Regular articleAltered effective connectivity during working memory performance in schizophrenia: a study with fMRI and structural equation modeling
Introduction
Aside from the characteristic psychopathology including positive and negative symptoms, deficits in a number of cognitive domains including working memory and executive functions have been described in schizophrenic patients (Goldberg and Gold, 1995). Initial fMRI studies demonstrated that schizophrenic patients exhibited less prefrontal cortical activation than normal controls while performing working memory and executive tasks Callicott et al 1998, Volz et al 1999, Stevens et al 1998. These findings suggesting “hypofrontality,” however, have recently been challenged by studies indicating a relatively increased prefrontal cortical activation pattern in schizophrenic patients in comparison to normal controls (Manoach et al., 1999). This difference between groups was present even when the schizophrenic patients were matched for (Manoach et al., 2000) or demonstrated impaired task performance (Callicott et al., 2000) relative to normal controls. The findings indicate a compensatory recruitment of neural circuits in the presence of impaired integrity of the main neural systems subserving working memory functions.
Whereas the major component of working memory has traditionally been ascribed to prefrontal brain areas (Goldman Rakic, 1994), functional brain imaging approaches provided information about widespread networks including frontal and parietal association areas involved in the performance of working memory tasks (Cohen, 1997).
Petrides 1994, Petrides 1996, Petrides 2000 has proposed a model of cortico-cortical functional interactions underlying mnemonic information processing according to which information is first processed in the parietal cortex. The posterior association areas are connected with the ventrolateral prefrontal cortex (VLPFC), where decision making, comparison, and reproduction of information held in working memory takes place. The dorsolateral prefrontal cortex (DLPFC), which is connected with the VLPFC, is additionally recruited when a higher order control of mnemonic processing such as monitoring and manipulation of information in working memory is required. This organization-by-process model could be confirmed in subsequent imaging studies. There is evidence that the dorsolateral prefrontal cortex is predominantly subserving executive control functions, whereas preprocessing and maintenance of information are mainly associated with activation of the ventrolateral frontal and also the parietal association cortex D’Esposito et al 2000, Postle et al 2000, Stern et al 2000.
Additionally, there is mounting evidence that, aside from cortical structures, cerebellar (Schmahmann and Pandya, 1997b) and thalamic (Alexander et al., 1986) areas are involved in higher cognitive functions in terms of a cortical-subcortical-cerebellar circuitry. It has been proposed that psychopathology and cognitive deficits in schizophrenia are associated with disruptions of defined information processing pathways in the brain Friston and Frith 1995, McGuire and Frith 1996. The concept of “cognitive dysmetria” has been introduced to characterize a disintegration at the system level of cortical-subcortical-cerebellar circuitry (Andreasen et al., 1998). A complex pattern of existing deficits and compensatory adaptation is likely to involve separate network components in a differential way.
Previous studies investigating schizophrenic patients with working memory tasks employed mostly univariate statistical approaches aiming at a functional segregation of distinct areas of activation. In dealing with issues of concerted network operations, the analysis of cerebral activation data provided by functional imaging has more recently moved to the search for connectivity within activated networks. Models of information processing for higher cognitive task have to address the interaction within large-scale neuronal networks. The notion of effective connectivity between brain areas involves model-based assumptions about the effect which one neural systems exerts over another (Mclntosh et al., 1999). Statistical approaches addressing effective connectivity such as structural equation modeling (SEM) have been successfully adopted to functional brain imaging data previously (Büchel and Friston, 1997).
In addition to existing pathological alterations, neuropsychological functions and related cognitive brain activation patterns can be modulated by antipsychotic treatment. Atypical antipsychotic agents appear to exert a more favorable effect on cognitive functions and in particular working memory than treatment with conventional neuroleptics (Green et al., 1997). In a comparison study of typical and atypical antipsychotic drugs, cognitive brain activation patterns as studied with fMRI and a working memory task were enhanced after switching from conventional to atypical agents (Honey et al., 1999). However, the effect of antipsychotic treatment on measures of effective connectivity has not been studied yet.
The present study was aimed to model information processing within defined cortical-subcortical-cerebellar circuitry subserving working memory functions. It was hypothesized that schizophrenic patients demonstrate a disruption in effective connectivity measures, possibly paralleled by the compensatory strengthening of other pathways, and that this dysfunction might be differentially modulated by treatment with typical and atypical antipsychotic drugs.
Section snippets
Subjects
Twelve patients were recruited from the inpatient service of a University Hospital. The patients fulfilled the diagnostic criteria for schizophrenia, either paranoid or undifferentiated type (DSM IV). All schizophrenic subjects were clinically stabilized with either typical or atypical antipsychotics for at least 2 weeks. In the typical antipsychotic group, 6 patients received haloperidol (mean dose = 4.67 mg/day) at the time of the fMRI investigation. Within the atypical antipsychotic group, 5
Psychometric performance
There was a significant difference in task performance A′ among the 3 groups (χ2(2) = 8.91, P < 0.05). The highest performance A′ was reached by the normal controls followed by the group of patients treated with atypical antipsychotics and finally the group receiving typical antipsychotics (Table 1). The difference in performance between controls and patients with typical antipsychotics received statistical significance (P < 0.05), whereas all other pairwise group differences were not
Discussion
The results of the present study suggest differences in connectivity strengths within a cortical-subcortical-cerebellar network between healthy controls and schizophrenic patients treated with typical and atypical antipsychotics. In patients treated with typical antipsychotics compared to normal controls, reduced connectivity strength could be observed in a left prefrontal/right cerebellar network and in interhemispheric connections whereas increased connectivity strength was found in
Acknowledgements
We thank the staff of the Department of Neuroradiology, University of Mainz, for performing the functional MRI scans.
References (71)
- et al.
Importance of pharmacologic control in PET studieseffects of thiothixene and haloperidol on cerebral glucose utilization in chronic schizophrenia
Psychiatry Res.
(1991) - et al.
Blockade of hippocampal dopamine (DA) receptorsa tool for antipsychotics with low extrapyramidal side effects
Prog. Neuropsychopharmacol. Biol. Psychiatry
(1988) - et al.
A parametric study of prefrontal cortex involvement in human working memory
NeuroImage
(1997) - et al.
How good is good enough in path analysis of fMRI data?
NeuroImage
(2000) - et al.
Functional magnetic resonance imaging brain mapping in psychiatrymethodological issues illustrated in a study of working memory in schizophrenia
Neuropsychopharmacology
(1998) Parametric analysis of fMRI data using linear systems methods
NeuroImage
(1997)- et al.
Noradrenergically mediated plasticity in a human attentional neuronal network
NeuroImage
(1999) - et al.
Abnormal cingulate modulation of fronto-temporal connectivity in schizophrenia
NeuroImage
(1999) - et al.
Event-related fMRIcharacterizing differential responses
NeuroImage
(1998) - et al.
Spatiotemporal dynamics of component processes in human working memory
Electroencephalogr. Clin. Neurophysiol.
(1993)
D(1) receptors in prefrontal cells and circuits
Brain Res. Brain Res. Rev.
Schizophrenic subjects show aberrant fMRI activation of dorsolateral prefrontal cortex and basal ganglia during working memory performance
Biol. Psychiatry
Schizophrenic subjects activate dorsolateral prefrontal cortex during a working memory task, as measured by fMRI
Biol. Psychiatry
Spatial pattern analysis of functional brain images using partial least squares
NeuroImage
Basal ganglia and cerebellar loopsmotor and cognitive circuits
Brain Res. Brain Res. Rev.
The assessment and analysis of handednessThe Edinburgh Inventory
Neuropsychologia
Using event-related fMRI to assess delay-period activity during performance of spatial and nonspatial working memory tasks
Brain Res. Brain Res. Protocols
Characterization of BOLD-fMRI signal during a verbal fluency paradigm in patients with intracerebral tumors affecting the frontal lobe
Magn. Reson. Imaging
Prefrontal cortex projections to the basilar pons in rhesus monkeyimplications for the cerebellar contribution to higher function
Neurosci. Lett.
The cerebrocerebellar system
Int. Rev. Neurobiol.
Activity in ventrolateral and mid-dorsolateral prefrontal cortex during nonspatial visual working memory processingevidence from functional magnetic resonance imaging
NeuroImage
Decreased frontal activation in schizophrenics during stimulation with the continuous performance test—a functional magnetic resonance imaging study
Eur. Psychiatry
Supporting evidence for the model of cognitive dysmetria in schizophrenia—a structural magnetic resonance imaging study using deformation-based morphometry
Schizophr. Res.
Clozapine preferentially increases dopamine release in the rhesus monkey prefrontal cortex compared with the caudate nucleus
Neuropsychopharmacology
Parallel organization of functionally segregated circuits linking basal ganglia and cortex
Annu. Rev. Neurosci.
Thalamic abnormalities in schizophrenia visualized through magnetic resonance image averaging
Science
“Cognitive dysmetria” as an integrative theory of schizophreniaa dysfunction in cortical-subcortical-cerebellar circuitry? Schizophr
Bull.
Effect of a haloperidol challenge on regional brain metabolism in neuroleptic-responsive and nonresponsive schizophrenic patients
Am. J. Psychiatry
Structural Equations with Latent Variables
Using Talairach atlas with the MNI template
NeuroImage
Modulation of connectivity in visual pathways by attentioncortical interactions evaluated with structural equation modelling and fMRI
Cereb. Cortex
Visualizing fronto-striatal circuitry and neuroleptic effects in schizophrenia
Acta Psychiatr. Scand. Suppl.
Age-related differences in effective neural connectivity during encoding and recall
NeuroReport
Physiological dysfunction of the dorsolateral prefrontal cortex in schizophrenia revisited
Cereb. Cortex
Cited by (244)
Neural correlates of cognitive deficits across developmental phases of schizophrenia
2019, Neurobiology of DiseaseImpaired prefrontal functional connectivity associated with working memory task performance and disorganization despite intact activations in schizophrenia
2019, Psychiatry Research - NeuroimagingVerbal working memory and functional large-scale networks in schizophrenia
2017, Psychiatry Research - NeuroimagingCitation Excerpt :Evidence comes from functional Magnetic Resonance Imaging (fMRI)1 studies including functional connectivity (FC) and effective connectivity (EC) studies in verbal working memory in patients with schizophrenia (SZ) and healthy controls (HC). Such studies repeatedly reported cortical dysconnectivity in SZ when compared to HC (Birnbaum and Weinberger, 2013; Dauvermann et al., 2014; Deserno et al., 2012; Glahn et al., 2005; Schlosser et al., 2003a, 2003b, 2006; Schmidt et al., 2013, 2014). Evidence from animal studies proposes that activity-dependent synaptic plasticity processes (Abbott et al., 1997; Rothman et al., 2009) are modulated via nonlinear effects.