I am a lecturer at the School of Film and Television and School of Neuroscience, Tel Aviv University, and a researcher at the Tel Aviv Center for Brain Functions, Sourasky Medical Center. My research agenda includes motion pictures and affective neuroscience, virtual reality, and brain-computer interfaces.
Research Items (45)
- Feb 2019
*** link: https://doi.org/10.1101/535377 *** Background: Deep neural networks have revolutionised machine learning, with unparalleled performance in object classification. However, in brain imaging (e.g. fMRI), the direct application of Convolutional Neural Networks (CNN) to decoding subject states or perception from imaging data seems impractical given the scarcity of available data. New method: In this work we propose a robust method to transfer information from deep learning (DL) features to brain fMRI data with the goal of decoding. By adopting Reduced Rank Regression with Ridge Regularisation we establish a multivariate link between imaging data and the fully connected layer (fc7) of a CNN. We exploit the reconstructed fc7 features by performing an object image classification task on two datasets: one of the largest fMRI databases, taken from different scanners from more than two hundred subjects watching different movie clips, and another with fMRI data taken while watching static images. Results: The fc7 features could be significantly reconstructed from the imaging data, and led to significant decoding performance. Comparison with existing methods: The decoding based on reconstructed fc7 outperformed the decoding based on imaging data alone. Conclusion: In this work we show how to improve fMRI-based decoding benefiting from the mapping between functional data and CNN features. The potential advantage of the proposed method is twofold: the extraction of stimuli representations by means of an automatic procedure (unsupervised) and the embedding of high-dimensional neuroimaging data onto a space designed for visual object discrimination, leading to a more manageable space from dimensionality point of view.
Real-time functional magnetic resonance imaging (rt-fMRI) has revived the translational perspective of neurofeedback (NF)¹. Particularly for stress management, targeting deeply located limbic areas involved in stress processing² has paved new paths for brain-guided interventions. However, the high cost and immobility of fMRI constitute a challenging drawback for the scalability (accessibility and cost-effectiveness) of the approach, particularly for clinical purposes³. The current study aimed to overcome the limited applicability of rt-fMRI by using an electroencephalography (EEG) model endowed with improved spatial resolution, derived from simultaneous EEG–fMRI, to target amygdala activity (termed amygdala electrical fingerprint (Amyg-EFP))4–6. Healthy individuals (n = 180) undergoing a stressful military training programme were randomly assigned to six Amyg-EFP-NF sessions or one of two controls (control-EEG-NF or NoNF), taking place at the military training base. The results demonstrated specificity of NF learning to the targeted Amyg-EFP signal, which led to reduced alexithymia and faster emotional Stroop, indicating better stress coping following Amyg-EFP-NF relative to controls. Neural target engagement was demonstrated in a follow-up fMRI-NF, showing greater amygdala blood-oxygen-level-dependent downregulation and amygdala–ventromedial prefrontal cortex functional connectivity following Amyg-EFP-NF relative to NoNF. Together, these results demonstrate limbic specificity and efficacy of Amyg-EFP-NF during a stressful period, pointing to a scalable non-pharmacological yet neuroscience-based training to prevent stress-induced psychopathology.
Impairments in social cognition and interactions are core psychopathologies in schizophrenia, often manifesting as an inability to appropriately relate to the intentions and feelings of others. Neuroimaging has helped to demarcate the dynamics of two distinct functional connectivity circuits underlying the social-affective processes related to mentalization (known as Theory of Mind, ToM) and somatic-affiliation (known as Embodied Simulation, ES). While evidence points to abnormal activation patterns within these networks among those suffering from schizophrenia, it is yet unclear however, if these patients exhibit this abnormal functional connectivity in the context of social-affective experiences. The current fMRI study, investigated functional connectivity dynamics within ToM and ES networks as subjects experienced evolving cinematic portrayals of fear. During scanning, schizophrenia patients and healthy controls passively watched a cinematic scene in which a mother and her son face various threatening events. Participants then provided a continuous and retrospective report of their fear intensity during a second viewing outside the scanner. Using network cohesion index (NCI) analysis, we examined modulations of ES-related and ToM-related functional connectivity dynamics and their relation to symptom severity and the continuous emotional ratings of the induced cinematic fear. Compared to patients, healthy controls showed higher ES-NCI and marginally lower ToM-NCI during emotional peaks. Cross-correlation analysis revealed an intriguing dynamic between NCI and the inter-group difference of reported fear. Schizophrenia patients rated their fear as lower relative to healthy controls, shortly after exhibiting lower ES connectivity. This increased difference in rating was also followed by higher ToM connectivity among schizophrenia patients. The clinical relevance of these findings is further highlighted by the following two results: (a) ToM-NCI was found to have a strong correlation with the severity of general symptoms during one of the two main emotional peaks (Spearman R = 0.77); and (b) k-mean clustering demonstrated that the networks' NCI dynamic during the social-affective context reliably differentiated between patients and controls. Together, these findings point to a possible neural marker for abnormal social-affective processing in schizophrenia, manifested as the disturbed balance between two functional networks involved in social-affective affiliation. This in turn suggests that exaggerated mentalization over somatic-affiliative processing, in response to another's' distress may underlie social-affective deficits in schizophrenia.
Emotion regulation is hypothesized to be mediated by the interactions between emotional reactivity and regulation networks during the dynamic unfolding of the emotional episode. Yet, it remains unclear how to delineate the effective relationships between these networks. In this study, we examined the aforementioned networks’ information flow hierarchy during viewing of an anger provoking movie excerpt. Anger regulation is particularly essential for averting individuals from aggression and violence, thus improving prosocial behavior. Using subjective ratings of anger intensity we differentiated between low and high anger periods of the film. We then applied the Dependency Network Analysis (DEPNA), a newly developed graph theory method to quantify networks’ node importance during the two anger periods. The DEPNA analysis revealed that the impact of the ventromedial prefrontal cortex (vmPFC) was higher in the high anger condition, particularly within the regulation network and on the connections between the reactivity and regulation networks. We further showed that higher levels of vmPFC impact on the regulation network were associated with lower subjective anger intensity during the high-anger cinematic period, and lower trait anger levels. Supporting and replicating previous findings, these results emphasize the previously acknowledged central role of vmPFC in modulating negative affect. We further show that the impact of the vmPFC relies on its correlational influence on the connectivity between reactivity and regulation networks. More importantly, the hierarchy network analysis revealed a link between connectivity patterns of the vmPFC and individual differences in anger reactivity and trait, suggesting its potential therapeutic role.
- Nov 2017
- Anger and Anxiety: Predictors, Coping Strategies, and Health Effects
Military combat training, where stoic-like pedagogy promotes anger regulation, offers a paradigmatic case study for examining response to angering provocations and internalization of regulatory strategies. To this end, we conducted a prospective neuroimaging study prior to and towards the end of intensive infantry training of combat soldiers recruited to the Paratroopers Brigade of the Israel Defense Forces. A control group was recruited consisting of age matched volunteers who took part in one-year pre-army civil service programs. At each time-point, participants played a modified Ultimatum Game (UG) to which anger was further infused using interpersonal insults, while their brains were scanned using functional Magnetic Resonance Imaging. We previously showed that within the 60 participants of the first time-point (38 soldiers and 22 civilians) there were no differences in any of the anger measures including behavior, emotional reports and brain activations between soldiers and civilians. Specifically, we showed that all participants in the first time-point were mostly angry and rejected more offers as those offers became more unequal. Additionally, it was found that participants who gained more money reported less anger and more positive feelings, and had more ventromedial Prefrontal Cortex (vmPFC) and less Locus Coeruleus (LC) activation, while the reverse pattern was found for those participants who gained less money. Congruent with previous findings associating the vmPFC with implicit emotion regulation and the LC with arousal and aggression, we asserted that these two neurobehavioral response patterns reflected a regulated and an unbalanced profile of anger, respectively. Here we examined the results of the second time-point (29 soldiers and 17 civilians), which demonstrated that soldiers with a priori unbalanced anger profile displayed an increase in monetary gain, an increase in reported positive emotions, and an increase in vmPFC activation in response to the anger-infused UG at the end of combat training, thus presenting a regulated anger profile. Soldiers with a priori regulated profile displayed a marginal decrease in monetary gain and an increase in anger, but generally showed no differences compared to their angry colleagues from the first time-point. The civilians control group displayed no changes in any of the anger related measures between time-points. Findings support the formulation of Stoic pedagogy in military practice as a program that empowers anger regulation and suggest that an intense socio-cultural practice such as combat training styles one's mind and body in a fashion that decreases individual variability and produces uniform and regulated responses to anger.
Major methodological advancements have been recently made in the field of neural decoding, which is concerned with the reconstruction of mental content from neuroimaging measures. However, in the absence of a large-scale examination of the validity of the decoding models across subjects and content, the extent to which these models can be generalized is not clear. This study addresses the challenge of producing generalizable decoding models, which allow the reconstruction of perceived audiovisual features from human magnetic resonance imaging (fMRI) data without prior training of the algorithm on the decoded content. We applied an adapted version of kernel ridge regression combined with temporal optimization on data acquired during film viewing (234 runs) to generate standardized brain models for sound loudness, speech presence, perceived motion, face-to-frame ratio, lightness, and color brightness. The prediction accuracies were tested on data collected from different subjects watching other movies mainly in another scanner. Substantial and significant (QFDR<0.05) correlations between the reconstructed and the original descriptors were found for the first three features (loudness, speech, and motion) in all of the 9 test movies (R¯=0.62, R¯ = 0.60, R¯ = 0.60, respectively) with high reproducibility of the predictors across subjects. The face ratio model produced significant correlations in 7 out of 8 movies (R¯=0.56). The lightness and brightness models did not show robustness (R¯=0.23, R¯ = 0). Further analysis of additional data (95 runs) indicated that loudness reconstruction veridicality can consistently reveal relevant group differences in musical experience. The findings point to the validity and generalizability of loudness, speech, motion, and face ratio models for complex cinematic stimuli (as well as for music in the case of loudness). While future research should further validate these models using controlled stimuli and explore the feasibility of extracting more complex models via this method, the reliability of our results indicates the potential usefulness of the approach and the resulting models in basic scientific and diagnostic contexts.
Invasive brain–computer interfaces (BCI) provide better signal quality in terms of spatial localization, frequencies and signal/noise ratio, in addition to giving access to deep brain regions that play important roles in cognitive or affective processes. Despite some anecdotal attempts, little work has explored the possibility of integrating such BCI input into more sophisticated interactive systems like those which can be developed with game engines. In this article, we integrated an amygdala depth electrode recorder with a virtual environment controlling a virtual crowd. Subjects were asked to down regulate their amygdala using the level of unrest in the virtual room as feedback on how successful they were. We report early results which suggest that users adapt very easily to this paradigm and that the timing and fluctuations of amygdala activity during self-regulation can be matched by crowd animation in the virtual room. This suggests that depth electrodes could also serve as high-performance affective interfaces, notwithstanding their strictly limited availability, justified on medical grounds only.
- Apr 2017
Parental empathy is a key component of sensitive parenting that supports children's social adaptation throughout life. Consistent with a two dissociable network perspective on empathy, we measured within- and between-network integrity of two empathy-related networks in the parental brain as predictors of children's social outcomes across the first six years of life. We focused on two empathy networks; embodied simulation, which supports parents' capacity to resonate with infant state and emotions and implicates cingulo-insulary structures, and mentalizing, which underpins parents' theory-of-mind and mental attributions via prefrontal-temporo-parietal circuit. We followed 87 first-time parents across the first six years of family formation, including heterosexual and homosexual parents. In infancy, parents' brain response to own versus unfamiliar infant stimuli was imaged; in preschool, children's cortisol production and emotion regulation were assessed; and at six years, children's behavior problems were reported. Parents' intra- and inter- network integrity increased when viewing their own infant compared to unfamiliar infant, suggesting that attachment stimuli increase network coherence in the parental brain. Functional connectivity within the parent's embodied simulation network in infancy predicted lower child cortisol production while inter-network connectivity among the embodied simulation and mentalizing networks was associated with more advanced child emotion regulation skills in preschool and lower internalizing problems at six years. Children's emotion regulation capacities mediated the link between inter-network integrity in the parental brain and internalizing symptoms. Our findings, the first to demonstrate that integrity of empathy-related networks in the parental brain shape children's long-term stress reactivity and emotional adaptation, highlight the brain component of the parental empathy attribute, suggest that increased coherence within the "parental caregiving network" marks a key feature of parent-infant attachment, and contribute to discussion on biobehavioral mechanisms underpinning the cross-generation transmission of human stress reactivity and sociality.
Identifying vulnerable individuals prone to develop post-traumatic stress symptoms (PTSS) is of paramount importance, especially in populations at high risk for stress exposure such as combat soldiers. While several neural and psychological risk factors are known, no post-traumatic stress disorder (PTSD) biomarker has yet progressed to clinical use. Here we present novel and clinically applicable anger-related neurobehavioral risk markers for military-related PTSS in a large cohort of Israeli soldiers. The psychological, electrophysiological and neural (Simultaneous recording of scalp electroencephalography [EEG] and functional magnetic resonance imaging [fMRI]) reaction to an anger-inducing film were measured prior to advanced military training and PTSS were recorded at 1-year follow-up. Limbic modulation was measured using a novel approach that monitors amygdala modulation using fMRI-inspired EEG, hereafter termed amygdala electrical fingerprint (amyg-EFP). Inter-subject correlation (ISC) analysis on fMRI data indicated that during movie viewing participants’ brain activity was synchronized in limbic regions including the amygdala. Self-reported state-anger and amyg-EFP modulation successfully predicted PTSS levels. State-anger significantly accounted for 20% of the variance in PTSS, and amyg-EFP signal modulation significantly accounted for additional 15% of the variance. Our study was limited by the moderate PTSS levels and lack of systematic baseline symptoms assessment. These results suggest that pre-stress neurobehavioral measures of anger may predict risk for later PTSS, pointing to anger-related vulnerability factors that can be measured efficiently and at a low cost before stress exposure. Possible mechanisms underlying the association between the anger response and risk for PTSS are discussed.
- Jan 2017
The ability to temporarily prioritize rapid and vigilant reactions over slower higher-order cognitive functions is essential for adaptive responding to threat. This reprioritization is believed to reflect shifts in resource allocation between large-scale brain networks that support these cognitive functions, including the salience and executive control networks. However, how changes in communication within and between such networks dynamically unfold as a function of threat-related arousal remains unknown. To address this issue, we collected functional MRI data and continuously assessed the heart rate from 120 healthy human adults as they viewed emotionally arousing and ecologically valid cinematographic material. We then developed an analysis method that tracks dynamic changes in large-scale network cohesion by quantifying the level of within-network and between-network interaction. We found a monotonically increasing relationship between heart rate, a physiological index of arousal, and within-network cohesion in the salience network, indicating that coordination of activity within the salience network dynamically tracks arousal. Strikingly, salience-executive control between-network cohesion peaked at moderate arousal. These findings indicate that at moderate arousal, which has been associated with optimal noradrenergic signaling, the salience network is optimally able to engage the executive control network to coordinate cognitive activity, but is unable to do so at tonically elevated noradrenergic levels associated with acute stress. Our findings extend neurophysiological models of the effects of stress-related neuromodulatory signaling at the cellular level to large-scale neural systems, and thereby explain shifts in cognitive functioning during acute stress, which may play an important role in the development and maintenance of stress-related mental disorders.
Deep neural networks have been developed drawing inspiration from the brain visual pathway, implementing an end-to-end approach: from image data to video object classes. However building an fMRI decoder with the typical structure of Convolutional Neural Network (CNN), i.e. learning multiple level of representations, seems impractical due to lack of brain data. As a possible solution, this work presents the first hybrid fMRI and deep features decoding approach: collected fMRI and deep learnt representations of video object classes are linked together by means of Kernel Canonical Correlation Analysis. In decoding, this allows exploiting the discriminatory power of CNN by relating the fMRI representation to the last layer of CNN (fc7). We show the effectiveness of embedding fMRI data onto a subspace related to deep features in distinguishing semantic visual categories based solely on brain imaging data. (https://arxiv.org/abs/1701.02133)
- Dec 2016
Significance statement: How does brain functioning change in arousing or stressful situations? Extant literature suggests that through global projections, arousal-related neuromodulatory changes can rapidly alter coordination of neural activity across brain-wide neural systems, or large-scale networks. Since it is unknown how such processes unfold, we developed a method to dynamically track levels of within- and between-network interaction. We applied this technique to human neuroimaging data acquired while participants watched realistic and emotionally arousing cinematographic material. Results demonstrate that cohesion within the salience network monotonically increases with arousal, while cohesion of this network with the executive control network peaks at moderate arousal. Our findings explain how cognitive performance shifts as a function of arousal, and provide new insights into vulnerability for stress-related psychopathology.
- Aug 2016
We introduce a novel method for delineating context-dependent functional brain networks whose connectivity dynamics are synchronized with the occurrence of a specific psychophysiological process of interest. In this method of context-related network dynamics analysis (CRNDA), a continuous psychophysiological index serves as a reference for clustering the whole-brain into functional networks. We applied CRNDA to fMRI data recorded during the viewing of a sadness-inducing film clip. The method reliably demarcated networks in which temporal patterns of connectivity related to the time series of reported emotional intensity. Our work successfully replicated the link between network connectivity and emotion rating in an independent sample group for seven of the networks. The demarcated networks have clear common functional denominators. Three of these networks overlap with distinct empathy-related networks, previously identified in distinct sets of studies. The other networks are related to sensorimotor processing, language, attention, and working memory. The results indicate that CRNDA, a data-driven method for network clustering that is sensitive to transient connectivity patterns, can productively and reliably demarcate networks that follow psychologically meaningful processes. Hum Brain Mapp, 2016. © 2016 Wiley Periodicals, Inc.
In the past decade neurofeedback (NF) has become the focus of a growing body of research. With real-time functional magnetic resonance imaging (fMRI) enabling online monitoring of emotion-related areas, such as the amygdala, many have begun testing its therapeutic benefits. However, most existing NF procedures still use monotonic uni-modal interfaces, thus possibly limiting user engagement and weakening learning efficiency. The current study tested a novel multi-sensory NF animated scenario (AS) aimed at enhancing user experience and improving learning. We examined whether relative to a simple uni-modal 2D interface, learning via an interface of complex multi-modal 3D scenario will result in improved NF learning. As a neural-probe, we used the recently developed fMRI-inspired EEG model of amygdala activity (“amygdala-EEG finger print”; amygdala-EFP), enabling low-cost and mobile limbic NF training. Amygdala-EFP was reflected in the AS by the unrest level of a hospital waiting room in which virtual characters become impatient, approach the admission desk and complain loudly. Successful downregulation was reflected as an ease in the room unrest level. We tested whether relative to a standard uni-modal 2D graphic thermometer (TM) interface, this AS could facilitate more effective learning and improve the training experience. Thirty participants underwent two separated NF sessions (1 week apart) practicing downregulation of the amygdala-EFP signal. In the first session, half trained via the AS and half via a TM interface. Learning efficiency was tested by three parameters: (a) effect size of the change in amygdala-EFP following training, (b) sustainability of the learned downregulation in the absence of online feedback, and (c) transferability to an unfamiliar context. Comparing amygdala-EFP signal amplitude between the last and the first NF trials revealed that the AS produced a higher effect size. In addition, NF via the AS showed better sustainability, as indicated by a no-feedback trial conducted in session 2 and better transferability to a new unfamiliar interface. Lastly, participants reported that the AS was more engaging and more motivating than the TM. Together, these results demonstrate the promising potential of integrating realistic virtual environments in NF to enhance learning and improve user’s experience.
- Jul 2016
Music is a powerful means for communicating emotions among individuals. Here we reveal that this continuous stream of affective information is commonly represented in the brains of different listeners and that particular musical attributes mediate this link. We examined participants' brain responses to two naturalistic musical pieces using functional Magnetic Resonance imaging (fMRI). Following scanning, as participants listened to the musical pieces for a second time, they continuously indicated their emotional experience on scales of valence and arousal. These continuous reports were used along with a detailed annotation of the musical features, to predict a novel index of Dynamic Common Activation (DCA) derived from ten large-scale data-driven functional networks. We found an association between the unfolding music-induced emotionality and the DCA modulation within a vast network of limbic regions. The limbic-DCA modulation further corresponded with continuous changes in two temporal musical features: beat-strength and tempo. Remarkably, this "collective limbic sensitivity" to temporal features was found to mediate the link between limbic-DCA and the reported emotionality. An additional association with the emotional experience was found in a left fronto-parietal network, but only among a sub-group of participants with a high level of musical experience (> 5years). These findings may indicate two processing-levels underlying the unfolding of common music emotionality; (1) a widely shared core-affective process that is confined to a limbic network and mediated by temporal regularities in music and (2) an experience based process that is rooted in a left fronto-parietal network that may involve functioning of the 'mirror-neuron system'.
Communication between and within brain regions is essential for information processing within functional networks. The current methods to determine the influence of one region on another are either based on temporal resolution, or require a predefined model for the connectivity direction. However these requirements are not always achieved, especially in fMRI studies, which have poor temporal resolution. We thus propose a new graph theory approach that focuses on the correlation influence between selected brain regions, entitled Dependency Network Analysis (DEPNA). Partial correlations are used to quantify the level of influence of each node during task performance. As a proof of concept, we conducted the DEPNA on simulated datasets and on two empirical motor and working memory fMRI tasks. The simulations revealed that the DEPNA correctly captures the network’s hierarchy of influence. Applying DEPNA to the functional tasks reveals the dynamics between specific nodes as would be expected from prior knowledge. To conclude, we demonstrate that DEPNA can capture the most influencing nodes in the network, as they emerge during specific cognitive processes. This ability opens a new horizon for example in delineating critical nodes for specific clinical interventions.
Recent theoretical and empirical work has highlighted the role of domain-general, large-scale brain networks in generating emotional experiences. These networks are hypothesized to process aspects of emotional experiences that are not unique to a specific emotional category (e.g., “sadness,” “happiness”), but rather that generalize across categories. In this article, we examined the dynamic interactions (i.e., changing cohesiveness) between specific domain-general networks across time while participants experienced various instances of sadness, fear, and anger. We used a novel method for probing the network connectivity dynamics between two salience networks and three amygdala-based networks. We hypothesized, and found, that the functional connectivity between these networks covaried with the intensity of different emotional experiences. Stronger connectivity between the dorsal salience network and the medial amygdala network was associated with more intense ratings of emotional experience across six different instances of the three emotion categories examined. Also, stronger connectivity between the dorsal salience network and the ventrolateral amygdala network was associated with more intense ratings of emotional experience across five out of the six different instances. Our findings demonstrate that a variety of emotional experiences are associated with dynamic interactions of domain-general neural systems.
- Feb 2016
Goal-conflict situations, involving the simultaneous presence of reward and punishment, occur commonly in real life, and reflect well-known individual differences in the behavioral tendency to approach or avoid. However, despite accumulating neural depiction of motivational processing, the investigation of naturalistic approach behavior and its interplay with individual tendencies is remarkably lacking. We developed a novel ecological interactive scenario which triggers motivational behavior under high- or low- goal-conflict conditions. 55 healthy subjects played the game during an fMRI scan. A machine-learning approach was applied to classify approach/avoidance behaviors during the game. To achieve an independent measure of individual tendencies, an integrative profile was composed from three established theoretical models. Results demonstrated that approach under high- relative to low- conflict involved increased activity in the VTA, PAG, VS and precuneus. Notably, only VS and VTA activations during high-conflict discriminated between approach/avoidance personality profiles, suggesting that the relationship between individual personality and naturalistic motivational tendencies is uniquely associated with the mesostriatal pathway. VTA-VS further demonstrated stronger coupling during high- vs. low-conflict. These findings are the first to unravel the multilevel relations between personality profile, approach tendencies in naturalistic set-up and their underlying neural manifestation; thus enabling new avenues for investigating approach-related psychopathologies.
This article reviews significant developments in affective neuroscience suggesting a refinement of the contemporary theoretical discourse on cinematic empathy. Accumulating evidence in the field points to a philogeneticontogenetic-neural boundary separating empathic processes driven by either cognitive or somato-visceral representations of others. Additional evidence suggests that these processes are linked with parasympathetically driven mitigation and proactive sympathetic arousal. It presents empirical findings from a functional magnetic resonance (fMRI) film viewing study, which are in line with this theoretical distinction. The findings are discussed in a proposed cinematographic framework of a general dichotomy between eso (inward-directed) and para (side by side with)—dramatic cinematic factors impinging on visceral representations of real-time occurrences or cognitive representations of another's mind, respectively. It demonstrates the significance of this dichotomy in elucidating the unsettling emotional experience elicited by Michael Haneke's Amour.
We report the development of an affective BCI based on frontal alpha asymmetry neurofeedback, which has previously been used in clinical experiments. Our results evidence a pattern of high-performance for some subjects, combined with high illiteracy, with 52.3% of subjects succeeding in the neurofeedback task. We suggest that individual asymmetry baseline values may be one of the factors explaining BCI illiteracy in this context.
Interactive Narrative is a form of digital entertainment based on AI techniques which support narrative generation and user interaction. Despite recent progress in the field, there is still a lack of unified models integrating narrative generation, user response and interaction. This paper addresses this issue by revisiting existing Interactive Narrative paradigms, granting explicit status to users' disposition towards story characters. We introduce a novel Brain-Computer Interface (BCI) design, which attempts to capture empathy for the main character in a way that is compatible with filmic theories of emotion. Results from two experimental studies with a fully-implemented system demonstrate the effectiveness of a neurofeedback-based approach, showing that subjects can successfully modulate their emotional support for a character who is confronted with challenging situations. A preliminary fMRI analysis also shows activation during user interaction, in regions of the brain associated with emotional control.
The recent development of Brain-Computer Interfaces (BCI) to Virtual World has resulted in a growing interest in realistic visual feedback. In this paper, we investigate the potential role of Virtual Agents in neurofeedback systems, which constitute an important paradigm for BCI. We discuss the potential impact of virtual agents on some important determinants of neurofeedback in the context of affective BCI. Throughout the paper, we illustrate our presentation with two fully implemented neurofeedback prototypes featuring virtual agents: the first is an interactive narrative in which the user empathises with the character through neurofeedback; the second recreates a natural environment in which crowd behaviour becomes a metaphor for arousal and the user engages in emotional regulation.
This paper discusses the potential of Brain-Computer Interfaces based on neurofeedback methods to support emotional control and pursue the goal of emotional control as a mechanism for human augmentation in specific contexts. We illustrate this discussion through two proof-ofconcept, fully-implemented experiments: one controlling disposition towards virtual characters using pre-frontal alpha asymmetry, and the other aimed at controlling arousal through activity of the amygdala. In the first instance, these systems are intended to explore augmentation technologies that would be incorporated into various media-based systems rather than permanently affect user behaviour.
Two empathy-related processes were recently distinguished neuroscientifically: automatic embodied-simulation (ES) based on visceromotor representation of another's affective state via cingulo-insulary circuit, and emotional sharing relying on cognitive "theory of mind" (ToM) via prefrontal-temporo-parietal circuit. Evidence that these regions are not only activated but also function as networks during empathic experience has yet to been shown. Employing a novel approach by analyzing fMRI fluctuations of network cohesion while viewing films portraying personal loss, this study demonstrates increased connectivity during empathic engagement (probed by behavioral and parasympathetic indices) both within these circuits, and between them and a set of limbic regions.Notably, this effect was context-dependent: when witnessing as a determined-loss presented as a future event, the ToM and ToM-limbic cohesion positively correlated with state- and trait-empathy indices. However, when the loss was presented as a probabilistic real-time occurrence, ToM cohesion negatively correlated with state-empathy index, which positively correlated with ES-limbic cohesion. The findings indicate a dichotomy between regulated empathy towards determined-loss and vicarious empathy towards a real-time occurrence.
Interactive Narrative is a form of digital entertainment heavily based on AI techniques to support narrative generation and user interaction, significant progress arriving with the adoption of planning techniques. However, there is a lack of unified models that integrate generation, user responses and interaction. This paper addresses this by revisiting existing Interactive Narrative paradigms, granting explicit status to users’ disposition towards story characters as part of narrative generation as well as adding support for new forms of interaction. We demonstrate this with a novel Brain-Computer Interface (BCI) design, incorporating empathy for a main character derived from brain signals within filmic conceptions of narrative which drives generation using planning techniques. Results from an experimental study with a fully-implemented system demonstrate the effectiveness of a EEG neurofeedback-based approach, showing that subjects can successfully modulate empathic support of a character in a medical drama. MRI analysis also shows activations in associated regions of the brain during expression of support.
- Jan 2012
Dynamic functional integration of distinct neural systems plays a pivotal role in emotional experience. We introduce a novel approach for studying emotion-related changes in the interactions within and between networks using fMRI. It is based on continuous computation of a network cohesion index (NCI), which is sensitive to both strength and variability of signal correlations between pre-defined regions. The regions encompass three clusters (namely limbic, medial prefrontal cortex (mPFC) and cognitive), each previously was shown to be involved in emotional processing. Two sadness-inducing film excerpts were viewed passively, and comparisons between viewer's rated sadness, parasympathetic, and inter-NCI and intra-NCI were obtained. Limbic intra-NCI was associated with reported sadness in both movies. However, the correlation between the parasympathetic-index, the rated sadness and the limbic-NCI occurred in only one movie, possibly related to a "deactivated" pattern of sadness. In this film, rated sadness intensity also correlated with the mPFC intra-NCI, possibly reflecting temporal correspondence between sadness and sympathy. Further, only for this movie, we found an association between sadness rating and the mPFC-limbic inter-NCI time courses. To the contrary, in the other film in which sadness was reported to commingle with horror and anger, dramatic events coincided with disintegration of these networks. Together, this may point to a difference between the cinematic experiences with regard to inter-network dynamics related to emotional regulation. These findings demonstrate the advantage of a multi-layered dynamic analysis for elucidating the uniqueness of emotional experiences with regard to an unguided processing of continuous and complex stimulation.
- Jul 2009
This review describes new developments in the study of transgenerational epigenetic inheritance, a component of epigenetics. We start by examining the basic concepts of the field and the mechanisms that underlie epigenetic inheritance. We present a comprehensive review of transgenerational cellular epigenetic inheritance among different taxa in the form of a table, and discuss the data contained therein. The analysis of these data shows that epigenetic inheritance is ubiquitous and suggests lines of research that go beyond present approaches to the subject. We conclude by exploring some of the consequences of epigenetic inheritance for the study of evolution, while also pointing to the importance of recognizing and understanding epigenetic inheritance for practical and theoretical issues in biology.
Some forty years after the publication of Hannah Arendt's controversial book Eichmann in Jerusalem: A Report on the Banality of Evil, the Israeli-born filmmaker Eyal Sivan released his documentary film The Specialist, explicitly referring to Arendt's work. Sivan took archive footage filmed in 1961, during the trial of Nazi criminal Adolf Eichmann in Jerusalem, and edited it to present a cinematic articulation of Arendt's book. The film discusses the fundamental flaws in the way the trial was conducted as well as the nature of Eichmann's crimes. This article analyzes Sivan's use of narrative, editing, visual, and auditory stylistic devices to expose the way the trial was used by the Zionist movement and to challenge its active role within Zionist collective memory. If interpreted as part of a more general post-Zionist artistic and intellectual production, The Specialist could be understood as deconstructing the accused / accuser dichotomy, and suggesting that the accusers and their contemporary heirs might themselves be guilty of some of the charges made against the defendant.