Modulation of the neural network involved in the processing of anger prosody: The role of task-relevance and social phobia

https://doi.org/10.1016/j.biopsycho.2008.01.014Get rights and content

Abstract

Individuals with social phobia display neural hyperactivation towards angry facial expressions. However, it is uncertain whether they also show abnormal brain responses when processing angry voices. In an event-related functional magnetic resonance imaging study, we investigated brain responses to neutral and angry voices in 12 healthy control participants and 12 individuals with social phobia when emotional prosody was either task-relevant or task-irrelevant. Regardless of task, both phobic and non-phobic participants recruited a network comprising frontotemporal regions, the amygdala, the insula, and the striatum, when listening to angry compared to neutral prosody. Across participants, increased activation in orbitofrontal cortex during task-relevant as compared to task-irrelevant emotional prosody processing was found. Compared to healthy controls, individuals with social phobia displayed significantly stronger orbitofrontal activation in response to angry versus neutral voices under both task conditions. These results suggest a disorder-associated increased involvement of the orbitofrontal cortex in response to threatening voices in social phobia.

Introduction

Individuals suffering from social phobia (SP) are excessively concerned about being negatively evaluated by others. Therefore, the hallmark of social phobia has been characterized as “a marked and persistent fear of social or performance situations in which embarrassment may occur” (APA, 2000, p. 450). Etiological models suggest that information-processing biases play a central role for the development and maintenance of the disorder (Beck et al., 1985, Clark and Wells, 1995). Indeed, it has been shown that individuals with SP display an increased sensitivity towards social cues of disapproval by others (for reviews see Heinrichs and Hofman, 2001, Hirsch et al., 2006). Consistent with this notion, functional imaging studies have repeatedly revealed enhanced neural activity to negative facial expressions, such as anger, contempt, disgust, or fear, in socially phobic compared to non-phobic individuals (Amir et al., 2005, Phan et al., 2006, Stein et al., 2002, Straube et al., 2004, Straube et al., 2005, Yoon et al., 2007). Interestingly, there is also initial evidence that brain activation towards threatening faces in social phobics differs most profoundly from healthy controls when facial emotional expressions are task-irrelevant, suggesting an automatic processing of facial anger cues in social phobia (Straube et al., 2004).

Although angry facial expressions are beyond doubt a prominent social cue to reveal a person's disapproval of others, corresponding inferences can also be drawn based on other, non-facial social cues, such as human voices. Surprisingly though, altered neural processing of interpersonal threat cues in social phobia has not yet been investigated with other expressive social signals than faces. It is well known, however, that whenever listeners perceive human voices they simultaneously process variations in duration, amplitude, pitch contour, tempo, and loudness of an utterance. These prosodic features are powerful signals during interpersonal contact since they enable the reliable discrimination and identification of a speaker's emotional state (Banse and Scherer, 1996, Frick, 1985). Angry voices, in particular, signal rejection and threat similar to angry faces (Banse and Scherer, 1996, Frick, 1985).

The neural network underlying the processing of emotional prosody has been studied intensively in previous studies. Originally, the comprehension of emotional prosody has been claimed to be a special domain of the right hemisphere (e.g., Ross, 1981), but growing evidence indicates that a network spatially distributed across both hemispheres contributes to the appropriate comprehension of emotional prosody (e.g., Adolphs et al., 2002, Imaizumi et al., 1997, Kotz et al., 2003, Schirmer and Kotz, 2006, Wildgruber et al., 2005). Above all, a bilateral fronto-temporal pattern of brain activation has been shown to be associated with processing emotional prosody from normal speech (Kotz et al., 2003). While it is assumed that the superior temporal region serves as an input region mainly involved in the fast analysis of acoustic features, subsequent processes which are more complex and integrative in nature such as labelling the affective tone and/or retrieving its reward value, are thought to be accomplished by bilateral frontal brain regions (Ethofer et al., 2006, Hornak et al., 1996, Wildgruber et al., 2004, Wildgruber et al., 2005, Kotz et al., 2003). Additionally, lesion studies suggest that not only damage to the temporal (Adolphs et al., 2002) and the orbitofrontal cortex (OFC, Hornak et al., 2003) is likely to impair the discrimination of emotional prosody, but also lesions of the basal ganglia (Cancelliere and Kertesz, 1990, Peper and Irle, 1997), and the amygdala (Scott et al., 1997, but see Adolphs and Tranel, 1999, Anderson and Phelps, 1998).

In line with these findings from lesion studies, the comprehension of anger prosody has been shown to rely on a neural network including the striatum (Calder et al., 2004), the rolandic operculum/insula region (Kotz et al., 2003), the amygdala (Sander et al., 2005, Scott et al., 1997), the superior temporal sulcus (STS, Grandjean et al., 2005) and the OFC (Sander et al., 2005). Initial evidence suggests that some areas such as the STS and the amygdala respond to anger prosody even when voices have to be ignored while the OFC seems to show increased activation when attention is focused to the utterances (Sander et al., 2005). Even more interestingly, there is also first evidence that the OFC response to anger prosody is correlated with interindividuell differences in proneness to anxiety reactions in general (Sander et al., 2005). Hence, although the neural network in the perception of anger prosody has not yet been investigated in individuals suffering from social phobia, it appears to be likely that an increased sensitivity towards cues of disapproval might be reflected in increased orbitofrontal activation during the comprehension of anger prosody.

Based on the above findings, the present study investigated the role of task-relevance of prosodic information and social phobia on brain activation to angry compared to neutral utterances. It was hypothesized that social phobia is associated with hyperactivation to angry prosody in the OFC, since OFC activation towards angry voices has formerly been shown to correlate with proneness to anxiety reactions in healthy controls (Sander et al., 2005). In addition, given earlier neuroimaging studies targeting angry facial expressions (Phan et al., 2006, Stein et al., 2002, Straube et al., 2004, Straube et al., 2005, Yoon et al., 2007), we also examined whether individuals with SP would show enhanced responses towards angry utterances in areas such as the insula and the amygdala. Finally, by means of task instructions we investigated whether task-relevance of emotional prosody affects differences in brain activation between socially phobic and non-phobic individuals. Compared to healthy controls, we expected individuals with SP to show increased brain responses especially under conditions in which emotional prosody is task-irrelevant. The latter result would support models assuming automatic biases in the processing of phobia-related cues.

Section snippets

Participants

Twelve Caucasian participants diagnosed with generalized social phobia (SP) and 12 Caucasian volunteers free of any psychiatric symptoms (healthy controls, HC) took part in the experiment. SP and HC were matched for age [HC = 24.00 vs. SP = 23.25, t(22) = .58, P > .05], gender (six men and six women per group) and level of education (student population only). All participants provided written informed consent for this study. The study was approved by the Ethics Committee of the University of Jena. For

Performance data

For analysis of accuracy and reaction times (see Table 1 for descriptive data) repeated measures analysis of variance (ANOVA) with two within factors (prosody and task) and one between factor (group) were computed. For accuracy, a main effect of task was found [F(1, 22) = 21.79, P < .05]. Participants were slightly better in the GEN than in the EMO task (99.22 ± 1.28% vs. 97.29 ± 2.86% correct answers), but correct identification was close to ceiling in both tasks (>95%). Further, there was a main

Discussion

In this event-related fMRI-study, we found increased activation to anger versus neutral prosody in several areas implicated in vocal and emotional processing. Both socially phobic and non-phobic participants recruited a network comprising frontotemporal regions, the insula, the amygdala, and the striatum when listening to angry compared to neutral emotional prosody regardless of task condition. Only the left OFC showed a modulation by task condition resulting in increased activation when

Conflict of interest

None.

Acknowledgments

The study was supported by grants of the Deutsche Forschungs Gemeinschaft and by grants from the federal state of Thuringia and the University of Jena awarded to Wolfgang H.R. Miltner and Thomas Straube.

References (59)

  • R.C. Oldfield

    The assessment and analysis of handedness: the Edinburgh inventory

    Neuropsychologia

    (1971)
  • M. Peper et al.

    Categorical and dimensional decoding of emotional intonations in patients with focal brain lesions

    Brain and Language

    (1997)
  • K.L. Phan et al.

    Association between amygdala hyperactivity to harsh faces and severity of social anxiety in generalized social phobia

    Biological Psychiatry

    (2006)
  • D. Sander et al.

    Selective attention modulates orbitofrontal response to emotional prosody

    Neuroimage

    (2005)
  • A. Schirmer et al.

    Beyond the right hemisphere: brain mechanisms mediating vocal emotional processing

    Trends in Cognitive Sciences

    (2006)
  • T. Straube et al.

    Effect of task conditions on brain responses to threatening faces in social phobics: an event-related functional magnetic resonance imaging study

    Biological Psychiatry

    (2004)
  • T. Straube et al.

    Neural mechanisms of automatic and direct processing of phobogenic stimuli in specific phobia

    Biological Psychiatry

    (2006)
  • S. Wiethoff et al.

    Cerebral processing of emotional prosody—influence of acoustic parameters and arousal

    Neuroimage

    (2008)
  • D. Wildgruber et al.

    Identification of emotional intonation evaluated by fMRI

    Neuroimage

    (2005)
  • D. Wildgruber et al.

    Cerebral processing of linguistic and emotional prosody: fMRI studies

    Progress in Brain Research

    (2006)
  • L.M. Williams et al.

    Distinct amygdala-autonomic arousal profiles in response to fear signals in healthy males and females

    Neuroimage

    (2005)
  • R. Adolphs et al.

    Neural systems for recognition of emotional prosody: a 3-D lesion study

    Emotion

    (2002)
  • A.K. Anderson et al.

    Intact recognition of vocal expressions of fear following bilateral lesions of the human amygdala

    Neuroreport

    (1998)
  • A.K. Anderson et al.

    Neural correlates of the automatic processing of threat facial signals

    The Journal of Neuroscience

    (2003)
  • American Psychiatric Association

    Diagnostic and Statistical Manual of Mental Disorders

    (2000)
  • R. Banse et al.

    Acoustic profiles in vocal emotion expression

    Journal of Personality and Social Psychology

    (1996)
  • A.T. Beck et al.

    An inventory for measuring depression

    Archives of General Psychiatry

    (1981)
  • A.T. Beck et al.

    Anxiety Disorders and Phobias: A Cognitive Perspective

    (1985)
  • R.J. Blair et al.

    Impaired social response reversal. A case of acquired sociopathy

    Brain

    (2000)
  • Cited by (79)

    • Tuned to voices and faces: Cerebral responses linked to social anxiety

      2019, NeuroImage
      Citation Excerpt :

      Moreover, brain areas related to attention and emotion regulation (i.e., dorsolateral prefrontal cortex, parietal cortex Brühl et al., 2014; Miskovic and Schmidt, 2012) and, finally, also regions implicated in the general processing of faces (i.e., occipital and occipito-temporal cortex; Brühl et al., 2014; Miskovic and Schmidt, 2012) exhibit increased responses to face cues. For socially threatening vocal expressions the limited number of comparable studies suggests a similar pattern with altered responses to social threat in the orbitofrontal cortex (Quadflieg et al., 2008; Simon et al., 2017), amygdala (Simon et al., 2017) and dorsolateral as well as dorsomedial prefrontal cortex (Kreifelts et al., 2014). However, it remains an open question if SA generally modulates the cerebral sensitivity to voices and faces irrespective of a socially threatening emotion, attitude or intention expressed in these signals.

    • Overestimation of threat from neutral faces and voices in social anxiety

      2017, Journal of Behavior Therapy and Experimental Psychiatry
      Citation Excerpt :

      The examination of decoding accuracy indicated that HSA individuals do not differ from their LSA counterparts for identifying angry and neutral expressions regardless of cue modality. This observation replicates results of prior face-based (Bell et al., 2011; Heuer et al., 2010; Philippot & Douilliez, 2005) and voice-based studies (Quadflieg et al., 2008, 2007). Further, expanding on these uni-modal studies, our results showed that HSA individuals exhibit a preserved ability to correctly decode emotions from facial or prosodic expressions but also from their combination.

    • Brain activation to task-irrelevant disorder-related threat in social anxiety disorder: The impact of symptom severity

      2017, NeuroImage: Clinical
      Citation Excerpt :

      Most functional imaging studies on brain responses during automatic processing of task-irrelevant threat stimuli in SAD presented emotional faces, which were judged with respect to emotion-irrelevant aspects such as gender discrimination (Blair, Shaywitz, et al., 2008; Campbell et al., 2007; Gentili et al., 2008; Stein et al., 2002; Straube et al., 2004). Other studies either used gender judgment on stimuli with emotional prosody (Quadflieg et al., 2008), disorder-related words in grammatical decision (Schmidt et al., 2010) or in an emotional Stroop task (Boehme et al., 2015). These studies particularly reported amygdala hyperactivation and less consistent hyperactivations in the insula, prefrontal regions (e.g. orbitofrontal cortex, dorsolateral prefrontal cortex (dlPFC)), anterior cingulate cortex (ACC), and superior temporal sulcus (STS) in SAD patients, suggesting emotional encoding even when the task does not focus on stimulus valence.

    • Looking at the self in front of others: Neural correlates of attentional bias in social anxiety

      2016, Journal of Psychiatric Research
      Citation Excerpt :

      Given that structural abnormalities of the lateral OFC in patients with SAD and adults with behaviorally inhibited temperament probably originate early in life (Schwartz et al., 2010; Talati et al., 2013), altered responsiveness in this region might begin early before repetitive exposures to social threats. Our findings of ACC and insula activations in patients during exposure to social threats are consistent with those of previous studies (Goldin et al., 2009; Quadflieg et al., 2008; Straube et al., 2005). Considering the role of the rostral ACC in the affective processing of negative information in SAD (Amir et al., 2005), this region would be responsible for distorted responses to perceived negative self-evaluations in socially anxious situations.

    View all citing articles on Scopus
    1

    Present address: School of Psychology, King's College, Old Aberdeen AB24 2UB, Scotland, United Kingdom. Tel.: +44 1224 272252; fax: +44 1224 273426.

    2

    Tel.: +49 3641 9 45 142; fax: +49 3641 9 45 151.

    3

    Tel.: +49 3641 9 35 358; fax: +49 3641 9 35 352.

    4

    Tel.: +49 3641 9 45 140; fax: +49 3641 9 45 142.

    View full text