- Email: [email protected]
Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study
SCHRES-06253; No of Pages 8 Schizophrenia Research xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study Johannes Pulkkinen a,b,⁎, Juha Nikkinen b,i,m, Vesa Kiviniemi b,i, Pirjo Mäki a,c,d,e,f,g,h, Jouko Miettunen a,c,i,j, Jenni Koivukangas a, Sari Mukkala a,j, Tanja Nordström l, Jennifer H. Barnett k,n, Peter B. Jones k, Irma Moilanen j, Graham K. Murray k, Juha Veijola a,c,i a
Department of Psychiatry, Institute of Clinical Medicine, University of Oulu, Oulu, Finland Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland c Department of Psychiatry, Oulu University Hospital, Oulu, Finland d Department of Psychiatry, Länsi-Pohja Healthcare District, Finland e Department of Psychiatry, The Middle Ostrobothnia Central Hospital, Kiuru, Finland f Mental Health Services, Joint Municipal Authority of Wellbeing in Raahe District, Finland g Mental Health Services, Basic Health Care District of Kallio, Finland h Visala Hospital, The Northern Ostrobothnia Hospital District, Finland i Medical Research Center Oulu, University of Oulu, Oulu University Hospital, Oulu, Finland j Department of Child Psychiatry, Institute of Clinical Medicine, University and University Hospital of Oulu, Oulu, Finland k Department of Psychiatry, University of Cambridge, Cambridge, UK l Institute of Health Sciences, University of Oulu, Oulu, Finland m Department of Oncology and Radiotherapy, Oulu University Hospital, University of Oulu, Oulu, Finland n Cambridge Cognition, UK b
a r t i c l e
i n f o
Article history: Received 10 June 2014 Received in revised form 26 January 2015 Accepted 27 January 2015 Available online xxxx Keywords: Birth cohort fMRI Facial expression stimulus Familial risk for psychosis Emotion recognition PPI Amygdala
a b s t r a c t Background: Social interaction requires mirroring to other people's mental state. Psychotic disorders have been connected to social interaction and emotion recognition impairment. We compared the brain activity between young adults with familial risk for psychosis (FR) and matched controls during visual exposure to emotional facial expression. We also investigated the role of the amygdala, the key region for social interaction and emotion recognition. Methods: 51 FR and 52 control subjects were drawn from the Northern Finland 1986 Birth Cohort (Oulu Brain and Mind Study). None of the included participants had developed psychosis. The FR group was defined as having a parent with psychotic disorder according to the Finnish Hospital Discharge Register. Participants underwent functional MRI (fMRI) using visual presentation of dynamic happy and fearful facial expressions. FMRI data were processed to produce maps of activation for happy and fearful facial expression, which were then compared between groups. Two spherical regions of interest (ROIs) in the amygdala were set to extract BOLD responses during happy and fearful facial expression. BOLD responses were then compared with subjects' emotion recognition, which was assessed after fMRI. Psychophysiological interaction (PPI) for the left and right amygdala during happy and fearful facial expression was conducted using the amygdala as seed regions. Results: FR subjects had increased activity in the left premotor cortex and reduced deactivation of medial prefrontal cortex structures during happy facial expression. There were no between-group differences during fearful facial expression. The FR group also showed a statistically significant linear correlation between mean amygdala BOLD response and facial expression recognition. PPI showed that there was a significant negative interaction between the amygdala and the dorsolateral prefrontal cortex (dlPFC) and superior temporal gyrus in FR subjects. Conclusions: Increased activations by positive valence in FR were in brain regions crucial to emotion recognition and social interaction. Increased activation of the premotor cortex may serve as a compensatory mechanism as FR subjects may have to exert more effort on processing the stimuli, as has been found earlier in schizophrenia. Failure to deactivate PFC structures may imply error in the default mode network. Abnormal PFC function in FR was also suggested by PPI, as the dlPFC showed decreased functional
⁎ Corresponding author at: Department of Psychiatry, P.O. Box 5000, 90014 University of Oulu, Finland. E-mail address: johannes.pulkkinen@oulu.fi (J. Pulkkinen).
http://dx.doi.org/10.1016/j.schres.2015.01.039 0920-9964/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
2
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
connectivity with the amygdala in the FR group. This may indicate that in FR subjects the amygdala have to take a greater role in emotion recognition and social functioning. This inference was supported by our discovery of statistically significant correlations between the amygdala BOLD response and emotion recognition in the FR group but not in controls. © 2015 Elsevier B.V. All rights reserved.
1. Introduction
2. Materials and methods
Social interaction requires recognition of other people's mental state. During social interaction we cannot avoid facial stimuli that activate several brain regions in a characteristic way. Perception of faces has been associated with activations of the occipital and occipitotemporal areas (Clark et al., 1996; Haxby et al., 1994), inferior frontal cortex (Nakamura et al., 1999), fusiform gyrus (Dolan et al., 1996), superior temporal sulcus (Allison et al., 2000), medial prefrontal cortex (Dolan et al., 1996; Nakamura et al., 1999) and the limbic system (Adolphs et al., 1994; Breiter et al., 1996; Phillips et al., 1997). Deficient emotion processing of others is a common feature in patients with psychotic disorders (Chan et al., 2010; Comparelli et al., 2013; Daros et al., 2014), including in the prodromal phase. Healthy subjects with psychotic-like symptoms have also been shown to have impaired emotion recognition (Pelletier et al., 2013). Previously in the general population-based Northern Finland 1986 Birth Cohort, self-reported difficulty or uncertainty in making contact with others in adolescence was shown to precede psychosis, but was not present in hospitalised nonpsychotic mental disorders or controls (Mäki et al., 2014). Differences in brain activation between the non-psychotic siblings of patients with schizophrenia and controls during processing of facial stimuli have been previously reported, including hyperactivation in the right superior frontal gyrus, right precentral gyrus (Li et al., 2012), hippocampus, medial prefrontal cortex, posterior and anterior cingulate cortex, medial temporal cortex (van Buuren et al., 2011) and the mid-brain (Barbour et al., 2012). A network study using dynamic causal modelling reported decreased frontolimbic connectivity (Diwadkar et al., 2012). However, previous familial risk (FR) studies are somewhat conflicting, with both hyperactivation (van Buuren et al., 2011) and hypoactivation (Habel et al., 2010) of amygdala reported. In addition one study reported no difference in amygdala activation (Rasetti et al., 2009). Because parental psychosis is a risk factor for later schizophrenia (Gottesman, 1991), some recent neuroimaging studies with facial stimuli have focused on activation patterns in patients with psychotic disorder and their healthy offspring. In previous facial affect fMRI studies patients with schizophrenia and their non-psychotic relatives showed limbic system hypoactivation similarities (Barbour et al., 2010; Habel et al., 2010). In one study, patients with schizophrenia and their healthy siblings reported similar reduced right hemisphere activation (de Achával et al., 2012). Here we used fMRI during visual exposure to dynamic happy and fearful facial expressions in a unique data set. The subjects were drawn from the Northern Finland 1986 Birth Cohort (NFBC 1986). In the NFBC 1986 we were able to study the offspring of parents with psychotic disorder and matched control subjects. The goal of the present study was to determine whether the offspring of parents with psychosis had deviant brain activation patterns, especially in regions connected to social interaction, when compared to control subjects. We focused on the amygdala, a central part of the limbic system with a wellestablished role in emotion recognition and interplay with other brain regions during social interaction (Adolphs, 2002; Adolphs et al., 1994). We wanted to investigate the relationship between bilateral amygdala responses and emotion recognition from facial expressions in both study groups, and whether the amygdala's functional connectivity with other brain regions during visual exposure to happy and fearful facial expression is deviant in subjects with FR for psychosis.
2.1. Setting of the fMRI study in the Northern Finland Birth Cohort 1986 (NFBC 1986) The participants of the NFBC 1986 were born between July 1st 1985 and June 30th 1986 in Oulu and Lapland provinces. NFBC 1986 included 99% of all births, comprising 9432 births, at the beginning of the study (Jarvelin et al., 1993). The Ethical Committee of the Northern Ostrobothnian Hospital District approved the MRI study design in 2006. The Oulu Brain and Mind Study, a substudy of NFBC 1986, was conducted between 2007 and 2010 and the methodology has been previously reported (Veijola et al., 2013). During the Oulu Brain and Mind Study, a facial affect fMRI study was conducted, along with cognitive tests, background questionnaires, Structured Interview for Prodromal Syndromes (SIPS) (McGlashan et al., 2001), Structured Clinical Interview for DSM-IV (First et al., 2002) and urine tests to assess drug exposure. Subjects were asked to avoid alcohol and drugs on the preceding night.
2.2. Familial risk group The familial risk (FR) group included those cohort members who had a parent treated during 1972–2005 in hospitals according to the Finnish Hospital Discharge Register (FHDR) for a psychotic disorder or an A-type personality disorder. The nationwide FHDR covers all mental and general hospitals, as well as beds in local health centres and private hospitals nationwide. Diagnoses of psychoses in the FHDR have been found to be reliable (Perala et al., 2007). International Classification of Diseases (ICD)-8, ICD-9 (codes: 295–299) and ICD-10 (codes F20–33 except non-psychotic mood disorders) were used. This method yielded 272 subjects of which 78 consented to participate in the Oulu Brain and Mind Study during 2007–2010 (Veijola et al., 2013). Of those 78 subjects, 54 underwent facial affect fMRI (lack of resources prevented scanning of a greater number). Of the 54 participants, one was excluded due to psychosis, one had a failed fMRI scan and one had excessive head motion during the fMRI procedure (N2 mm absolute mean displacement). Thus the total FR group comprised 51 individuals (Fig. 1). Of the 51 subjects, 16 had a parent with schizophrenia and 35 a parent diagnosed with psychosis other than schizophrenia (schizoaffective disorder n = 4, delusional disorder n = 2, psychotic depression n = 8, psychotic bipolar disorder n = 5 and other psychotic disorders, n = 16) Five subjects from the FR group had current prodromal symptoms one had current major depression episode according to the SCID.
2.3. Control group Controls were drawn randomly from the NFBC 1986. They were screened according to FHDR against any psychotic episode. 80 controls initially participated in the Oulu Brain and Mind Study and of those, 59 underwent the facial affect fMRI procedure. Five controls were excluded due to a current prodromal syndrome according to the SIPS, one was excluded due to excessive head motion during the fMRI procedure and one was excluded due to psychosis. Thus, the control group comprised 52 individuals (Fig. 1.).
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
3
Fig. 1. Flow chart of study participation.
IQ assessed with Vocabulary and Matrix Reasoning from the Wechsler Adult Intelligence Scale Test III (WAIS-III) (Mukkala et al., 2011) Finnish Edition. Basic education was divided into three categories: less than nine years education, comprehensive school (nine years of education) and matriculation (12 years of education). Age, handedness and gender were also determined. The SIPS interview was used to assess positive and negative symptoms and global assessment of functioning (GAF).
(FOV) was 25.6 cm × 25.6 cm, matrix size 128 × 128 and flip angle 90°. T1 weighted anatomical scans were co-registered to Montreal Neurological Institute (MNI) standard images using 3D Fast Spoiled Gradient echo (FSPGR) using TR 12.4 ms and TE 5.2 ms. Thickness of oblique axial slices was 1 mm; FOV was 24.0 cm × 24.0 cm, matrix size 256 × 256 and flip angle 20°. The stimulus used in the fMRI was a block design created by Helsinki University of Technology, which is described in detail elsewhere (Kätsyri, 2006). The stimulus consisted of four happy and four fearful facial expression blocks in semi-random order separated by dynamic
2.5. FMRI imaging procedure
Table 1 Demographics of the familial risk group and the control group.
2.4. Background variables
The fMRI imaging method was identical to another earlier study conducted in Oulu, Finland (Rahko et al., 2010). Participants protected their hearing with earplugs and minimized head motion using soft pads fitted over the ears. During the fMRI scan subjects were instructed to lay still and relax and watch stimuli on screen through a mirror system installed in the head coil. Functional imaging was executed with 1.5-T General Electric Signa HDx equipped with an 8-channel head coil employing parallel imaging with an acceleration factor of 2.0. TR (repetition time) was 3200 ms and TE (echo time) 45 ms with whole brain coverage using 37 2.9 mm oblique axial slices between 0.3 mm spaces. The first three functional scans were excluded due to magnetic equilibrium effects. Field of view
Variable Age [M (SD)] Sex, male Handedness, right [M (SD)] Education b9 school years Comprehensive (9 school years) Matriculation (12 school years) IQ [M (SD)]
FR-group
Controls
22.4 (0.76) 20 (39,2) 44 (86.3)
22.3 (0.72) 20 (38.5) 51 (98.1)
1 (2.0) 23 (45.1) 27 (52.9) 110.3 (23.0)
0 (0) 17 (33.3) 34 (66.7) 107.1 (20.8)
Statistical testing (p-values) 0.38b 0.94a 0.03c 0.27c
0.46b
SD = standard deviation. a χ2-Test. b t-Test. c Fisher's exact test.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
4
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Table 2 Positive symptoms, negative symptoms and functioning according to the SIPS interview and emotion recognition in FR and control groups. Variable
FR-group
SIPS [M (SD)] Positive symptoms 0.804 (1.06) Negative symptoms 1.16 (1.30) GAF 82.1 (8.30) Emotion recognition [M (SD)] 5.80 (0.40) Happyb Fearfulc 3.96 (1.11)
Controls
Statistical testing (p-values)
0.615 (0.72) 1.04 (1.22) 81.4 (9.22)
0.29a 0.64a 0.66a
5.86 (0.46) 3.59 (1.31)
0.55a 0.15a
impression. Emotional facial expressions (happy and fearful) were dynamic and the neutral facial expression was a static figure. Five options were given in each facial expression: happy, fearful, angry, surprised and neutral. 2.7. Data preprocessing and statistical analyses
SD = standard deviation. SIPS = Structured interview for prodromal syndromes. GAF = Global assessment of functioning. a t-Test. b Max score was 6. c Max score was 5.
mosaic baseline blocks. The whole imaging session lasted 8 min and 10 s. Happy and fearful facial expression blocks consisted of 12 × 2.5 s semi-randomly ordered facial expressions. Happy facial expression blocks consisted of six persons' two times randomized facial expressions. Fearful blocks consisted of five persons' two or three times randomized facial expressions. 2.6. Emotion recognition After the fMRI procedure all participants completed an emotion recognition test. The test consisted of viewing 16 facial expressions on a computer screen: six happy, five fearful and five neutral facial expressions in a semi-random order. Facial expressions in the test were the same as used in the fMRI stimuli. Facial expressions were seen on a computer screen and subjects were instructed to answer based on their first
For data preprocessing and analysis of the fMRI data FSL (FMRIB Centre, University of Oxford, www.fmrib.ox.uk/fsl) versions 4.14 and 5.07 were used. For statistical analyses R (http://cran.r-project.org) version 3.1.1 was used. For structural data preprocessing BET (Smith, 2002) was used for extracting non-brain tissue. The effect of head motion (translations and rotations) in functional scans was corrected with MCFLIRT (Jenkinson et al., 2002). Absolute mean displacement was set to 2 mm as an exclusion criterion. Registration of functional scans to highresolution structural and MNI-152 standard space images was carried out using FLIRT (Jenkinson and Smith, 2001; Jenkinson et al., 2002). Spatial smoothing was executed using a Gaussian kernel of FWHM 4.0 mm. For analysing fMRI data FSL's software tool FEAT was used. In the subject-level analysis happy and fearful positive (activation) and negative (deactivation) blood oxygen level dependent (BOLD) responses were assessed using FILM (Woolrich et al., 2004). Dynamic mosaic blocks were modelled as baseline stimuli in the design matrix. Standard motion parameters were used as covariates in the model. In group level analysis study groups were analysed both alone and together using FLAME 1. Z (Gaussianised T/F), statistic images were thresholded using clusters determined by default Z N 2.3 and cluster significance was thresholded p = 0.05 (corrected) (Worsley et al., 2001). We then extracted percent (%) BOLD responses from the clusters in the group deviation contrast of parameter estimate (COPE) in order to further determine results as activation or deactivation deviations.
Fig. 2. Positive (activation) and negative (deactivation) BOLD responses during happy and fearful facial expressions in study groups (thresholded at Z N 2.3).
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
5
Fig. 3. BOLD response deviation clusters during happy facial expressions. Blue cluster represents between group deviations in the anterior paracingulate cortex (thresholded at Z N 2.3) and the scattergram with group means on the top right % BOLD responses for both study groups within that cluster (t-test, t = 3.19, p-value = 0.002). Red cluster represents between-group deviations in the left superior frontal gyrus (including the left premotor cortex) (thresholded at Z N 2.3) and the scattergram with group means on the bottom right % BOLD responses for both study groups within that cluster (t-test, t = 4.32, p-value b 0.0001).
For analysing the amygdala the Harvard–Oxford Subcortical Structural Atlas supplied with FSL was used in order to set regions of interest (ROIs) corresponding the highest probability of the amygdala. Spherical ROIs with a radius of 5 mm were centred at − 24, − 4, and − 18 (MNI-coordinates) for the left amygdala and 26, − 4, and − 18 (MNI-coordinates) for the right amygdala. Then each subject's left and right amygdala's BOLD time courses and % BOLD responses for happy and fearful facial expression were extracted. In order to assess how the left and right amygdala interact with other brain regions during visual exposure to happy and fearful facial expressions we carried out a psychophysiological interaction (PPI) (Friston et al., 1997) using FEAT. PPI investigates task-specific relationships between seed region and the whole brain. A design matrix with four explanatory variables (EV) was constructed for each subject: (i) happy or fearful facial expression stimuli convolved with a hemodynamic response function (HRF); (ii) a time series regressor containing the left or right amygdala BOLD time course; (iii) regressor for positive and negative interaction between the first and second EV; and (iv) the remaining facial expression (either happy or fearful) convolved with a HRF. Z statistic images were thresholded using clusters determined by Z N 2.3 and cluster significance was thresholded p = 0.05 (corrected) (Worsley et al., 2001).
non-participating FR subjects, 29% had a parent with schizophrenia according to the FDHR, while of the participants 31% had a parent with schizophrenia (Pearson chi square test, p = 0.76). In the FR group 7.3% (N = 16) of the non-participants as well as 5.6% (N = 3) of the participants had been treated in a hospital due to a psychiatric disorder during 2001–2005 (Fisher's exact test, p-value = 0.78). None of the participating controls and 1.5% (N = 2) of the non-participating controls had been treated in a hospital due to a non-psychotic psychiatric disorder (Fisher's exact test, p-value = 1).
2.8. Attrition analysis
The results for the emotion recognition test and SIPS are presented in Table 2. Emotion recognition data were missing for five of the 51 FR (9.8%) subjects and two of the 52 control subjects (3.8%). Only 6 (11.8%) of the FR group and 7 (11.5%) of the control group had fully
In the FR group 54 of 272 FR subjects (20%), and 59 of 193 control subjects (31%) participated in the present fMRI study (Fig. 1). Of the
3. Results 3.1. Demographic data Demographic data and statistical testing for both study groups are presented in Table 1. By chance both groups had the same sex distribution. Groups did not differ statistically significantly for age, IQ and education level. However there were 7 left handed people in the FR group and only 1 left handed person in the control group which were a statistically significant difference. 3.2. Emotion recognition and SIPS
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
6
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Fig. 4. Green spheres represent ROIs used as amygdala. a) Represents reduced functional connectivity in PPI with the left (blue cluster) and right (light-blue cluster) amygdala BOLD responses in the FR group (thresholded at Z N 2.3) during happy expressions. b) Represents reduced functional connectivity in PPI with left amygdala BOLD responses in the FR group (thresholded at Z N 2.3) during fearful expressions. c) Represents the linear relationship between mean amygdala BOLD responses (happy and fearful) and emotion recognition (Pearson's product-moment correlation).
correct answers. None of the subjects confused happy and fearful facial expressions. The emotion recognition for happy and fearful facial expressions did not differ significantly between groups. The FR and control groups did not differ significantly for current positive and negative symptoms according to SIPS interview, nor in GAF scores. 3.3. FMRI imaging data The fMRI results are presented in Figs. 2–4 and Table 3. Fig. 2 shows positive (activation) and negative (deactivation) BOLD responses for both groups during happy and fearful expressions. The BOLD response deviations in study groups during happy facial expressions are presented in Fig. 3. The FR group showed increased activation in the left premotor cortex and superior frontal gyrus and decreased deactivation in the anterior paracingulate cortex and during happy facial expressions. There
were no between-group differences in BOLD response during fearful facial expressions. Fig. 4 shows PPI results and correlations between mean amygdala BOLD response and emotion recognition. In PPI the FR group showed decreased functional connectivity between both the left and right amygdala and right dorsolateral prefrontal cortex (dlPFC) during happy facial expressions and between the left amygdala and left superior temporal gyrus during fearful facial expressions. There was a statistically significant (p b 0.05) correlation between mean % BOLD response during fearful facial expressions in the amygdala and emotion recognition in the FR group. No such relationship was present in controls. 4. Discussion The present study found abnormal function of PFC structures in the FR group during happy facial expressions, specifically increased
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx Table 3 Between group deviation clusters during happy facial expression. Anatomical region(s) corresponding cluster
BA
Superior frontal gyrus L/premotor cortex L Anterior paracingulate cortex
MNI-coordinates
Voxels
Z-scores
46
431
3.76
22
218
3.8
X
Y
Z
6/8
0
46
32
4
40
BA = Brodmann's area.
activation in the superior frontal gyrus and decreased deactivation in the anterior paracingulate cortex. It has been reported that dynamic happy faces are vivid at the very early stage of visual perception (Becker et al., 2011). This serves a useful feature as during social interaction we recognize other people's good intentions immediately. Our results therefore indicate that FR subjects may not have the same vivid response to dynamic happy faces, which may be a risk factor for social withdrawal due to a lack of enjoyment of social interactions. Our discovery of increased activation of the premotor cortex in the FR group has support in a previous facial affect FR study (Li et al., 2012). As the left premotor cortex serves motor planning and executional tasks (Rushworth et. al., 2003), elevated activity may be best understood as a trait of increased effort for emotion recognition in the FR group during this task. The anterior paracingulate cortex has been previously associated with theory of mind and emotional tasks (Brunet-Gouet and Decety, 2006; Gallagher and Frith, 2003). Structural MRI studies with subjects at genetic risk for psychosis have reported cortical thinning in this region (Byun et al., 2012; Rosso et al., 2010) which could be consistent with our discovery, as cortical thinning may cause, or follow, altered function. Abnormal PFC function during cognitive tasks is a well-reported finding in previous studies with patients with psychotic disorder. Both hypoactivation (Hill et al., 2004) and hyperactivation of prefrontal structures (Callicott et al., 2003; Manoach et al., 1999) have been reported, depending on the difficulty of the task. This has led to the proposition of an inverted U-shaped working load curve for the PFC, in which patients with psychosis reach capacity earlier, followed by hypoactivation of the PFC (Whalley et al., 2005). PFC function may be also affected by factors like the difficulty of the task. One previous FR study using emotional figures detected hyper-activation of the medial prefrontal cortex (mPFC) (van Buuren et al., 2011). Differences between our results and theirs may be explained by differences in the valence used. Negative symptoms and other features of schizophrenia may also affect the working load curve of the PFC. Hypo-activation of the PFC has been reported in FR subjects with depression (Whalley et al., 2008) and in FR subjects expressing negative symptoms of schizophrenia (Barbour et al., 2012). In our study only five FR subjects had prodromal symptoms according to SIPS interview, and one FR subject and three controls had current major depression according to the SCID interview. Previous studies with patients with schizophrenia have reported inefficient task-induced deactivation in prefrontal structures (Guerrero-Pedraza et al., 2012; Schneider et al., 2011). There is also evidence that patients with schizophrenia and their relatives have an inability to switch from the default mode network, which is a resting state network, to a task positive network (Whitfield-Gabrieli et al., 2009). This may be the background to our discovery of decreased deactivation of the anterior paracingulate cortex. Our study found decreased functional connectivity between the amygdala and PFC. Similar results were found in another FR study (Diwadkar et al., 2012). Impaired interplay of the amygdala and PFC during emotional input also supports the notion of abnormal PFC function in the FR group. This may lead to functional compensations in other brain regions as they take a greater role in processing emotions from facial expressions, which might have been the cause for the statistically significant correlation between mean amygdala response and emotion recognition in the FR group but not controls.
7
There were several strengths of our study. First we had a unique data setting: participants were born in the same area and were the same age at the time of the field study. Subjects were in their early twenties which is the peak age for developing schizophrenia, which may increase our sensitivity to detect abnormalities in the at-risk population (Hafner et al., 1998; Paus et al., 2008). Subjects also had a similar ethnic background in each group and subjects' family background was explicitly determined by FHDR. The FHDR was used for exclusion criteria among other criteria in the cohort members with psychoses too. Attrition analysis did not indicate any major bias. Subjects were scanned with the same fMRI device and groups had same sex distribution, avoiding sex-related activation differences. There were also some limitations in our study. First, participation rates were relatively low (20% of the FR group and 29% of the control group). The number of subjects with FR for schizophrenia was also relatively low which may have reduced brain activation differences between study groups. The emotion recognition test data was missing for five FR subjects and two controls. Also, the emotion recognition test showed that both groups did not recognize fearful facial expression as well as they did happy expressions. This may mean that fMRI stimuli for fearful facial expressions were somewhat ambiguous. Handedness did differ significantly between study groups although there were still very few left-handed subjects in both groups. A further limitation is that subjects who had prodromal symptoms according to SIPS were excluded from the control group but not from FR group. This was done to guarantee a control group without psychotic-like symptoms. Our findings support conceptualizations of schizophrenia that involve dysfunction of the brain processes underpinning social interaction. Subjects with FR for psychosis showed abnormal BOLD responses in PFC structures during happy facial expressions. FR subjects also had abnormal functional connectivity between the amygdala and PFC structures, supporting the notion of an abnormally-functioning PFC in the FR group. This may lead to compensations of other brain regions, as they may have to take a greater role in FR subjects. This may explain our discovery of a significant relationship between mean amygdala response and emotion recognition in the FR group but not in controls. Role of funding source The study was funded by grants from the Academy of Finland, (#124257, #212818, #214273) the Sigrid Juselius Foundation and the Signe and Ane Gyllenberg Foundation, Finland.
Contributors Authors Vesa Kiviniemi, Jouko Miettunen, Jenny Barnett, Graham K. Murray, Peter Jones, Irma Moilanen, Sari Mukkala, Pirjo Mäki and Juha Veijola planned the study. Authors Juha Nikkinen and Vesa Kiviniemi conducted the brain imaging. Authors Sari Mukkala, Pirjo Mäki, Jouko Miettunen, Juha Veijola and Jenni Koivukangas performed the data collection. Author Johannes Pulkkinen conducted statistical analyses, analysed the fMRI data and wrote the first draft of the manuscript. All authors have accepted the final version of the manuscript. Conflict of interest All the authors declared no conflicts of interest.
Acknowledgement The facial stimuli used in present study were contributed by professor Mikko Sams (Department of Biomedical Engineering and Computational Science (BECS), the Aalto University School of Science and Technology, Helsinki, Finland).
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.schres.2015.01.039. References Adolphs, R., 2002. Neural systems for recognizing emotion. Curr. Opin. Neurobiol. 12 (2), 169–177.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
8
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Adolphs, R., Tranel, D., Damasio, H., Damasio, A., 1994. Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372 (6507), 669–672. Allison, T., Puce, A., McCarthy, G., 2000. Social perception from visual cues: role of the STS region. Trends Cogn. Sci. 4 (7), 267–278 (Regul.Ed.). Barbour, T., Murphy, E., Pruitt, P., Eickhoff, S.B., Keshavan, M.S., Rajan, U., Zajac-Benitez, C., Diwadkar, V.A., 2010. Reduced intra-amygdala activity to positively valenced faces in adolescent schizophrenia offspring. Schizophr. Res. 123 (2–3), 126–136. Barbour, T., Pruitt, P., Diwadkar, V.A., 2012. fMRI responses to emotional faces in children and adolescents at genetic risk for psychiatric illness share some of the features of depression. J. Affect. Disord. 136 (3), 276–285. Becker, D.V., Neel, R., Srinivasan, N., Neufeld, S., Kumar, D., Fouse, S., 2011. The vividness of happiness in dynamic facial displays of emotion. PLoS ONE 7 (1), e26551. Breiter, H.C., Etcoff, N.L., Whalen, P.J., Kennedy, W.A., Rauch, S.L., Buckner, R.L., Strauss, M.M., Hyman, S.E., Rosen, B.R., 1996. Response and habituation of the human amygdala during visual processing of facial expression. Neuron 17 (5), 875–887. Brunet-Gouet, E., Decety, J., 2006. Social brain dysfunctions in schizophrenia: a review of neuroimaging studies. Psychiatry Res. Neuroimaging 148 (2–3), 75–92. Byun, M.S., Kim, J.S., Jung, W.H., Jang, J.H., Choi, J., Kim, S.N., Choi, C., Chung, C.K., An, S.K., Kwon, J.S., 2012. Regional cortical thinning in subjects with high genetic loading for schizophrenia. Schizophr. Res. 141 (2–3), 197–203. Callicott, J.H., Mattay, V.S., Verchinski, B.A., Marenco, S., Egan, M.F., Weinberger, D.R., 2003. Complexity of prefrontal cortical dysfunction in schizophrenia: more than up or down. Am. J. Psychiatry 160 (12), 2209–2215. Chan, R.C.K., Li, H., Cheung, E.F.C., Gong, Q., 2010. Impaired facial emotion perception in schizophrenia: a meta-analysis. Psychiatry Res. 178 (2), 381–390. Clark, V.P., Keil, K., Maisog, J.M., Courtney, S., Ungerleider, L.G., Haxby, J.V., 1996. Functional magnetic resonance imaging of human visual cortex during face matching: a comparison with positron emission tomography. Neuroimage 4 (1), 1–15. Comparelli, A., Corigliano, V., De Carolis, A., Mancinelli, I., Trovini, G., Ottavi, G., Dehning, J., Tatarelli, R., Brugnoli, R., Girardi, P., 2013. Emotion recognition impairment is present early and is stable throughout the course of schizophrenia. Schizophr. Res. 143 (1), 65–69. Daros, A.R., Ruocco, A.C., Reilly, J.L., Harris, M.S.H., Sweeney, J.A., 2014. Facial emotion recognition in first-episode schizophrenia and bipolar disorder with psychosis. Schizophr. Res. 153 (1–3), 32–37. de Achával, D., Villarreal, M.F., Costanzo, E.Y., Douer, J., Castro, M.N., Mora, M.C., Nemeroff, C.B., Chu, E., Bär, K., Guinjoan, S.M., 2012. Decreased activity in right-hemisphere structures involved in social cognition in siblings discordant for schizophrenia. Schizophr. Res. 134 (2–3), 171–179. Diwadkar, V.A., Wadehra, S., Pruitt, P., Keshavan, M.S., Rajan, U., Zajac-Benitez, C., Eickhoff, S.B., 2012. Disordered corticolimbic interactions during affective processing in children and adolescents at risk for schizophrenia revealed by functional magnetic resonance imaging and dynamic causal modeling. Arch. Gen. Psychiatry 69 (3), 231–242. Dolan, R.J., Fletcher, P., Morris, J., Kapur, N., Deakin, J.F.W., Frith, C.D., 1996. Neural activation during covert processing of positive emotional facial expressions. Neuroimage 4 (3), 194–200. First, M., Spitzer, R., Gibbon, M., Williams, J., 2002. Structured Clinical Interview for DSMIV-TR Axis I Disorders, Research Version, Patient Edition With Psychotic Screen (SCID-I/P W/PSYSCREEN). Biometrics Research, New York State Psychiatric Institute, New York. Friston, K.J., Buechel, C., Fink, G.R., Morris, J., Rolls, E., Dolan, R.J., 1997. Psychophysiological and modulatory interactions in neuroimaging. Neuroimage. 6 (3), 218–229. Gallagher, H.L., Frith, C.D., 2003. Functional imaging of ‘theory of mind’. Trends Cogn. Sci. 7 (2), 77–83 (Regul.Ed.). Gottesman, I.I., 1991. Schizophrenia Genesis. The Origins of Madness. W. H. Freeman and Company, New York, NY. Guerrero-Pedraza, A., McKenna, P.J., Gomar, J.J., Sarro, S., Salvador, R., Amann, B., Carrion, M.I., Landin-Romero, R., Blanch, J., Pomarol-Clotet, E., 2012. First-episode psychosis is characterized by failure of deactivation but not by hypo- or hyperfrontality. Psychol. Med. 42 (1), 73–84. Habel, U., Chechko, N., Pauly, K., Koch, K., Backes, V., Seiferth, N., Shah, N.J., Stöcker, T., Schneider, F., Kellermann, T., 2010. Neural correlates of emotion recognition in schizophrenia. Schizophr. Res. 122 (1–3), 113–123. Hafner, H., an der Heiden, W., Behrens, S., Gattaz, W.F., Hambrecht, M., Loffler, W., Maurer, K., Munk-Jorgensen, P., Nowotny, B., Riecher-Rossler, A., Stein, A., 1998. Causes and consequences of the gender difference in age at onset of schizophrenia. Schizophr. Bull. 24 (1), 99–113. Haxby, J.V., Horwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P., Grady, C.L., 1994. The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. J. Neurosci. 14 (11 Pt 1), 6336–6353. Hill, K., Mann, L., Laws, K.R., Stephenson, C.M., Nimmo-Smith, I., McKenna, P.J., 2004. Hypofrontality in schizophrenia: a meta-analysis of functional imaging studies. Acta Psychiatr. Scand. 110 (4), 243–256. Jarvelin, M.R., Hartikainen-Sorri, A.L., Rantakallio, P., 1993. Labour induction policy in hospitals of different levels of specialisation. Br. J. Obstet. Gynaecol. 100 (4), 310–315. Jenkinson, M., Smith, S., 2001. A global optimisation method for robust affine registration of brain images. Med. Image Anal. 5 (2), 143–156. Jenkinson, M., Bannister, P., Brady, M., Smith, S., 2002. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 17 (2), 825–841.
Kätsyri, J., 2006. Human Recognition of Basic Emotions From Posed and Animated Dynamic Facial Expressions. Helsinki University of Technology (Dissertation, http:// lib.tkk.fi/Diss/2006/isbn951228538X/isbn951228538X.pdf). Li, H., Chan, R.C.K., Gong, Q., Liu, Y., Liu, S., Shum, D., Ma, Z., 2012. Facial emotion processing in patients with schizophrenia and their non-psychotic siblings: a functional magnetic resonance imaging study. Schizophr. Res. 134 (2–3), 143–150. Mäki, P., Koskela, S., Murray, G.K., Nordström, T., Miettunen, J., Jääskeläinen, E., Veijola, J.M., 2014. Difficulty in making contact with others and social withdrawal as early signs of psychosis in adolescents — the Northern Finland Birth Cohort 1986. Eur. Psychiatry 29 (6), 345–351. Manoach, D.S., Press, D.Z., Thangaraj, V., Searl, M.M., Goff, D.C., Halpern, E., Saper, C.B., Warach, S., 1999. Schizophrenic subjects activate dorsolateral prefrontal cortex during a working memory task, as measured by fMRI. Biol. Psychiatry 45 (9), 1128–1137. McGlashan, T., Miller, T.J., Woods, S.W., Rosen, J.L., Hoffman, R.E., Davidson, L., 2001. Structured Interview for Prodromal Syndromes — Version for Present Prodromal Syndromes. Version 3 ed. Yale School, New Haven, CT. Mukkala, S., Ilonen, T., Nordstrom, T., Miettunen, J., Loukkola, J., Barnett, J.H., Murray, G.K., Jaaskelainen, E., Maki, P., Taanila, A., Moilanen, I., Jones, P.B., Heinimaa, M., Veijola, J., 2011. Different vulnerability indicators for psychosis and their neuropsychological characteristics in the Northern Finland 1986 Birth Cohort. J. Clin. Exp. Neuropsychol. 33 (4), 385–394. Nakamura, K., Kawashima, R., Ito, K., Sugiura, M., Kato, T., Nakamura, A., Hatano, K., Nagumo, S., Kubota, K., Fukuda, H., Kojima, S., 1999. Activation of the Right Inferior Frontal Cortex During Assessment of Facial Emotion. J. Neurophysiol. 82 (3), 1610–1614. Paus, T., Keshavan, M., Giedd, J.N., 2008. Why do many psychiatric disorders emerge during adolescence? Nat. Rev. Neurosci. 9 (12), 947–957. Pelletier, A.L., Dean, D.J., Lunsford-Avery, J.R., Smith, A.K., Orr, J.M., Gupta, T., Millman, Z.B., Mittal, V.A., 2013. Emotion recognition and social/role dysfunction in non-clinical psychosis. Schizophr. Res. 143 (1), 70–73. Perala, J., Suvisaari, J., Saarni, S.I., Kuoppasalmi, K., Isometsa, E., Pirkola, S., Partonen, T., Tuulio-Henriksson, A., Hintikka, J., Kieseppa, T., Harkanen, T., Koskinen, S., Lonnqvist, J., 2007. Lifetime prevalence of psychotic and bipolar I disorders in a general population. Arch. Gen. Psychiatry 64 (1), 19–28. Phillips, M.L., Young, A.W., Senior, C., Brammer, M., Andrew, C., Calder, A.J., Bullmore, E.T., Perrett, D.I., Rowland, D., Williams, S.C.R., Gray, J.A., David, A.S., 1997. A specific neural substrate for perceiving facial expressions of disgust. Nature 389 (6650), 495–498. Rahko, J., Paakki, J.J., Starck, T., Nikkinen, J., Remes, J., Hurtig, T., Kuusikko-Gauffin, S., Mattila, M.L., Jussila, K., Jansson-Verkasalo, E., Katsyri, J., Sams, M., Pauls, D., Ebeling, H., Moilanen, I., Tervonen, O., Kiviniemi, V., 2010. Functional mapping of dynamic happy and fearful facial expression processing in adolescents. Brain Imaging Behav. 4 (2), 164–176. Rasetti, R., Mattay, V.S., Wiedholz, L.M., Kolachana, B.S., Hariri, A.R., Callicott, J.H., MeyerLindenberg, A., Weinberger, D.R., 2009. Evidence that altered amygdala activity in schizophrenia is related to clinical state and not genetic risk. Am. J. Psychiatry 166 (2), 216–225. Rosso, I.M., Makris, N., Thermenos, H.W., Hodge, S.M., Brown, A., Kennedy, D., Caviness, V.S., Faraone, S.V., Tsuang, M.T., Seidman, L.J., 2010. Regional prefrontal cortex gray matter volumes in youth at familial risk for schizophrenia from the Harvard Adolescent High Risk Study. Schizophr. Res. 123 (1), 15–21. Rushworth, M.F.S., Johansen-Berg, H., Göbel, S.M., Devlin, J.T., 2003. The left parietal and premotor cortices: motor attention and selection. Neuroimage 20, S89–S100 (Suppl. 1). Schneider, F.C., Royer, A., Grosselin, A., Pellet, J., Barral, F., Laurent, B., Brouillet, D., Lang, F., 2011. Modulation of the default mode network is task-dependant in chronic schizophrenia patients. Schizophr. Res. 125 (2–3), 110–117. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 17 (3), 143–155. van Buuren, M., Vink, M., Rapcencu, A.E., Kahn, R.S., 2011. Exaggerated brain activation during emotion processing in unaffected siblings of patients with schizophrenia. Biol. Psychiatry 70 (1), 81–87. Veijola, J., Maki, P., Jaaskelainen, E., Koivukangas, J., Moilanen, I., Taanila, A., Nordstrom, T., Hurtig, T., Kiviniemi, V., Mukkala, S., Heinimaa, M., Lindholm, P., Jones, P.B., Barnett, J.H., Murray, G.K., Miettunen, J., 2013. Young people at risk for psychosis: case finding and sample characteristics of the Oulu Brain and Mind Study. Early Interv. Psychiatry 7 (2), 146–154. Whalley, H.C., Whyte, M.C., Johnstone, E.C., Lawrie, S.M., 2005. Neural correlates of enhanced genetic risk for schizophrenia. Neuroscientist 11 (3), 238–249. Whalley, H.C., Mowatt, L., Stanfield, A.C., Hall, J., Johnstone, E.C., Lawrie, S.M., McIntosh, A.M., 2008. Hypofrontality in subjects at high genetic risk of schizophrenia with depressive symptoms. J. Affect. Disord. 109 (1–2), 99–106. Whitfield-Gabrieli, S., Thermenos, H.W., Milanovic, S., Tsuang, M.T., Faraone, S.V., McCarley, R.W., Shenton, M.E., Green, A.I., Nieto-Castanon, A., LaViolette, P., Wojcik, J., Gabrieli, J.D., Seidman, L.J., 2009. Hyperactivity and hyperconnectivity of the default network in schizophrenia and in first-degree relatives of persons with schizophrenia. Proc. Natl. Acad. Sci. U. S. A. 106 (4), 1279–1284. Woolrich, M.W., Behrens, T.E.J., Beckmann, C.F., Jenkinson, M., Smith, S.M., 2004. Multilevel linear modelling for FMRI group analysis using Bayesian inference. Neuroimage. 21 (4), 1732–1747. Worsley, K.J., Jezzard, P., Matthews, P.M., Smith, S.M., 2001. Statistical analysis of activation images. Ch 14. Functional MRI: An Introduction to Methods.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study Johannes Pulkkinen a,b,⁎, Juha Nikkinen b,i,m, Vesa Kiviniemi b,i, Pirjo Mäki a,c,d,e,f,g,h, Jouko Miettunen a,c,i,j, Jenni Koivukangas a, Sari Mukkala a,j, Tanja Nordström l, Jennifer H. Barnett k,n, Peter B. Jones k, Irma Moilanen j, Graham K. Murray k, Juha Veijola a,c,i a
Department of Psychiatry, Institute of Clinical Medicine, University of Oulu, Oulu, Finland Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland c Department of Psychiatry, Oulu University Hospital, Oulu, Finland d Department of Psychiatry, Länsi-Pohja Healthcare District, Finland e Department of Psychiatry, The Middle Ostrobothnia Central Hospital, Kiuru, Finland f Mental Health Services, Joint Municipal Authority of Wellbeing in Raahe District, Finland g Mental Health Services, Basic Health Care District of Kallio, Finland h Visala Hospital, The Northern Ostrobothnia Hospital District, Finland i Medical Research Center Oulu, University of Oulu, Oulu University Hospital, Oulu, Finland j Department of Child Psychiatry, Institute of Clinical Medicine, University and University Hospital of Oulu, Oulu, Finland k Department of Psychiatry, University of Cambridge, Cambridge, UK l Institute of Health Sciences, University of Oulu, Oulu, Finland m Department of Oncology and Radiotherapy, Oulu University Hospital, University of Oulu, Oulu, Finland n Cambridge Cognition, UK b
a r t i c l e
i n f o
Article history: Received 10 June 2014 Received in revised form 26 January 2015 Accepted 27 January 2015 Available online xxxx Keywords: Birth cohort fMRI Facial expression stimulus Familial risk for psychosis Emotion recognition PPI Amygdala
a b s t r a c t Background: Social interaction requires mirroring to other people's mental state. Psychotic disorders have been connected to social interaction and emotion recognition impairment. We compared the brain activity between young adults with familial risk for psychosis (FR) and matched controls during visual exposure to emotional facial expression. We also investigated the role of the amygdala, the key region for social interaction and emotion recognition. Methods: 51 FR and 52 control subjects were drawn from the Northern Finland 1986 Birth Cohort (Oulu Brain and Mind Study). None of the included participants had developed psychosis. The FR group was defined as having a parent with psychotic disorder according to the Finnish Hospital Discharge Register. Participants underwent functional MRI (fMRI) using visual presentation of dynamic happy and fearful facial expressions. FMRI data were processed to produce maps of activation for happy and fearful facial expression, which were then compared between groups. Two spherical regions of interest (ROIs) in the amygdala were set to extract BOLD responses during happy and fearful facial expression. BOLD responses were then compared with subjects' emotion recognition, which was assessed after fMRI. Psychophysiological interaction (PPI) for the left and right amygdala during happy and fearful facial expression was conducted using the amygdala as seed regions. Results: FR subjects had increased activity in the left premotor cortex and reduced deactivation of medial prefrontal cortex structures during happy facial expression. There were no between-group differences during fearful facial expression. The FR group also showed a statistically significant linear correlation between mean amygdala BOLD response and facial expression recognition. PPI showed that there was a significant negative interaction between the amygdala and the dorsolateral prefrontal cortex (dlPFC) and superior temporal gyrus in FR subjects. Conclusions: Increased activations by positive valence in FR were in brain regions crucial to emotion recognition and social interaction. Increased activation of the premotor cortex may serve as a compensatory mechanism as FR subjects may have to exert more effort on processing the stimuli, as has been found earlier in schizophrenia. Failure to deactivate PFC structures may imply error in the default mode network. Abnormal PFC function in FR was also suggested by PPI, as the dlPFC showed decreased functional
⁎ Corresponding author at: Department of Psychiatry, P.O. Box 5000, 90014 University of Oulu, Finland. E-mail address: johannes.pulkkinen@oulu.fi (J. Pulkkinen).
http://dx.doi.org/10.1016/j.schres.2015.01.039 0920-9964/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
2
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
connectivity with the amygdala in the FR group. This may indicate that in FR subjects the amygdala have to take a greater role in emotion recognition and social functioning. This inference was supported by our discovery of statistically significant correlations between the amygdala BOLD response and emotion recognition in the FR group but not in controls. © 2015 Elsevier B.V. All rights reserved.
1. Introduction
2. Materials and methods
Social interaction requires recognition of other people's mental state. During social interaction we cannot avoid facial stimuli that activate several brain regions in a characteristic way. Perception of faces has been associated with activations of the occipital and occipitotemporal areas (Clark et al., 1996; Haxby et al., 1994), inferior frontal cortex (Nakamura et al., 1999), fusiform gyrus (Dolan et al., 1996), superior temporal sulcus (Allison et al., 2000), medial prefrontal cortex (Dolan et al., 1996; Nakamura et al., 1999) and the limbic system (Adolphs et al., 1994; Breiter et al., 1996; Phillips et al., 1997). Deficient emotion processing of others is a common feature in patients with psychotic disorders (Chan et al., 2010; Comparelli et al., 2013; Daros et al., 2014), including in the prodromal phase. Healthy subjects with psychotic-like symptoms have also been shown to have impaired emotion recognition (Pelletier et al., 2013). Previously in the general population-based Northern Finland 1986 Birth Cohort, self-reported difficulty or uncertainty in making contact with others in adolescence was shown to precede psychosis, but was not present in hospitalised nonpsychotic mental disorders or controls (Mäki et al., 2014). Differences in brain activation between the non-psychotic siblings of patients with schizophrenia and controls during processing of facial stimuli have been previously reported, including hyperactivation in the right superior frontal gyrus, right precentral gyrus (Li et al., 2012), hippocampus, medial prefrontal cortex, posterior and anterior cingulate cortex, medial temporal cortex (van Buuren et al., 2011) and the mid-brain (Barbour et al., 2012). A network study using dynamic causal modelling reported decreased frontolimbic connectivity (Diwadkar et al., 2012). However, previous familial risk (FR) studies are somewhat conflicting, with both hyperactivation (van Buuren et al., 2011) and hypoactivation (Habel et al., 2010) of amygdala reported. In addition one study reported no difference in amygdala activation (Rasetti et al., 2009). Because parental psychosis is a risk factor for later schizophrenia (Gottesman, 1991), some recent neuroimaging studies with facial stimuli have focused on activation patterns in patients with psychotic disorder and their healthy offspring. In previous facial affect fMRI studies patients with schizophrenia and their non-psychotic relatives showed limbic system hypoactivation similarities (Barbour et al., 2010; Habel et al., 2010). In one study, patients with schizophrenia and their healthy siblings reported similar reduced right hemisphere activation (de Achával et al., 2012). Here we used fMRI during visual exposure to dynamic happy and fearful facial expressions in a unique data set. The subjects were drawn from the Northern Finland 1986 Birth Cohort (NFBC 1986). In the NFBC 1986 we were able to study the offspring of parents with psychotic disorder and matched control subjects. The goal of the present study was to determine whether the offspring of parents with psychosis had deviant brain activation patterns, especially in regions connected to social interaction, when compared to control subjects. We focused on the amygdala, a central part of the limbic system with a wellestablished role in emotion recognition and interplay with other brain regions during social interaction (Adolphs, 2002; Adolphs et al., 1994). We wanted to investigate the relationship between bilateral amygdala responses and emotion recognition from facial expressions in both study groups, and whether the amygdala's functional connectivity with other brain regions during visual exposure to happy and fearful facial expression is deviant in subjects with FR for psychosis.
2.1. Setting of the fMRI study in the Northern Finland Birth Cohort 1986 (NFBC 1986) The participants of the NFBC 1986 were born between July 1st 1985 and June 30th 1986 in Oulu and Lapland provinces. NFBC 1986 included 99% of all births, comprising 9432 births, at the beginning of the study (Jarvelin et al., 1993). The Ethical Committee of the Northern Ostrobothnian Hospital District approved the MRI study design in 2006. The Oulu Brain and Mind Study, a substudy of NFBC 1986, was conducted between 2007 and 2010 and the methodology has been previously reported (Veijola et al., 2013). During the Oulu Brain and Mind Study, a facial affect fMRI study was conducted, along with cognitive tests, background questionnaires, Structured Interview for Prodromal Syndromes (SIPS) (McGlashan et al., 2001), Structured Clinical Interview for DSM-IV (First et al., 2002) and urine tests to assess drug exposure. Subjects were asked to avoid alcohol and drugs on the preceding night.
2.2. Familial risk group The familial risk (FR) group included those cohort members who had a parent treated during 1972–2005 in hospitals according to the Finnish Hospital Discharge Register (FHDR) for a psychotic disorder or an A-type personality disorder. The nationwide FHDR covers all mental and general hospitals, as well as beds in local health centres and private hospitals nationwide. Diagnoses of psychoses in the FHDR have been found to be reliable (Perala et al., 2007). International Classification of Diseases (ICD)-8, ICD-9 (codes: 295–299) and ICD-10 (codes F20–33 except non-psychotic mood disorders) were used. This method yielded 272 subjects of which 78 consented to participate in the Oulu Brain and Mind Study during 2007–2010 (Veijola et al., 2013). Of those 78 subjects, 54 underwent facial affect fMRI (lack of resources prevented scanning of a greater number). Of the 54 participants, one was excluded due to psychosis, one had a failed fMRI scan and one had excessive head motion during the fMRI procedure (N2 mm absolute mean displacement). Thus the total FR group comprised 51 individuals (Fig. 1). Of the 51 subjects, 16 had a parent with schizophrenia and 35 a parent diagnosed with psychosis other than schizophrenia (schizoaffective disorder n = 4, delusional disorder n = 2, psychotic depression n = 8, psychotic bipolar disorder n = 5 and other psychotic disorders, n = 16) Five subjects from the FR group had current prodromal symptoms one had current major depression episode according to the SCID.
2.3. Control group Controls were drawn randomly from the NFBC 1986. They were screened according to FHDR against any psychotic episode. 80 controls initially participated in the Oulu Brain and Mind Study and of those, 59 underwent the facial affect fMRI procedure. Five controls were excluded due to a current prodromal syndrome according to the SIPS, one was excluded due to excessive head motion during the fMRI procedure and one was excluded due to psychosis. Thus, the control group comprised 52 individuals (Fig. 1.).
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
3
Fig. 1. Flow chart of study participation.
IQ assessed with Vocabulary and Matrix Reasoning from the Wechsler Adult Intelligence Scale Test III (WAIS-III) (Mukkala et al., 2011) Finnish Edition. Basic education was divided into three categories: less than nine years education, comprehensive school (nine years of education) and matriculation (12 years of education). Age, handedness and gender were also determined. The SIPS interview was used to assess positive and negative symptoms and global assessment of functioning (GAF).
(FOV) was 25.6 cm × 25.6 cm, matrix size 128 × 128 and flip angle 90°. T1 weighted anatomical scans were co-registered to Montreal Neurological Institute (MNI) standard images using 3D Fast Spoiled Gradient echo (FSPGR) using TR 12.4 ms and TE 5.2 ms. Thickness of oblique axial slices was 1 mm; FOV was 24.0 cm × 24.0 cm, matrix size 256 × 256 and flip angle 20°. The stimulus used in the fMRI was a block design created by Helsinki University of Technology, which is described in detail elsewhere (Kätsyri, 2006). The stimulus consisted of four happy and four fearful facial expression blocks in semi-random order separated by dynamic
2.5. FMRI imaging procedure
Table 1 Demographics of the familial risk group and the control group.
2.4. Background variables
The fMRI imaging method was identical to another earlier study conducted in Oulu, Finland (Rahko et al., 2010). Participants protected their hearing with earplugs and minimized head motion using soft pads fitted over the ears. During the fMRI scan subjects were instructed to lay still and relax and watch stimuli on screen through a mirror system installed in the head coil. Functional imaging was executed with 1.5-T General Electric Signa HDx equipped with an 8-channel head coil employing parallel imaging with an acceleration factor of 2.0. TR (repetition time) was 3200 ms and TE (echo time) 45 ms with whole brain coverage using 37 2.9 mm oblique axial slices between 0.3 mm spaces. The first three functional scans were excluded due to magnetic equilibrium effects. Field of view
Variable Age [M (SD)] Sex, male Handedness, right [M (SD)] Education b9 school years Comprehensive (9 school years) Matriculation (12 school years) IQ [M (SD)]
FR-group
Controls
22.4 (0.76) 20 (39,2) 44 (86.3)
22.3 (0.72) 20 (38.5) 51 (98.1)
1 (2.0) 23 (45.1) 27 (52.9) 110.3 (23.0)
0 (0) 17 (33.3) 34 (66.7) 107.1 (20.8)
Statistical testing (p-values) 0.38b 0.94a 0.03c 0.27c
0.46b
SD = standard deviation. a χ2-Test. b t-Test. c Fisher's exact test.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
4
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Table 2 Positive symptoms, negative symptoms and functioning according to the SIPS interview and emotion recognition in FR and control groups. Variable
FR-group
SIPS [M (SD)] Positive symptoms 0.804 (1.06) Negative symptoms 1.16 (1.30) GAF 82.1 (8.30) Emotion recognition [M (SD)] 5.80 (0.40) Happyb Fearfulc 3.96 (1.11)
Controls
Statistical testing (p-values)
0.615 (0.72) 1.04 (1.22) 81.4 (9.22)
0.29a 0.64a 0.66a
5.86 (0.46) 3.59 (1.31)
0.55a 0.15a
impression. Emotional facial expressions (happy and fearful) were dynamic and the neutral facial expression was a static figure. Five options were given in each facial expression: happy, fearful, angry, surprised and neutral. 2.7. Data preprocessing and statistical analyses
SD = standard deviation. SIPS = Structured interview for prodromal syndromes. GAF = Global assessment of functioning. a t-Test. b Max score was 6. c Max score was 5.
mosaic baseline blocks. The whole imaging session lasted 8 min and 10 s. Happy and fearful facial expression blocks consisted of 12 × 2.5 s semi-randomly ordered facial expressions. Happy facial expression blocks consisted of six persons' two times randomized facial expressions. Fearful blocks consisted of five persons' two or three times randomized facial expressions. 2.6. Emotion recognition After the fMRI procedure all participants completed an emotion recognition test. The test consisted of viewing 16 facial expressions on a computer screen: six happy, five fearful and five neutral facial expressions in a semi-random order. Facial expressions in the test were the same as used in the fMRI stimuli. Facial expressions were seen on a computer screen and subjects were instructed to answer based on their first
For data preprocessing and analysis of the fMRI data FSL (FMRIB Centre, University of Oxford, www.fmrib.ox.uk/fsl) versions 4.14 and 5.07 were used. For statistical analyses R (http://cran.r-project.org) version 3.1.1 was used. For structural data preprocessing BET (Smith, 2002) was used for extracting non-brain tissue. The effect of head motion (translations and rotations) in functional scans was corrected with MCFLIRT (Jenkinson et al., 2002). Absolute mean displacement was set to 2 mm as an exclusion criterion. Registration of functional scans to highresolution structural and MNI-152 standard space images was carried out using FLIRT (Jenkinson and Smith, 2001; Jenkinson et al., 2002). Spatial smoothing was executed using a Gaussian kernel of FWHM 4.0 mm. For analysing fMRI data FSL's software tool FEAT was used. In the subject-level analysis happy and fearful positive (activation) and negative (deactivation) blood oxygen level dependent (BOLD) responses were assessed using FILM (Woolrich et al., 2004). Dynamic mosaic blocks were modelled as baseline stimuli in the design matrix. Standard motion parameters were used as covariates in the model. In group level analysis study groups were analysed both alone and together using FLAME 1. Z (Gaussianised T/F), statistic images were thresholded using clusters determined by default Z N 2.3 and cluster significance was thresholded p = 0.05 (corrected) (Worsley et al., 2001). We then extracted percent (%) BOLD responses from the clusters in the group deviation contrast of parameter estimate (COPE) in order to further determine results as activation or deactivation deviations.
Fig. 2. Positive (activation) and negative (deactivation) BOLD responses during happy and fearful facial expressions in study groups (thresholded at Z N 2.3).
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
5
Fig. 3. BOLD response deviation clusters during happy facial expressions. Blue cluster represents between group deviations in the anterior paracingulate cortex (thresholded at Z N 2.3) and the scattergram with group means on the top right % BOLD responses for both study groups within that cluster (t-test, t = 3.19, p-value = 0.002). Red cluster represents between-group deviations in the left superior frontal gyrus (including the left premotor cortex) (thresholded at Z N 2.3) and the scattergram with group means on the bottom right % BOLD responses for both study groups within that cluster (t-test, t = 4.32, p-value b 0.0001).
For analysing the amygdala the Harvard–Oxford Subcortical Structural Atlas supplied with FSL was used in order to set regions of interest (ROIs) corresponding the highest probability of the amygdala. Spherical ROIs with a radius of 5 mm were centred at − 24, − 4, and − 18 (MNI-coordinates) for the left amygdala and 26, − 4, and − 18 (MNI-coordinates) for the right amygdala. Then each subject's left and right amygdala's BOLD time courses and % BOLD responses for happy and fearful facial expression were extracted. In order to assess how the left and right amygdala interact with other brain regions during visual exposure to happy and fearful facial expressions we carried out a psychophysiological interaction (PPI) (Friston et al., 1997) using FEAT. PPI investigates task-specific relationships between seed region and the whole brain. A design matrix with four explanatory variables (EV) was constructed for each subject: (i) happy or fearful facial expression stimuli convolved with a hemodynamic response function (HRF); (ii) a time series regressor containing the left or right amygdala BOLD time course; (iii) regressor for positive and negative interaction between the first and second EV; and (iv) the remaining facial expression (either happy or fearful) convolved with a HRF. Z statistic images were thresholded using clusters determined by Z N 2.3 and cluster significance was thresholded p = 0.05 (corrected) (Worsley et al., 2001).
non-participating FR subjects, 29% had a parent with schizophrenia according to the FDHR, while of the participants 31% had a parent with schizophrenia (Pearson chi square test, p = 0.76). In the FR group 7.3% (N = 16) of the non-participants as well as 5.6% (N = 3) of the participants had been treated in a hospital due to a psychiatric disorder during 2001–2005 (Fisher's exact test, p-value = 0.78). None of the participating controls and 1.5% (N = 2) of the non-participating controls had been treated in a hospital due to a non-psychotic psychiatric disorder (Fisher's exact test, p-value = 1).
2.8. Attrition analysis
The results for the emotion recognition test and SIPS are presented in Table 2. Emotion recognition data were missing for five of the 51 FR (9.8%) subjects and two of the 52 control subjects (3.8%). Only 6 (11.8%) of the FR group and 7 (11.5%) of the control group had fully
In the FR group 54 of 272 FR subjects (20%), and 59 of 193 control subjects (31%) participated in the present fMRI study (Fig. 1). Of the
3. Results 3.1. Demographic data Demographic data and statistical testing for both study groups are presented in Table 1. By chance both groups had the same sex distribution. Groups did not differ statistically significantly for age, IQ and education level. However there were 7 left handed people in the FR group and only 1 left handed person in the control group which were a statistically significant difference. 3.2. Emotion recognition and SIPS
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
6
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Fig. 4. Green spheres represent ROIs used as amygdala. a) Represents reduced functional connectivity in PPI with the left (blue cluster) and right (light-blue cluster) amygdala BOLD responses in the FR group (thresholded at Z N 2.3) during happy expressions. b) Represents reduced functional connectivity in PPI with left amygdala BOLD responses in the FR group (thresholded at Z N 2.3) during fearful expressions. c) Represents the linear relationship between mean amygdala BOLD responses (happy and fearful) and emotion recognition (Pearson's product-moment correlation).
correct answers. None of the subjects confused happy and fearful facial expressions. The emotion recognition for happy and fearful facial expressions did not differ significantly between groups. The FR and control groups did not differ significantly for current positive and negative symptoms according to SIPS interview, nor in GAF scores. 3.3. FMRI imaging data The fMRI results are presented in Figs. 2–4 and Table 3. Fig. 2 shows positive (activation) and negative (deactivation) BOLD responses for both groups during happy and fearful expressions. The BOLD response deviations in study groups during happy facial expressions are presented in Fig. 3. The FR group showed increased activation in the left premotor cortex and superior frontal gyrus and decreased deactivation in the anterior paracingulate cortex and during happy facial expressions. There
were no between-group differences in BOLD response during fearful facial expressions. Fig. 4 shows PPI results and correlations between mean amygdala BOLD response and emotion recognition. In PPI the FR group showed decreased functional connectivity between both the left and right amygdala and right dorsolateral prefrontal cortex (dlPFC) during happy facial expressions and between the left amygdala and left superior temporal gyrus during fearful facial expressions. There was a statistically significant (p b 0.05) correlation between mean % BOLD response during fearful facial expressions in the amygdala and emotion recognition in the FR group. No such relationship was present in controls. 4. Discussion The present study found abnormal function of PFC structures in the FR group during happy facial expressions, specifically increased
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx Table 3 Between group deviation clusters during happy facial expression. Anatomical region(s) corresponding cluster
BA
Superior frontal gyrus L/premotor cortex L Anterior paracingulate cortex
MNI-coordinates
Voxels
Z-scores
46
431
3.76
22
218
3.8
X
Y
Z
6/8
0
46
32
4
40
BA = Brodmann's area.
activation in the superior frontal gyrus and decreased deactivation in the anterior paracingulate cortex. It has been reported that dynamic happy faces are vivid at the very early stage of visual perception (Becker et al., 2011). This serves a useful feature as during social interaction we recognize other people's good intentions immediately. Our results therefore indicate that FR subjects may not have the same vivid response to dynamic happy faces, which may be a risk factor for social withdrawal due to a lack of enjoyment of social interactions. Our discovery of increased activation of the premotor cortex in the FR group has support in a previous facial affect FR study (Li et al., 2012). As the left premotor cortex serves motor planning and executional tasks (Rushworth et. al., 2003), elevated activity may be best understood as a trait of increased effort for emotion recognition in the FR group during this task. The anterior paracingulate cortex has been previously associated with theory of mind and emotional tasks (Brunet-Gouet and Decety, 2006; Gallagher and Frith, 2003). Structural MRI studies with subjects at genetic risk for psychosis have reported cortical thinning in this region (Byun et al., 2012; Rosso et al., 2010) which could be consistent with our discovery, as cortical thinning may cause, or follow, altered function. Abnormal PFC function during cognitive tasks is a well-reported finding in previous studies with patients with psychotic disorder. Both hypoactivation (Hill et al., 2004) and hyperactivation of prefrontal structures (Callicott et al., 2003; Manoach et al., 1999) have been reported, depending on the difficulty of the task. This has led to the proposition of an inverted U-shaped working load curve for the PFC, in which patients with psychosis reach capacity earlier, followed by hypoactivation of the PFC (Whalley et al., 2005). PFC function may be also affected by factors like the difficulty of the task. One previous FR study using emotional figures detected hyper-activation of the medial prefrontal cortex (mPFC) (van Buuren et al., 2011). Differences between our results and theirs may be explained by differences in the valence used. Negative symptoms and other features of schizophrenia may also affect the working load curve of the PFC. Hypo-activation of the PFC has been reported in FR subjects with depression (Whalley et al., 2008) and in FR subjects expressing negative symptoms of schizophrenia (Barbour et al., 2012). In our study only five FR subjects had prodromal symptoms according to SIPS interview, and one FR subject and three controls had current major depression according to the SCID interview. Previous studies with patients with schizophrenia have reported inefficient task-induced deactivation in prefrontal structures (Guerrero-Pedraza et al., 2012; Schneider et al., 2011). There is also evidence that patients with schizophrenia and their relatives have an inability to switch from the default mode network, which is a resting state network, to a task positive network (Whitfield-Gabrieli et al., 2009). This may be the background to our discovery of decreased deactivation of the anterior paracingulate cortex. Our study found decreased functional connectivity between the amygdala and PFC. Similar results were found in another FR study (Diwadkar et al., 2012). Impaired interplay of the amygdala and PFC during emotional input also supports the notion of abnormal PFC function in the FR group. This may lead to functional compensations in other brain regions as they take a greater role in processing emotions from facial expressions, which might have been the cause for the statistically significant correlation between mean amygdala response and emotion recognition in the FR group but not controls.
7
There were several strengths of our study. First we had a unique data setting: participants were born in the same area and were the same age at the time of the field study. Subjects were in their early twenties which is the peak age for developing schizophrenia, which may increase our sensitivity to detect abnormalities in the at-risk population (Hafner et al., 1998; Paus et al., 2008). Subjects also had a similar ethnic background in each group and subjects' family background was explicitly determined by FHDR. The FHDR was used for exclusion criteria among other criteria in the cohort members with psychoses too. Attrition analysis did not indicate any major bias. Subjects were scanned with the same fMRI device and groups had same sex distribution, avoiding sex-related activation differences. There were also some limitations in our study. First, participation rates were relatively low (20% of the FR group and 29% of the control group). The number of subjects with FR for schizophrenia was also relatively low which may have reduced brain activation differences between study groups. The emotion recognition test data was missing for five FR subjects and two controls. Also, the emotion recognition test showed that both groups did not recognize fearful facial expression as well as they did happy expressions. This may mean that fMRI stimuli for fearful facial expressions were somewhat ambiguous. Handedness did differ significantly between study groups although there were still very few left-handed subjects in both groups. A further limitation is that subjects who had prodromal symptoms according to SIPS were excluded from the control group but not from FR group. This was done to guarantee a control group without psychotic-like symptoms. Our findings support conceptualizations of schizophrenia that involve dysfunction of the brain processes underpinning social interaction. Subjects with FR for psychosis showed abnormal BOLD responses in PFC structures during happy facial expressions. FR subjects also had abnormal functional connectivity between the amygdala and PFC structures, supporting the notion of an abnormally-functioning PFC in the FR group. This may lead to compensations of other brain regions, as they may have to take a greater role in FR subjects. This may explain our discovery of a significant relationship between mean amygdala response and emotion recognition in the FR group but not in controls. Role of funding source The study was funded by grants from the Academy of Finland, (#124257, #212818, #214273) the Sigrid Juselius Foundation and the Signe and Ane Gyllenberg Foundation, Finland.
Contributors Authors Vesa Kiviniemi, Jouko Miettunen, Jenny Barnett, Graham K. Murray, Peter Jones, Irma Moilanen, Sari Mukkala, Pirjo Mäki and Juha Veijola planned the study. Authors Juha Nikkinen and Vesa Kiviniemi conducted the brain imaging. Authors Sari Mukkala, Pirjo Mäki, Jouko Miettunen, Juha Veijola and Jenni Koivukangas performed the data collection. Author Johannes Pulkkinen conducted statistical analyses, analysed the fMRI data and wrote the first draft of the manuscript. All authors have accepted the final version of the manuscript. Conflict of interest All the authors declared no conflicts of interest.
Acknowledgement The facial stimuli used in present study were contributed by professor Mikko Sams (Department of Biomedical Engineering and Computational Science (BECS), the Aalto University School of Science and Technology, Helsinki, Finland).
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.schres.2015.01.039. References Adolphs, R., 2002. Neural systems for recognizing emotion. Curr. Opin. Neurobiol. 12 (2), 169–177.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039
8
J. Pulkkinen et al. / Schizophrenia Research xxx (2015) xxx–xxx
Adolphs, R., Tranel, D., Damasio, H., Damasio, A., 1994. Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372 (6507), 669–672. Allison, T., Puce, A., McCarthy, G., 2000. Social perception from visual cues: role of the STS region. Trends Cogn. Sci. 4 (7), 267–278 (Regul.Ed.). Barbour, T., Murphy, E., Pruitt, P., Eickhoff, S.B., Keshavan, M.S., Rajan, U., Zajac-Benitez, C., Diwadkar, V.A., 2010. Reduced intra-amygdala activity to positively valenced faces in adolescent schizophrenia offspring. Schizophr. Res. 123 (2–3), 126–136. Barbour, T., Pruitt, P., Diwadkar, V.A., 2012. fMRI responses to emotional faces in children and adolescents at genetic risk for psychiatric illness share some of the features of depression. J. Affect. Disord. 136 (3), 276–285. Becker, D.V., Neel, R., Srinivasan, N., Neufeld, S., Kumar, D., Fouse, S., 2011. The vividness of happiness in dynamic facial displays of emotion. PLoS ONE 7 (1), e26551. Breiter, H.C., Etcoff, N.L., Whalen, P.J., Kennedy, W.A., Rauch, S.L., Buckner, R.L., Strauss, M.M., Hyman, S.E., Rosen, B.R., 1996. Response and habituation of the human amygdala during visual processing of facial expression. Neuron 17 (5), 875–887. Brunet-Gouet, E., Decety, J., 2006. Social brain dysfunctions in schizophrenia: a review of neuroimaging studies. Psychiatry Res. Neuroimaging 148 (2–3), 75–92. Byun, M.S., Kim, J.S., Jung, W.H., Jang, J.H., Choi, J., Kim, S.N., Choi, C., Chung, C.K., An, S.K., Kwon, J.S., 2012. Regional cortical thinning in subjects with high genetic loading for schizophrenia. Schizophr. Res. 141 (2–3), 197–203. Callicott, J.H., Mattay, V.S., Verchinski, B.A., Marenco, S., Egan, M.F., Weinberger, D.R., 2003. Complexity of prefrontal cortical dysfunction in schizophrenia: more than up or down. Am. J. Psychiatry 160 (12), 2209–2215. Chan, R.C.K., Li, H., Cheung, E.F.C., Gong, Q., 2010. Impaired facial emotion perception in schizophrenia: a meta-analysis. Psychiatry Res. 178 (2), 381–390. Clark, V.P., Keil, K., Maisog, J.M., Courtney, S., Ungerleider, L.G., Haxby, J.V., 1996. Functional magnetic resonance imaging of human visual cortex during face matching: a comparison with positron emission tomography. Neuroimage 4 (1), 1–15. Comparelli, A., Corigliano, V., De Carolis, A., Mancinelli, I., Trovini, G., Ottavi, G., Dehning, J., Tatarelli, R., Brugnoli, R., Girardi, P., 2013. Emotion recognition impairment is present early and is stable throughout the course of schizophrenia. Schizophr. Res. 143 (1), 65–69. Daros, A.R., Ruocco, A.C., Reilly, J.L., Harris, M.S.H., Sweeney, J.A., 2014. Facial emotion recognition in first-episode schizophrenia and bipolar disorder with psychosis. Schizophr. Res. 153 (1–3), 32–37. de Achával, D., Villarreal, M.F., Costanzo, E.Y., Douer, J., Castro, M.N., Mora, M.C., Nemeroff, C.B., Chu, E., Bär, K., Guinjoan, S.M., 2012. Decreased activity in right-hemisphere structures involved in social cognition in siblings discordant for schizophrenia. Schizophr. Res. 134 (2–3), 171–179. Diwadkar, V.A., Wadehra, S., Pruitt, P., Keshavan, M.S., Rajan, U., Zajac-Benitez, C., Eickhoff, S.B., 2012. Disordered corticolimbic interactions during affective processing in children and adolescents at risk for schizophrenia revealed by functional magnetic resonance imaging and dynamic causal modeling. Arch. Gen. Psychiatry 69 (3), 231–242. Dolan, R.J., Fletcher, P., Morris, J., Kapur, N., Deakin, J.F.W., Frith, C.D., 1996. Neural activation during covert processing of positive emotional facial expressions. Neuroimage 4 (3), 194–200. First, M., Spitzer, R., Gibbon, M., Williams, J., 2002. Structured Clinical Interview for DSMIV-TR Axis I Disorders, Research Version, Patient Edition With Psychotic Screen (SCID-I/P W/PSYSCREEN). Biometrics Research, New York State Psychiatric Institute, New York. Friston, K.J., Buechel, C., Fink, G.R., Morris, J., Rolls, E., Dolan, R.J., 1997. Psychophysiological and modulatory interactions in neuroimaging. Neuroimage. 6 (3), 218–229. Gallagher, H.L., Frith, C.D., 2003. Functional imaging of ‘theory of mind’. Trends Cogn. Sci. 7 (2), 77–83 (Regul.Ed.). Gottesman, I.I., 1991. Schizophrenia Genesis. The Origins of Madness. W. H. Freeman and Company, New York, NY. Guerrero-Pedraza, A., McKenna, P.J., Gomar, J.J., Sarro, S., Salvador, R., Amann, B., Carrion, M.I., Landin-Romero, R., Blanch, J., Pomarol-Clotet, E., 2012. First-episode psychosis is characterized by failure of deactivation but not by hypo- or hyperfrontality. Psychol. Med. 42 (1), 73–84. Habel, U., Chechko, N., Pauly, K., Koch, K., Backes, V., Seiferth, N., Shah, N.J., Stöcker, T., Schneider, F., Kellermann, T., 2010. Neural correlates of emotion recognition in schizophrenia. Schizophr. Res. 122 (1–3), 113–123. Hafner, H., an der Heiden, W., Behrens, S., Gattaz, W.F., Hambrecht, M., Loffler, W., Maurer, K., Munk-Jorgensen, P., Nowotny, B., Riecher-Rossler, A., Stein, A., 1998. Causes and consequences of the gender difference in age at onset of schizophrenia. Schizophr. Bull. 24 (1), 99–113. Haxby, J.V., Horwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P., Grady, C.L., 1994. The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. J. Neurosci. 14 (11 Pt 1), 6336–6353. Hill, K., Mann, L., Laws, K.R., Stephenson, C.M., Nimmo-Smith, I., McKenna, P.J., 2004. Hypofrontality in schizophrenia: a meta-analysis of functional imaging studies. Acta Psychiatr. Scand. 110 (4), 243–256. Jarvelin, M.R., Hartikainen-Sorri, A.L., Rantakallio, P., 1993. Labour induction policy in hospitals of different levels of specialisation. Br. J. Obstet. Gynaecol. 100 (4), 310–315. Jenkinson, M., Smith, S., 2001. A global optimisation method for robust affine registration of brain images. Med. Image Anal. 5 (2), 143–156. Jenkinson, M., Bannister, P., Brady, M., Smith, S., 2002. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 17 (2), 825–841.
Kätsyri, J., 2006. Human Recognition of Basic Emotions From Posed and Animated Dynamic Facial Expressions. Helsinki University of Technology (Dissertation, http:// lib.tkk.fi/Diss/2006/isbn951228538X/isbn951228538X.pdf). Li, H., Chan, R.C.K., Gong, Q., Liu, Y., Liu, S., Shum, D., Ma, Z., 2012. Facial emotion processing in patients with schizophrenia and their non-psychotic siblings: a functional magnetic resonance imaging study. Schizophr. Res. 134 (2–3), 143–150. Mäki, P., Koskela, S., Murray, G.K., Nordström, T., Miettunen, J., Jääskeläinen, E., Veijola, J.M., 2014. Difficulty in making contact with others and social withdrawal as early signs of psychosis in adolescents — the Northern Finland Birth Cohort 1986. Eur. Psychiatry 29 (6), 345–351. Manoach, D.S., Press, D.Z., Thangaraj, V., Searl, M.M., Goff, D.C., Halpern, E., Saper, C.B., Warach, S., 1999. Schizophrenic subjects activate dorsolateral prefrontal cortex during a working memory task, as measured by fMRI. Biol. Psychiatry 45 (9), 1128–1137. McGlashan, T., Miller, T.J., Woods, S.W., Rosen, J.L., Hoffman, R.E., Davidson, L., 2001. Structured Interview for Prodromal Syndromes — Version for Present Prodromal Syndromes. Version 3 ed. Yale School, New Haven, CT. Mukkala, S., Ilonen, T., Nordstrom, T., Miettunen, J., Loukkola, J., Barnett, J.H., Murray, G.K., Jaaskelainen, E., Maki, P., Taanila, A., Moilanen, I., Jones, P.B., Heinimaa, M., Veijola, J., 2011. Different vulnerability indicators for psychosis and their neuropsychological characteristics in the Northern Finland 1986 Birth Cohort. J. Clin. Exp. Neuropsychol. 33 (4), 385–394. Nakamura, K., Kawashima, R., Ito, K., Sugiura, M., Kato, T., Nakamura, A., Hatano, K., Nagumo, S., Kubota, K., Fukuda, H., Kojima, S., 1999. Activation of the Right Inferior Frontal Cortex During Assessment of Facial Emotion. J. Neurophysiol. 82 (3), 1610–1614. Paus, T., Keshavan, M., Giedd, J.N., 2008. Why do many psychiatric disorders emerge during adolescence? Nat. Rev. Neurosci. 9 (12), 947–957. Pelletier, A.L., Dean, D.J., Lunsford-Avery, J.R., Smith, A.K., Orr, J.M., Gupta, T., Millman, Z.B., Mittal, V.A., 2013. Emotion recognition and social/role dysfunction in non-clinical psychosis. Schizophr. Res. 143 (1), 70–73. Perala, J., Suvisaari, J., Saarni, S.I., Kuoppasalmi, K., Isometsa, E., Pirkola, S., Partonen, T., Tuulio-Henriksson, A., Hintikka, J., Kieseppa, T., Harkanen, T., Koskinen, S., Lonnqvist, J., 2007. Lifetime prevalence of psychotic and bipolar I disorders in a general population. Arch. Gen. Psychiatry 64 (1), 19–28. Phillips, M.L., Young, A.W., Senior, C., Brammer, M., Andrew, C., Calder, A.J., Bullmore, E.T., Perrett, D.I., Rowland, D., Williams, S.C.R., Gray, J.A., David, A.S., 1997. A specific neural substrate for perceiving facial expressions of disgust. Nature 389 (6650), 495–498. Rahko, J., Paakki, J.J., Starck, T., Nikkinen, J., Remes, J., Hurtig, T., Kuusikko-Gauffin, S., Mattila, M.L., Jussila, K., Jansson-Verkasalo, E., Katsyri, J., Sams, M., Pauls, D., Ebeling, H., Moilanen, I., Tervonen, O., Kiviniemi, V., 2010. Functional mapping of dynamic happy and fearful facial expression processing in adolescents. Brain Imaging Behav. 4 (2), 164–176. Rasetti, R., Mattay, V.S., Wiedholz, L.M., Kolachana, B.S., Hariri, A.R., Callicott, J.H., MeyerLindenberg, A., Weinberger, D.R., 2009. Evidence that altered amygdala activity in schizophrenia is related to clinical state and not genetic risk. Am. J. Psychiatry 166 (2), 216–225. Rosso, I.M., Makris, N., Thermenos, H.W., Hodge, S.M., Brown, A., Kennedy, D., Caviness, V.S., Faraone, S.V., Tsuang, M.T., Seidman, L.J., 2010. Regional prefrontal cortex gray matter volumes in youth at familial risk for schizophrenia from the Harvard Adolescent High Risk Study. Schizophr. Res. 123 (1), 15–21. Rushworth, M.F.S., Johansen-Berg, H., Göbel, S.M., Devlin, J.T., 2003. The left parietal and premotor cortices: motor attention and selection. Neuroimage 20, S89–S100 (Suppl. 1). Schneider, F.C., Royer, A., Grosselin, A., Pellet, J., Barral, F., Laurent, B., Brouillet, D., Lang, F., 2011. Modulation of the default mode network is task-dependant in chronic schizophrenia patients. Schizophr. Res. 125 (2–3), 110–117. Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 17 (3), 143–155. van Buuren, M., Vink, M., Rapcencu, A.E., Kahn, R.S., 2011. Exaggerated brain activation during emotion processing in unaffected siblings of patients with schizophrenia. Biol. Psychiatry 70 (1), 81–87. Veijola, J., Maki, P., Jaaskelainen, E., Koivukangas, J., Moilanen, I., Taanila, A., Nordstrom, T., Hurtig, T., Kiviniemi, V., Mukkala, S., Heinimaa, M., Lindholm, P., Jones, P.B., Barnett, J.H., Murray, G.K., Miettunen, J., 2013. Young people at risk for psychosis: case finding and sample characteristics of the Oulu Brain and Mind Study. Early Interv. Psychiatry 7 (2), 146–154. Whalley, H.C., Whyte, M.C., Johnstone, E.C., Lawrie, S.M., 2005. Neural correlates of enhanced genetic risk for schizophrenia. Neuroscientist 11 (3), 238–249. Whalley, H.C., Mowatt, L., Stanfield, A.C., Hall, J., Johnstone, E.C., Lawrie, S.M., McIntosh, A.M., 2008. Hypofrontality in subjects at high genetic risk of schizophrenia with depressive symptoms. J. Affect. Disord. 109 (1–2), 99–106. Whitfield-Gabrieli, S., Thermenos, H.W., Milanovic, S., Tsuang, M.T., Faraone, S.V., McCarley, R.W., Shenton, M.E., Green, A.I., Nieto-Castanon, A., LaViolette, P., Wojcik, J., Gabrieli, J.D., Seidman, L.J., 2009. Hyperactivity and hyperconnectivity of the default network in schizophrenia and in first-degree relatives of persons with schizophrenia. Proc. Natl. Acad. Sci. U. S. A. 106 (4), 1279–1284. Woolrich, M.W., Behrens, T.E.J., Beckmann, C.F., Jenkinson, M., Smith, S.M., 2004. Multilevel linear modelling for FMRI group analysis using Bayesian inference. Neuroimage. 21 (4), 1732–1747. Worsley, K.J., Jezzard, P., Matthews, P.M., Smith, S.M., 2001. Statistical analysis of activation images. Ch 14. Functional MRI: An Introduction to Methods.
Please cite this article as: Pulkkinen, J., et al., Functional mapping of dynamic happy and fearful facial expressions in young adults with familial risk for psychosis — Oulu Brain and Mind Study, Schizophr. Res. (2015), http://dx.doi.org/10.1016/j.schres.2015.01.039