Nina Y. Seiferth

Test-retest reliability of resting-state connectivity network characteristics using fMRI and graph theoretical measures.

Characterizing the brain connectome using neuroimaging data and measures derived from graph theory emerged as a new approach that has been applied to brain maturation, cognitive function and neuropsychiatric disorders. For a broad application of this method especially for clinical populations and longitudinal studies, the reliability of this approach and its robustness to confounding factors need to be explored. Here we investigated test-retest reliability of graph metrics of functional networks derived from functional magnetic resonance imaging (fMRI) recorded in 33 healthy subjects during rest. We constructed undirected networks based on the Anatomic-Automatic-Labeling (AAL) atlas template and calculated several commonly used measures from the field of graph theory, focusing on the influence of different strategies for confound correction. For each subject, method and session we computed the following graph metrics: clustering coefficient, characteristic path length, local and global efficiency, assortativity, modularity, hierarchy and the small-worldness scalar. Reliability of each graph metric was assessed using the intraclass correlation coefficient (ICC). Overall ICCs ranged from low to high (0 to 0.763) depending on the method and metric. Methodologically, the use of a broader frequency band (0.008-0.15 Hz) yielded highest reliability indices (mean ICC=0.484), followed by the use of global regression (mean ICC=0.399). In general, the second order metrics (small-worldness, hierarchy, assortativity) studied here, tended to be more robust than first order metrics. In conclusion, our study provides methodological recommendations which allow the computation of sufficiently robust markers of network organization using graph metrics derived from fMRI data at rest.

Gender differences in the cognitive control of emotion: An fMRI study.

The interaction of emotion and cognition has become a topic of major interest. However, the influence of gender on the interplay between the two processes, along with its neural correlates have not been fully analysed so far. In this functional magnetic resonance imaging (fMRI) study we induced negative emotion using negative olfactory stimulation while male (n=21) and female (n=19) participants performed an n-back verbal working memory task. Based on findings indicating increased emotional reactivity in women, we expected the female participants to exhibit stronger activation in characteristically emotion-associated areas during the interaction of emotional and cognitive processing in comparison to the male participants. Both groups were found to be significantly impaired in their working memory performance by negative emotion induction. However, fMRI analysis revealed distinct differences in neuronal activation between groups. In men, cognitive performance under negative emotion induction was associated with extended activation patterns in mainly prefrontal and superior parietal regions. In women, the interaction between emotion and working memory yielded a significantly stronger response in the amygdala and the orbitofrontal cortex (OFC) compared to their male counterparts. Our data suggest that in women the interaction of verbal working memory and negative emotion is associated with relative hyperactivation in more emotion-associated areas whereas in men regions commonly regarded as important for cognition and cognitive control are activated. These results provide new insights in gender-specific cerebral mechanisms.

Increased neural response related to neutral faces in individuals at risk for psychosis.

OBJECTIVEnThe reliable discrimination of emotional expressions in faces is essential for adequate social interaction. Deficits in facial emotion processing are an important impairment in schizophrenia with major consequences for social functioning and subjective well-being. Whether neural circuits underlying emotion processing are already altered before illness onset is yet unclear. Investigating neural correlates of emotion processing in individuals clinically at risk for psychosis offers the possibility to examine neural processes unchanged by the manifest disorder and to study trait aspects of emotion dysfunctions.nnnMATERIAL AND METHODSnTwelve subjects clinically at risk for psychosis and 12 matched control subjects participated in this study. fMRI data were acquired during an emotion discrimination task consisting of standardized photographs of faces displaying different emotions (happiness, sadness, anger, fear) as well as faces with neutral facial expression.nnnRESULTSnThere were no group differences in behavioral performance. Emotion discrimination was associated with hyperactivations in high-risk subjects in the right lingual and fusiform gyrus as well as the left middle occipital gyrus. Further, high-risk compared to control subjects exhibited stronger activation related to neutral faces relative to emotional faces in the inferior and superior frontal gyri, the cuneus, the thalamus and the hippocampus.nnnCONCLUSIONSnThe present study indicates that individuals clinically at risk for psychosis show differences in brain activation associated with processing of emotional and--more pronounced--neutral facial expressions despite an adequate behavioral performance. The proneness to attribute salience to neutral stimuli might indicate a biological risk marker for psychosis.

The neural basis of video gaming

Video game playing is a frequent recreational activity. Previous studies have reported an involvement of dopamine-related ventral striatum. However, structural brain correlates of video game playing have not been investigated. On magnetic resonance imaging scans of 154 14-year-olds, we computed voxel-based morphometry to explore differences between frequent and infrequent video game players. Moreover, we assessed the Monetary Incentive Delay (MID) task during functional magnetic resonance imaging and the Cambridge Gambling Task (CGT). We found higher left striatal grey matter volume when comparing frequent against infrequent video game players that was negatively correlated with deliberation time in CGT. Within the same region, we found an activity difference in MID task: frequent compared with infrequent video game players showed enhanced activity during feedback of loss compared with no loss. This activity was likewise negatively correlated with deliberation time. The association of video game playing with higher left ventral striatum volume could reflect altered reward processing and represent adaptive neural plasticity.

Neuronal Correlates of Facial Emotion Discrimination in Early Onset Schizophrenia

Emotion discrimination deficits represent a well-established finding in schizophrenia. Although imaging studies addressed the cerebral dysfunctions underlying emotion perception in adult patients, the question of trait vs state characteristics is still unresolved. The investigation of juvenile patients offers the advantage of studying schizophrenia at an age where influences of illness course and long-term medication are minimized. This may enable a more detailed characterization of emotion discrimination impairments and their cerebral correlates with respect to their appearance and exact nature. A total of 12 juvenile patients with early onset schizophrenia and matched healthy juveniles participated in this study. fMRI data were acquired during an emotion discrimination task consisting of standardized photographs of faces displaying happy, sad, angry, fearful, or neutral facial expression. Similar to findings in adult patients, juvenile patients exhibited reduced performance specificity whereas sensitivity was unaffected. Independent of the valence, their processing of emotional faces was associated with hypoactivations in both fusiform gyri and in the left inferior occipital gyrus. In addition, hyperactivations in patients were found in the right cuneus common to happy, angry, and fearful faces. Further, most distinct changes were present in juvenile patients when processing sad faces. These results point to a dysfunction in cerebral circuits relevant for emotion processing already prominent in adolescent schizophrenia patients. Regions affected by a decrease in activation are related to visual and face processing, similar to deficits reported in adult patients. These changes are accompanied by hyperactivations in areas related to emotion regulation and attribution, possibly reflecting compensatory mechanisms.

Neural correlates of emotion recognition in schizophrenia

The following fMRI study aimed to characterize the neural correlates of explicit emotion discrimination in 17 patients with schizophrenia and 17 matched healthy controls. In patients, emotion recognition impairments were found to be paralleled by cerebral dysfunctions in the affective division of the anterior cingulate cortex, the bilateral dorsomedial prefrontal cortex, the right superior temporal gyrus and the right fusiform gyrus. While the patients responses to emotional faces were characterized predominantly by hypoactivations, the neutral faces elicited hyperactivations mainly in the frontal and cingulate areas, and the basal ganglia, along with misattribution errors. The decreased activation in the fusiform face area during responses to both emotional and neutral stimuli may be indicative of general face processing deficits. Similar although less pronounced deficits have been observed in subjects at high risk of psychosis as well as in patients with early onset. In adult schizophrenia, the evidence of an imbalanced cerebral network appears early in the course of the illness, with the dysfunctions, as indicated by correlations here, becoming more pronounced in patients with longer illness duration.

Cerebral dysfunctions of emotion-cognition interactions in adolescent-onset schizophrenia.

OBJECTIVEnSchizophrenia is among the most severe of psychiatric disorders, leading to impairments of affective and cognitive abilities. These dysfunctions affect each other mutually. Adolescent-onset schizophrenia (AOS) constitutes a particularly severe form of the disorder. In this study, possible dysfunctions of the neural correlates underlying the interaction of negative emotion and working memory in AOS were investigated.nnnMETHODnDuring functional magnetic resonance imaging, 12 patients with AOS and 12 non-AOS adolescents performed a verbal n-back task. Intermittently, negative and neutral emotions were induced by olfactory stimulation. Group differences in working memory, emotion, and their interaction were evaluated.nnnRESULTSnIn patients with AOS, lower performance sensitivity was observed, along with dorsolateral prefrontal, anterior cingulate, and inferior parietal hypoactivation during working memory demands. For negative versus neutral emotion induction, patients with AOS mainly showed increased brain activation compared with control subjects in widespread brain regions including the left orbitofrontal cortex and the medial frontal gyrus. Finally, during the interaction of emotion and cognition, altered patterns of activation in patients with AOS were found in the thalamocortical network, including the angular and the middle cingulate gyri extending to the precuneus. These activation differences were further decomposed by parameter estimates.nnnCONCLUSIONSnOur results provide new insights into the neural correlates underlying the mutual influence of affective and cognitive symptoms in AOS. During the n-back task, areas typically associated with working memory performance were found hypoactivated in patients relative to the control subjects, including the dorsolateral prefrontal and parietal cortex and the anterior cingulate. However, patients with AOS mainly demonstrated increased activation in key areas of emotion processing, such as the left orbitofrontal cortex and medial frontal areas, during negative emotion induction. A dysfunctional thalamocortical network during the interaction mainly included regions involved in the integration of converging information--either on the subcortical (thalamus) or on a higher-order cortical level (comprising the angular gyrus). These findings point to dysfunctional emotion-cognition interactions in AOS, which may explain its poor prognosis.

The influence of olfactory-induced negative emotion on verbal working memory: individual differences in neurobehavioral findings.

The influence of emotion on cognition plays an important role in peoples everyday life as well as in psychiatric and neurological disorders. The present study used fMRI to examine the neural correlates of cognitive-emotional interactions and its inter-individual differences. Twenty-one healthy males performed a 0-back/2-back task while negative or neutral emotion was induced by negative/neutral olfactory stimulation. Subjects revealed a differential effect of emotion on cognition; in 9 subjects, negative odor had a deteriorating influence on verbal working memory (affected group, AG) while in 12 subjects, performance was not affected in a negative way (unaffected group, UAG). Although no brain activation differences emerged during the working memory task, the interaction of working memory and emotion yielded significant differences between the AG and the UAG. The latter showed greater activation in the fronto-parieto-cerebellar working memory (WM) network including the precuneus while the AG demonstrated stronger activation in more emotional areas (mainly the temporal and medial frontal cortex) as well as compensatory activations in prefrontal regions known to be essential for the cognitive down-regulation of emotions. Hence, the UAG may have been better able to counteract the detrimental influence of negative stimulation during the 2-back task and to effectively sustain or even increase activation in the task-relevant WM network. Correlation analyses for the whole group supported this interpretation; reduced working memory performance during negative stimulation was accompanied by higher activation in the inferior frontal gyrus whereas less performance impairment was related to higher activation in the precuneus. Results confirm the importance of incorporating individual differences in emotion processing and its interaction with cognitive functions in neuroimaging.

Self-face recognition in schizophrenia

OBJECTIVEnAltered self-awareness might be a core feature of schizophrenia. Facial self-recognition in children and non-human primates has been linked to the emergence of self-awareness. In this study, the ability to recognize the own face as an indicator of certain aspects of self-awareness was investigated in patients with schizophrenia.nnnMETHODSnStandardized facial pictures of the participants (20 patients with DSM-IV schizophrenia and 20 healthy controls), of close same-sex relatives of the participants and of unknown persons were taken. These stimuli were presented on a computer screen serially in three forced choice identity recognition experiments: facial identities were presented (I) for 5 s centrally on the screen; (II) in the participants left and right visual hemifields for 100 ms; (III) as morphed blendings between the identities, centrally for 5 s.nnnRESULTSnThere was no interaction between group and facial identity in experiments I and III. However, in experiment II an interaction between hemifield and identity emerged in the patients (p=.002). They exhibited higher error rates for their own face presented to the right hemifield (p=.003), whereas there was no effect for the control subjects. Additionally, self-face recognition (reaction time in experiment I; p=.0009 and error rates in II; p=.0006) was related to hallucinations in the patients.nnnCONCLUSIONSnThese results support the notion of a specific self-face processing dysfunction in schizophrenia. This might be related to altered self-awareness in schizophrenia.

Altered amygdala activation in schizophrenia patients during emotion processing

Dysfunctional emotion processing in patients suffering from schizophrenia is a prominent clinical feature of great importance for social functioning and subjective well-being. The neurobiological underpinnings are still poorly understood. Here we investigated a large sample of schizophrenia patients and matched healthy controls with an event-related fMRI task during emotion processing using emotional pictures from the International Affective Picture System (IAPS). Schizophrenia patients revealed stronger right amygdala activation during negative and attenuated response during positive affective picture processing compared to healthy controls. Further analysis indicated that medication status influences activation of the ventral anterior cingulate cortex during negative affective stimuli processing. These results might represent a correlate of altered emotional experience in schizophrenia patients who are known to report less positive and more negative affective states in daily life situations.