- Email: [email protected]
Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study
SCHRES-07159; No of Pages 7 Schizophrenia Research xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study Lisa Buchy a,⁎, Mariapaola Barbato a, Carolina Makowski b, Signe Bray c,d, Frank P. MacMaster e,f, Stephanie Deighton a, Jean Addington a a
Hotchkiss Brain Institute, Department of Psychiatry, University of Calgary, Alberta, Canada McGill Centre for Integrative Neuroscience, McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal, Québec, Canada Department of Radiology and Paediatrics, University of Calgary, Alberta, Canada d Child and Adolescent Imaging Research (CAIR) Program, Alberta Children's Hospital Research Institute, Alberta, Canada e Departments of Psychiatry and Pediatrics, University of Calgary, Alberta, Canada f Strategic Clinical Network for Addictions and Mental Health, Alberta, Canada b c
a r t i c l e
i n f o
Article history: Received 28 November 2016 Received in revised form 24 January 2017 Accepted 27 January 2017 Available online xxxx Keywords: Connectivity Facial affect First-episode schizophrenia Magnetic resonance imaging Social cognition
a b s t r a c t People with psychosis show deficits recognizing facial emotions and disrupted activation in the underlying neural circuitry. We evaluated associations between facial emotion recognition and cortical thickness using a correlation-based approach to map structural covariance networks across the brain. Fifteen people with an early psychosis provided magnetic resonance scans and completed the Penn Emotion Recognition and Differentiation tasks. Fifteen historical controls provided magnetic resonance scans. Cortical thickness was computed using CIVET and analyzed with linear models. Seed-based structural covariance analysis was done using the mapping anatomical correlations across the cerebral cortex methodology. To map structural covariance networks involved in facial emotion recognition, the right somatosensory cortex and bilateral fusiform face areas were selected as seeds. Statistics were run in SurfStat. Findings showed increased cortical covariance between the right fusiform face region seed and right orbitofrontal cortex in controls than early psychosis subjects. Facial emotion recognition scores were not significantly associated with thickness in any region. A negative effect of Penn Differentiation scores on cortical covariance was seen between the left fusiform face area seed and right superior parietal lobule in early psychosis subjects. Results suggest that facial emotion recognition ability is related to covariance in a temporal-parietal network in early psychosis. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Social cognition is defined as the mental processes involved in understanding, observing, and interpreting information in one's social environment. Research has established that people with psychosis show difficulties in social cognition, most prominently in recognizing facial emotions (Kohler et al., 2010). These deficits are present at all stages of the illness including the first-episode (Barkl et al., 2014) and are related to poorer social and occupational functioning (Couture et al., 2006; Fett et al., 2011; Irani et al., 2012). Functional magnetic resonance imaging (fMRI) studies in nonclinical subjects have shown that processing of emotional faces in humans activates a network of regions that includes visual (fusiform face area), limbic (amygdala), temporal-parietal, and prefrontal brain areas (Fusar-Poli et al., 2009). Evidence from lesion studies (Adolphs ⁎ Corresponding author at: Mathison Centre for Mental Health Research & Education, University of Calgary, 3280 Hospital Drive NW, Calgary, Alberta T2N 4Z6, Canada. E-mail address: [email protected] (L. Buchy).
et al., 2000; Adolphs et al., 2003) and experiments using transcranial magnetic stimulation (Pitcher et al., 2008) have demonstrated that the face region of the right somatosensory cortex is also critical for accurate facial emotion recognition. A growing literature indicates that people with psychosis show abnormal activation in these brain regions when making judgements about facial emotional expressions, and these deviant activation patterns are thought to contribute to impairments in recognizing facial emotions (Gur et al., 2007; Li et al., 2010; Pinkham et al., 2011). However, the relationship between facial emotion recognition ability and neural structure in people with psychosis is largely understudied, and should be investigated using sophisticated imaging analyses and at differing illness stages such as soon after the onset of a psychosis. Functional imaging data suggests that the “face network” may have developed atypically in this population, which may also be reflected in aberrant structure or structural networks. Recent developments in structural imaging analyses provide an opportunity to study associations between neuroanatomy and behavioral measures, including cognition (Dziobek et al., 2010; Lerch et al., 2006). Cortical thickness measures can be extracted through a fully-automated
http://dx.doi.org/10.1016/j.schres.2017.01.054 0920-9964/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
2
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
measurement of magnetic resonance (MR) images at a subvoxel resolution, and are believed to primarily reflect morphometric gray matter features such as the size, density or arrangement of cells. (Lerch and Evans, 2005; Parent and Carpenter, 1995). Structural covariance analysis of MRbased cortical thickness data can be used to further map inter-regional anatomical networks. This approach allows measurement of cortical thickness/gray matter volume in which areas of the cortex correlate with one another, allowing evaluation of anatomical relationships in the context of large-scale networks. Covariations in gray matter are thought to result from mutually trophic and maturational influences, and have been shown to partially reflect underlying white matter tracts and functional connectivity networks (Alexander-Bloch et al., 2013; Mechelli et al., 2005; Raznahan et al., 2011). A combined approach involving cortical thickness and structural covariance network analyses has the potential to provide information about underlying patterns of structure and network relationships that are associated with facial emotion recognition abilities in people with psychosis, which is lacking in the current literature. The aims of the current study were fourfold. Our first aim was to compare cortical thickness in an early psychosis sample to a historical control group. In line with meta-analytic results (Bora et al., 2011) we hypothesized reduced cortical thickness in early psychosis subjects compared to controls in frontal and temporal regions. Our second aim was to assess structural covariance within intrinsic networks of facial emotion recognition, using the right somatosensory cortex (Pitcher et al., 2008) and bilateral fusiform face areas (Fusar-Poli et al., 2009) as seed regions of interest, in early psychosis subjects vs. controls. We hypothesized that early psychosis subjects would show alterations in network properties compared with controls. Our third aim was to evaluate associations between facial emotion recognition and cortical thickness across the cortical mantle in the early psychosis group, using the right somatosensory cortex and bilateral fusiform face regions as seeds. In line with functional imaging data showing that poorer facial emotion abilities are associated with altered activation in widespread cortical regions in this population (Gur et al., 2007), we hypothesized that early psychosis subjects would show different patterns of associations between cortical thickness and facial emotion recognition ability than controls. Our fourth aim was to evaluate the modulation of facial emotion recognition on structural covariance between thickness in right somatosensory cortex and bilateral fusiform face area seeds and thickness across the brain in early psychosis subjects. Given the limited published research on structural covariance analyses, aim four was exploratory and no hypothesis was put forth.
based on age first and secondly on sex. Unfortunately, the dataset that the historical control group was drawn from was comprised mostly of females over the age of 18 and males under the age of 18. Because we elected to first match on age, this resulted in an over-representation of females in the historical control group. MR data of controls was used in the current study to compare cortical covariance with our early psychosis patients. All participants provided written informed consent and the study was approved by the University of Calgary Conjoint Health Research Ethics Board. 2.2. Measures Demographics recorded included age, sex, years of educations and handedness. Antipsychotic dosage was recorded and computed as chlorpromazine equivalent dosage. Symptom severity was assessed using the positive and negative subscales of the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987). IQ was assessed with a 2subtest form of the Wechsler Abbreviated Scale of Intelligence comprising the Vocabulary and Block Design subtests (Wechsler, 1999). Facial emotion recognition was assessed with the Penn Emotion Recognition task (ER40; Gur et al., 2002) and the Penn Emotion Differentiation task (EDF40; Silver et al., 2002). In these tasks, pictures representing facial expressions are shown in color, with an equal number of male and female faces, and four races represented (Caucasian, African-American, Asian and Hispanic). In the ER40, faces are presented one at a time and participants choose the emotion that is represented from a list of five possibilities (anger, fear, neutral, happy and sad). In the EDF40, two faces are shown and participants indicate which shows an emotion (either happiness or sadness) more intensely. The ER40 yields a total score ranging from 0 to 40, and individual sub-scores for happy, sad, angry, fearful and neutral facial expressions. The EDF40 provides a total score ranging from 0 to 40, and two sub-scores for happy and sad facial expressions. For the current manuscript, we report only total accuracy scores for each of these tasks. 2.3. Study procedures PANSS ratings were conducted by experienced research clinicians. MR scans were performed on the same day that ER40/EDF40 task performance was collected. Participants received monetary remuneration for their participation. 2.4. MRI acquisition
2. Materials and methods 2.1. Participants Fifteen participants with a early psychosis were recruited through the Early Psychosis Treatment Service at Foothills Hospital in Calgary, Alberta, Canada, and all provided MR scans. For this study an early psychosis was defined as being within the first 3 years of receiving an initial diagnosis of psychosis, which was confirmed through chart records. Exclusion criteria were history of neurological disorder, loss of consciousness for more than 5 min, or presence of metal in the body. Twelve participants were taking anti-psychotic medications at the time of the study and three were unmedicated. Although a control group was not recruited for this study, data was available for 15 historical controls from previous studies conducted at the University of Calgary in the research program of the senior author (J.A.). Control subjects could not meet criteria for any prodromal syndrome, any current or past psychotic disorder or a Cluster A personality disorder diagnosis, not have a family history (in first-degree relatives) of any psychotic disorder or any other disorder involving psychotic symptoms. They could not be currently using psychotropic medication. The historical control group was matched to the early psychosis group
MRI scanning was conducted on a 3 Tesla GE Signa scanner with an 8-channel head coil at the Seaman Family MR Research Centre at the University of Calgary. All participants underwent a high-resolution anatomical scan (3D SPGR, 180 slices, FOV = 25.6 cm, 1 × 1 × 1 mm, flip angle = 12°). 2.5. Measurement of cortical thickness A quality control (QC) procedure was carried out by one rater on all raw T1-weighted images to ensure no visible motion artefacts or poor resolution of gray/white matter contrast. The high quality MRIs were then submitted to the CIVET processing pipeline (Version 2.0.0) (http://www.bic.mni.mcgill.ca/ServicesSoftware/CIVET) (Ad-Dab'bagh et al., 2006; Zijdenbos et al., 2002), using the CBRAIN platform (Sherif et al., 2014). Native T1-weighted images were first registered to the ICBM152 template using a linear transformation (Collins et al., 1994; Grabner et al., 2006) and simultaneously corrected for non-uniformity artefacts using N3 (Sled et al., 1998). The transformed images were then segmented into gray matter, white matter, cerebral spinal fluid and background using a neural net classifier (INSECT) (Zijdenbos et al., 2002). Gray matter and white matter surfaces were extracted
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
using the CLASP algorithm (Kabani et al., 2001; MacDonald et al., 2000). A spherical-mesh deformation algorithm was used to produce a surface mesh of 81,924 polygons (40,962 nodes or vertices) for each hemisphere. Nonlinear registration of both cortical surfaces to a highresolution average surface template generated from the ICBM152 data-set was performed to establish inter-subject correspondence of vertices (Lyttelton et al., 2007; Robbins, 2004). Reverse linear transformation of volumes was performed to allow vertex-based corticometric measurements in native space for each subject's MRI (Ad-Dab'bagh et al., 2005). The deformation algorithm first fits the white matter surface and then expands to the outer gray matter and cerebral spinal fluid intersection. From these surfaces, cortical thickness was computed in native space using the t-link method (Lerch and Evans, 2005), which determines the linked distance between the inner and outer cortical surfaces at each of 81,924 vertices. Each participant's cortical thickness map was subsequently blurred using a 30-mm full-width at halfmaximum surface-based diffusion smoothing kernel (Chung et al., 2003). Resultant gray and white matter surfaces were further manually checked for quality and corrected where possible. Specifically, inaccurate extraction of gray/white matter surfaces in close proximity to the ventricles due to a gradient error were corrected in three subjects via in-house scripts. All subjects passed QC procedures. Surfaces were then re-run through specific stages of CIVET corresponding to the stage at which the error was found. Statistics were performed across all 81,924 vertices using the SurfStat toolbox (http://www.math.mcgill.ca/keith/surfstat/) within MATLAB www.mathworks.com, to assess differences in cortical thickness. For each vertex a linear model was fitted to calculate: a) Group differences in cortical thickness Y ~ intercept + β1(Group) + β2(Seed) + β3(Age) + β4(Sex) b) Modulation of cortical thickness with facial emotion recognition measures in early psychosis subjects only Y ~ intercept + β1(ER40/EDF40) + β2(Age) Y represents cortical thickness, and β values represent regression coefficients (where β1 is the regressor of interest, and remaining β are covariates). Sex was not entered as a covariate in model b) above as the early psychosis sample was comprised of a very low proportion of females (13%). For all analyses, a cluster forming threshold was used with random field theory (RFT) applied over a p-uncorrected = 0.005 map (Worsley et al., 2004). This procedure is implemented within SurfStat and limits the chance of reporting a false positive finding to below 0.05. 2.6. Structural covariance analysis Seed-based analysis of structural covariance was conducted using the mapping correlations across cerebral cortex (MACACC) methodology (Lerch et al., 2006), whereby Pearson correlation coefficients were performed between mean cortical thickness of each seed and cortical thickness at all other vertices across the entire brain. To map structural covariance networks involved in facial emotion recognition, the right somatosensory cortex, left fusiform face area and right fusiform face area were selected as seed regions. The right somatosensory cortex region was selected based on published literature showing that repetitive transcranial magnetic stimulation to the face area of the right somatosensory cortex impairs facial emotion recognition (Pitcher et al., 2008). The fusiform face region was selected based on results of a meta-analysis (Fusar-Poli et al., 2009), which a) showed that processing of facial emotions is associated with robust increased activation within the fusiform face region, and b) identified detailed neurofunctional maps for use as normative references in fMRI studies of emotional face processing. Coordinates taken from this meta-analysis and used here were x = 40, y = − 50, z = − 18 for the right fusiform face region, and the left fusiform face region was identified at x = −40, y = −50,
3
z = − 18. The right somatosensory cortex was located at x = 44, y = −9, z = −45 (Pitcher et al., 2008). Seed regions of interest were defined as spheres of 10 mm radius around the coordinates in MNI space. Representative images are shown in Fig. 1. All statistics were run in SurfStat via MATLAB. For each of the three seeds, a linear model was fitted to calculate: a) Group differences in seed-based structural covariance Y ~ intercept + β1(Group) + β2(Seed) + β3(Age) + β4(Sex) + β5(Group ∗ Seed) b) Modulation of seed-based structural covariance by facial emotion recognition in early psychosis only Y ~ intercept + β1(ER40/EDF40) + β2(Seed) + β3(Age) + β4(ER40/ EDF40 ∗ Seed) Each of these models were run separately using the mean cortical thickness of each of the three seed regions (corresponding to “β2(Seed)” in the model). We note that although the current sample size is small, this sample size has been shown to be sensitive to cortical thickness and structural covariance using the same methodologies reported here (Bernhardt et al., 2014). 3. Results 3.1. Sample characteristics Table 1 displays demographic characteristics of the early psychosis and historical control samples, as well as mean ER40 and EDF40 total accuracy scores of the early psychosis group. Control subjects had significantly greater years of education and number of females than early psychosis subjects. 3.2. Group differences in cortical thickness Surface-based measurement of cortical thickness revealed that early psychosis subjects showed significantly thinner cortex in a cluster in the right frontal cortex encompassing the orbitofrontal, medial and middle frontal cortices (p = 0.004) than historical control subjects, and this is shown in Fig. 2. There were no group differences in early psychosis subjects N controls. 3.3. Group differences in seed-based structural covariance Reduced covariance was found overall in early psychosis subjects compared with controls between cortical thickness in the right fusiform face region seed region of interest and cortical thickness in the right orbitofrontal cortex (p = 0.008), and this is shown in Fig. 3. No control N early psychosis group differences were seen using the left fusiform face region or the right somatosensory cortex. No group differences were seen for early psychosis N controls for any seed region. 3.4. Modulation of cortical thickness by facial emotion recognition ability in early psychosis subjects ER40 and EDF40 total accuracy scores were not significantly associated with cortical thickness in any region. 3.5. Modulation of structural covariance networks by facial emotion recognition in early psychosis subjects A negative effect of EDF40 total accuracy scores was seen on structural covariance between thickness in the left fusiform seed and thickness in right supramarginal gyrus (p = 0.046), as shown in Fig. 4. EDF40 total accuracy did not modulate covariance in thickness between either the right fusiform face area or right somatosensory seeds and any other cortical locus.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
4
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
Fig. 1. Representative images of masks of the three seed regions of interest. R = right, D = dorsal, V = ventral. A) right somatosensory cortex, and B) left and right fusiform face area regions. Masks are shown on an inflated view of the brain.
ER40 scores did not significantly modulate covariance between thickness in right somatosensory or bilateral fusiform face region seeds and any cortical region. 4. Discussion In the current study, our early psychosis subjects showed thinner cortex in right orbitofrontal cortex extending into the medial and middle frontal gyri compared to controls. Group differences in seed-based structural MRI covariance showed that early psychosis subjects showed reduced covariation between thickness in the right fusiform face region and thickness in right orbitofrontal cortex compared to controls. In early psychosis subjects, facial emotion recognition ability was not significantly associated with thickness in any cortical area. Lastly, when evaluating the modulation of structural covariance networks by facial emotion recognition ability in early psychosis subjects, a negative effect of EDF40 total accuracy scores was seen in covariance between thickness in the left fusiform face area and thickness in right supramarginal gyrus. Compared to controls, early psychosis subjects showed thinner cortex in right orbitofrontal cortex and medial/middle frontal gyri at a slightly reduced statistical threshold. This finding is consistent with several meta-analyses showing that people with psychosis show reduced
frontal cortical volume (Bora et al., 2011; Ellison-Wright and Bullmore, 2010; Olabi et al., 2011; Palaniyappan et al., 2012; Radua et al., 2012). However, other cortical regions have been implicated in the pathophysiology of psychosis such as the anterior cingulate, superior temporal cortex and insula, that were not seen here. This lack of sensitivity may, in part, relate to the small sample size used in the current study. Covariance networks in the current study were centered on three key regions involved in facial emotion recognition: right somatosensory cortex, left fusiform face area and right fusiform face area. Results indicated that early psychosis subjects showed reduced covariation between thickness in the right fusiform face region and thickness in the right orbitofrontal cortex compared to controls. This network pattern is in agreement with data from previous animal tract-tracing studies (Averbeck and Seo, 2008), and with functional connectivity analyses showing that face processing comprises a distributed cortical network that includes the right fusiform face area and orbitofrontal cortex (Li et al., 2009). Our inter-regional structural covariance data suggests that early psychosis may be characterized by an underlying reduction in brain network integration in facial emotion recognition networks. Our finding is in accord with data from a functional MRI analysis showing that the orbitofrontal cortex is engaged when participants view facial emotion expressions (Faivre et al., 2012). In addition to extending our understanding of structural networks centered on the right fusiform
Table 1 Characteristics of the first-episode psychosis and historical control samples. Variable
Early psychosis (n = 15)
Historical controls (n = 15)
Test statistic
p-Value
Mean (SD) Age (years) Education (years) WASI PANSS positive PANSS negative Antipsychotic dosagea ER40 total accuracy EDF40 total accuracy
22.7 (2.6) 13.4 (2.2) 108.5 (11.0) 14.7 (7.5) 12.6 (3.5) 288.7 (511.9) 31.9 (3.4) 21.5 (5.5)
24.0 (3.6) 15.9 (2.5) 114.3 (12.8) – – – – –
t = 1.15 t = 3.03 t = 1.31
p = 0.27 p = 0.005 p = 0.20
Ʃ2 = 8.89
p = 0.003
13 (86.7) 2 (13.3)
5 (33.3) 10 (66.7) Ʃ2 = 0.00
p = 1.00
1 (6.7) 14 (93.3)
1 (6.7) 14 (93.3)
9 (60) 1 (6.7) 1 (6.7) 4 (26.6)
– – – –
Frequency (%) Sex Male Female Handedness Left Right Diagnosis Schizophrenia Delusional disorder Brief psychotic disorder Psychosis not otherwise specified a
Calculated as chlorpromazine equivalent dosage.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
5
Fig. 2. Random field theory (RFT) corrected map depicting group differences in cortical thickness in early psychosis subjects and historical controls. Participants with an early psychosis showed cortical thinning in right orbitofrontal cortex extending into both medial and middle frontal gyri at a reduced threshold of p = 0.01 uncorrected (df = 28, 1909 vertices, p = 0.004). We note that the left parahippocampal finding is non-significant after RFT correction. No group differences were seen for early psychosis subjects N historical controls.
face region, results suggest that people with a early psychosis show reduced covariance within nodes of the facial emotion processing network. This result may reflect atypical maturational processes (Averbeck and Seo, 2008), which may be secondary to a developmental disturbance of inter-regional brain networks. Cortical thickness did not appear to be associated with facial emotion recognition ability in early psychosis subjects. Most of the existing literature on the neural correlates of facial emotion recognition has employed functional neuroimaging analyses, revealing robust activation in cortical and subcortical foci during successful facial emotion recognition (Fusar-Poli et al., 2009). This combination of observations may suggest that this aspect of cognition may be better captured in a fluid process such as functional MRI or other physiologically based imaging procedures. Seed-based structural covariance analysis showed that poorer facial emotion recognition scores were associated with greater network covariance in thickness between the left fusiform face area seed and
thickness in right superior parietal lobule. This result is concordant with functional neuroimaging analyses showing that the superior parietal lobule is part of a neural network that activates during the processing of human emotional faces (Fusar-Poli et al., 2010; van de Riet et al., 2009). The direction of the effect complements functional MRI data showing that people with schizophrenia show greater neural activation in fusiform gyrus and amygdala during misidentification of facial emotion expressions (Gur et al., 2007). Our results lead to the postulate that greater network covariance between the fusiform face area and superior parietal cortex is coupled with deteriorating performance or failure to discriminate an emotion. The EDF40 captures one's ability to discriminate which of two facial emotion expressions is shown more intensely, whereas the ER40 taps correct identification of an emotion. Thus, a mechanism by which greater covariance could emerge is through mutual activation over a period of time, and this could be related to dysfunctional co-engagement patterns. The covariance pattern in left fusiform face area and superior right parietal cortex may reflect
Fig. 3. Random field theory (RFT) corrected map depicting seed-based structural covariance networks comparing early psychosis and control groups. Reduced covariance was found overall in early psychosis subjects compared with controls between cortical thickness in the right fusiform face region seed and thickness in right orbitofrontal cortex, (df = 24, 995 vertices, p = 0.008). L = lateral, V = ventral, M = medial.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
6
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
Fig. 4. Random field theory (RFT) corrected map depicting modulation of cortical structure by EDF40 total accuracy scores in subjects with early psychosis. Poorer facial emotion recognition abilities, as measured by EDF40 total accuracy scores, were related to increased covariation in thickness between the left fusiform face area seed and thickness in right superior parietal lobule (df = 10, 103 vertices, p = 0.046). L = lateral, C = caudal.
long-term synchronized developmental patterns within the facial emotion recognition network. Limitations include a very low sample size and over-representation of females in the historical control group, and results should be considered rather preliminary. However, the results can be informative for future work by demonstrating that structural covariance analyses are sensitive to robust effects in small samples, and can therefore inform sample size selection for future research. Although we used a seed based structural covariance analysis, which may have introduced bias, this was a hypothesis-driven approach, which has the advantage of reducing the number of multiple comparisons or spatial correlation concerns, increasing the statistical power of the tests. Control participants were recruited from a historical control group, and facial emotion recognition scores were not available for our historical controls. A betweengroup interaction analysis would be able to determine whether the effects in patients are significantly more negative, or whether the ER40/ EDF40 scores have a general negative effect on the covariance irrespective of the diagnostic category. Future studies could benefit from including a measure to control for medication effects, which are suggested to have neurotrophic properties that could affect cortical thickness (Lieberman et al., 2005), and consider how comorbidity with other non-psychotic psychiatric disorders impacts results. In spite of these limitations, these findings have important implications for understanding the neural underpinnings of poor facial emotion recognition in people with psychotic disorders. Conflict of interest All authors declare no conflict of interest. Contributors The first author performed behavioral and imaging analysis and wrote the first version of the manuscript. The second and sixth authors assisted in data collection and organization. The third author assisted with imaging analyses. The fourth and fifth authors assisted in conceptualizing the study. The seventh author oversaw the project from conception to completion and was responsible for all clinical components. All authors have contributed to the writing of the manuscript and approved the final version.
Role of the funding source The Mathison Centre had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Acknowledgements This study was supported by a Mathison Centre Pilot Research Fund Program awarded to Jean Addington, Signe Bray and Frank McMaster. Lisa Buchy is supported by a CIHR fellowship; Mariapaola Barbato is supported by a Mathison Centre postdoctoral fellowship; Carolina Makowski is supported by an FRSQ studentship. The authors thank staff at Seaman's Family MR Research Centre for assistance with data collection. We are thankful for all the people who participated in the study.
References Ad-Dab'bagh, Y., Singh, V., Robbins, S., Lerch, J., Lyttelton, O., Fombonne, E., Evans, A., 2005. Native-space cortical thickness measurement and the absence of correlation to cerebral volume. 11th Annual Meeting of the Organization for Human Brain Mapping (OHBM), Toronto. Ad-Dab'bagh, Y., Einarson, D., Lyttelton, O., Muehlboeck, J.S., Mok, K., Ivanov, O., Vincent, R., Lepage, C., Lerch, J., Fombonne, E., Evans, A.C., 2006. The CIVET image-processing environment: a fully automated comprehensive pipeline for anatomical neuroimaging research. The 12th Annual Meeting of the Organization for Human Brain Mapping (OHBM), Florence, Italy. Adolphs, R., Damasio, H., Tranel, D., Cooper, G., Damasio, A.R., 2000. A role for somatosensory cortices in the visual recognition of emotion as revealed by three-dimensional lesion mapping. J. Neurosci. 20 (7), 2683–2690. Adolphs, R., Tranel, D., Damasio, A.R., 2003. Dissociable neural systems for recognizing emotions. Brain Cogn. 52 (1), 61–69. Alexander-Bloch, A., Raznahan, A., Bullmore, E., Giedd, J., 2013. The convergence of maturational change and structural covariance in human cortical networks. J. Neurosci. 33 (7), 2889–2899. Averbeck, B.B., Seo, M., 2008. The statistical neuroanatomy of frontal networks in the macaque. PLoS Comput. Biol. 4 (4), e1000050. Barkl, S.J., Lah, S., Harris, A.W., Williams, L.M., 2014. Facial emotion identification in earlyonset and first-episode psychosis: a systematic review with meta-analysis. Schizophr. Res. 159 (1), 62–69. Bernhardt, B.C., Valk, S.L., Silani, G., Bird, G., Frith, U., Singer, T., 2014. Selective disruption of sociocognitive structural brain networks in autism and alexithymia. Cerebral Cortex 24 (12), 3258–3267. Bora, E., Fornito, A., Radua, J., Walterfang, M., Seal, M., Wood, S.J., Yucel, M., Velakoulis, D., Pantelis, C., 2011. Neuroanatomical abnormalities in schizophrenia: a multimodal voxelwise meta-analysis and meta-regression analysis. Schizophr. Res. 127 (1–3), 46–57. Chung, M.K., Worsley, K.J., Robbins, S., Paus, T., Taylor, J., Giedd, J.N., Rapoport, J.L., Evans, A.C., 2003. Deformation-based surface morphometry applied to gray matter deformation. NeuroImage 18 (2), 198–213. Collins, D., Peters, T., Evans, A., 1994. An automated 3D non-linear deformation procedure for determination of gross morphometric variability in the human brain. Visualization in biomedical computing. Proc. SPIE 2359, 180–194. Couture, S.M., Penn, D.L., Roberts, D.L., 2006. The functional significance of social cognition in schizophrenia: a review. Schizophr. Bull. 32 (Suppl. 1), S44–S63. Dziobek, I., Bahnemann, M., Convit, A., Heekeren, H.R., 2010. The role of the fusiformamygdala system in the pathophysiology of autism. Arch. Gen. Psychiatry 67 (4), 397–405. Ellison-Wright, I., Bullmore, E., 2010. Anatomy of bipolar disorder and schizophrenia: a meta-analysis. Schizophr. Res. 117 (1), 1–12.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx Faivre, N., Charron, S., Roux, P., Lehericy, S., Kouider, S., 2012. Nonconscious emotional processing involves distinct neural pathways for pictures and videos. Neuropsychologia 50 (14), 3736–3744. Fett, A.K., Viechtbauer, W., Dominguez, M.D., Penn, D.L., van Os, J., Krabbendam, L., 2011. The relationship between neurocognition and social cognition with functional outcomes in schizophrenia: a meta-analysis. Neurosci. Biobehav. Rev. 35 (3), 573–588. Fusar-Poli, P., Placentino, A., Carletti, F., Landi, P., Allen, P., Surguladze, S., Benedetti, F., Abbamonte, M., Gasparotti, R., Barale, F., Perez, J., McGuire, P., Politi, P., 2009. Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. J. Psychiatry Neurosci. 34 (6), 418–432. Fusar-Poli, P., Bhattacharyya, S., Allen, P., Crippa, J.A., Borgwardt, S., Martin-Santos, R., Seal, M., O'Carroll, C., Atakan, Z., Zuardi, A.W., McGuire, P., 2010. Effect of image analysis software on neurofunctional activation during processing of emotional human faces. J. Clin. Neurosci. 17 (3), 311–314. Grabner, G., Janke, A.L., Budge, M.M., Smith, D., Pruessner, J., Collins, D.L., 2006. Symmetric atlasing and model based segmentation: an application to the hippocampus in older adults. Med. Image Comput. Comput. Assist. Interv. 9 (Pt 2), 58–66. Gur, R.C., Sara, R., Hagendoorn, M., Marom, O., Hughett, P., Macy, L., Turner, T., Bajcsy, R., Posner, A., Gur, R.E., 2002. A method for obtaining 3-dimensional facial expressions and its standardization for use in neurocognitive studies. J. Neurosci. Methods 115 (2), 137–143. Gur, R.E., Loughead, J., Kohler, C.G., Elliott, M.A., Lesko, K., Ruparel, K., Wolf, D.H., Bilker, W.B., Gur, R.C., 2007. Limbic activation associated with misidentification of fearful faces and flat affect in schizophrenia. Arch. Gen. Psychiatry 64 (12), 1356–1366. Irani, F., Seligman, S., Kamath, V., Kohler, C., Gur, R.C., 2012. A meta-analysis of emotion perception and functional outcomes in schizophrenia. Schizophr. Res. 137 (1–3), 203–211. Kabani, N., Le Goualher, G., MacDonald, D., Evans, A.C., 2001. Measurement of cortical thickness using an automated 3-D algorithm: a validation study. NeuroImage 13 (2), 375–380. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Kohler, C.G., Walker, J.B., Martin, E.A., Healey, K.M., Moberg, P.J., 2010. Facial emotion perception in schizophrenia: a meta-analytic review. Schizophr. Bull. 36 (5), 1009–1019. Lerch, J.P., Evans, A.C., 2005. Cortical thickness analysis examined through power analysis and a population simulation. NeuroImage 24 (1), 163–173. Lerch, J.P., Worsley, K., Shaw, W.P., Greenstein, D.K., Lenroot, R.K., Giedd, J., Evans, A.C., 2006. Mapping anatomical correlations across cerebral cortex (MACACC) using cortical thickness from MRI. NeuroImage 31 (3), 993–1003. Li, J., Liu, J., Liang, J., Zhang, H., Zhao, J., Huber, D.E., Rieth, C.A., Lee, K., Tian, J., Shi, G., 2009. A distributed neural system for top-down face processing. Neurosci. Lett. 451 (1), 6–10. Li, H., Chan, R.C., McAlonan, G.M., Gong, Q.Y., 2010. Facial emotion processing in schizophrenia: a meta-analysis of functional neuroimaging data. Schizophr. Bull. 36 (5), 1029–1039. Lieberman, J.A., Tollefson, G.D., Charles, C., Zipursky, R., Sharma, T., Kahn, R.S., Keefe, R.S., Green, A.I., Gur, R.E., McEvoy, J., Perkins, D., Hamer, R.M., Gu, H., Tohen, M., Group, H.S., 2005. Antipsychotic drug effects on brain morphology in first-episode psychosis. Arch. Gen. Psychiatry 62 (4), 361–370.
7
Lyttelton, O., Boucher, M., Robbins, S., Evans, A., 2007. An unbiased iterative group registration template for cortical surface analysis. NeuroImage 34 (4), 1535–1544. MacDonald, D., Kabani, N., Avis, D., Evans, A.C., 2000. Automated 3-D extraction of inner and outer surfaces of cerebral cortex from MRI. NeuroImage 12 (3), 340–356. Mechelli, A., Friston, K.J., Frackowiak, R.S., Price, C.J., 2005. Structural covariance in the human cortex. J. Neurosci. 25 (36), 8303–8310. Olabi, B., Ellison-Wright, I., McIntosh, A.M., Wood, S.J., Bullmore, E., Lawrie, S.M., 2011. Are there progressive brain changes in schizophrenia? A meta-analysis of structural magnetic resonance imaging studies. Biol. Psychiatry 70 (1), 88–96. Palaniyappan, L., Balain, V., Liddle, P.F., 2012. The neuroanatomy of psychotic diathesis: a meta-analytic review. J. Psychiatr. Res. 46 (10), 1249–1256. Parent, A., Carpenter, M.B., 1995. Human Neuroanatomy. Williams & Wilkins, Baltimore, MD. Pinkham, A.E., Loughead, J., Ruparel, K., Overton, E., Gur, R.E., Gur, R.C., 2011. Abnormal modulation of amygdala activity in schizophrenia in response to direct- and averted-gaze threat-related facial expressions. Am. J. Psychiat. 168 (3), 293–301. Pitcher, D., Garrido, L., Walsh, V., Duchaine, B.C., 2008. Transcranial magnetic stimulation disrupts the perception and embodiment of facial expressions. J. Neurosci. 28 (36), 8929–8933. Radua, J., Borgwardt, S., Crescini, A., Mataix-Cols, D., Meyer-Lindenberg, A., McGuire, P.K., Fusar-Poli, P., 2012. Multimodal meta-analysis of structural and functional brain changes in first episode psychosis and the effects of antipsychotic medication. Neurosci. Biobehav. Rev. 36 (10), 2325–2333. Raznahan, A., Lerch, J.P., Lee, N., Greenstein, D., Wallace, G.L., Stockman, M., Clasen, L., Shaw, P.W., Giedd, J.N., 2011. Patterns of coordinated anatomical change in human cortical development: a longitudinal neuroimaging study of maturational coupling. Neuron 72 (5), 873–884. Robbins, S.M., 2004. Anatomical Standardization of the Human Brain in Euclidean 3Space and on the Cortical 2-Manifold. School of Computer Science. McGill University, Montreal, p. 315. Sherif, T., Rioux, P., Rousseau, M.E., Kassis, N., Beck, N., Adalat, R., Das, S., Glatard, T., Evans, A.C., 2014. CBRAIN: a web-based, distributed computing platform for collaborative neuroimaging research. Front. Neuroinf. 8 (24). Silver, H., Shlomo, N., Turner, T., Gur, R.C., 2002. Perception of happy and sad facial expressions in chronic schizophrenia: evidence for two evaluative systems. Schizophr. Res. 55 (1–2), 171–177. Sled, J.G., Zijdenbos, A.P., Evans, A.C., 1998. A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans. Med. Imaging 17 (1), 87–97. van de Riet, W.A., Grezes, J., de Gelder, B., 2009. Specific and common brain regions involved in the perception of faces and bodies and the representation of their emotional expressions. Soc. Neurosci. 4 (2), 101–120. Wechsler, D., 1999. Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, San Antonio, TX. Worsley, K.J., Taylor, J.E., Tomaiuolo, F., Lerch, J., 2004. Unified univariate and multivariate random field theory. NeuroImage 23 (Suppl. 1), S189–S195. Zijdenbos, A.P., Forghani, R., Evans, A.C., 2002. Automatic “pipeline” analysis of 3-D MRI data for clinical trials: application to multiple sclerosis. IEEE Trans. Med. Imaging 21 (10), 1280–1291.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study Lisa Buchy a,⁎, Mariapaola Barbato a, Carolina Makowski b, Signe Bray c,d, Frank P. MacMaster e,f, Stephanie Deighton a, Jean Addington a a
Hotchkiss Brain Institute, Department of Psychiatry, University of Calgary, Alberta, Canada McGill Centre for Integrative Neuroscience, McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal, Québec, Canada Department of Radiology and Paediatrics, University of Calgary, Alberta, Canada d Child and Adolescent Imaging Research (CAIR) Program, Alberta Children's Hospital Research Institute, Alberta, Canada e Departments of Psychiatry and Pediatrics, University of Calgary, Alberta, Canada f Strategic Clinical Network for Addictions and Mental Health, Alberta, Canada b c
a r t i c l e
i n f o
Article history: Received 28 November 2016 Received in revised form 24 January 2017 Accepted 27 January 2017 Available online xxxx Keywords: Connectivity Facial affect First-episode schizophrenia Magnetic resonance imaging Social cognition
a b s t r a c t People with psychosis show deficits recognizing facial emotions and disrupted activation in the underlying neural circuitry. We evaluated associations between facial emotion recognition and cortical thickness using a correlation-based approach to map structural covariance networks across the brain. Fifteen people with an early psychosis provided magnetic resonance scans and completed the Penn Emotion Recognition and Differentiation tasks. Fifteen historical controls provided magnetic resonance scans. Cortical thickness was computed using CIVET and analyzed with linear models. Seed-based structural covariance analysis was done using the mapping anatomical correlations across the cerebral cortex methodology. To map structural covariance networks involved in facial emotion recognition, the right somatosensory cortex and bilateral fusiform face areas were selected as seeds. Statistics were run in SurfStat. Findings showed increased cortical covariance between the right fusiform face region seed and right orbitofrontal cortex in controls than early psychosis subjects. Facial emotion recognition scores were not significantly associated with thickness in any region. A negative effect of Penn Differentiation scores on cortical covariance was seen between the left fusiform face area seed and right superior parietal lobule in early psychosis subjects. Results suggest that facial emotion recognition ability is related to covariance in a temporal-parietal network in early psychosis. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Social cognition is defined as the mental processes involved in understanding, observing, and interpreting information in one's social environment. Research has established that people with psychosis show difficulties in social cognition, most prominently in recognizing facial emotions (Kohler et al., 2010). These deficits are present at all stages of the illness including the first-episode (Barkl et al., 2014) and are related to poorer social and occupational functioning (Couture et al., 2006; Fett et al., 2011; Irani et al., 2012). Functional magnetic resonance imaging (fMRI) studies in nonclinical subjects have shown that processing of emotional faces in humans activates a network of regions that includes visual (fusiform face area), limbic (amygdala), temporal-parietal, and prefrontal brain areas (Fusar-Poli et al., 2009). Evidence from lesion studies (Adolphs ⁎ Corresponding author at: Mathison Centre for Mental Health Research & Education, University of Calgary, 3280 Hospital Drive NW, Calgary, Alberta T2N 4Z6, Canada. E-mail address: [email protected] (L. Buchy).
et al., 2000; Adolphs et al., 2003) and experiments using transcranial magnetic stimulation (Pitcher et al., 2008) have demonstrated that the face region of the right somatosensory cortex is also critical for accurate facial emotion recognition. A growing literature indicates that people with psychosis show abnormal activation in these brain regions when making judgements about facial emotional expressions, and these deviant activation patterns are thought to contribute to impairments in recognizing facial emotions (Gur et al., 2007; Li et al., 2010; Pinkham et al., 2011). However, the relationship between facial emotion recognition ability and neural structure in people with psychosis is largely understudied, and should be investigated using sophisticated imaging analyses and at differing illness stages such as soon after the onset of a psychosis. Functional imaging data suggests that the “face network” may have developed atypically in this population, which may also be reflected in aberrant structure or structural networks. Recent developments in structural imaging analyses provide an opportunity to study associations between neuroanatomy and behavioral measures, including cognition (Dziobek et al., 2010; Lerch et al., 2006). Cortical thickness measures can be extracted through a fully-automated
http://dx.doi.org/10.1016/j.schres.2017.01.054 0920-9964/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
2
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
measurement of magnetic resonance (MR) images at a subvoxel resolution, and are believed to primarily reflect morphometric gray matter features such as the size, density or arrangement of cells. (Lerch and Evans, 2005; Parent and Carpenter, 1995). Structural covariance analysis of MRbased cortical thickness data can be used to further map inter-regional anatomical networks. This approach allows measurement of cortical thickness/gray matter volume in which areas of the cortex correlate with one another, allowing evaluation of anatomical relationships in the context of large-scale networks. Covariations in gray matter are thought to result from mutually trophic and maturational influences, and have been shown to partially reflect underlying white matter tracts and functional connectivity networks (Alexander-Bloch et al., 2013; Mechelli et al., 2005; Raznahan et al., 2011). A combined approach involving cortical thickness and structural covariance network analyses has the potential to provide information about underlying patterns of structure and network relationships that are associated with facial emotion recognition abilities in people with psychosis, which is lacking in the current literature. The aims of the current study were fourfold. Our first aim was to compare cortical thickness in an early psychosis sample to a historical control group. In line with meta-analytic results (Bora et al., 2011) we hypothesized reduced cortical thickness in early psychosis subjects compared to controls in frontal and temporal regions. Our second aim was to assess structural covariance within intrinsic networks of facial emotion recognition, using the right somatosensory cortex (Pitcher et al., 2008) and bilateral fusiform face areas (Fusar-Poli et al., 2009) as seed regions of interest, in early psychosis subjects vs. controls. We hypothesized that early psychosis subjects would show alterations in network properties compared with controls. Our third aim was to evaluate associations between facial emotion recognition and cortical thickness across the cortical mantle in the early psychosis group, using the right somatosensory cortex and bilateral fusiform face regions as seeds. In line with functional imaging data showing that poorer facial emotion abilities are associated with altered activation in widespread cortical regions in this population (Gur et al., 2007), we hypothesized that early psychosis subjects would show different patterns of associations between cortical thickness and facial emotion recognition ability than controls. Our fourth aim was to evaluate the modulation of facial emotion recognition on structural covariance between thickness in right somatosensory cortex and bilateral fusiform face area seeds and thickness across the brain in early psychosis subjects. Given the limited published research on structural covariance analyses, aim four was exploratory and no hypothesis was put forth.
based on age first and secondly on sex. Unfortunately, the dataset that the historical control group was drawn from was comprised mostly of females over the age of 18 and males under the age of 18. Because we elected to first match on age, this resulted in an over-representation of females in the historical control group. MR data of controls was used in the current study to compare cortical covariance with our early psychosis patients. All participants provided written informed consent and the study was approved by the University of Calgary Conjoint Health Research Ethics Board. 2.2. Measures Demographics recorded included age, sex, years of educations and handedness. Antipsychotic dosage was recorded and computed as chlorpromazine equivalent dosage. Symptom severity was assessed using the positive and negative subscales of the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987). IQ was assessed with a 2subtest form of the Wechsler Abbreviated Scale of Intelligence comprising the Vocabulary and Block Design subtests (Wechsler, 1999). Facial emotion recognition was assessed with the Penn Emotion Recognition task (ER40; Gur et al., 2002) and the Penn Emotion Differentiation task (EDF40; Silver et al., 2002). In these tasks, pictures representing facial expressions are shown in color, with an equal number of male and female faces, and four races represented (Caucasian, African-American, Asian and Hispanic). In the ER40, faces are presented one at a time and participants choose the emotion that is represented from a list of five possibilities (anger, fear, neutral, happy and sad). In the EDF40, two faces are shown and participants indicate which shows an emotion (either happiness or sadness) more intensely. The ER40 yields a total score ranging from 0 to 40, and individual sub-scores for happy, sad, angry, fearful and neutral facial expressions. The EDF40 provides a total score ranging from 0 to 40, and two sub-scores for happy and sad facial expressions. For the current manuscript, we report only total accuracy scores for each of these tasks. 2.3. Study procedures PANSS ratings were conducted by experienced research clinicians. MR scans were performed on the same day that ER40/EDF40 task performance was collected. Participants received monetary remuneration for their participation. 2.4. MRI acquisition
2. Materials and methods 2.1. Participants Fifteen participants with a early psychosis were recruited through the Early Psychosis Treatment Service at Foothills Hospital in Calgary, Alberta, Canada, and all provided MR scans. For this study an early psychosis was defined as being within the first 3 years of receiving an initial diagnosis of psychosis, which was confirmed through chart records. Exclusion criteria were history of neurological disorder, loss of consciousness for more than 5 min, or presence of metal in the body. Twelve participants were taking anti-psychotic medications at the time of the study and three were unmedicated. Although a control group was not recruited for this study, data was available for 15 historical controls from previous studies conducted at the University of Calgary in the research program of the senior author (J.A.). Control subjects could not meet criteria for any prodromal syndrome, any current or past psychotic disorder or a Cluster A personality disorder diagnosis, not have a family history (in first-degree relatives) of any psychotic disorder or any other disorder involving psychotic symptoms. They could not be currently using psychotropic medication. The historical control group was matched to the early psychosis group
MRI scanning was conducted on a 3 Tesla GE Signa scanner with an 8-channel head coil at the Seaman Family MR Research Centre at the University of Calgary. All participants underwent a high-resolution anatomical scan (3D SPGR, 180 slices, FOV = 25.6 cm, 1 × 1 × 1 mm, flip angle = 12°). 2.5. Measurement of cortical thickness A quality control (QC) procedure was carried out by one rater on all raw T1-weighted images to ensure no visible motion artefacts or poor resolution of gray/white matter contrast. The high quality MRIs were then submitted to the CIVET processing pipeline (Version 2.0.0) (http://www.bic.mni.mcgill.ca/ServicesSoftware/CIVET) (Ad-Dab'bagh et al., 2006; Zijdenbos et al., 2002), using the CBRAIN platform (Sherif et al., 2014). Native T1-weighted images were first registered to the ICBM152 template using a linear transformation (Collins et al., 1994; Grabner et al., 2006) and simultaneously corrected for non-uniformity artefacts using N3 (Sled et al., 1998). The transformed images were then segmented into gray matter, white matter, cerebral spinal fluid and background using a neural net classifier (INSECT) (Zijdenbos et al., 2002). Gray matter and white matter surfaces were extracted
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
using the CLASP algorithm (Kabani et al., 2001; MacDonald et al., 2000). A spherical-mesh deformation algorithm was used to produce a surface mesh of 81,924 polygons (40,962 nodes or vertices) for each hemisphere. Nonlinear registration of both cortical surfaces to a highresolution average surface template generated from the ICBM152 data-set was performed to establish inter-subject correspondence of vertices (Lyttelton et al., 2007; Robbins, 2004). Reverse linear transformation of volumes was performed to allow vertex-based corticometric measurements in native space for each subject's MRI (Ad-Dab'bagh et al., 2005). The deformation algorithm first fits the white matter surface and then expands to the outer gray matter and cerebral spinal fluid intersection. From these surfaces, cortical thickness was computed in native space using the t-link method (Lerch and Evans, 2005), which determines the linked distance between the inner and outer cortical surfaces at each of 81,924 vertices. Each participant's cortical thickness map was subsequently blurred using a 30-mm full-width at halfmaximum surface-based diffusion smoothing kernel (Chung et al., 2003). Resultant gray and white matter surfaces were further manually checked for quality and corrected where possible. Specifically, inaccurate extraction of gray/white matter surfaces in close proximity to the ventricles due to a gradient error were corrected in three subjects via in-house scripts. All subjects passed QC procedures. Surfaces were then re-run through specific stages of CIVET corresponding to the stage at which the error was found. Statistics were performed across all 81,924 vertices using the SurfStat toolbox (http://www.math.mcgill.ca/keith/surfstat/) within MATLAB www.mathworks.com, to assess differences in cortical thickness. For each vertex a linear model was fitted to calculate: a) Group differences in cortical thickness Y ~ intercept + β1(Group) + β2(Seed) + β3(Age) + β4(Sex) b) Modulation of cortical thickness with facial emotion recognition measures in early psychosis subjects only Y ~ intercept + β1(ER40/EDF40) + β2(Age) Y represents cortical thickness, and β values represent regression coefficients (where β1 is the regressor of interest, and remaining β are covariates). Sex was not entered as a covariate in model b) above as the early psychosis sample was comprised of a very low proportion of females (13%). For all analyses, a cluster forming threshold was used with random field theory (RFT) applied over a p-uncorrected = 0.005 map (Worsley et al., 2004). This procedure is implemented within SurfStat and limits the chance of reporting a false positive finding to below 0.05. 2.6. Structural covariance analysis Seed-based analysis of structural covariance was conducted using the mapping correlations across cerebral cortex (MACACC) methodology (Lerch et al., 2006), whereby Pearson correlation coefficients were performed between mean cortical thickness of each seed and cortical thickness at all other vertices across the entire brain. To map structural covariance networks involved in facial emotion recognition, the right somatosensory cortex, left fusiform face area and right fusiform face area were selected as seed regions. The right somatosensory cortex region was selected based on published literature showing that repetitive transcranial magnetic stimulation to the face area of the right somatosensory cortex impairs facial emotion recognition (Pitcher et al., 2008). The fusiform face region was selected based on results of a meta-analysis (Fusar-Poli et al., 2009), which a) showed that processing of facial emotions is associated with robust increased activation within the fusiform face region, and b) identified detailed neurofunctional maps for use as normative references in fMRI studies of emotional face processing. Coordinates taken from this meta-analysis and used here were x = 40, y = − 50, z = − 18 for the right fusiform face region, and the left fusiform face region was identified at x = −40, y = −50,
3
z = − 18. The right somatosensory cortex was located at x = 44, y = −9, z = −45 (Pitcher et al., 2008). Seed regions of interest were defined as spheres of 10 mm radius around the coordinates in MNI space. Representative images are shown in Fig. 1. All statistics were run in SurfStat via MATLAB. For each of the three seeds, a linear model was fitted to calculate: a) Group differences in seed-based structural covariance Y ~ intercept + β1(Group) + β2(Seed) + β3(Age) + β4(Sex) + β5(Group ∗ Seed) b) Modulation of seed-based structural covariance by facial emotion recognition in early psychosis only Y ~ intercept + β1(ER40/EDF40) + β2(Seed) + β3(Age) + β4(ER40/ EDF40 ∗ Seed) Each of these models were run separately using the mean cortical thickness of each of the three seed regions (corresponding to “β2(Seed)” in the model). We note that although the current sample size is small, this sample size has been shown to be sensitive to cortical thickness and structural covariance using the same methodologies reported here (Bernhardt et al., 2014). 3. Results 3.1. Sample characteristics Table 1 displays demographic characteristics of the early psychosis and historical control samples, as well as mean ER40 and EDF40 total accuracy scores of the early psychosis group. Control subjects had significantly greater years of education and number of females than early psychosis subjects. 3.2. Group differences in cortical thickness Surface-based measurement of cortical thickness revealed that early psychosis subjects showed significantly thinner cortex in a cluster in the right frontal cortex encompassing the orbitofrontal, medial and middle frontal cortices (p = 0.004) than historical control subjects, and this is shown in Fig. 2. There were no group differences in early psychosis subjects N controls. 3.3. Group differences in seed-based structural covariance Reduced covariance was found overall in early psychosis subjects compared with controls between cortical thickness in the right fusiform face region seed region of interest and cortical thickness in the right orbitofrontal cortex (p = 0.008), and this is shown in Fig. 3. No control N early psychosis group differences were seen using the left fusiform face region or the right somatosensory cortex. No group differences were seen for early psychosis N controls for any seed region. 3.4. Modulation of cortical thickness by facial emotion recognition ability in early psychosis subjects ER40 and EDF40 total accuracy scores were not significantly associated with cortical thickness in any region. 3.5. Modulation of structural covariance networks by facial emotion recognition in early psychosis subjects A negative effect of EDF40 total accuracy scores was seen on structural covariance between thickness in the left fusiform seed and thickness in right supramarginal gyrus (p = 0.046), as shown in Fig. 4. EDF40 total accuracy did not modulate covariance in thickness between either the right fusiform face area or right somatosensory seeds and any other cortical locus.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
4
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
Fig. 1. Representative images of masks of the three seed regions of interest. R = right, D = dorsal, V = ventral. A) right somatosensory cortex, and B) left and right fusiform face area regions. Masks are shown on an inflated view of the brain.
ER40 scores did not significantly modulate covariance between thickness in right somatosensory or bilateral fusiform face region seeds and any cortical region. 4. Discussion In the current study, our early psychosis subjects showed thinner cortex in right orbitofrontal cortex extending into the medial and middle frontal gyri compared to controls. Group differences in seed-based structural MRI covariance showed that early psychosis subjects showed reduced covariation between thickness in the right fusiform face region and thickness in right orbitofrontal cortex compared to controls. In early psychosis subjects, facial emotion recognition ability was not significantly associated with thickness in any cortical area. Lastly, when evaluating the modulation of structural covariance networks by facial emotion recognition ability in early psychosis subjects, a negative effect of EDF40 total accuracy scores was seen in covariance between thickness in the left fusiform face area and thickness in right supramarginal gyrus. Compared to controls, early psychosis subjects showed thinner cortex in right orbitofrontal cortex and medial/middle frontal gyri at a slightly reduced statistical threshold. This finding is consistent with several meta-analyses showing that people with psychosis show reduced
frontal cortical volume (Bora et al., 2011; Ellison-Wright and Bullmore, 2010; Olabi et al., 2011; Palaniyappan et al., 2012; Radua et al., 2012). However, other cortical regions have been implicated in the pathophysiology of psychosis such as the anterior cingulate, superior temporal cortex and insula, that were not seen here. This lack of sensitivity may, in part, relate to the small sample size used in the current study. Covariance networks in the current study were centered on three key regions involved in facial emotion recognition: right somatosensory cortex, left fusiform face area and right fusiform face area. Results indicated that early psychosis subjects showed reduced covariation between thickness in the right fusiform face region and thickness in the right orbitofrontal cortex compared to controls. This network pattern is in agreement with data from previous animal tract-tracing studies (Averbeck and Seo, 2008), and with functional connectivity analyses showing that face processing comprises a distributed cortical network that includes the right fusiform face area and orbitofrontal cortex (Li et al., 2009). Our inter-regional structural covariance data suggests that early psychosis may be characterized by an underlying reduction in brain network integration in facial emotion recognition networks. Our finding is in accord with data from a functional MRI analysis showing that the orbitofrontal cortex is engaged when participants view facial emotion expressions (Faivre et al., 2012). In addition to extending our understanding of structural networks centered on the right fusiform
Table 1 Characteristics of the first-episode psychosis and historical control samples. Variable
Early psychosis (n = 15)
Historical controls (n = 15)
Test statistic
p-Value
Mean (SD) Age (years) Education (years) WASI PANSS positive PANSS negative Antipsychotic dosagea ER40 total accuracy EDF40 total accuracy
22.7 (2.6) 13.4 (2.2) 108.5 (11.0) 14.7 (7.5) 12.6 (3.5) 288.7 (511.9) 31.9 (3.4) 21.5 (5.5)
24.0 (3.6) 15.9 (2.5) 114.3 (12.8) – – – – –
t = 1.15 t = 3.03 t = 1.31
p = 0.27 p = 0.005 p = 0.20
Ʃ2 = 8.89
p = 0.003
13 (86.7) 2 (13.3)
5 (33.3) 10 (66.7) Ʃ2 = 0.00
p = 1.00
1 (6.7) 14 (93.3)
1 (6.7) 14 (93.3)
9 (60) 1 (6.7) 1 (6.7) 4 (26.6)
– – – –
Frequency (%) Sex Male Female Handedness Left Right Diagnosis Schizophrenia Delusional disorder Brief psychotic disorder Psychosis not otherwise specified a
Calculated as chlorpromazine equivalent dosage.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
5
Fig. 2. Random field theory (RFT) corrected map depicting group differences in cortical thickness in early psychosis subjects and historical controls. Participants with an early psychosis showed cortical thinning in right orbitofrontal cortex extending into both medial and middle frontal gyri at a reduced threshold of p = 0.01 uncorrected (df = 28, 1909 vertices, p = 0.004). We note that the left parahippocampal finding is non-significant after RFT correction. No group differences were seen for early psychosis subjects N historical controls.
face region, results suggest that people with a early psychosis show reduced covariance within nodes of the facial emotion processing network. This result may reflect atypical maturational processes (Averbeck and Seo, 2008), which may be secondary to a developmental disturbance of inter-regional brain networks. Cortical thickness did not appear to be associated with facial emotion recognition ability in early psychosis subjects. Most of the existing literature on the neural correlates of facial emotion recognition has employed functional neuroimaging analyses, revealing robust activation in cortical and subcortical foci during successful facial emotion recognition (Fusar-Poli et al., 2009). This combination of observations may suggest that this aspect of cognition may be better captured in a fluid process such as functional MRI or other physiologically based imaging procedures. Seed-based structural covariance analysis showed that poorer facial emotion recognition scores were associated with greater network covariance in thickness between the left fusiform face area seed and
thickness in right superior parietal lobule. This result is concordant with functional neuroimaging analyses showing that the superior parietal lobule is part of a neural network that activates during the processing of human emotional faces (Fusar-Poli et al., 2010; van de Riet et al., 2009). The direction of the effect complements functional MRI data showing that people with schizophrenia show greater neural activation in fusiform gyrus and amygdala during misidentification of facial emotion expressions (Gur et al., 2007). Our results lead to the postulate that greater network covariance between the fusiform face area and superior parietal cortex is coupled with deteriorating performance or failure to discriminate an emotion. The EDF40 captures one's ability to discriminate which of two facial emotion expressions is shown more intensely, whereas the ER40 taps correct identification of an emotion. Thus, a mechanism by which greater covariance could emerge is through mutual activation over a period of time, and this could be related to dysfunctional co-engagement patterns. The covariance pattern in left fusiform face area and superior right parietal cortex may reflect
Fig. 3. Random field theory (RFT) corrected map depicting seed-based structural covariance networks comparing early psychosis and control groups. Reduced covariance was found overall in early psychosis subjects compared with controls between cortical thickness in the right fusiform face region seed and thickness in right orbitofrontal cortex, (df = 24, 995 vertices, p = 0.008). L = lateral, V = ventral, M = medial.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
6
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx
Fig. 4. Random field theory (RFT) corrected map depicting modulation of cortical structure by EDF40 total accuracy scores in subjects with early psychosis. Poorer facial emotion recognition abilities, as measured by EDF40 total accuracy scores, were related to increased covariation in thickness between the left fusiform face area seed and thickness in right superior parietal lobule (df = 10, 103 vertices, p = 0.046). L = lateral, C = caudal.
long-term synchronized developmental patterns within the facial emotion recognition network. Limitations include a very low sample size and over-representation of females in the historical control group, and results should be considered rather preliminary. However, the results can be informative for future work by demonstrating that structural covariance analyses are sensitive to robust effects in small samples, and can therefore inform sample size selection for future research. Although we used a seed based structural covariance analysis, which may have introduced bias, this was a hypothesis-driven approach, which has the advantage of reducing the number of multiple comparisons or spatial correlation concerns, increasing the statistical power of the tests. Control participants were recruited from a historical control group, and facial emotion recognition scores were not available for our historical controls. A betweengroup interaction analysis would be able to determine whether the effects in patients are significantly more negative, or whether the ER40/ EDF40 scores have a general negative effect on the covariance irrespective of the diagnostic category. Future studies could benefit from including a measure to control for medication effects, which are suggested to have neurotrophic properties that could affect cortical thickness (Lieberman et al., 2005), and consider how comorbidity with other non-psychotic psychiatric disorders impacts results. In spite of these limitations, these findings have important implications for understanding the neural underpinnings of poor facial emotion recognition in people with psychotic disorders. Conflict of interest All authors declare no conflict of interest. Contributors The first author performed behavioral and imaging analysis and wrote the first version of the manuscript. The second and sixth authors assisted in data collection and organization. The third author assisted with imaging analyses. The fourth and fifth authors assisted in conceptualizing the study. The seventh author oversaw the project from conception to completion and was responsible for all clinical components. All authors have contributed to the writing of the manuscript and approved the final version.
Role of the funding source The Mathison Centre had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Acknowledgements This study was supported by a Mathison Centre Pilot Research Fund Program awarded to Jean Addington, Signe Bray and Frank McMaster. Lisa Buchy is supported by a CIHR fellowship; Mariapaola Barbato is supported by a Mathison Centre postdoctoral fellowship; Carolina Makowski is supported by an FRSQ studentship. The authors thank staff at Seaman's Family MR Research Centre for assistance with data collection. We are thankful for all the people who participated in the study.
References Ad-Dab'bagh, Y., Singh, V., Robbins, S., Lerch, J., Lyttelton, O., Fombonne, E., Evans, A., 2005. Native-space cortical thickness measurement and the absence of correlation to cerebral volume. 11th Annual Meeting of the Organization for Human Brain Mapping (OHBM), Toronto. Ad-Dab'bagh, Y., Einarson, D., Lyttelton, O., Muehlboeck, J.S., Mok, K., Ivanov, O., Vincent, R., Lepage, C., Lerch, J., Fombonne, E., Evans, A.C., 2006. The CIVET image-processing environment: a fully automated comprehensive pipeline for anatomical neuroimaging research. The 12th Annual Meeting of the Organization for Human Brain Mapping (OHBM), Florence, Italy. Adolphs, R., Damasio, H., Tranel, D., Cooper, G., Damasio, A.R., 2000. A role for somatosensory cortices in the visual recognition of emotion as revealed by three-dimensional lesion mapping. J. Neurosci. 20 (7), 2683–2690. Adolphs, R., Tranel, D., Damasio, A.R., 2003. Dissociable neural systems for recognizing emotions. Brain Cogn. 52 (1), 61–69. Alexander-Bloch, A., Raznahan, A., Bullmore, E., Giedd, J., 2013. The convergence of maturational change and structural covariance in human cortical networks. J. Neurosci. 33 (7), 2889–2899. Averbeck, B.B., Seo, M., 2008. The statistical neuroanatomy of frontal networks in the macaque. PLoS Comput. Biol. 4 (4), e1000050. Barkl, S.J., Lah, S., Harris, A.W., Williams, L.M., 2014. Facial emotion identification in earlyonset and first-episode psychosis: a systematic review with meta-analysis. Schizophr. Res. 159 (1), 62–69. Bernhardt, B.C., Valk, S.L., Silani, G., Bird, G., Frith, U., Singer, T., 2014. Selective disruption of sociocognitive structural brain networks in autism and alexithymia. Cerebral Cortex 24 (12), 3258–3267. Bora, E., Fornito, A., Radua, J., Walterfang, M., Seal, M., Wood, S.J., Yucel, M., Velakoulis, D., Pantelis, C., 2011. Neuroanatomical abnormalities in schizophrenia: a multimodal voxelwise meta-analysis and meta-regression analysis. Schizophr. Res. 127 (1–3), 46–57. Chung, M.K., Worsley, K.J., Robbins, S., Paus, T., Taylor, J., Giedd, J.N., Rapoport, J.L., Evans, A.C., 2003. Deformation-based surface morphometry applied to gray matter deformation. NeuroImage 18 (2), 198–213. Collins, D., Peters, T., Evans, A., 1994. An automated 3D non-linear deformation procedure for determination of gross morphometric variability in the human brain. Visualization in biomedical computing. Proc. SPIE 2359, 180–194. Couture, S.M., Penn, D.L., Roberts, D.L., 2006. The functional significance of social cognition in schizophrenia: a review. Schizophr. Bull. 32 (Suppl. 1), S44–S63. Dziobek, I., Bahnemann, M., Convit, A., Heekeren, H.R., 2010. The role of the fusiformamygdala system in the pathophysiology of autism. Arch. Gen. Psychiatry 67 (4), 397–405. Ellison-Wright, I., Bullmore, E., 2010. Anatomy of bipolar disorder and schizophrenia: a meta-analysis. Schizophr. Res. 117 (1), 1–12.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054
L. Buchy et al. / Schizophrenia Research xxx (2017) xxx–xxx Faivre, N., Charron, S., Roux, P., Lehericy, S., Kouider, S., 2012. Nonconscious emotional processing involves distinct neural pathways for pictures and videos. Neuropsychologia 50 (14), 3736–3744. Fett, A.K., Viechtbauer, W., Dominguez, M.D., Penn, D.L., van Os, J., Krabbendam, L., 2011. The relationship between neurocognition and social cognition with functional outcomes in schizophrenia: a meta-analysis. Neurosci. Biobehav. Rev. 35 (3), 573–588. Fusar-Poli, P., Placentino, A., Carletti, F., Landi, P., Allen, P., Surguladze, S., Benedetti, F., Abbamonte, M., Gasparotti, R., Barale, F., Perez, J., McGuire, P., Politi, P., 2009. Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. J. Psychiatry Neurosci. 34 (6), 418–432. Fusar-Poli, P., Bhattacharyya, S., Allen, P., Crippa, J.A., Borgwardt, S., Martin-Santos, R., Seal, M., O'Carroll, C., Atakan, Z., Zuardi, A.W., McGuire, P., 2010. Effect of image analysis software on neurofunctional activation during processing of emotional human faces. J. Clin. Neurosci. 17 (3), 311–314. Grabner, G., Janke, A.L., Budge, M.M., Smith, D., Pruessner, J., Collins, D.L., 2006. Symmetric atlasing and model based segmentation: an application to the hippocampus in older adults. Med. Image Comput. Comput. Assist. Interv. 9 (Pt 2), 58–66. Gur, R.C., Sara, R., Hagendoorn, M., Marom, O., Hughett, P., Macy, L., Turner, T., Bajcsy, R., Posner, A., Gur, R.E., 2002. A method for obtaining 3-dimensional facial expressions and its standardization for use in neurocognitive studies. J. Neurosci. Methods 115 (2), 137–143. Gur, R.E., Loughead, J., Kohler, C.G., Elliott, M.A., Lesko, K., Ruparel, K., Wolf, D.H., Bilker, W.B., Gur, R.C., 2007. Limbic activation associated with misidentification of fearful faces and flat affect in schizophrenia. Arch. Gen. Psychiatry 64 (12), 1356–1366. Irani, F., Seligman, S., Kamath, V., Kohler, C., Gur, R.C., 2012. A meta-analysis of emotion perception and functional outcomes in schizophrenia. Schizophr. Res. 137 (1–3), 203–211. Kabani, N., Le Goualher, G., MacDonald, D., Evans, A.C., 2001. Measurement of cortical thickness using an automated 3-D algorithm: a validation study. NeuroImage 13 (2), 375–380. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Kohler, C.G., Walker, J.B., Martin, E.A., Healey, K.M., Moberg, P.J., 2010. Facial emotion perception in schizophrenia: a meta-analytic review. Schizophr. Bull. 36 (5), 1009–1019. Lerch, J.P., Evans, A.C., 2005. Cortical thickness analysis examined through power analysis and a population simulation. NeuroImage 24 (1), 163–173. Lerch, J.P., Worsley, K., Shaw, W.P., Greenstein, D.K., Lenroot, R.K., Giedd, J., Evans, A.C., 2006. Mapping anatomical correlations across cerebral cortex (MACACC) using cortical thickness from MRI. NeuroImage 31 (3), 993–1003. Li, J., Liu, J., Liang, J., Zhang, H., Zhao, J., Huber, D.E., Rieth, C.A., Lee, K., Tian, J., Shi, G., 2009. A distributed neural system for top-down face processing. Neurosci. Lett. 451 (1), 6–10. Li, H., Chan, R.C., McAlonan, G.M., Gong, Q.Y., 2010. Facial emotion processing in schizophrenia: a meta-analysis of functional neuroimaging data. Schizophr. Bull. 36 (5), 1029–1039. Lieberman, J.A., Tollefson, G.D., Charles, C., Zipursky, R., Sharma, T., Kahn, R.S., Keefe, R.S., Green, A.I., Gur, R.E., McEvoy, J., Perkins, D., Hamer, R.M., Gu, H., Tohen, M., Group, H.S., 2005. Antipsychotic drug effects on brain morphology in first-episode psychosis. Arch. Gen. Psychiatry 62 (4), 361–370.
7
Lyttelton, O., Boucher, M., Robbins, S., Evans, A., 2007. An unbiased iterative group registration template for cortical surface analysis. NeuroImage 34 (4), 1535–1544. MacDonald, D., Kabani, N., Avis, D., Evans, A.C., 2000. Automated 3-D extraction of inner and outer surfaces of cerebral cortex from MRI. NeuroImage 12 (3), 340–356. Mechelli, A., Friston, K.J., Frackowiak, R.S., Price, C.J., 2005. Structural covariance in the human cortex. J. Neurosci. 25 (36), 8303–8310. Olabi, B., Ellison-Wright, I., McIntosh, A.M., Wood, S.J., Bullmore, E., Lawrie, S.M., 2011. Are there progressive brain changes in schizophrenia? A meta-analysis of structural magnetic resonance imaging studies. Biol. Psychiatry 70 (1), 88–96. Palaniyappan, L., Balain, V., Liddle, P.F., 2012. The neuroanatomy of psychotic diathesis: a meta-analytic review. J. Psychiatr. Res. 46 (10), 1249–1256. Parent, A., Carpenter, M.B., 1995. Human Neuroanatomy. Williams & Wilkins, Baltimore, MD. Pinkham, A.E., Loughead, J., Ruparel, K., Overton, E., Gur, R.E., Gur, R.C., 2011. Abnormal modulation of amygdala activity in schizophrenia in response to direct- and averted-gaze threat-related facial expressions. Am. J. Psychiat. 168 (3), 293–301. Pitcher, D., Garrido, L., Walsh, V., Duchaine, B.C., 2008. Transcranial magnetic stimulation disrupts the perception and embodiment of facial expressions. J. Neurosci. 28 (36), 8929–8933. Radua, J., Borgwardt, S., Crescini, A., Mataix-Cols, D., Meyer-Lindenberg, A., McGuire, P.K., Fusar-Poli, P., 2012. Multimodal meta-analysis of structural and functional brain changes in first episode psychosis and the effects of antipsychotic medication. Neurosci. Biobehav. Rev. 36 (10), 2325–2333. Raznahan, A., Lerch, J.P., Lee, N., Greenstein, D., Wallace, G.L., Stockman, M., Clasen, L., Shaw, P.W., Giedd, J.N., 2011. Patterns of coordinated anatomical change in human cortical development: a longitudinal neuroimaging study of maturational coupling. Neuron 72 (5), 873–884. Robbins, S.M., 2004. Anatomical Standardization of the Human Brain in Euclidean 3Space and on the Cortical 2-Manifold. School of Computer Science. McGill University, Montreal, p. 315. Sherif, T., Rioux, P., Rousseau, M.E., Kassis, N., Beck, N., Adalat, R., Das, S., Glatard, T., Evans, A.C., 2014. CBRAIN: a web-based, distributed computing platform for collaborative neuroimaging research. Front. Neuroinf. 8 (24). Silver, H., Shlomo, N., Turner, T., Gur, R.C., 2002. Perception of happy and sad facial expressions in chronic schizophrenia: evidence for two evaluative systems. Schizophr. Res. 55 (1–2), 171–177. Sled, J.G., Zijdenbos, A.P., Evans, A.C., 1998. A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans. Med. Imaging 17 (1), 87–97. van de Riet, W.A., Grezes, J., de Gelder, B., 2009. Specific and common brain regions involved in the perception of faces and bodies and the representation of their emotional expressions. Soc. Neurosci. 4 (2), 101–120. Wechsler, D., 1999. Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, San Antonio, TX. Worsley, K.J., Taylor, J.E., Tomaiuolo, F., Lerch, J., 2004. Unified univariate and multivariate random field theory. NeuroImage 23 (Suppl. 1), S189–S195. Zijdenbos, A.P., Forghani, R., Evans, A.C., 2002. Automatic “pipeline” analysis of 3-D MRI data for clinical trials: application to multiple sclerosis. IEEE Trans. Med. Imaging 21 (10), 1280–1291.
Please cite this article as: Buchy, L., et al., Mapping structural covariance networks of facial emotion recognition in early psychosis: A pilot study, Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.01.054