Understanding why patients with schizophrenia do not perceive the hollow-mask illusion using dynamic causal modelling
Introduction
In order to perceive the environment as meaningful the interaction between bottom-up and top-down processing has to be intact (Wallbott and Ricci-Bitti, 1993, Cauller, 1995). Visual illusions provide a useful tool to study the mechanisms by which top-down and bottom-up processes interact in perception; they can occur when the brain interprets sensory information on the basis of contextual information and previous experience, resulting in a percept that diverges substantially from the true sensory input. In this study we used the ‘hollow-mask illusion’ (Gregory, 1973) to investigate such an interaction. The hollow-mask illusion occurs when a hollow-mask is perceived (incorrectly) as a normal face. It is understood to be a process that involves the generation of hypotheses about the three-dimensional shape of faces by interpreting the bottom-up signals received from the eyes using conceptual and perceptual knowledge (top-down processing), as well as general rules of perception, such as Gestalt laws of organisation and perspective (Yellott, 1981, Ramachandran, 1988, Hill and Bruce, 1993, Gregory, 1998).
Almost a century ago Bleuler (1911) coined the term schizophrenia to represent the ‘splitting’ of different mental domains. This idea is still influential, but in recent years has been recast in terms of pathological connectivity between brain areas. In this framework, the symptoms of schizophrenia are not considered as a single deficit but can be seen as ‘resulting form the abnormal integration of two or more processes…and are expressed when two or more regions interact’ (Friston, 1998). Similarly, Emrich (1989) proposed that the pathogenesis of schizophrenia can be described as a functional disequilibrium within the human brain, and that an impairment of the bottom-up and top-down interaction may be a plausible explanation for the disintegrative and reality-impairing properties of psychotic disorders. Frith and Done, 1988, Frith and Done, 1989 and Malenka et al. (1982) suggested that internal correcting systems may be deficient in psychotic states, and that an imbalance occurs in systems responsible for concept formation, suggesting that schizophrenics are forced to rely on stimulus-driven processing, whereby fragments of stimuli are pieced together without reference to an expected or stored model (Hemsley, 1987).The insusceptibility of patients with schizophrenia to visual illusions is consistent with such theories. For example, it has been demonstrated that patients suffering from schizophrenia do not experience the hollow-mask illusion, i.e. the hollow stimulus is correctly perceived as hollow (Schneider et al., 1996, Scheider et al., 2002, Emrich et al., 1997), consistent with weakened top-down influences in schizophrenia.
While clear hypotheses relating to the integration of top-down and bottom-up processes arise from the theoretical positions discussed above, no study to date has investigated the neural mechanisms underpinning the failure to perceive visual illusions in schizophrenia. Understanding the interaction between top-down and bottom-up processes in schizophrenic patients is important in further understanding the pathology of schizophrenia. The primary aim of this study was therefore to use measures of effective connectivity arising from dynamic causal modelling (DCM) to explain differences in both the perception of hollow faces and associated neural responses between patients with schizophrenia and controls. We hypothesised that top-down influences from the fronto-parietal network give rise to the hollow-mask illusion in controls, and that normal or strengthened bottom-up influences from visual areas in the absence of top-down input from the fronto-parietal network prevent the patients from experiencing the illusion.
Section snippets
Subjects
Thirteen patients and 16 healthy controls matched for age, gender and educational level participated in the study (see Table 1). All schizophrenic patients fulfilled DSM-IV and ICD-10 criteria for schizophrenia and were taking atypical antipsychotic medication. Schizophrenic patients with other psychiatric disorders, including drug and or alcohol abuse and neurological disorders, were excluded. The Positive and Negative Syndrome Scale (PANSS) was used to evaluate the current symptomatology of
Behavioural data
We initially established that, as expected, patients and controls differed in terms of the susceptibility to the inverted-face illusion. Post-scan questionnaires revealed that none from the controls reported seeing a face as ‘hollow’, while all patients did (Fisher's exact test, p < 0.0001). Analysis of the response data collected during the task revealed that, due to the illusion, controls occasionally erroneously classified a 3D inverted face as flat (2D) (mean = 8.3% and std = 6.8%). Notably, the
Discussion
In this study we demonstrated that schizophrenic patients and healthy controls differ in terms of the modulation of neural connectivity during the presentation of illusory stimuli. More precisely, the data of the control group were best explained by a model where the dynamic modulation of connectivity according to face-type (normal or inverted) was placed on the backwards connections from IPS to LOC. This finding is consistent with the hypothesis that top-down influences from the
Acknowledgments
This research work was funded by a Marie Curie Early Stage Training Fellowship of the European Community's Sixth Framework Programme under the contract number MEST-CT-2005-021014. We thank our colleagues for helpful discussions, the Institute of Diagnostic and Interventional Neuroradiology, Medical School Hannover, for using their facilities to perform the fMRI scans, and Karl Friston for valuable comments on the manuscript as well as guidance in statistical and fMRI analysis. Finally, we would
References (39)
- et al.
Parietal cortex and attention
Curr. Opin. Neurobiol.
(2004) Layer I of primary sensory neocortex: where top-down converges upon bottom-up
Behav. Brain Res.
(1995)- et al.
Reduced binocular depth inversion as an indicator of cannabis-induced censorship impairment
Pharmacol. Biochem. Behav.
(1991) - et al.
Towards a cannabinoid hypothesis of schizophrenia: cognitive impairments due to a dysregulation of the endogenous cannabinoid system
Pharmacol. Biochem. Behav.
(1997) The disconnection hypothesis
Schizophr. Res.
(1998)- et al.
Dynamic causal modeling
NeuroImage
(2003) - et al.
Dynamic causal modelling of evoked potentials: a reproducibility study
NeuroImage
(2007) - et al.
Different effects of nabilone and cannabidol on binocular depth inversion in man
Pharmacol. Biochem. Behav.
(2000) The assessment and analysis of handedness: the Edinburgh Inventory
Neuropsychologia
(1971)- et al.
Reduced binocular depth inversion in regular cannabis users
Pharmacol. Biochem. Behav.
(2003)
Synaptic plasticity and dysconnection in schizophrenia
Biol. Psychiatry
A three-component-system-hypothesis of psychosis. Impairment of binocular depth inversion as an indicator of functional dysequilibrium
Br. J. Psychiatry
Systems-theory of psychosis — the relevance of ‘internal censorship’
Pharmacopsychiatry
Dysfunctional connectivity in schzophrenia
Am. J. Psychiatry
Towards a neuropsychology of schizophrenia
Br. J. Psychiatry
Experiences of alien control in schizophrenia reflect a disorder in central monitoring of action
Psychol. Med.
Isolation, structure and partial synthesis of an active constituent of hashish
J. Am. Chem. Soc.
Evoked brain responses are generated by feedback loops
Proc. Natl. Acad. Sci. U. S. A.
Cited by (146)
Aberrant Effective Connectivity During Eye Gaze Processing Is Linked to Social Functioning and Symptoms in Schizophrenia
2023, Biological Psychiatry: Cognitive Neuroscience and NeuroimagingVisual illusions in young people reporting psychotic-like experiences
2023, Journal of Behavior Therapy and Experimental PsychiatrySystematic review of visual illusions in schizophrenia
2023, Schizophrenia ResearchThe Hermann grid illusion fails to fool patients with schizophrenia: Experimental support for a reduced lateral inhibition hypothesis
2022, Psychiatry Research Communications