Elsevier

Schizophrenia Research

Volume 202, December 2018, Pages 129-137
Schizophrenia Research

Neural correlates of dynamic emotion perception in schizophrenia and the influence of prior expectations

https://doi.org/10.1016/j.schres.2018.06.015Get rights and content

Abstract

Impaired emotion perception is a well-established and stable deficit in schizophrenia; however, there is limited knowledge about the underlying aberrant cognitive and brain processes that result in emotion perception deficits. Recent influential work has shown that perceptual deficits in schizophrenia may result from aberrant precision in prior expectations, associated with disrupted activity in frontal regions. In the present study, we investigated the perception of dynamic, multisensory emotion, the influence of prior expectations and the underlying aberrant brain processes in schizophrenia. During a functional Magnetic Resonance Imaging scan, participants completed the Dynamic Emotion Perception task, which induces prior expectations with emotion instruction cues. We delineated neural responses and functional connectivity in whole-brain large-scale networks underlying emotion perception. Compared to healthy individuals, schizophrenia patients had lower accuracy specifically for emotions that were congruent with prior expectations. At the neural level, schizophrenia patients had less engagement of right inferior frontal and parietal regions, as well as right amygdala dysconnectivity during discrimination of emotions congruent with prior expectations. The results indicate that individuals with schizophrenia may have aberrant prior expectations about emotional expressions, associated with under-activity in inferior frontoparietal regions and right amygdala dysconnectivity, which results in impaired perception of emotion.

Introduction

Emotion perception is impaired in schizophrenia (SCZ) with a significant impact on overall functional outcome (Irani et al., 2012). Currently, there is inadequate knowledge of the cognitive and neural processes underlying emotion perception difficulties in SCZ. The ‘Predictive Coding’ theory of perception proposes that our brain continually generates predictive models of the world, based on prior expectations (generated from previous experiences) and new sensory information (Friston and Kiebel, 2009; Friston et al., 2006). Under this theory, emotional information that is congruent with prior expectations is processed more efficiently, as prior expectations direct attentional focus and decrease processing resources (Barbalat et al., 2013; Brown and Brune, 2012). For example, prior expectations have been found to improve speed and accuracy during emotion discrimination in healthy individuals (Barbalat et al., 2013; Dzafic et al., 2016). Within the Predictive Coding theory, SCZ has been conceptualized as a disorder of aberrant precision (certainty) in prior expectations (Sterzer et al., 2018). However, there is conflicting evidence whether patients with SCZ have reduced (Adams et al., 2016; Chambon et al., 2011; Dima et al., 2010) or increased precision in prior expectations (Alderson-Day et al., 2017; Powers et al., 2017; Teufel et al., 2015). Reduced precision in prior expectations can lead to inefficient directing of attention and nosier incoming sensory information. In contrast, increased precision in prior expectations has been implicated in psychotic experiences, such as hallucinations. In the current study, we investigated whether aberrancy in prior expectations leads to impaired recognition of dynamic, audio-visual emotion in patients with SCZ.

Processing of emotions that are congruent with prior expectations is associated with activity in frontal areas (Barbalat et al., 2013) and the amygdala (Dzafic et al., 2016). Aberrant precision in prior expectations in SCZ has been proposed to reflect impaired activity in frontal regions, resulting in aberrant inhibitory top-down influence over primary sensory regions (Adams et al., 2016). This is compatible with the considerable evidence for frontal dysfunction in SCZ (Fan et al., 2013; He et al., 2013; Huang et al., 2010) and hyper-connectivity in sensory regions (Anticevic et al., 2014). In addition to frontal dysfunction in SCZ, several converging lines of evidence have found that deficits in emotion perception are associated with dysconnectivity in functional networks involving the amygdala (Bjorkquist et al., 2016; Das et al., 2007; Mukherjee et al., 2012). However, no study to date has directly explored the neural circuitry underlying aberrant prior expectations in SCZ during emotion perception.

In summary, our aim was to investigate the influence of prior expectations on naturalistic emotion perception in SCZ, and the underlying distinct patterns of brain activity, and functional connectivity with the rAMY. The effect of prior expectations on emotion perception in SCZ has only been investigated using static emotion displays (Barbalat et al., 2012), despite that sensory information in emotional expressions is dynamic in nature and often changes rapidly in social situations. In line with previous studies (Hargreaves et al., 2016; Johnston et al., 2010) we predicted that SCZ patients would have deficits in discriminating dynamic emotion perception in general, with greater deficits compared to healthy controls when emotion perception relies on prior expectations (Chambon et al., 2011); in other words, detecting emotions congruent with prior expectations. At the neural level, we predicted reduced activation in frontal regions in patients with SCZ during emotion perception that is congruent with prior expectations (Anticevic et al., 2014; Barbalat et al., 2013). Finally, we predicted that SCZ patients would have greater difficulty using prior expectation to facilitate efficient emotion perception, as indexed by increased response times, and this would be associated with rAMY dysconnectivity.

Section snippets

Participants

Sixteen, right-handed patients with chronic SCZ (age range = 30–57; mean age = 46.40, SD = 9.43) were recruited from the Queensland Centre for Mental Health Research (QCMHR). Sixteen age and sex matched, right-handed healthy controls (HC; age range = 34–58; mean age = 45.19, SD = 7.92) were recruited from a National Health and Medical Research Council (NHMRC)-funded, population-based Australian sample of individuals as controls for the SCZ participants. The SCZ patients were comprehensively

Behavioural analyses

Discriminability was calculated using d′ scores (d′ = z(Hits) − z(False Alarms)) (Macmillan and Creelman, 1990) to assess emotion discriminability for each participant. For this calculation, we adjusted d′ according to Corwin (1994), where Hit rate = 1 or False alarm = 0. We conducted a two-way factorial ANOVA to investigate differences between SCZ and HC in discriminability for each emotional video (angry, happy, and neutral).

Mean reaction times (RTs) and accuracy percentage from all responses

Behavioural findings – emotion discriminability

A factorial ANOVA 3 (emotion video) × 2 (group) on discriminability d′ showed a significant main effect of group, F(1, 28) = 11.30, p = 0.002, indicating that SCZ have a lower discriminability compared with HC across different emotions. There was also a trend for an interaction between group and emotion, F(2, 56) = 2.65, p = 0.079 (see Fig. 1c).

Behavioural findings – prior expectations on emotion perception

A factorial ANOVA 3 (emotion video) × 2 (congruency) × 2 (group) on percentage accuracy revealed a main effect of group, F(1, 27) = 7.88, p = 0.009, SCZ

Discussion

In the current study, we investigated dynamic emotion perception in patients with SCZ and the influence of prior expectations at the behavioural and neural levels. We identified reduced ability in SCZ patients to identify emotions that were congruently cued (i.e., congruent with prior expectations), as evidenced by poorer accuracy. At the neural level, we found reduced activity in right inferior frontal gyrus (IFG) and inferior parietal lobule, as well as rAMY dysconnectivity during congruent

Funding

This work was supported by the Australian National Health and Medical Research Council [Grant number: 631671] awarded to BM.

Contributors

Author ID designed the paradigm, analysed the data and wrote the first draft of the manuscript. All authors assisted in the design of the paradigm and contributed to and have approved the final manuscript.

Conflict of interest

The authors declare no competing financial interests.

Acknowledgements

The authors thank the participants for their time and acknowledge the invaluable practical support provided by the imaging staff at the Centre for Advanced Imaging.

References (56)

  • M.M. Garvert et al.

    Subcortical amygdala pathways enable rapid face processing

    NeuroImage

    (2014)
  • M.F. Green et al.

    Approaching a consensus cognitive battery for clinical trials in schizophrenia: the NIMH-MATRICS conference to select cognitive domains and test criteria

    Biol. Psychiatry

    (2004)
  • A. Hargreaves et al.

    Detecting facial emotion recognition deficits in schizophrenia using dynamic stimuli of varying intensities

    Neurosci. Lett.

    (2016)
  • X.Q. Huang et al.

    Localization of cerebral functional deficits in treatment-naive, first-episode schizophrenia using resting-state fMRI

    NeuroImage

    (2010)
  • C.L.C. Huang et al.

    Are there differential deficits in facial emotion recognition between paranoid and non-paranoid schizophrenia? A signal detection analysis

    Psychiatry Res.

    (2013)
  • F. Irani et al.

    A meta-analysis of emotion perception and functional outcomes in schizophrenia

    Schizophr. Res.

    (2012)
  • A.R. McIntosh et al.

    Spatial pattern analysis of functional brain images using partial least squares

    NeuroImage

    (1996)
  • P. Mukherjee et al.

    Lower effective connectivity between amygdala and parietal regions in response to fearful faces in schizophrenia

    Schizophr. Res.

    (2012)
  • D.L. Robins et al.

    Superior temporal activation in response to dynamic audio-visual emotional cues

    Brain Cogn.

    (2009)
  • D.V. Sheehan et al.

    The validity of the Mini International Neuropsychiatric Interview (MINI) according to the SCID-P and its reliability

    Eur. Psychiatry

    (1997)
  • B.K. Suarez et al.

    Genomewide linkage scan of 409 European-Ancestry and African American families with schizophrenia: suggestive evidence of linkage at 8p23.3-p21.2 and 11p13.1-q14.1 in the combined sample

    Am. J. Hum. Genet.

    (2006)
  • C. Summerfield et al.

    A neural representation of prior information during perceptual inference

    Neuron

    (2008)
  • R. Underwood et al.

    Psychobiology of threat appraisal in the context of psychotic experiences: a selective review

    Eur. Psychiatry

    (2015)
  • R.A. Adams et al.

    Computational psychiatry: towards a mathematically informed understanding of mental illness

    J. Neurol. Neurosurg. Psychiatry

    (2016)
  • B. Alderson-Day et al.

    Distinct processing of ambiguous speech in people with non-clinical auditory verbal hallucinations

    Brain

    (2017)
  • A. Anticevic et al.

    Characterizing thalamo-cortical disturbances in schizophrenia and bipolar illness

    Cereb. Cortex

    (2014)
  • G. Barbalat et al.

    The influence of prior expectations on facial expression discrimination in schizophrenia

    Psychol. Med.

    (2012)
  • G. Barbalat et al.

    The influence of prior expectations on emotional face perception in adolescence

    Cereb. Cortex

    (2013)
  • Cited by (2)

    • Neural Correlates of Sensory Abnormalities Across Developmental Disabilities

      2018, International Review of Research in Developmental Disabilities
      Citation Excerpt :

      Comparison of the available findings resulted in the discovery of consistent reports of amygdala overgrowth in ASD, while those with ADHD tend to present with structurally normal amygdalae. On the other hand, other studies performed with just subjects with ADHD, as well as SCHZ, have shown amygdala abnormalities (e.g., Cocchi et al., 2012; Dzafic, Burianová, Martin, & Mowry, 2018). It is possible that amygdala deficiencies are common across DDs, but that they are more pronounced in ASD.

    View full text