Long-term stability of frontal electroencephalographic asymmetry in adults with a history of depression and controls

https://doi.org/10.1016/j.ijpsycho.2005.02.008Get rights and content

Abstract

We investigated the stability in resting EEG across a 1- to 3-year interval in 49 adults (33 female and 16 male) with a history of unipolar depression (first onset prior to the age of 14) and 50 controls (33 female and 17 male) with no history of major psychopathology. Current depressive symptoms were quantified by self-report at both assessments. For the entire sample, EEG asymmetry in the alpha range was moderately stable (intraclass correlations between 0.39 and 0.61). Sex, history of depression, depressive symptom severity at Time 2, and change in symptom severity between Time 1 and Time 2 were unrelated to stability of EEG asymmetry. These findings support the view that resting frontal EEG asymmetry reflects a moderately stable individual difference in adults, irrespective of sex and history of depression.

Introduction

Resting frontal EEG asymmetry, the relative imbalance of bioelectrical cerebral activity in the alpha frequency band, recorded from sites on the left and the right frontal scalp, is believed to reflect certain affective styles and motivational biases of individuals (Tomarken and Keener, 1998, Sutton and Davidson, 1997, Davidson et al., 1990, Henriques and Davidson, 1990, Tomarken et al., 1990, Davidson and Tomarken, 1989). Specifically, it has been proposed that individuals exhibiting left frontal EEG asymmetry (greater left versus right frontal brain activity) are more likely to display behaviors associated with approach motivation and positive affect, while those exhibiting right frontal EEG asymmetry (greater right versus left frontal brain activity) are more likely to display behaviors associated with withdrawal and negative affect.

This proposition has received support in studies of both child temperament and adult affective style. Infants exhibiting right frontal EEG asymmetry (in comparison with those with left frontal EEG asymmetry) are more likely to cry and fuss in response to stress (Davidson and Fox, 1989) and are more likely to be described as having behavior problems (Fox et al., 1996). Adults exhibiting greater right versus left frontal brain activity are more likely to rate video segments in a negative fashion, in comparison with subjects displaying greater left versus right frontal brain activity (Tomarken et al., 1990). According to another study, subjects exhibiting greater left versus right frontal brain activity rate themselves higher in behavioral approach relative to behavioral inhibition (Sutton and Davidson, 1997). In both the developmental and adult studies, EEG was recorded during a resting, non-challenge situation (often referred to as baseline).

There also are indications that frontal EEG asymmetry is sensitive to concurrent affective disorder in adults and may characterize youngsters at risk for mood disorder (Tomarken and Keener, 1998). Right frontal EEG asymmetry during baseline has been found in adults with current (Henriques and Davidson, 1991, Miller et al., 2002) and past depression (Henriques and Davidson, 1990, Miller et al., 2002), adolescent children of depressed mothers (Tomarken et al., 2004), and infants of mothers with depressive symptoms (Dawson et al., 1992, Dawson et al., 1999, Field et al., 1995, Jones et al., 1997). Henriques and Davidson, 1990, Henriques and Davidson, 1991 found that right frontal EEG asymmetry persisted even after remission from an acute depressive episode. Thus, the indications are that right frontal EEG asymmetry reflects both current affective state and stable individual differences in the tendency to experience negative affect.

Implicit to the research utilizing frontal EEG asymmetry is the assumption that these measures reflect a stable individual characteristic, or trait. Stability is defined as consistency across longer periods of time such that one assumes to evaluate trait characteristic of the respective measure. However, to the best of our knowledge, few studies have examined the stability of frontal EEG asymmetry.

In those studies that have examined stability, 4 months was the longest test–retest interval (Allen et al., 2004, Hagemann et al., 2002). Across a 4-month retest interval, stability was around 0.60 for frontal EEG asymmetry (re-retest correlations, Hagemann et al., 2002; intraclass correlations, Allen et al., 2004); 60% was attributable to trait-specific effects (Hagemann et al., 2002). In other studies, in which time intervals ranged from 1 to 6 weeks, stability of EEG asymmetry was reported to be in the moderate range (0.51 to 0.74) (Ehrlichman and Wiener, 1979, Tomarken et al., 1992b, Papousek and Schulter, 1998, Debener et al., 2000, Sutton and Davidson, 1997). In this literature, stability was quantified using intraclass correlation (Allen et al., 2004, Sutton and Davidson, 1997, Tomarken et al., 1992b), Pearson's moment correlation coefficient (Debener et al., 2000, Tomarken et al., 1992b, Ehrlichman and Wiener, 1979), and Spearman rank correlation (Papousek and Schulter, 1998).

Studies in which stability of frontal EEG asymmetry was estimated in clinically depressed subjects have yielded contradictory results (Allen et al., 2004, Debener et al., 2000). Debener et al. (2000) reported frontal EEG asymmetry to be unstable (Pearson's correlations: − 0.22 to 0.09) across 3 weeks in a mixed sex sample. In contrast, Allen et al. (2004) found moderate stability (intraclass correlations: 0.61 to 0.72) in frontal EEG asymmetry across 16 weeks in women. Because these relatively brief test–retest durations are well within the mean length of a depressive episode, unreported by Debener et al. (2000), it is possible that stability of symptoms may have contributed to stability of EEG asymmetry. The one study to consider change in clinical state (Allen et al., 2004) found that it was unrelated to change in EEG asymmetry.

Factors that may affect stability include medication, sex of subjects, and handedness. The one study that assessed medicated individuals (Debener et al., 2000) did not investigate stability of EEG asymmetry and medication status. Handedness is an important factor due to the organization of the brain, which led to the evaluation of mean EEG asymmetry in mostly right-handed people. Sex has been shown to influence mean EEG asymmetry (Miller et al., 2002). Among five studies that included both women and men (Ehrlichman and Wiener, 1979, Papousek and Schulter, 1998, Debener et al., 2000, Sutton and Davidson, 1997), only one examined sex differences and found them unrelated to stability (Hagemann et al., 2002).

The goal of the current study was to examine the longer term stability of EEG asymmetry among adults, its relationship to clinical history as well as changes in depressive symptomatology, and possible effects of salient covariates. We tested whether EEG asymmetry represents a stable individual characteristic, which is robust to change in level of depressive symptoms. The current study was part of a larger, ongoing multidisciplinary program project investigating correlates of and risk factors for very early onset mood disorders in young adults. To estimate long-term stability, we measured frontal EEG asymmetry on two occasions across a period of at least 1 year. Following previous literature, both sessions consisted of two resting baseline conditions.

Section snippets

Participants

One hundred and six subjects were selected from a pool of young adults who were enrolled in a longitudinal, multidisciplinary program project on correlates of and risk factors for childhood onset mood disorders. Only those subjects were selected for the present study who participated in the psychophysiological assessment on two occasions at least 1 year apart and had either childhood onset unipolar depression (n = 53) or were normal controls (n = 53). From this initial sample, seven subjects (3

Symptom severity

For COD subjects, depressive symptom severity, assessed by the BDI (see Table 2) ranged from 0 to 45 at Time 1 and from 0 to 27 at Time 2. For controls, BDI scores ranged from 0 to 25 at Time 1, and from 0 to 12 at Time 2. Change in depressive symptom severity (BDI score at Time 2 minus BDI score at Time 1) ranged from − 30 to 8 (MEAN =  3.5) among CODs, and from − 17 to 10 (MEAN =  0.8) among controls. As expected, CODs had significantly higher BDI scores and a wider range in change of BDI than did

Discussion

EEG asymmetry across a 1- to 3-year interval was found to be moderately stable in all regions of interest (mid-frontal, lateral frontal, and parietal region) for participants with and without a history of depression. The magnitude of stability found in the current study is comparable to what has been reported in the literature across short time intervals (e.g., Allen et al., 2004, Tomarken et al., 1992b).

Lack of association with depressive symptom severity found in the current study is

Acknowledgements

The authors owe thanks to the staff of the Affect Analysis Group at the Department of Psychology, University of Pittsburgh, for the quality of psychophysiological data they carefully collected and processed. The authors also extend their gratitude to the staff of the program project MH-56193 for their assistance with participant scheduling, data collection, and data management. Special thanks to Dr. Erika E. Forbes, Dr. Vera Bulaevskaya, and Dr. Tingting “Rachel” Chung for their useful comments

References (48)

  • J.J.B. Allen et al.

    The stability of resting frontal electroencephalographic asymmetry in depression

    Psychophysiology

    (2004)
  • American Electroencephalographic Society

    Guideline thirteen: guidelines for standard electrode position nomenclature

    Journal of Clinical Neurophysiology

    (1994)
  • American Psychiatric Association

    Diagnostic and Statistical Manual of Mental Disorders

    (1980)
  • American Psychiatric Association

    Diagnostic and Statistical Manual of Mental Disorders

    (1994)
  • Y. Benjamini et al.

    Controlling the false discover rate: a practical and powerful approach to multiple testing

    Journal of the Royal Statistical Society. Series B

    (1995)
  • J.C. Cappelleri et al.

    A modified large-sample approach to approximate interval estimation for a particular intraclass correlation coefficient

    Statistics in Medicine

    (2003)
  • J.A. Coan et al.

    Frontal EEG asymmetry as a moderator and mediator of emotion

    Biological Psychology

    (2004)
  • R.J. Davidson

    Cerebral asymmetry, emotion, and affective style

  • R.J. Davidson et al.

    Frontal brain asymmetry predicts infants' response to maternal separation

    Journal of Abnormal Psychology

    (1989)
  • R.J. Davidson et al.

    Laterality and emotion: an electrophysiological approach

  • R.J. Davidson et al.

    Approach-withdrawal and cerebral asymmetry: emotional expression and brain physiology: I

    Journal of Personality and Social Psychology

    (1990)
  • G. Dawson et al.

    Frontal lobe activity and affective behavior of infants of mothers with depressive symptoms

    Child Development

    (1992)
  • G. Dawson et al.

    Infants of depressed mothers exhibit atypical frontal electrical brain activity during interactions with mother and with a familiar, non-depressed adult

    Child Development

    (1999)
  • S. Debener et al.

    Is resting anterior EEG alpha asymmetry a trait marker for depression? Findings for healthy adults and clinically depressed patients

    Neuropsychobiology

    (2000)
  • Cited by (94)

    • Resting frontal EEG asymmetry in adolescents with major depression: Impact of disease state and comorbid anxiety disorder

      2018, Clinical Neurophysiology
      Citation Excerpt :

      Of note, if the EEG was referenced to the linked mastoids (LM) or the average of all EEG leads (AVG), no group differences emerged, indicating the importance of reporting different reference schemes when assessing FAA (Stewart et al., 2010). Further supporting the trait-hypothesis, studies have found no association between FAA and the severity of acute depressive symptoms in patients with a history of depression (Nusslock et al., 2018; Vuga et al., 2006). In this context, it is important to mention that one of these studies reported that the results were influenced by whether or not patients with a history of depression suffered from a comorbid anxiety disorder: Women with a history of MD without comorbid anxiety disorder demonstrated a greater relative right-frontal activity compared to HCs, while this difference was not revealed in women with a history of MD plus comorbid anxiety disorder (Nusslock et al., 2018).

    • Cortical alpha asymmetry at central and posterior – but not anterior – sites is associated with individual differences in behavioural loss aversion

      2018, Personality and Individual Differences
      Citation Excerpt :

      EEG studies have attempted to broadly localise asymmetries in participants' resting state EEG recordings to frontal, central, or posterior sites. Resting state EEG has been found to be a relatively stable marker of behavioural approach and avoidance over a period of months to years (e.g., Brooker, Canen, Davidson, & Goldsmith, 2017; Davidson, 2004; Jones, Field, Davalos, & Pickens, 1997; Tomarken et al., 1992; Vuga et al., 2006). EEG research on behavioural loss aversion, a putative avoidance behaviour, is scarce and reports mixed findings both in terms of the oscillations studied and the locations in which the asymmetries are observed.

    • The effects of gender and depression severity on the association between alpha asymmetry and depression across four brain regions

      2017, Behavioural Brain Research
      Citation Excerpt :

      FLA is defined as the relative difference between the amounts of electrical activity in the left and right frontal lobes during measurements of resting EEG [5,6]. As opposed to healthy non-depressed individuals, who demonstrate approximately equal amounts of electrical activity in both frontal lobes, depressed individuals have been observed to show relative greater electrical activity in the right frontal lobe than in the left frontal lobe [7–11,6,12]. This finding of FLA has been explained by the “approach withdrawal” model [5], in which it is hypothesized that electrical activity in the left prefrontal cortex reflects the behavioural approach system (responsible for engaging with pleasant stimuli) [5] and the activation of the right prefrontal cortex represents the behavioural withdrawal system (responsible for disengaging from or avoiding aversive stimuli) [13].

    View all citing articles on Scopus
    View full text