Elsevier

Biological Psychiatry

Volume 53, Issue 6, 15 March 2003, Pages 474-481
Biological Psychiatry

Cerebral function in posttraumatic stress disorder during verbal working memory updating: a positron emission tomography study

https://doi.org/10.1016/S0006-3223(02)01505-6Get rights and content

Abstract

Background

This study examined cerebral function in posttraumatic stress disorder (PTSD) during the updating of working memory to trauma-neutral, verbal information.

Methods

Ten PTSD and matched control subjects completed a visuoverbal target detection task involving continuous updating (Variable target condition) or no updating (Fixed target condition) of target identity, with updating activity estimated by condition comparison.

Results

Normal updating activity using this paradigm involved bilateral activation of the dorsolateral prefrontal cortex (DLPFC) and inferior parietal lobe. The PTSD group lacked this activation in the left hemisphere and was significantly different from control subjects in this regard, but showed additional activation in the superior parietal lobe, bilaterally.

Conclusions

The pattern of parietal activation suggests a dependence on visuospatial coding for working memory representation of trauma-neutral, verbal information. Group differences in the relative involvement of the DLPFC indicate less dependence in PTSD on the executive role normally attributed to the left DLPFC for monitoring and manipulation of working memory content in posterior regions of the brain.

Introduction

Memory functioning is central to understanding the psychopathology of posttraumatic stress disorder (PTSD). The intrusive recall of memories for the traumatic event is one central feature of the disorder. Similarly, cognitive difficulties not related directly to traumatic material have also been noted in studies of PTSD. In general, these difficulties involve learning and memory capabilities that fall broadly within the domain of working memory and related executive function (e.g., Barrett et al 1996, Bremner et al 1993, Moradi et al 1999, Sutker et al 1995, Uddo et al 1993, Yehuda et al 1995. These include abnormalities of attention, controlled learning, verbal fluency, the reaction to novelty, and the monitoring and regulation of memories. Working memory systems are central to the ongoing management of information, being critical to everyday living. They involve the ability to initiate, control, or discontinue action, use information flexibly, make reasonable inferences, think abstractly, respond to novel situations and information, and determine the capacity to direct behavior in a goal-oriented manner (Banich 1997).

Functional neuroimaging studies clearly show that working memory involves the coordinated activity of multiple cortical regions (Clark et al 2000), including the supplementary motor area, inferior parietal lobe (IPL), superior parietal lobe (SPL), dorsolateral prefrontal cortex (DLPFC), anterior cingulate, cerebellum, premotor area, and Broca’s area. The regions most consistently activated include the middle frontal gyrus (MFG) of the DLPFC Braver et al 1997, Cohen et al 1997, McCarthy et al 1993, Paulesu et al 1995, Smith et al 1996 and the supramarginal gyrus (SMG) of the IPL Awh et al 1996, Jonides et al 1997, Paulesu et al 1993, Smith et al 1996. The MFG is consistently involved when active manipulation and monitoring of information is required for the purposes of planned action Owen 1997, Petrides 1995, Petrides et al 1993. The left IPL appears to be important for the updating of verbal perceptual working memory and has been associated with phonological storage Awh et al 1996, Jonides et al 1997, Paulesu et al 1993.

In our previous work in PTSD Galletly et al 2001, McFarlane et al 1993 we used event-related potentials (ERPs) to examine working memory processing of neutral information. These studies have used auditory, oddball-style paradigms that index endogenous (i.e., cognitive) ERP components generated by and indexing attention and related working memory processes in the brain (Hillyard and Kutas 1983). The earlier of these studies (McFarlane et al 1993) demonstrated abnormalities in scalp components (N2 and P3) associated with working memory function. These findings were confirmed in subsequent studies using variants of this or related paradigms Attias et al 1996, Charles et al 1995, Metzger et al 1997. The later study from our group (Galletly et al 2001) specifically examined activity during the updating of working memory and found abnormal frontal slow-wave activity during such activity. This result suggested changes in the assimilation of new information into working memory stores. The present study pursues this line of enquiry with positron emission tomography (PET), using a visuoverbal variant of our earlier paradigms that specifically targeted working memory updating processes to trauma-neutral stimuli.

Interestingly, there has been little neuroimaging work specifically examining the pattern of cerebral activity to such neutral stimuli. Most have addressed cerebral activity in response to emotional challenge and/or symptom provocation and have generally identified abnormal limbic or paralimbic activity in PTSD (e.g., Bremner et al 1999, Liberzon et al 1999, Rauch et al 2000, Sachinvala et al 2000, Semple et al 2000, Shin et al 1997, Shin et al 2001; see also Pitman et al 2001, Villarreal and King 2001). Similarly, there has been little neuroimaging work specifically investigating working memory function to everyday verbal material, though one study (Semple et al 1996) has identified reduced activity in right frontal and parietal cortex during an auditory sustained attention task. The study, however, focused on conjoint PTSD and substance abuse rather than PTSD alone.

The experimental paradigm for the present study was designed to evaluate the processes associated with the updating of working memory content related to trauma-neutral verbal stimuli. There are a number of reasons for selecting verbal stimuli. There are indications of decreased everyday verbal memory abilities for the disorder and an increased dependence on nonverbal working memory brain areas to compensate (Bremner et al 1999). It has also been suggested that memories in PTSD do not have the same narrative quality, tending to be stored more as somatosensory components Rauch et al 1996, van der Kolk and Fisler 1995. It has been shown that the processing of scripts in PTSD is associated with decreased activation of the language area of Broca in the left hemisphere (Rauch et al 1996). From a broader perspective, the verbal modality tends to predominate in moment-to-moment human working memory function and is far more appropriate ecologically than the pure tone stimuli used in ours and a number of other earlier studies of PTSD assessing working memory function.

The paradigm employed was a visuoverbal, target-detection task that permits comparison of activity obtained when target identity changes frequently (Variable condition) with activation obtained when target identity remains invariant (Fixed condition). Both task conditions are identical in terms of stimulus demand, response demand, and target probability, but the Variable condition additionally requires the updating in working memory of target identity following each task stimulus. Thus, an analysis of condition differences arguably taps activity exclusive to this updating process. Analysis of condition differences obtained from normal subjects (Clark et al 1998) has associated significant increases in ERP activity over frontal and parietal regions with the updating of working memory content. With PET, we have shown the source of this activity to be in the supramarginal gyrus of the IPL and the DLPFC regions, bilaterally (Clark et al 2000). In the present study, the primary issue of interest was whether these cortical sources would vary in PTSD, and perhaps reflect the decreased everyday verbal memory abilities clinically evident for PTSD that interfere with the ability to deal efficiently with common, everyday environments.

Section snippets

Subjects

Subjects were 10 participants suffering from PTSD and 10 healthy control subjects recruited from the outpatient departments at the Austin and Repatriation Medical Centre, Melbourne, and the Queen Elizabeth Hospital, Adelaide. All subjects were right handed (Annett 1970) with normal visual acuity and color vision. Patients were matched with control subjects on age, gender, premorbid intelligence (National Adult Reading Test; Nelson 1992), and years of education. All PTSD subjects had a current

Clinical

All patients had chronic PTSD as a result of trauma experienced from motor vehicle accidents (three patients), police duty (three patients), war service (two patients), a farm accident (one patient), or a natural disaster (one patient), and with higher general psychological distress (GHQ), depressive symptoms (CIDI), and trait anxiety (Spielberger Trait) than the control subjects (Table 1). The mean time (± SD) since trauma was 9.9 ± 8.8 years.

The PTSD subjects had high levels of PTSD

Discussion

This study revealed differences in the brain regions recruited by PTSD subjects and their control subjects when challenged with a visuoverbal task requiring the updating of working memory content. Areas recruited by control subjects included the supramarginal gyrus of the IPL, bilaterally, and bilateral regions of the DLPFC. Areas recruited by PTSD subjects included the right supramarginal gyrus of the IPL, the right MFG, the posterior SPL, bilaterally, and in the right supplementary motor

Acknowledgements

This research was supported by Veterans Affairs Grant 952929 and the National Health and Medical Research Council Neuroimaging Consortium.

The authors thank Professor Graham Burrows and the Austin Repatriation Hospital for facilitating the collection of PET data.

References (67)

  • L.M. Shin et al.

    An fMRI study of anterior cingulate function in posttraumatic stress disorder

    Biol Psychiatry

    (2001)
  • T.J. Anderson et al.

    Cortical control of saccades and fixation in man. A PET study

    Brain

    (1994)
  • N.C. Andreassen et al.

    Short term and long term verbal memoryA positron emission tomography study

    Proc Natl Acad Sci U S A

    (1995)
  • M. Annett

    A classification of hand preference by association analysis

    Br J Psychol

    (1970)
  • J. Attias et al.

    Classification of veterans with post-traumatic stress disorder using visual brain evoked P3s to traumatic stimuli

    Br J Psychiatry

    (1996)
  • E. Awh et al.

    Dissociation of storage and rehearsal in verbal working memory

    Psychol Sci

    (1996)
  • M.T. Banich

    NeuropsychologyThe neural bases of mental function

    (1997)
  • D.H. Barrett et al.

    Cognitive functioning and posttraumatic stress disorder

    Am J Psychiatry

    (1996)
  • A.T. Beck et al.

    Beck Depression Inventory Manual

    (1987)
  • D.D. Blake et al.

    A clinician rating scale for assessing current and lifetime PTSDThe CAPS-1

    Behav Ther

    (1990)
  • J.D. Bremner et al.

    Deficits in short-term memory in posttraumatic stress disorder

    Am J Psychiatry

    (1993)
  • K.O. Bushara et al.

    Modality-specific frontal and parietal areas for auditory and visual spatial location in humans

    Nat Neurosci

    (1999)
  • G. Charles et al.

    P300 in posttraumatic stress disorder

    Neuropsychobiology

    (1995)
  • C.R. Clark et al.

    Updating working memory for wordsA PET activation study

    Hum Brain Mapp

    (2000)
  • C.R. Clark et al.

    Working memory updating to visual verbal stimuliA high resolution ERP study

  • J. Clarke et al.

    “R2D2,” a bedside [oxygen-15] water infuser

  • J.D. Cohen et al.

    Temporal dynamics of brain activation during a working memory task

    Nature

    (1997)
  • M. Cohen et al.

    Changes in cortical activity during mental rotationA mapping study using functional MRI

    Brain

    (1996)
  • S.M. Courtney et al.

    Transient and sustained activity in a distributed neural system for human working memory

    Nature

    (1997)
  • A. Daniel

    Power, Privilege and Prestige: Occupations in Australia

    (1983)
  • K.J. Friston et al.

    Statistical parametric maps in functional imagingA general linear approach

    Hum Brain Mapp

    (1995)
  • C.A. Galletly et al.

    Working memory in posttraumatic stress disorder—An event-related potential study

    J Trauma Stress

    (2001)
  • I.M. Harris et al.

    Selective right parietal lobe activation during mental rotationA parametric PET study

    Brain

    (2000)
  • Cited by (111)

    • Association between spatial working memory and Re-experiencing symptoms in PTSD

      2022, Journal of Behavior Therapy and Experimental Psychiatry
    • Emotion-driven impulsivity in PTSD

      2020, Emotion in Posttraumatic Stress Disorder: Etiology, Assessment, Neurobiology, and Treatment
    View all citing articles on Scopus
    View full text