O2.5. DISENTANGLING FALSE PERCEPTIONS ELICITED BY WHITE NOISE IN PEOPLE WITH AUDITORY HALLUCINATIONS: THE ROLE OF SOUND FREQUENCIES AND EXPECTATIONS. (9th April 2019)

• Record Type: Journal Article
• Authors: Laloyaux, Julien; Specht, Karsten; Hugdahl, Kenneth; Larøi, Frank
• Is Part Of: Schizophrenia bulletin. Volume 45(2019)Supplement 2
• Page Start: S164
• Page End: S164
• Publication Date: 2019-04-09
• Subjects: Schizophrenia -- Periodicals; Schizophrenia -- Research -- Periodicals; 616.898005
• DOI: 10.1093/schbul/sbz021.189
• Languages: English
• ISSNs: 0586-7614

Abstract

Abstract: Background: Auditory Hallucinations (AH) are prevalent in many psychopathologies as well as in a minority of the healthy general population. The literature suggests that people with AH present a signal detection bias, that is, in conditions of uncertainty, they tend to perceive signals that are not actually there– or false perceptions. For example, when listening to white noise, they perceive voices that are not actually in the noise. However, the literature is inconsistent regarding the relation between the number of false perceptions and hallucination proneness. In particular, whereas some studies have reported a significant correlation, other have not. Such inconsistencies may be explained by the lack of understanding of the white noise paradigm and in particular the role of the noise and of the level of expectation. The aim of the present study was thus to disentangle the white noise paradigm and to explore the specific relations with AH and other cognitive mechanisms or personality traits. Methods: Twenty-three healthy individuals with a high proneness to AH and 20 with a low proneness were recruited. They were asked to carry out a semantic signal detection task where participants are required to listen to recordings of sentences from which the last word was replaced by a burst of noise alone or by a burst of noise with the last word of the sentence embedded in it. After each sentence, participants were asked to report whether or not they heard the last word. The sentences were created to induce either a high or a low level of expectation regarding the last word. Two different types of noise (that replaced the last word) were created. The first one was composed of white noise in which the language-related frequencies (formants) were predominant – or human noise. The second noise was composed of white noise minus the language-related frequencies – or non-human noise. Participants were also required to complete questionnaires assessing hallucination proneness, fantasy proneness, dissociation, social desirability, mind wandering, suggestibility, aberrant salience, and encoding style. Results: Results revealed a significant interaction effect between the level of expectation, the type of noise, and the level of hallucination proneness. In particular, when confronted with both human noise and a high level of expectation, participants with a high proneness towards AH showed a greater signal detection bias than people with a low proneness, i.e., they reported more false alarms. On the contrary, no significant difference was found between the groups for the non-human noise or for a low level of expectation. Finally, regression analyses revealed that hallucination proneness was the best predictor of the signal detection bias. Discussion: For the first time in the literature, the present study demonstrated that the signal detection bias frequently observed in people with AH is driven by both the specific language-related frequencies contained in white noise and by a high level of expectation. Moreover, the results showed that the signal detection bias was specifically predicted by a high proneness towards AH. These results are in agreement with the latest cognitive models of AH that claim that AH arise from an interaction between top-down (the human noise) and bottom-up (the high level of expectation) cognitive mechanisms. Futures studies are needed to replicate these findings, nevertheless, the present results suggest that the human noise should be used instead of white noise in order to improve the specificity of the white noise effect. One implication of these findings is that future studies could examine the specificity of this paradigm in terms of detection of ultra-high-risk individuals for psychosis.