Elsevier

NeuroImage

Volume 121, 1 November 2015, Pages 217-226
NeuroImage

Affective state and locus of control modulate the neural response to threat

https://doi.org/10.1016/j.neuroimage.2015.07.034Get rights and content

Highlights

  • Individual differences in the learning-related threat response were investigated.

  • The peripheral emotional response varied with threat predictability.

  • Learning-related ventromedial PFC activity varied with locus of control.

  • Threat-related neural activity varied with affectivity in a curvilinear fashion.

Abstract

The ability to regulate the emotional response to threat is critical to healthy emotional function. However, the response to threat varies considerably from person-to-person. This variability may be partially explained by differences in emotional processes, such as locus of control and affective state, which vary across individuals. Although the basic neural circuitry that mediates the response to threat has been described, the impact individual differences in affective state and locus of control have on that response is not well characterized. Understanding how these factors influence the neural response to threat would provide new insight into processes that mediate emotional function. Therefore, the present study used a Pavlovian conditioning procedure to investigate the influence individual differences in locus of control, positive affect, and negative affect have on the brain and behavioral responses to predictable and unpredictable threats. Thirty-two healthy volunteers participated in a fear conditioning study in which predictable and unpredictable threats (i.e., unconditioned stimulus) were presented during functional magnetic resonance imaging (fMRI). Locus of control showed a linear relationship with learning-related ventromedial prefrontal cortex (PFC) activity such that the more external an individual's locus of control, the greater their differential response to predictable versus unpredictable threat. In addition, positive and negative affectivity showed a curvilinear relationship with dorsolateral PFC, dorsomedial PFC, and insula activity, such that those with high or low affectivity showed reduced regional activity compared to those with an intermediate level of affectivity. Further, activity within the PFC, as well as other regions including the amygdala, were linked with the peripheral emotional response as indexed by skin conductance and electromyography. The current findings demonstrate that the neural response to threat within brain regions that mediate the peripheral emotional response is modulated by an individual's affective state as well as their perceptions of an event's causality.

Introduction

The ability to effectively respond to threats in the environment is critical for healthy emotional function. The response to a threat, however, can vary depending on the circumstances in which the threat occurs. For example, an unpredictable threat elicits a larger emotional response than a predictable threat (Knight et al., 2011). The response to threat also varies considerably from one person to another and appears to be influenced by individual differences in emotion-related processes. For example, the emotional response to threat varies with aspects of anxiety (Grillon et al., 1993, Knight et al., 2011, Wood et al., 2012, Wood et al., 2013). Anxious behavior, however, is influenced by a number of characteristics that vary across individuals. For example, locus of control (i.e., the degree to which an individual believes events are internally versus externally controlled) and affective state (i.e., the degree to which an individual experiences positive and negative emotions in daily life) also vary from person-to-person and appear to influence anxious behavior (Chorpita and Barlow, 1998, Gros et al., 2007, Rotter, 1966, Watson et al., 1988). Therefore, individual differences in locus of control and affective state may also explain variability in the response to threat. However, there is limited prior research on whether inter-subject variability in locus of control, positive affect, and negative affect influences the neural processes that mediate expression of the emotional response. Thus, determining whether individual differences in these attributes influence the response to threat would provide important insight into emotion-related processes.

Pavlovian conditioning is a procedure often used to investigate emotional learning and memory processes. During Pavlovian fear conditioning, an originally innocuous cue (conditioned stimulus; CS) is typically paired with an innately aversive stimulus (unconditioned stimulus; UCS) that produces a reflexive response (unconditioned response; UCR). Repeated pairing of the CS and UCS then comes to elicit a conditioned response (CR) in anticipation of the UCS. Expression of a CR is typically used to index learning, while the UCR is often considered an unlearned response. The UCR, however, also shows learning-related modulation. For instance, prior Pavlovian conditioning research has demonstrated learning-related changes in brain and behavioral responses to predictable compared to unpredictable threat (Baxter, 1966, Canli et al., 1992, Dunsmoor et al., 2008). These studies have demonstrated a diminished UCR once the CS–UCS relationship has been learned, a process known as conditioned UCR diminution (Dunsmoor et al., 2008, Kimmel, 1967, Knight et al., 2010, Knight et al., 2011, Wood et al., 2012). Thus, as the threat (i.e., the UCS) becomes predictable, the response to the threat (i.e., the UCR) is modulated (Bouton et al., 2001, Domjan, 2005, Wagner and Brandon, 1989). Interestingly, these studies have also found a relationship between UCR expression and negative affect indexed by the State-Trait Anxiety Inventory (STAI; Spielberger, 1983, Knight et al., 2011, Wood et al., 2012) suggesting that individual differences in emotional disposition modulate the response to threat. However, questions regarding whether the emotional response to threat also varies with locus of control and positive affect, independent of negative affect, remain unanswered.

Previous research has often taken a categorical approach to the study of anxious behavior. For example, clinical research has traditionally investigated groups with versus without anxiety disorders. This line of research has demonstrated important differences in both brain and behavior between patient and healthy control groups (Monk et al., 2006, Phan et al., 2005, Thayer et al., 1996, Zhao et al., 2007). Even in relatively healthy samples, prior work has often separated participants into “high” and “low” anxiety groups. These studies have typically found a larger emotional response in “high” compared to “low” anxiety participants (Cook et al., 1992, Grillon et al., 1993, Smith et al., 2005). This categorical division of “high” and “low” anxiety groups however may not fully capture subtleties in the degree to which individual differences in anxiety impact the emotional response.

Therefore, more recent research has focused on differences between individuals, and has demonstrated that brain and behavioral data vary with indices of anxiety in a graded, rather than all or none, manner (Carre et al., 2013, Knight et al., 2011, Sehlmeyer et al., 2011, Wood et al., 2012). For example, functional magnetic resonance imaging (fMRI) research has shown graded changes in the blood-oxygen-level-dependent (BOLD) response that varies with the negative affect indexed by the STAI (Bishop, 2008, Bishop et al., 2004, Wood et al., 2012, Vytal et al., 2014). Specifically, the BOLD response within prefrontal cortex (PFC), inferior parietal lobule (IPL), and amygdala often shows a linear relationship to STAI scores (Bishop, 2008, Etkin et al., 2004, Klumpp et al., 2011, Wood et al., 2012). Further, functional connectivity studies have found that the connectivity strength between areas that include the PFC and amygdala varies with STAI scores (Wheelock et al., 2014, Vytal et al., 2014). This prior work suggests that negative affect influences brain processes that mediate the emotional response. Prior research, however, has given limited attention to positive affect which varies independently of negative affect, and may also influence the emotional response (Brown et al., 1998, Gros et al., 2007). Thus, determining the impact of individual differences in positive and negative affectivity on BOLD fMRI and behavioral responses to threat may provide new insight into neural processes that mediate the emotional response.

Most prior work has focused on identifying linear relationships between the brain and behavior. However, non-linear brain–behavior relationships have also been observed. For example, prior emotion research has demonstrated a curvilinear relationship between emotional stimuli, psychophysiological responses, and the BOLD response (Bradley et al., 2001a, Bradley et al., 2003, Lang et al., 1998, Wood et al., 2014). It is also well established that there is a curvilinear relationship between emotional arousal and many aspects of cognitive and behavioral performance (Dickman, 2002, Yerkes and Dodson, 1908). In addition, individuals with an “internal” locus of control show increased responses to uncontrollable threats and decreased responses to controllable threats, while those with an “external” locus of control show the opposite pattern (Bollini et al., 2004, Lundberg and Frankenhaeuser, 1978). Further, brain structure also varies with locus of control. For example, hippocampal volume increases as locus of control increases (Pruessner et al., 2005). This suggests that the fMRI signal response may also be influenced by an individual's locus of control. Taken together, these findings suggest that locus of control may also modulate brain and behavioral responses to threat. Thus, brain regions that mediate expression and regulation of emotion may show linear or non-linear relationships with locus of control and affective state.

The present study used a Pavlovian conditioning procedure to investigate the effect of individual differences in locus of control, positive affect, and negative affect on the emotional (i.e., brain and behavioral) response to predictable and unpredictable threats. Previous work has demonstrated differences in brain and behavioral responses to predictable and unpredictable threats (Dunsmoor et al., 2008, Knight et al., 2010, Knight et al., 2011, Wood et al., 2012). Individual differences in locus of control and affective state may influence learning-related changes in the brain and behavioral responses to threat. However, these differences may also impact the response to threat independent of learning. Therefore, the present study focused on both learning-related changes in the response to threat and the response to threat in general. Given the role of the amygdala, hippocampus, PFC, IPL, and insula in emotional processes, we hypothesized that the fMRI signal response would vary linearly or curvilinearly with locus of control, positive affect, and negative affect. Further, we hypothesized that learning-related changes in the neural response would vary (linearly and/or curvilinearly) with individual differences in locus of control, positive affect, and negative affect.

Section snippets

Participants

Thirty-six healthy participants were recruited from the Birmingham-Metropolitan area. All participants provided written informed consent as approved by the University of Alabama at Birmingham Institutional Review Board. Four participants were excluded from all analyses for issues affecting general data quality (e.g., failure to follow instructions, non-responsiveness, and excessive movement). Thus, 32 healthy participants were included in the final analyses (12 females, 20 males; 14 Caucasian,

Self-report questionnaires

PANAS and Rotter's Locus of Control questionnaires were included in linear mixed effects analyses on brain and behavioral data to assess the influence of locus of control, positive affect, and negative affect on these measures. All correlations between locus of control, positive affect, and negative affect were nonsignificant (all p > 0.05; Table S1). Prior to analyses, all predictor variables were mean-centered around zero. Polynomial terms were then created from the mean-centered data. One

Discussion

The ability to effectively respond to threats in our environment is important for healthy emotional function. However, the response to threat varies considerably from person-to-person. This variability may be explained, in part, by individual differences in factors that influence emotional processes, such as locus of control and affective state. However, the neural substrates that mediate the effect of locus of control and affective state on the response to threat are not well characterized.

Conclusions

The present study demonstrates that cognitive-affective predispositions (i.e., locus of control and affective state) modulate threat-related activity within brain regions that mediate the emotional response to threat. Specifically, perceptions of control influenced learning-related activity within the vmPFC. Further, positive and negative affectivity varied with threat-related activation within dlPFC and dmPFC, such that a diminished neural response was observed in those with relatively low and

Acknowledgments

The authors would like to thank Edwin W. Cook III for statistical assistance as well as Josh Shumen for aid in collecting the data. This research was funded by the National Institute of Mental Health of the National Institutes of Health under award number R01-MH098348 (S.M. & D.C.K.) and the University of Alabama at Birmingham, Office for Equity and Diversity's CMFSDP Fellowship (N.G.H.).

References (89)

  • C. Grillon et al.

    Fear-potentiated startle: relationship to the level of state/trait anxiety in healthy subjects

    Biol. Psychiatry

    (1993)
  • F.J. Helmstetter

    Contribution of the amygdala to learning and performance of conditional fear

    Physiol. Behav.

    (1992)
  • R. Kalisch et al.

    Making a mountain out of a molehill: on the role of the rostral dorsal anterior cingulate and dorsomedial prefrontal cortex in conscious threat appraisal, catastrophizing, and worrying

    Neurosci. Biobehav. Rev.

    (2014)
  • L.A. Kilpatrick et al.

    Sex-related differences in amygdala functional connectivity during resting conditions

    NeuroImage

    (2006)
  • D.C. Knight et al.

    The role of the human amygdala in the production of conditioned fear responses

    NeuroImage

    (2005)
  • D.C. Knight et al.

    Learning-related diminution of unconditioned SCR and fMRI signal responses

    Neuroimage

    (2010)
  • U. Lundberg et al.

    Psychophysiological reactions to noise as modified by personal control over noise intensity

    Biol. Psychol.

    (1978)
  • K. Mogg et al.

    A cognitive-motivational analysis of anxiety

    Behav. Res. Ther.

    (1998)
  • K.N. Ochsner et al.

    The cognitive control of emotion

    Trends Cogn. Sci.

    (2005)
  • E.A. Phelps et al.

    Extinction learning in humans: role of the amygdala and vmPFC

    Neuron

    (2004)
  • J.C. Pruessner et al.

    Self-esteem, locus of control, hippocampal volume, and cortisol regulation in young and old adulthood

    NeuroImage

    (2005)
  • O.J. Robinson et al.

    Towards a mechanistic understanding of pathological anxiety: the dorsal medial prefrontal–amygdala ‘aversive amplification’ circuit in unmedicated generalized and social anxiety disorders

    Lancet Psychiatry

    (2014)
  • J.C. Smith et al.

    State anxiety and affective physiology: effects of sustained exposure to affective pictures

    Biol. Psychiatry

    (2005)
  • J.F. Thayer et al.

    Autonomic characteristics of generalized anxiety disorder and worry

    Biol. Psychiatry

    (1996)
  • M.D. Wheelock et al.

    Threat-related learning relies on distinct dorsal prefrontal cortex network connectivity

    NeuroImage

    (2014)
  • K.H. Wood et al.

    Neural mechanisms underlying the conditioned diminution of the unconditioned fear response

    NeuroImage

    (2012)
  • X.H. Zhao et al.

    Altered default mode network activity in patient with anxiety disorders: an fMRI study

    Eur. J. Radiol.

    (2007)
  • J.J. Allen et al.

    Manipulation of frontal EEG asymmetry through biofeedback alters self-reported emotional responses and facial EMG

    Psychophysiology

    (2001)
  • R. Baxter

    Diminution and recovery of the UCR in delayed and trace classical GSR conditioning

    J. Exp. Psychol.

    (1966)
  • S.J. Bishop

    Neural mechanisms underlying selective attention to threat

    Ann. N. Y. Acad. Sci.

    (2008)
  • S.J. Bishop et al.

    Prefrontal cortical function and anxiety: controlling attention to threat-related stimuli

    Nat. Neurosci.

    (2004)
  • T.D. Blumenthal et al.

    Committee report: guidelines for human startle eyeblink electromyographic studies

    Psychophysiology

    (2005)
  • M.E. Bouton et al.

    A modern learning theory perspective on the etiology of panic disorder

    Psychol. Rev.

    (2001)
  • M.M. Bradley et al.

    Emotion and motivation I: defensive and appetitive reactions in picture processing

    Emotion

    (2001)
  • M.M. Bradley et al.

    Emotion and motivation II: sex differences in picture processing

    Emotion

    (2001)
  • M.M. Bradley et al.

    Activation of the visual cortex in motivated attention

    Behav. Neurosci.

    (2003)
  • T.A. Brown et al.

    Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal

    J. Abnorm. Psychol.

    (1998)
  • J.M. Burger

    Desire for control, locus of control, and proneness to depression

    J. Pers.

    (1984)
  • T. Canli et al.

    Potentiation or diminution of discrete motor unconditioned responses (rabbit eyeblink) to an aversive Pavlovian unconditioned stimulus by two associate processes: conditioned fear and a conditioned diminution of unconditioned stimulus processing

    Behav. Neurosci.

    (1992)
  • A. Carre et al.

    Linear association between social anxiety symptoms and neural activations to angry faces: from subclinical to clinical levels

    Soc. Cogn. Affect. Neurosci.

    (2013)
  • S. Chatterjee et al.

    Influential observations, high leverage points, and outliers in linear regression

    Stat. Sci.

    (1986)
  • D.T. Cheng et al.

    Functional MRI of human amygdala activity during Pavlovian fear conditioning: stimulus processing versus response expression

    Behav. Neurosci.

    (2003)
  • D.T. Cheng et al.

    Human amygdala activity during the expression of fear responses

    Behav. Neurosci.

    (2006)
  • D.T. Cheng et al.

    Activity in the human amygdala corresponds to early, rather than late period autonomic responses to a signal for shock

    Learn. Mem.

    (2007)
  • Cited by (23)

    • Thalamic volume and fear extinction interact to predict acute posttraumatic stress severity

      2021, Journal of Psychiatric Research
      Citation Excerpt :

      Among participants experiencing posttraumatic stress symptoms, two distinct brain-behavior profiles emerged, indicative that neither GMV nor FPS data alone sufficiently characterized the risk for posttraumatic symptom development. Our findings that subcortical GMV did not vary with autonomic reactivity are inconsistent with prior human functional (Cheng et al., 2006; Harnett et al., 2015; Knight et al., 2004) and structural (Hartley et al., 2011) MRI research demonstrating relationships between the amygdala and hippocampus and skin conductance responses (SCRs) during fear acquisition (Pohlack et al., 2012). Although both SCR and FPS index psychophysiological arousal, they have differing neural substrates that may lead to different brain-behavior relationships (Abend et al., 2020; Glover et al., 2011; Lindner et al., 2015; Young et al., 2018), potentially underlying the discrepant findings.

    • Anticipatory prefrontal cortex activity underlies stress-induced changes in Pavlovian fear conditioning

      2018, NeuroImage
      Citation Excerpt :

      We hypothesized that the effect of learning on UCR diminution would be differentially mediated by stress reactivity. 120 right-handed volunteers (53 males, 67 females, mean age = 20.15 years, age range 18–23 years) participated in a functional magnetic resonance imaging (fMRI) study using a variant of the Montreal Imaging Stress Task [MIST; (Dedovic et al., 2005; Goodman et al., 2016; Wheelock et al., 2016)] followed by a Pavlovian fear conditioning task (Harnett et al., 2015). All participants provided written informed consent as approved by the University of Alabama at Birmingham Institutional Review Board.

    • Negative affect is related to reduced differential neural responses to social and non-social stimuli in 5-to-8-month-old infants: A functional near-infrared spectroscopy-study

      2018, Developmental Cognitive Neuroscience
      Citation Excerpt :

      In addition, high Negative Affect has been associated with infants being less frequently involved in joint attention, which is considered to be a key factor in infants' social-cognitive development (Salley and Dixon, 2007; Todd and Dixon, 2010; Vaughan et al., 2003; Vaughan Van Hecke et al., 2007). A considerable number of studies have linked temperament, including Negative Affect, to differential activation in, and connectivity between subcortical and cortical brain regions which are part of the “social brain network”, including the amygdala, ACC, medial PFC and posterior STS/TPJ (Ball et al., 2012; Davey et al., 2015; Harnett et al., 2015; Perlman et al., 2015; Whittle et al., 2006; Vrticka et al., 2013). However, these studies almost exclusively involve school-age children, adolescents and adults.

    View all citing articles on Scopus
    View full text