The effect of emotional context on cognitive inhibition and attentional processing in dissociative identity disorder

https://doi.org/10.1016/j.brat.2004.03.011Get rights and content

Abstract

In light of previous research, the current study tested the hypothesis that dissociative identity disorder (DID) would be characterised by effective cognitive inhibitory functioning when tested in a neutral context, but weakened inhibition when tested in an emotionally negative context. Using a negative priming task (i.e. the flanker task) to assess inhibitory ability 11 DID, 11 depressed, and 11 general population participants were tested in the two differing experimental contexts. The contexts were manipulated by instructions and word stimuli, and following the completion of this task participants completed the Dissociative Experiences Scale and the Schizotypal Personality Scale. DID participants displayed a greater degree of self-reported anxiety in the negative context and as expected displayed a reduction in inhibition in this context but not in the neutral context. The degree of negative priming for the depressed and general population samples remained stable across contexts as did their anxiety levels. The DID sample displayed slower response times to negative compared to neutral words but this attentional bias was not evident for the two comparison groups. The relationship between increased arousal, inhibitory functioning, dissociation and information processing in DID is discussed.

Introduction

Being broadly characterised as breakdowns in integrated psychobiological processing and operating (American Psychiatric Association, 2000; Nijenhuis, van der Hart, & Steele, 2002), episodes of dissociation (e.g., dissociative symptoms) would be expected to have an impact on executive functioning in working memory. Executive functions are involved in attention, coordination of cognitive systems, planning, volition and action (Baddeley, 1996; Lezak, 1995). A fundamental question is whether pathological dissociative symptoms, and those people with a dissociative disorder, display generalised and stable anomalies in executive functioning, or whether given the association between trauma and dissociative symptoms anomalous executive functioning is more state dependent, vacillating between typical and atypical functioning depending on variables such as emotional state or emotional context.

Previous work has shown a relationship between aspects of executive functioning in working memory and dissociation when tested with neutral (Freyd, Martorello, Alvarado, Hayes, & Christman, 1998) and emotional stimuli (e.g., DePrince & Freyd, 1999; Waller et al., 1995). This work has utilised an incongruent colour-word task (i.e., Stroop, 1935) and word naming tasks to examining interference and attentional bias in information processing in non-clinical dissociators. Initial findings showed evidence of general and stable deficits in executive functioning in high non-clinical dissociators in the form of greater Stroop processing interference (using traditional neutral Stroop stimuli) and more attentional bias (i.e., slower processing) to threat words. In a later study of Stroop interference it was revealed that high non-clinical dissociators displayed less interference compared to low dissociators when performing two simultaneous tasks (i.e., dividing attention), and thereby loading working memory (DePrince & Freyd, 1999). Together (see Freyd & DePrince, 2001, for review) these findings suggest that while dissociation may be related to executive deficits when tested under selective attention conditions, it is related to enhanced performance when tested under divided attention conditions. In short, interference effects in high and low non-clinical dissociators display state-like characteristics that are dependent on environmental demands.

Studies with dissociative identity disorder (DID) samples examining cognitive inhibition, an executive process involved in selective attention, have demonstrated differing performances using differing experimental stimuli. With reference to cognitive inhibition, Baddeley (1996) argues that “one important role of the central executive should be to act as an attentional controller, selecting certain streams of incoming information and rejecting others” (p. 8). In line with this function, cognitive inhibition refers to the ability to withhold the mental representations of distracting stimuli from attention so that target stimuli can be fully processed with only limited interference from competing information (Neill, 1977; Tipper, 1985). Cognitive inhibition is a central process in the ability to engage in selective attention. Johnson and Dark (1986, p. 44) describe selective attention as the “differential processing of simultaneous sources of information”. For differential processing of information at least two processes are required: the activation of relevant information and the inhibition of irrelevant information (Kane, May, Hasher, Rahhal, & Stoltzfus, 1997). In the laboratory cognitive inhibitory ability is often assessed via tasks of negative priming. Negative priming refers to the retardation in response time compared to baseline for a target stimulus presented as a distractor in the previous trial.

Using a negative priming task (the flanker task—see method for overview) to infer cognitive inhibitory ability, Dorahy, Irwin, and Middleton (2002, study 2) found no evidence of negative priming (i.e., weakened cognitive inhibition) in a sample of DID participants when words were used as experimental stimuli. However, in studies that used single digit numbers as test stimuli (Dorahy, Irwin, & Middleton, 2004; Dorahy, Middleton, & Irwin, in press), DID samples displayed significant negative priming (i.e., effective cognitive inhibition). Dorahy, Irwin, and Middleton (in press) suggest that word stimuli may create a more anxiety provoking experimental context than number stimuli because of the uncertainty surrounding the possible threatening quality of even emotionally neutral words for individuals with DID. The currently study was designed to directly address this issue by presenting participants with a single digit number negative priming task in two distinct emotional contexts. Single digit numbers were used in the negative priming task because, as mentioned above, DID samples have shown effective cognitive inhibition when using them. The emotional contexts were manipulated through specific experimental instructions and the use of neutral and emotionally charged negative words. In the first test session participants were told they would see neutral words and were presented with a single neutral word between a set of negative priming trials. In the second test session participants completed the same task but half the neutral stimuli were replaced with emotionally charged words and they were told that negative words would be presented. Along with the DID sample, a control sample and non-dissociative psychiatric sample were utilised to provide more experimental control. Given the theoretical and empirical link between reduced negative priming and both dissociation (Dorahy et al., 2002) and schizotypy (e.g., Beech & Claridge, 1987), these constructs were also assessed.

The study aimed to examine cognitive inhibitory ability in both a neutral and emotionally aversive experimental context. From the suggestive but inconclusive indications of previous studies, it was predicted that the DID sample would display significant negative priming (i.e., effective cognitive inhibition) in the neutral context, but no evidence of negative priming (i.e., weakened inhibition) in the negative context. In addition, following the results from Waller et al. (1995) in a non-clinical sample, it was expected that the DID group would display an attentional bias (i.e., slower response times) to negative words compared to neutral words.

Section snippets

Participants

The DID sample comprised 11 participants (10 females; 1 male) with an Axis I diagnosis of DID from both a psychiatrist and an independent structured clinical interview assessment (i.e., Dissociative Disorders Interview Schedule; Ross et al., 1989). Ten of the 11 participants had taken part in at least one other study of information processing conducted several months before. Participants were aged between 20 and 46 years of age (M=33.7 years, SD=9.1) and had been in treatment for DID from

Results

There were no significant differences between groups for age [F(2,30)=1.64] or days between testing [F(2,30)=.96]. In addition, the total daily dosage of psychotropic medication did not differ for the DID and psychiatric comparison samples [t(20)=.07]. A two-way mixed ANOVA (context by group) on anxiety scores showed no main effect for context [F(1,30)=.94] or group [F(2,30)=.07]. However, the interaction between context and group was significant [F(2,30)=6.84, p<.01]. Post-hoc analyses show

Discussion

It was predicted that cognitive inhibitory processes in DID would display state-like characteristics when tested under differing emotional contexts. More specifically, it was expected that a DID sample would display effective inhibitory performance in a neutral, non-threatening context, but when tested in an emotionally negative context inhibitory ability would weakened. Support for this hypothesis was found in the form of significant negative priming in the DID sample in the neutral context,

Acknowledgements

This work was completed while Martin Dorahy was at the School of Psychology, University of New England, Armidale, N.S.W., Australia. The authors would like to thank Dean Davidson for his help in computer programming and an anonymous reviewer for their comments on an earlier draft of this paper.

References (33)

  • M.J. Dorahy et al.

    Assessing markers of working memory function in dissociative identity disorder using neutral stimuliA comparison with clinical and general population samples

    Australian and New Zealand Journal of Psychiatry

    (2004)
  • Dorahy, M. J., Middleton, W., & Irwin, H. J. (in press). Investigating cognitive inhibition in dissociative identity...
  • E.B. Foa et al.

    Processing of threat-related information in rape victims

    Journal of Abnormal Psychology

    (1991)
  • W.N. Francis et al.

    Frequency analysis of English usage

    (1982)
  • J.J. Freyd et al.

    Perspectives on memory for trauma and cognitive processes associated with dissociative tendencies

    Journal of Aggression, Maltreatment, & Trauma

    (2001)
  • J.J. Freyd et al.

    Cognitive environments and dissociative tendenciesPerformance on the standard Stroop task for high versus low dissociators

    Applied Cognitive Psychology

    (1998)
  • Cited by (39)

    • Response inhibition in patients with functional neurological symptom disorder

      2018, Journal of Clinical Neuroscience
      Citation Excerpt :

      Our hypothesis, therefore, was that abnormal inhibition would be found in FNSD and that it might be decreased or increased depending on the phenomenology. We studied response inhibition in patients with FNSD with PNES, FW, or a combination of the two and compared them to an age and sex matched sample of healthy controls using a go-nogo task, a stop-signal reaction time (SSRT) task [15] and a negative priming flanker task [16]. Psychometric tests of depression, anxiety and dissociation were assessed as potential covariates, because depression and anxiety are frequent comorbidities, and FNSD may be a form of dissociative disorder [17–20].

    • Leave everything to the imagination: Cognitive functioning of individuals high in dissociative absorption and imaginative involvement

      2018, Journal of Research in Personality
      Citation Excerpt :

      Specifically, these obstacles may reflect a disruption in the ability to perform a new operation in the face of proactive interference or negative priming. Miyake’s conceptualization echoes the findings of Dorahy et al. (2005, 2006) regarding disruption in the ability to overcome negative priming among individuals with dissociative identity disorder. Hypothesis 3, regarding somewhat “sloppier” performance of absorbers in CF tasks, was supported, in that they tended to have more errors in several tasks.

    View all citing articles on Scopus
    View full text