Research reportDeficient inhibition of emotional information in depression
Introduction
The identification of vulnerability factors for the development, maintenance, and recurrence of Major Depressive Disorder (MDD) is an important challenge to psychiatry. This should not be considered merely of importance within the framework of prevention and treatment of single episodes of depression, but is also of high importance in addressing the problem of recurring depression. After all, despite the existing pharmacological and psychological treatments, relapse rates remain very high: over 50% of the depressed patients relapse within 2 years of recovery and over 80% of the depressed patients experience more than one depressive episode (Gotlib et al., 1997, Hollon and Shelton, 2001).
Cognitive psychology has a research tradition of several decades investigating vulnerability for MDD. Cognitive theories postulate that information-processing biases influence the aetiology and maintenance of depression. Traditionally, the presence and automatic re-activation of negative self-referent schemas about the self, the world and the future is often conceived to be of major importance regarding vulnerability. ‘Schemas’ can be seen as memory structures that are developed throughout life based on unique experiences (Beck et al., 1979). The idea that those depressogenic schemas play a major role in the way new information will be processed led to considerable research on the identification of information processing biases in depression (Mathews and MacLeod, 2005). The majority of those information-processing studies have primarily focused on attention and memory. Hereby, it is hypothesized that depressed individuals attend selectively to and have a better memory for schema-congruent as compared to for schema-incongruent information. This assumption is in line with the associative network theory of Bower (1981), in which it is stated that affect acts as a prime by which mood-congruent information is activated and retrieved from memory.
In a meta-analysis of studies examining memory biases in depression, Matt et al. (1992) concluded that depressed people retrieve negative information better than positive information, whereas for never-depressed individuals a better recall of positive stimuli appears to be part of the normative pattern of memory performance.
Concerning attentional biases in MDD, the research data are rather mixed. Research has failed to find automatic, early attentional bias for negative information in depression. However, recent findings suggest that depression is associated with attentional biases if the task that is used involves long stimulus exposure (Mogg and Bradley, 2005). This might indicate that attentional bias in depression appears during later, elaborative stages of information processing, once that information has become the focus of attention (difficulty to disengage from negative information) (Mogg and Bradley, 2005, Koster et al., 2005). At present, there is little explanation for the mechanisms underlying these information-processing characteristics in MDD.
A potentially important construct with high explanatory value for information-processing characteristics in MDD that has received only limited attention is inhibitory functioning with regard to emotional information. More specifically, it has been argued that attention cannot be seen as a unitary concept: Several models suggest that selective attention involves multiple components consisting of at least two different interrelated mechanisms: active selection of relevant information and active inhibition of irrelevant stimuli (Hasher and Zacks, 1988). Until now, there has been no systematic study in which the contribution of each of those components is been investigated in major depression. Yet, the delineation of the underlying mechanism(s) is necessary to improve current treatments taking into account specific aspects of information processing. The concept of inhibition provides useful links between the cognitive theories of depression, general cognitive impairments, and neurological impairments.
It is hypothesized that depression is characterized by dysfunctional inhibition toward negative information, based on several reasons. First, in normal circumstances, the anterior cingulate (AC) regulates the attention–emotion balance by signalling the dorsolateral prefrontal cortex (DLPFC) the need for attention control over task-related competing decisions (MacDonald et al., 2000, Bush et al., 2000). However, depressed individuals show lowered activation in these regions (Rogers et al., 2004), which might relate to the fact that they are not able to gain attentional control over emotional interference. Second, a more clinical indication for this dysfunctional attention inhibition hypothesis can be found in the strong relationship between depression and rumination. Depressive rumination indicates the tendency to think and rethink the causes and the possible consequences of the present affective and physical state (Nolen-Hoeksema, 1991a). To date, rumination is considered the best-proven causal factor related to relapse (Spasojevic and Alloy, 2001). There are theoretical reasons and some empirical data to suggest that impaired inhibitory function might be one of the underlying mechanisms of rumination (Hertel, 1997, Joormann, 2004, Linville, 1996).
The present study is among the first to examine impaired inhibition of negative affect in a clinically depressed sample. In addition, by including a formerly depressed group, this study assessed whether this cognitive factor could be observed in remitted patients and can, thus, be regarded as a vulnerability factor for recurrent depression. Previous research has examined inhibitory function by using the emotional modification of the popular colour naming Stroop task. In this task, the participant's assignment is to identify the colour of a word, while ignoring its semantic content. Patients are often slower to name the colours of words associated with concerns relevant to their clinical condition, which is assumed to reflect the extent to which emotional stimuli ‘capture attentional resources' (Williams et al., 1996). However, an important argument against this task is the fact that it is not possible to differentiate the sub-processes of stimulus selection and subsequent inhibition because the task-relevant (colour of the word) as well as the task-irrelevant information (semantic content) are presented within the same stimulus presentation (Mogg and Bradley, 2005).
An improved paradigm to measure inhibition capacity towards emotional information is an affective modification of the negative priming task (NAP). In this multi-stimulus task, a complete trial includes two separate trials: a prime trial and a probe trial. Prime as well as probe trial consists of a simultaneously presented distracter and target. In the prime as in the probe, participants are required to respond to the target (e.g., the picture in a black frame) by evaluating it as negative or positive, while ignoring (inhibiting) the distracter (e.g., the picture in a grey frame). In the experimental condition, the valence of the distracter in the prime trial corresponds with the valence of the probe-trial target. In the control condition, there is no such similarity between prime and probe. The effect of negative affective priming can be indexed by the degree to which responding to the probe target would be slowed down by the previous prime distracter having the same valence, i.e., responding to the probe target ‘negative face’ would be slower if the previous prime distracter had been a negative face (experimental condition) compared to if the previous prime distracter had been a positive face (control condition). This slowdown is called NAP effect and can be considered as a valid index of inhibitory function towards affective material (Wentura, 1999). Table 1 provides an overview of the different conditions used in NAP.
In a recent study, Joormann (2004) used this task with emotional words as stimuli in a dysphoric and formerly depressed student population. Results were indeed indicative for inhibitory deficits in the processing of emotional material in dysphoria. Moreover, another study on the role of affective interference and inhibition revealed that dysphoric participants exhibited both elevated interference and an inhibitory bias for negative but not for positive distracters (Gotlib et al., 2005).
The aim of our study was to further explore the inhibitory mechanisms in hospitalized patients with MDD and formerly depressed individuals. Based on the negative priming literature and taking into account methodological drawbacks of earlier NAP studies (Joormann, 2004, Wentura, 1999), we developed a modified, pictorial NAP paradigm. We hypothesized that, as compared with a control population, depressed participants will show a reduced NAP effect on negative trials, whereas a normal NAP effect is expected on positive trials. This means that for the control group our a priori hypothesis involves a normal slowdown in response time in the experimental condition as compared with the control condition for both negative and positive trials. For the depressed group, our hypothesis involves no difference or a faster response for negative trials in the experimental condition as compared with the control condition, whereas for positive trials the normal response time slowdown in the experimental condition as compared with the control condition is expected. Moreover, assuming that dysfunctional inhibitory processing of affective material might be a vulnerability factor for depression, the same pattern as hypothesized in the depressed group is expected in the formerly depressed group. Furthermore, we wanted to investigate the relationship between inhibitory dysfunction and rumination, hypothesizing that inhibitory function for negative material and the degree of rumination would be inversely correlated. In addition, we assessed the predictive value of the NAP effect and rumination tendency with regard to the severity of depressive symptoms.
Section snippets
Participants
Participants were 60 Dutch-speaking adults between the ages of 19 and 67 years (mean = 43.2; S.D. = 12.0): 20 of them were in-patients of a psychiatric department of a university hospital, diagnosed by a psychiatrist according to the DSM-IV-TR criteria for a MDD. Furthermore, 20 formerly depressed patients were recruited through psychotherapists. All participants in this group met two criteria: They experienced at least two depressive episodes in the past, with the latest occurrence at least 8
Group characteristics
The depressed, formerly depressed and control group did not differ significantly with respect to age (F = 2.21, df = 2, 58, p = 0.118) or male/female ratio, (χ2 = 0.57, df = 2, 58, p = 0.75). As expected, one-way ANOVAs yielded significant differences among the three groups on the BDI-II (F = 38.9, df = 2, 58, p < 0.001), RRS (F = 15.8, df = 2, 58, p < 0.001) and STAI-II scores, (F = 27.1, df = 2, 58, p < 0.001). See Table 2 for an overview.
Statistical NAP analysis
The responses to the prime and the probe trials were recorded, but only the
Discussion
The major aim of this study was to explore inhibitory mechanisms towards affective material in formerly depressed and in hospitalized patients with MDD. The following effects were observed: First, we demonstrated that never-depressed individuals were able to successfully inhibit affective information. As predicted, depressed patients showed a reduced and even reversed NAP effect on negative trials, indicating a less effective inhibition towards negative information. This effect was specific for
Acknowledgments
This work was funded by the Special Research Fund of Ghent University (Belgium) and partially supported by ‘Zorgkantoor Waardenland/Midden Holland’, Gorinchem (the Netherlands).
The authors thank Dr. K. Roose from the Jan-Palfijn Depression Clinic in Ghent (Belgium), Dr. H. D'Haenen, Dr. C. Baeken and I. Van Mulders from the Academical Hospital of the Free University of Brussels (Belgium) and Dr. P. Schoof and J. Molenaar from the Albert Schweitzer Hospital in Dordrecht (the Netherlands) for
References (40)
- et al.
Cognitive and emotional influences in anterior cingulate cortex
Trends Cogn. Sci.
(2000) - et al.
Working memory, comprehension and aging: a review and a new view
- et al.
Treatment guidelines for major depressive disorder
Behav. Ther.
(2001) - et al.
Mood-congruent recall of affectively toned stimuli: A meta-analytic review
Clin. Psychol. Rev.
(1992) - et al.
MINI International Neuropsychiatric Schedule: clinical utility and patient acceptance
Eur. Psychiatr.
(2003) - et al.
Executive and prefrontal dysfunction in unipolar depression: a review of neuropsychological and imaging evidence
Neurosci. Res.
(2004) - et al.
Cognitive Therapy of Depression
(1979) - et al.
Manual for the Beck Depression Inventory
(1996) Mood and memory
Am. Psychol.
(1981)- et al.
A simple and sensitive method to measure timing accuracy
Behav. Res. Meth. Instrum. Comput.
(2003)