Auditory verbal hallucinations (AVH) are prevalent in schizophrenia and are often distressing. However, relatively little is known about their linguistic structure, although a number of authors have commented that they tend take the form of short, syntactically simple sentences (Tovar et al Schizophr Res 2019; 206 111-117; Corona-Hernández et al Schizophr Res 2022; 241 210-217). It has been suggested that these features may be related to the high frequency with which AVH feature insults and commands (which are normally short and simple).

We aimed to quantify sentence length and complexity of AVH in schizophrenia patients, and to examine how far length reductions were attributable to presence of insults and commands. We also examined the same variables in real speech from patients with schizophrenia and healthy controls.

We transcribed verbatim AVH from 11 patients with very frequent AVH following a previously used protocol (Fuentes-Claramonte et al Sci Rep 2021; 23 18890). Mean sentence length and mean dependency distance (a measure of syntactic complexity) were calculated using the udpipe package in R. Insults and commands were also coded. For comparison, (real) speech samples were collected and transcribed from patients with schizophrenia (N=14) and healthy controls (N=15). All groups were matched for age, sex and estimated premorbid IQ.

We found that AVH sentences were on average significantly shorter (t(37)= -6.51, p < 0.001; see Fig. 1A) and syntactically simpler (t(37)= -4.37, p < 0.001; see Fig. 1B) than in the (real) speech of healthy controls. AVH sentences were also shorter (Fig. 1A) and simpler (Fig. 1B) than the speech of schizophrenia patients, although the latter comparison only approached significance (t(37)= -4.09, p < 0.001 and t(37)= -2.31, p = .08, respectively). After insults and commands were removed from the analysis, AVH sentences were still shorter (t(37)= -6.09, p < 0.001) and simpler (t(37)= -3.89, p < 0.001) than those in the speech of controls, and shorter (t(37)= -3.68, p < 0.01) than those in the speech of patients, but not simpler (t(37)= -1.86, p = 0.213).

Image 1:

From our data AVH mainly (though not exclusively) take the form of short and simple sentences. These features are not explained by presence of insults and commands.

None Declared

An influential current theory suggests that auditory verbal hallucinations (AVH) result from abnormal activity in the auditory cortex. Against this, however, Fuentes-Claramonte et al (Sci Rep 2021; 11 18890) recently found that AVH did not activate the primary or secondary auditory cortex, although there were activations in other areas including Broca’s area. At the level of brain structure, sulcal depth has been increasingly linked to brain function (Natu et al Cereb Cortex 2021; 31 48-61; De Vareilles et al Dev Cogn Neurosci 2023; 61 101249 ). Accordingly, exploring brain structural-functional associations for AVH may provide new insights in their biological basis.

To assess the relationship between sulcal depth and brain activity during AVH and perception of real speech.

Functional (fMRI) and structural (sMRI) 3T scans were obtained from 14 patients with schizophrenia who experienced near-continuous AVH. During fMRI, participants pressed a button when they experienced AVH or heard real speech similar in form to their AVH. Standard fMRI analysis was conducted with FSL, while sMRI images were processed using Freesurfer’s recon-all pipeline to measure sulcal depth. Cross-modal registration aligned whole-brain fMRI activation maps to corresponding structural data and correlations between sulcal depth and brain activity were calculated for each vertex; age, sex and estimated premorbid IQ were covaried for. Cluster-based correction was applied for multiple comparisons.

During real speech, a positive correlation was found between brain activations and sulcal depth in the left superior temporal sulcus (STS, BA 22), and negative correlations in the middle temporal (BA 21), frontal (BA 46), and parietal cortex. On the right, positive correlations were seen in the superior and middle temporal cortex (BA 38, 20, 42), while negative correlations were found in the STS (BA 22), pars triangularis (BA45), and precentral (figure a). During AVH, there was a negative correlation in the left pars triangularis (BA 45) only, including Broca’s area, with no significant correlations in the right hemisphere (figure b).

Image 1:

The left STS, along with frontal and temporoparietal areas, appear structurally and functionally linked to perception of real speech. In contrast, AVH primarily engages Broca’s area and adjacent left inferior frontal regions.

None Declared

Schizophrenia is a severe psychiatric disorder characterized by positive (delusions and hallucinations), negative and disorganization symptoms. According to the influential dopamine hypothesis, positive symptoms of schizophrenia are linked to increased dopamine transmission in subcortical regions, particularly the striatum. Kapur’s aberrant salience theory further suggests that hyperdopaminergia leads to increased and disorganized reward prediction error signalling, leading to the misattribution of significance to irrelevant stimuli, which contributes to the development of delusions. Negative symptoms have been argued to reflect reduced reward prediction error signalling.

To design an optimized monetary incentive delay (MID) task for use in fMRI studies of schizophrenia patients. For this, a pilot study was conducted in order to test the effects of monetary incentives on task performance in schizophrenia patients and healthy controls.

Nine healthy controls and seven patients with DSM-5 schizophrenia completed the MID task, including a training and test phase. This task evaluates reward processing by presenting cues that predict either rewarding or non-rewarding outcomes before a target that requires a speeded response. We investigated the effect of these cues on task performance (accuracy and reaction times) through a repeated measures ANOVA that compared rewarding and non-rewarding cues with a between-subjects factor for diagnosis.

The patients with schizophrenia exhibited slower RT and lower accuracy compared to healthy controls (main effect of group, RT p = 0.008, accuracy p = 0.047). There was also a significant main effect of condition, with better accuracy and shorter RT in the rewarded condition in both the patients and controls (accuracy p = 0.01, RT p = 0.003). However, there was no significant group*condition interaction. Both the patients and the controls showed significant improvements in task performance when rewards were offered, compared to when no rewards were provided.

Our MID task shows expected performance effects in patients with schizophrenia, producing effects comparable to those observed in healthy controls. This MID task design is therefore suitable for examining and understanding symptom profiles associated with reward processing, such as delusions and negative symptoms.

None Declared

Delusions in schizophrenia have been theoretically linked to probabilistic reasoning bias (‘jumping to conclusions’, JTC), although experimental support has been mixed (Garety et al BJP 2013; 203 327-333). Ward and Garety (Schiz Res 2019; 203 80-87) recently proposed a reformulation of the theory in terms of Kahneman’s concepts of ‘fast’ and ‘slow’ thinking. This proposes that decision-making involves two cognitive processes: a fast, heuristic-based approach which is prone to errors, and a slow, deliberate process that carefully evaluates all the relevant evidence. According to this view, an overreliance on fast thinking and/or reduced engagement of slow thinking underlies the initial development of delusional interpretations of everyday events and also makes them harder to be corrected.

Our aim was to develop a novel task to investigate the fast and slow thinking hypothesis of delusions in patients with schizophrenia, for use in behavioural and functional imaging studies. As a preliminary step, we tested this task on healthy participants.

A battery of 137 experimental questions (where fast thinking leads to incorrect answers) was generated from multiple sources, including examples of the base rate and conjunction fallacies, the cognitive reflection test (CRT), trick questions, and syllogisms. Example questions included: If it takes 5 machines 5 minutes to make 5 widgets, how long would it take 100 machines to make 100 widgets? [Correct answer = 5 minutes; intuitive answer = 100 minutes; category: cognitive reflection] A farmer had 15 sheep and all but 8 died. How many are left? [intuitive answer: 7; correct answer: 8, category: trick question]. 137 control questions (where both fast and slow thinking give the correct answer) were adapted from the experimental questions. The questions were administered online to 176 healthy volunteers using PsychoPy software, with 15 experimental and 15 control questions randomly assigned to each participant.

The sample had a mean age of 40.3 years (range 17-77 years); 55.1% were female and 65.9% had a university education. Correct answers to experimental questions were markedly fewer than answers to control questions in all categories (overall p < 0.001). Response latency for the experimental questions was slightly higher than for the control questions, apart from in one category (CRT) (overall p = 0.004).

Results from a large sample of healthy participants indicate that a battery of questions can be feasibly developed to reliably detect fast thinking.

None Declared

Auditory verbal hallucinations (AVH) are one of the primary symptoms of schizophrenia, but the biological mechanisms underlying them remain uncertain (1,2). Theoretical approaches have proposed that AVH are caused by abnormal activity in the auditory cortex; or that they represent misinterpreted cognitive activity such as inner speech. Recently, our group found, using a symptom capture task, that AVH did not trigger activity in the auditory cortex, but instead in language-related areas, thus shifting the focus towards cognitive theories of AVH (3). To date, cognitive approaches have only been preliminarily investigated, and mostly in psychological studies (1,2).

Our aim was to test the theory that a disturbance in inner speech processes underlie AVH. We used conjunction analysis to examine common activation patterns between the experience of AVH and phonological encoding.

Eleven patients meeting DSM-5 criteria for schizophrenia or schizoaffective disorder with near-continuous AVH underwent fMRI during symptom capture and during a phonological encoding task. In the symptom capture task, the patients were instructed to press their left index finger when they begin to hear an AVH, wait three minutes, mentally repeat what they heard, and then press their right index finger. The phonological encoding task required them to indicate, via button press, whether the names of two objects shown in line drawings rhymed.

Pre-processing and analyses were carried out with FSL software using linear models. Activation maps were thresholded at p<0.05, cluster-corrected for multiple comparisons. To find regions of common activation between the two tasks, the activation maps from the contrasts of interest were binarized and entered into a conjunction analysis. Regions showing significant activation in both tasks simultaneously were considered activated in the conjunction analysis.

The conjunction analysis showed common activation in several regions involved in phonological encoding, such as Broca’s area and its right homologue, supplementary motor area bilaterally, Wernicke’s area and cerebellum, in patients with AVH.

These results support a non-perceptual origin of AVH and link them to brain areas related to the phonological loop and working memory in schizophrenia.

None Declared

Psychedelic drugs are a focus of interest in the treatment of depression and other disorders but there are longstanding concerns about possible adverse psychiatric consequences. Because the relevant literature is largely informal, the seriousness of these risks is difficult to evaluate.

Searches were made for case reports of schizophrenia-spectrum, affective or other psychiatric disorders after use of psychedelic drugs. Case reports of flashbacks were also searched for. Individuals with recent use of other drugs (apart from cannabis and alcohol) and/or a previous history of major psychiatric disorder were excluded. Symptoms were tabulated using the Syndrome Check List of the Present State Examination (PSE-9).

We found 17 case reports of schizophrenia spectrum disorder, 17 of affective disorder (depression, mania, or both), 3 cases of anxiety, 1 of depersonalization, and 1 of unclassifiable illness. The states could develop after a single use of the drug (5/17 schizophrenia; 6/17 affective disorder), and duration was highly variable. Recovery was the rule in cases of affective disorder but not in schizophrenia spectrum disorder. Twelve of 29 cases of flashbacks showed psychiatric symptomatology definitely outlasting the attacks, mainly anxiety (5 cases) and depression (8 cases). Flashback symptoms resolved within twelve months in approximately half of the cases but in a few persisted for years.

Reliable descriptions of schizophrenia spectrum disorder and major affective disorder after psychedelic drug use disorder exist but are relatively uncommon. Flashbacks are sometimes but not always associated with psychiatric symptomatology, mainly anxiety or depression.

Autobiographical memory is known to be disturbed in schizophrenia. In addition, a leading theory of auditory hallucinations (AVH) is that they are intrusive – typically negative – autobiographical memories that are misinterpreted as perceptions.

The aim of this study was to examine the brain functional correlates of recall of negatively emotionally valanced autobiographical memories in patients with schizophrenia, with a longer term aim of comparing patients with and without AVH.

11 patients meeting DSM-5 criteria for schizophrenia or schizoaffective disorder and 10 age, sex and estimated premorbid IQ-matched healthy controls have so far taken part.

Participants underwent functional MRI in a 3T scanner while performing a task requiring them to recall autobiographical memories in response to individually tailored pairs of cue words. The cue words were based on autobiographical memories previously elicited in an interview with each patient and were designed to evoke the same memory. The cue words were presented in 10 20-second blocks interspersed with blocks where the subjects viewed cue words that did not evoke autobiographical memories. Brain activations were examined in three contrasts of interest: memory evoking words vs baseline, neutral words vs baseline and memory evoking vs neutral words.

Pre-processing and analysis were carried out with the FEAT module included in the FSL software. Statistical analysis was performed by means of a General Linear Model (GLM) approach.

In the memory evoking vs baseline contrast the patients showed hypoactivation in the medial frontal cortex compared to the healthy controls (Figure 1). There were no differences in activation between the patients and the controls comparing the memory evoking and neutral cues.

Image:

The finding of hypoactivation in the medial frontal cortex compared to low level baseline in patients with schizophrenia suggests dysfunction in the default mode network, which is known to activate during recall of autobiographical memories.

These preliminary results suggest that recall of negative autobiographical memories in patients with schizophrenia is associated with reduced activity in the default mode network. A planned larger sample of patients and controls will be used to examine activations in patients with and without AVH.

None Declared

One hypothesis proposed to underlie formal thought disorder (FTD), the incoherent speech is seen in some patients with schizophrenia, is that it reflects impairment in frontal/executive function. While this proposal has received support in neuropsychological studies, it has been relatively little tested using functional imaging. This study aimed to examine brain activations associated with FTD, and its two main factor-analytically derived subsyndromes, during the performance of a working memory task.

Seventy patients with schizophrenia showing a full range of FTD scores and 70 matched healthy controls underwent fMRI during the performance of the 2-back version of the n-back task. Whole-brain corrected, voxel-based correlations with FTD scores were examined in the patient group.

During 2-back performance the patients showed clusters of significant inverse correlation with FTD scores in the inferior frontal cortex and dorsolateral prefrontal cortex bilaterally, the left temporal cortex and subcortically in the basal ganglia and thalamus. Further analysis revealed that these correlations reflected an association only with ‘alogia’ (poverty of speech, poverty of content of speech and perseveration) and not with the ‘fluent disorganization’ component of FTD.

This study provides functional imaging support for the view that FTD in schizophrenia may involve impaired executive/frontal function. However, the relationship appears to be exclusively with alogia and not with the variables contributing to fluent disorganization.