words. Auditory hallucinations were assessed by the Psychotic Symptom Rating Scales - Auditory Hallucinations Scale. Results: SZ did not show a significant increase in familiarity of real words after priming (P = .96); where HC were prone to “error” after the priming to the real words (P = .01). Conversely, with pseudo-words, SZ were prone to phonological priming of ambiguous stimuli and experienced a significant increase in their rating of familiarity to the otherwise meaningless pseudo-words (P = .006), whereas HC did not (P = .10). More severe auditory hallucinations correlated highly with increased familiarity to ambiguous stimuli after active phonological priming (r = .58, P = .001). Conclusion: The paradigm aims to generate behavioral readouts of priming effects to both familiar words and ambiguous pseudo-words under passive vs. active association conditions. The results indicate that schizophrenia patients are specifically vulnerable to active association to ambiguous stimuli and this vulnerability is associated with more severe auditory hallucinations.