Asymmetrical frontal cortical activity associated with differential risk for mood and anxiety disorder symptoms: An RDoC perspective

Highlights

• We examine frontal EEG asymmetry and mood/anxiety symptoms from an RDoC perspective.

• Decreased left frontal activity is predicted to relate to anhedonia.

• Elevated left frontal activity is predicted to relate to approach hypo/manic symptoms.

• Frontal EEG asymmetry in anxious-arousal versus anxious-apprehension is discussed.

• A motivational, as opposed to valence, based framework is proposed.

Abstract

The recently launched NIMH Research Domain Criteria (RDoC) initiative aims to examine the relationship between core biobehavioral dimensions and symptom profiles that either cut across traditional disorder categories or that are unique to specific clinical phenomenon. A biobehavioral construct that has received considerable attention and that is directly relevant to the Positive Valence Systems domain of the RDoC initiative is approach motivation. One way approach motivation is frequently operationalized is left versus right frontal electroencephalographic (EEG) activity, with greater relative left frontal EEG activity reflecting increased approach motivation and decreased relative left frontal EEG activity reflecting decreased approach motivation or increased withdrawal tendencies. The objective of the present review paper is to examine the relationship between relative left frontal EEG activity and mood and anxiety related symptoms from an RDoC perspective. We first provide an overview of the approach–withdrawal motivational model of frontal EEG asymmetry. Second, we review evidence that relative left frontal EEG activity is associated with a differential risk for unipolar depression versus bipolar disorder. Third, and in line with the mission statement of the RDoC, we move beyond considering mood and anxiety disorders as unitary constructs or homogenous disorders and instead propose that individual differences in relative left frontal EEG activity may be uniquely associated with specific symptom clusters of depression (i.e., anhedonia), hypomania/mania (i.e., symptoms characterized by excessive approach motivation), and anxiety (i.e., anxious apprehension versus anxious arousal). Identifying the relationship between relative left frontal EEG activity and specific mood and anxiety-related symptom clusters has important implications for clinical science, assessment, and treatment.

Introduction

The Diagnostic and Statistical Manual of Mental Disorders (DSM-5th ed.; American Psychiatric Association, 2013) is based on clinical observations and self-reported symptoms. The development of this system, however, predates breakthroughs in neurophysiology and neuroscience, and our reliance on this system may have negative implications for both the assessment and treatment of psychiatric illness. For example, epidemiological data indicate that it takes an average of 6 to 10 years for an individual with bipolar disorder to receive a correct diagnosis and appropriate treatment (Ghaemi et al., 1999, Ghaemi et al., 2000). Those who are misdiagnosed consult an average of four physicians prior to receiving an accurate diagnosis, and close to 60% of individuals with bipolar disorder are initially misclassified as having MDD (Hirschfeld et al., 2003, Nusslock and Frank, 2011). Furthermore, sole reliance on DSM may be impeding research into the pathophysiology of psychiatric symptoms given it is unlikely that the mechanisms underlying these symptoms cleanly map onto DSM classifications.

To help address this issue, the National Institute of Mental Health (NIMH) recently launched the Research Domain Criteria (RDoC) initiative, which calls for the development of new ways of classifying psychiatric illness based on core brain-behavior dimensions (Insel et al., 2010). Rather than start with an illness definition based on clinical observations and then seek its neurophysiological or neurobiological underpinnings, RDoC begins with our current understanding of physiological mechanisms and aims to link these mechanisms to clinical phenomena. The intention of RDoC is to eventually generate a classification system for psychiatric illness that is grounded in contemporary neuroscience. It is argued that this classification system may help generate empirically-derived, biological markers of psychiatric illness that can increase the precision and reliability of psychiatric assessment.¹

In its present form, the RDoC framework involves five domains or dimensions reflecting contemporary knowledge about major systems of cognition, motivation, and behavior. These domains are Negative Valence Systems, Positive Valence Systems, Cognitive Systems, Systems for Social Processes, and Arousal/Regulatory Systems. RDoC specifies multiple Units of Analysis that can be used to examine these domains, including, but not limited to, genes, circuits, physiology, and behavior. One stated goal of RDoC is to identify pathophysiological mechanisms that cut across, or are common to, multiple psychiatric disorders. As an example, elevated threat processing (Negative Valence Systems) is observed across multiple psychiatric disorders, including unipolar depression (Hamilton et al., 2012), bipolar disorder (Phillips and Vieta, 2007, Almeida et al., 2010), and anxiety disorders (Etkin and Wager, 2007). Thus, elevated threat processing may reflect a risk factor for transdiagnostic symptoms that are common across multiple psychiatric conditions.

Another stated goal of RDoC, however, is to identify mechanisms that are unique to specific psychiatric symptoms, and that reflect biosignatures of differential risk for these distinct symptom profiles. Relevant to this goal is growing evidence that certain psychiatric disorders are characterized by distinct and opposite profiles of activation within the Positive Valence Systems. For example, unipolar depression (without a history of hypo/mania) has been associated with abnormally reduced positive emotion or approach motivation (Forbes, 2009, Pizzagalli et al., 2008, Thibodeau et al., 2006), bipolar disorder has been associated with abnormally elevated positive emotion or approach motivation (Alloy and Abramson, 2010, Johnson, 2005, Nusslock et al., 2012a), and certain anxiety symptoms may be associated with elevated or maintained approach motivation (Heller et al., 1997, Mathersul et al., 2008, Nitschke et al., 2009, Guyer et al., 2006). Thus, if one were to look for mechanisms of differential risk for specific psychiatric symptoms, we argue that the Positive Valance Systems are an appropriate target.

A central construct of the Positive Valence Systems domain within the RDoC framework is approach motivation (Insel et al., 2010). Approach motivation involves mechanisms and processes that regulate the direction and maintenance of approach-related behavior. Approach motivation may simply reflect the impulse to go toward and it may be associated with positive or negative emotions, as we discuss later (e.g., Harmon-Jones et al., 2013). Approach behavior can be directed toward innate or acquired cues (i.e., unconditioned vs. learned stimuli), external (e.g., an opportunity for promotion) or internal (e.g., expectancy of winning an award) stimuli, or toward the removal of goal-obstruction.

To date, one of the most reliable neurophysiological indices of approach motivation involves asymmetrical activity in the alpha frequency band over the frontal cortex (Coan and Allen, 2004, Allen et al., 2004, Davidson, 1995, Davidson, 1998a, Davidson, 1998b, Harmon-Jones et al., 2010). Alpha power is typically operationalized as power between 8 and 13 Hz in adults, although lower frequencies have been examined in children, as these lower frequencies in the developing brain are assumed to be equivalent to adult alpha (see Coan and Allen, 2004 for review). A guiding assumption underlying the interpretation of frontal EEG alpha asymmetry is that alpha power is inversely related to cortical activity; such that greater alpha power is indicative of less neuronal activity and reduced alpha power is indicative of elevated neuronal activity (see Allen et al., 2004, Davidson, 1998b for review). In line with this assumption is research documenting that sensory input shows modality-specific blocking of alpha activity at cortical regions involved in processing such input. For example, whereas visual stimuli block alpha over the occipital cortex, a region that is central to processing visual stimuli, auditory stimuli block alpha more so over the auditory cortex (see Allen et al., 2004, Davidson, 1988 for review).² Investigators conducting frontal EEG asymmetry research often use a difference score or asymmetry index (ln(right) − ln(left) alpha power) to conveniently summarize the relative activity at homologous right hemisphere and left hemisphere electrodes (Allen et al., 2004). Given the inverse relationship between alpha power and cortical activity (Allen et al., 2004, Larson et al., 1998), this asymmetry index provides a unidimensional scale in which greater values indicate increased relative left hemispheric cortical activity and lower values indicate decreased relative left hemispheric cortical activity.

The approach–withdrawal motivational model of frontal EEG asymmetry posits that increased relative left frontal activity indicates a propensity to approach or engage a stimulus, whereas decreased relative left frontal activity indicates a propensity toward reduced approach-related motivation or increased withdrawal motivation (Coan and Allen, 2004, Davidson, 1995, Davidson, 1998a, Davidson, 1998b, Harmon-Jones, 2003a). Furthermore, growing evidence indicates that relative left frontal EEG activity may be associated with differential risk for unipolar depression versus bipolar disorder. Specifically, unipolar depression is associated with reduced approach motivation and decreased relative left frontal activity and bipolar disorder is associated with elevated approach motivation and increased relative left frontal activity (Harmon-Jones et al., 2002, Harmon-Jones et al., 2008, Nusslock et al., 2011, Nusslock et al., 2012b, Thibodeau et al., 2006). By contrast, anxiety appears to be characterized by either elevated or reduced relative left frontal activity, depending on the specific symptom cluster (anxious-apprehension versus anxious-arousal) (Heller et al., 1997, Mathersul et al., 2008, Nitschke et al., 2009, Guyer et al., 2006).

The objective of the present review paper is to examine the relationship between relative left frontal EEG activity and mood and anxiety disorder symptoms from an RDoC perspective. In addition, we also extend previous reviews of frontal EEG asymmetry and psychopathology (e.g., Allen and Reznik, in press, Shankman and Klein, 2003) by examining the relationship between relative left frontal EEG activity and mood disorder symptoms across the entire mood spectrum, from unipolar depression to bipolar disorder. We first provide an overview of the approach–withdrawal motivational model of frontal EEG asymmetry. Second, we review evidence that relative left frontal EEG activity is associated with a differential risk for unipolar depression versus bipolar disorder. Our review of the existing literature on relative left frontal activity in unipolar depression versus bipolar disorder focuses on individuals with a DSM diagnosis given that most of the research to date on this topic has been conducted on diagnosed mood disorder samples. As part of this second aim, we briefly review complimentary neuroimaging research to highlight the fact that at multiple units of analysis, biological indices of approach motivation are associated with differential risk for unipolar depression versus bipolar disorder. Third, we move beyond considering mood and anxiety disorders as unitary constructs or homogenous disorders and instead propose that individual differences in frontal EEG asymmetry may be useful in identifying differential risk for specific clusters of mood and anxiety-related symptoms. This third aim is directly in line with one of the stated goals of the RDoC initiative, which is to identify mechanisms that are uniquely related to specific psychiatric symptoms and that reflect biosignatures of differential risk for these distinct symptom profiles (Insel et al., 2010). Specifically, we predict that a) decreased relative left frontal EEG activity will be most strongly associated with the unipolar depressive symptom of anhedonia; b) elevated relative left frontal EEG activity will be most strongly associated with a cluster of hypomanic/manic symptoms characterized by excessive approach motivation (i.e., elevated energy, increased goal-directed activity, decreased need for sleep, increased confidence, and irritability when goal-pursuit is thwarted); and c) anxious-apprehension and anxious-arousal are characterized by distinct and opposite profiles of relative left frontal EEG activity. Finally, we argue that a motivational based framework organized around whether mechanisms facilitate approach versus withdrawal/inhibitory tendencies may be superior to the valence based framework currently employed by the RDoC initiative, which focuses on whether mechanisms facilitate positive versus negative emotions.