Inefficient executive cognitive control in schizophrenia is preceded by altered functional activation during information encoding: An fMRI study

https://doi.org/10.1016/j.neuropsychologia.2007.07.006Get rights and content

Abstract

Working memory deficits are a core feature of schizophrenia. Previous working memory studies suggest a load dependent storage deficit. However, explicit studies of higher executive working memory processes are limited. Moreover, few studies have examined whether subcomponents of working memory such as encoding and maintenance of information are differentially affected by these deficits.

Therefore, the aim of the present study was to examine the neural substrates of working memory subprocesses requiring stimulus encoding, maintenance and higher executive processing.

Using functional magnetic resonance imaging a modified Sternberg working memory task involving verbal stimulus material was applied. The event-related design enabled the segregation of encoding, active maintenance and executive manipulation of information. Forty-one patients with schizophrenia and 41 healthy subjects were included.

Relative to normal controls, schizophrenic patients demonstrated a significantly stronger activation pattern in a fronto-parietal network during executive information manipulation. Additionally, significant relative hypoactivity was detectable in the thalamus. Conversely, during stimulus encoding the patients demonstrated lower activation relative to controls in the prefrontal cortex and the anterior cingulate gyrus.

The present findings indicate a pronounced prefrontal functional hyperactivation within the neural network subserving higher executive working memory control processes in schizophrenia. Moreover, they suggest that these altered activations during executive control are related to a preceding abnormality of information encoding. During encoding, a reduced activation in mainly dorsolateral prefrontal and anterior cingulate regions was observed. These results could be explained by increased top-down control processing from prefrontal cortex as a compensation for functional deficits occurring during encoding.

Introduction

Converging lines of evidence suggest that cognitive deficits are a core feature of schizophrenia. In particular, schizophrenic patients exhibit dysfunctions in complex working memory (WM) tasks demanding a high amount of cognitive control. WM enables the short-term storage of information and manipulation of this information while being kept in storage. In the classical WM model of Baddeley a central executive is residing above a verbal and a visuo-spatial sketchpad (Repovs & Baddeley, 2006). According to this model, the sketchpad “slave” systems mainly support short-term maintenance of information. The central executive exerts cognitive control and allows for higher mnemonic processes such as manipulation of information.

Functional imaging studies have provided evidence that the neural substrates underlying short-term maintenance might be partially dissociable from those systems responsible for executive control. Maintenance of verbal information has been shown to be associated with mainly ventrolateral prefrontal cortex activation while executive processing has been ascribed to predominantly dorsolateral regions of the PFC (for a review, see Smith & Jonides, 1999). However, several other studies which found the VLPFC also to be involved in processes of stimulus manipulation and executive control challenge this clear-cut functional differentiation (Glahn et al., 2002; Veltman, Rombouts, & Dolan, 2003; Wagner, Maril, Bjork, & Schacter, 2001).

Although activation abnormalities during working memory processing in mainly dorsolateral and ventrolateral prefrontal regions have frequently been reported in patients with schizophrenia (Callicott et al., 2003, Manoach, 2003; Perlstein, Carter, Noll, & Cohen, 2001; Schneider et al., 2007) few working memory studies have specifically examined functional activation patterns during executive control processes in patients (Cannon et al., 2005; Kim, Glahn, Nuechterlein, & Cannon, 2004; Perry et al., 2001; Tan, Choo, Fones, & Chee, 2005). Applying a delayed-response task, Kim et al. (2004) found that simultaneous maintenance and manipulation of verbal and spatial material was associated with significantly worse performance than mere maintenance in patients with schizophrenia. Cannon et al. (2005) used a Sternberg item recognition task (Sternberg, 1966) to compare maintenance and manipulation of spatial stimuli with mere maintenance processing. In this study, patients exhibited significantly stronger deficits in the combined task in association with activation abnormalities in mainly dorsolateral prefrontal regions. Similar results have been reported in a verbal working memory study by Tan et al. (2005) who contrasted maintenance and manipulation related activation with mere maintenance. Tan et al. (2005) detected relative hyperactivity in the left VLPFC (BA 45, 44) in combination with a significantly reduced signal in the anterior cingulate cortex (ACC; BA 32) and the left dorsolateral prefrontal cortex (DLPFC; BA 9) in patients. The authors concluded from their findings that manipulation of information or executive processing is the subprocess which is most seriously impaired in schizophrenia.

It is, however, still a matter of debate which temporal subcomponents of WM are predominantly impaired by the disorder. Recent results suggest that working memory deficits in patients might occur particularly during the encoding phase of WM (Johnson et al., 2006). Utilizing a Sternberg design which allowed the separate analysis of encoding and retrieval processing Johnson et al. found activation abnormalities in terms of significantly reduced activations in the ACC, the left DLPFC, and several subcortical regions in patients already during encoding of verbal stimulus material. These results add to the vast amount of episodic memory studies which show deficits in verbal encoding in association with a reduced cerebral activation in patients (Hofer et al., 2003, Jessen et al., 2003, Ragland et al., 2001, Ragland et al., 2004). Hence, there is reason to assume that working memory deficits in patients might manifest already during encoding of information.

The present fMRI study focused on the executive cognitive control functions in schizophrenia and sought to differentially examine which of the WM subprocesses are affected in schizophrenia. To achieve this goal, a modified Sternberg task arranged as an event-related design differentiating between encoding, active maintenance and executive processing was employed. In order to acquire stable and representative data, relatively large samples of patients and controls were examined. In correspondence with results by Johnson et al. (2006) we expected to find activation differences in terms of hypoactivations in task-related regions (predominantly anterior cingulate and dorsolateral prefrontal) in the patient group already during encoding of stimulus material.

We hypothesized that these encoding-related activation abnormalities, predominantly in the ACC, might be related to the integrity of the subsequent manipulation-associated activation during executive control processing.

Section snippets

Subjects

The study included 41 right-handed patients (28 male, 13 female) with a DSM-IV diagnosis of schizophrenia and 41 right-handed healthy subjects (27 male, 14 female). Handedness was assessed by the Edinburgh Inventory (Oldfield, 1971). On average, patients were 30.17 ± 9.46 years old and had a mean own education of 11.12 ± 1.98 years and a mean parental education of 10.87 ± 1.83 years. In the healthy controls mean age was 29.17 ± 8.85, mean own education 12.15 ± 1.46 years and mean parental education 11.05 ±

Behavioral data

The repeated measures ANOVA on the mean reaction times yielded a significant main effect for group (F(1, 80) = 13.10, p = 0.001) indicating significantly worse reaction time in patients, no significant main effect for task (F(1, 80) = 0.78, p = 0.38), and no significant interaction between task and group (F(1, 80) = 0.03, p = 0.87). The non-parametric Mann–Whitney test for the percentage of correct reactions revealed a significantly worse performance in patients as compared to controls for both task

Discussion

The aim of the present study was to examine the functional neuroarchitecture subserving processes of executive control in schizophrenic patients and to investigate how these activation patterns relate to the preceding encoding phase.

Acknowledgments

This work was supported by an IZKF program grant of the Friedrich-Schiller-University of Jena (to RS and HS)/TMWFK (B307-04004) and the Bundesministerium für Bildung und Forschung BMBF (Core-Unit MRTMethodik, 01ZZ0405). IN was supported by a Junior Scientist Grant (Nachwuchsförderung) of the IZKF (Interdisziplinäres Zentrum für Klinische Forschung), Jena.

References (57)

  • R.C. Oldfield

    The assessment and analysis of handedness: The Edinburgh inventory

    Neuropsychologia

    (1971)
  • G. Repovs et al.

    The multi-component model of working memory: Explorations in experimental cognitive psychology

    Neuroscience

    (2006)
  • R. Schlosser et al.

    Altered effective connectivity during working memory performance in schizophrenia: A study with fMRI and structural equation modeling

    Neuroimage

    (2003)
  • R. Schlosser et al.

    White matter abnormalities and brain activation in schizophrenia: A combined DTI and fMRI study

    Schizophrenia Research

    (2007)
  • R. Schlosser et al.

    Assessing the working memory network: Studies with functional magnetic resonance imaging and structural equation modeling

    Neuroscience

    (2006)
  • F. Schneider et al.

    Neural correlates of working memory dysfunction in first-episode schizophrenia patients: An fMRI multi-center study

    Schizophrenia Research

    (2007)
  • H.W. Thermenos et al.

    The effect of working memory performance on functional MRI in schizophrenia

    Schizophrenia Research

    (2005)
  • D.J. Veltman et al.

    Maintenance versus manipulation in verbal working memory revisited: An fMRI study

    Neuroimage

    (2003)
  • A.D. Wagner et al.

    Prefrontal contributions to executive control: fMRI evidence for functional distinctions within lateral Prefrontal cortex

    Neuroimage

    (2001)
  • T.S. Woodward et al.

    Functional connectivity reveals load dependent neural systems underlying encoding and maintenance in verbal working memory

    Neuroscience

    (2006)
  • N.C. Andreasen et al.

    Schizophrenia and cognitive dysmetria: A positron-emission tomography study of dysfunctional prefrontal–thalamic–cerebellar circuitry

    Proceedings of the National Academy of Sciences of the United States of America

    (1996)
  • J. Andrews et al.

    Abnormalities of thalamic activation and cognition in schizophrenia

    American Journal of Psychiatry

    (2006)
  • M. Brett et al.

    The problem of functional localization in the human brain

    Nature Reviews Neuroscience

    (2002)
  • R.L. Buckner et al.

    Frontal cortex contributes to human memory formation

    Nature Neuroscience

    (1999)
  • J.H. Callicott et al.

    Physiological dysfunction of the dorsolateral prefrontal cortex in schizophrenia revisited

    Cerebral Cortex

    (2000)
  • J.H. Callicott et al.

    Complexity of prefrontal cortical dysfunction in schizophrenia: More than up or down

    American Journal of Psychiatry

    (2003)
  • T.D. Cannon et al.

    Dorsolateral prefrontal cortex activity during maintenance and manipulation of information in working memory in patients with schizophrenia

    Archives of General Psychiatry

    (2005)
  • B. Crespo-Facorro et al.

    Recalling word lists reveals “cognitive dysmetria” in schizophrenia: A positron emission tomography study

    American Journal of Psychiatry

    (1999)
  • Cited by (71)

    • Frontotemporal thalamic connectivity in schizophrenia and schizotypal personality disorder

      2022, Psychiatry Research - Neuroimaging
      Citation Excerpt :

      Our results are broadly consistent with prior studies indicating that individuals with schizophrenia demonstrate lower thalamo-frontal functional connectivity (Anticevic et al., 2014; Woodward and Heckers, 2016; Hua et al., 2019) and a recent activation likelihood meta-analysis reporting thalamocortical hypoconnectivity specifically with middle frontal regions (Ramsay, 2019). A disruption in thalamo-frontal functional connectivity in schizophrenia from our study is consistent with the hypothesis of disturbances in functions associated with top-down control (Schlösser et al., 2008), which could serve as a target for cognitive remediation (Ramsay et al., 2017, 2020). Individuals with schizophrenia had significantly lower functional connectivity from the pulvinar to the transverse temporal cortex (i.e., Heschl's gyrus) compared with HCs.

    • Disrupted modulation of thalamus activation and thalamocortical connectivity during dual task performance in schizophrenia

      2019, Schizophrenia Research
      Citation Excerpt :

      Thalamic hypofunction was more prominent during tasks with higher cognitive demand, such as tasks requiring mental updating or manipulation. In a modified Sternberg task, thalamic hypofunction was observed when manipulation of maintained information was required, but not while passively maintaining information (Schlösser et al., 2008). A recent study found thalamic hypofunction during encoding and maintenance, but not retrieval of spatial information (Choi et al., 2012).

    • Resting-state brain entropy in schizophrenia

      2019, Comprehensive Psychiatry
    • Verbal working memory and functional large-scale networks in schizophrenia

      2017, Psychiatry Research - Neuroimaging
      Citation Excerpt :

      It is also conceivable that the different functional large-scale network used by SZ may reflect a compensatory ‘network’ mechanism which explains the equally high behavioral performance level compared to HC. This interpretation of findings at the network level extends the widely shared notion that reduced DLPFC BOLD response during working memory in SZ may resemble cortical dysfunction (Schlosser et al., 2008) or a compensation mechanism to impaired cognitive function (Tan et al., 2006). Additionally, it is likely that antipsychotic medication may have affected the task-dependent and nonlinear EC findings.

    View all citing articles on Scopus
    View full text