Research projects

A unique experimental model will be used to study the neurobiological basis of aggressive animal behavior and translate novel findings about its cellular and circuit underpinnings into human phenotypes of pathological aggression. Specifically, the post-weaning social isolation model of early-life adverse experience allows experimental access to dissect the ‘switch’ from functional adaptive aggression to excessive pathological aggressive behavior. In male mice, conspecific chemostimuli trigger innate aggressive behavior. The relevant aggression-promoting circuits along the(sensory) input to (aggressive) output axis comprise the accessory olfactory bulb (AOB), a hub for processing of social chemosignals, and the medial amygdala (MeA), a crucial control centre for regulation of aggressive behavior. Therefore, this project addresses the principles that govern aggression-promoting information transfer along the AOB-to-MeA signaling pathway.

The focus will be on the influences of a provocative context on social threat processing in AMD under different testosterone levels. Specifically, the project aims to analyze the modulating function of context under testosterone application versus suppression on threat sensitivity in healthy controls as well as patient groups. Additionally, we will determine the influence of endogenous hormone variations (testosterone, oxytocin, estrogen and cortisol) on NVS in high versus low aggressive patients in a large group of patients recruited in Q01. With this sample, we will try to identify multidimensional biosignatures based on hormonal levels in combination with fMRI measures of amygdala and amygdala-prefrontal connectivity, NVS measures by questionnaires, aggression measures and psychopathological data.

The neural and neurochemical patterns of acute threat as modulators of aggression in BPD will be investigated in this project. The modulation of aggressive responses under acute threat is induced by the threat-of-shock paradigm. The main translational research question is if and how aggressive responses are modulated by threat, and which neurofunctional and neurochemical patterns underlie these responses during safe and threat conditions. MR spectroscopy will be used in patients to assess glutamate and GABA levels. In a further translational approach, the least and the most aggressive/impulsive recombinant inbred mouse lines identified in C01 in Frankfurt will be tested in Mannheim with animal MR spectroscopy at 9.4T to determine the relationship between glutamate, GABA, impulsivity, and aggression in these mouse lines as well as in comparable brain regions assessing neurofunctional and neurochemical patterns.

We make use of threat-related chemosensory stimuli, namely body odor, acquired during aggressive behavior (boxing) and unconsciously perceived, to investigate heightened amygdala responses to threat stimuli in aggressive patients. Body odors have the major advantage of being directly projected into the amygdala, circumventing cortical preprocessing, thereby enabling the differentiation of mechanisms between bottom-up altered limbic processing and top-down modulated altered cognitive evaluation. We investigate the potential of such body odors to bias responses to ambiguous visual social cues towards threat and their effects during peripersonal space (PPS) violation where they may be especially relevant.

Peripersonal space, the representation of the space immediately surrounding the body, will be studied as an underlying factor for threat experience. Early-life stressors and daily-life stressors will be tested as factors influencing PPS processing and associated specific brain activation patterns. Location tracking and geoinformatics mapping, virtual reality (VR) experiments, physiological stress markers, and brain function during the processing of PPS violations in healthy at- risk individuals will be used to identify predictive biomarkers related to psychiatric risk, enhanced neural behavioral sensitivity to PPS interference and reactive aggression in daily life.

The project employs fMRI and functional near-infrared spectroscopy (fNIRS) hyperscanning techniques to explore how brain-to-brain synchrony and dynamic processes within peer dyads facilitate or inhibit aggressive behavior under diverse levels of provocation in adolescent patients and controls. In two fully interactive tasks, we will probe aggressive behavior towards a task partner, and quantify the building of interpersonal trust/distrust applying a social interaction and economic exchange paradigm. These paradigms will be employed within dyads in fMRI hyperscanning settings and extended by group-based fNIRS methods in triads to study effects of peers, social exclusion, and coalitions on aggressive behavior in semi-naturalistic interactions. Between-brain neural synchrony will be computed and related to everyday social experiences and individual predispositions to identify markers for the prediction of aggressive behavior.

This translational project investigates sex-dependent behavioral effects of faecal microbiota transplantation to microbiome-depleted mice from AMD patients (selected based on their aggressive and impulsive traits from Q01), as well as healthy controls. Impulsivity will be assessed via the continuous performance test and responses towards acute threat via the escalated resident intruder test. The goal is to determine the sex-dependent effects of faecal transplantation on selected readouts involved in the transfer of the patient’s phenotype to the mice, such as immune parameters, sex hormones, neuronal activity (and morphology, e.g., neurite outgrowth, spines, etc.), and gene expression (e.g., Rbfox1 from prior studies and novel candidates from C01 and C04).

Beta-hydroxy-butyrate (BHB), a ketone body, is negatively associated with aggressive behavior. BHB is a metabolite and an active signaling substrate involved in epigenetic regulation of e.g., neurotrophic factor genes in the brain. Of the three main ketone bodies, acetone, acetoacetate and BHB, acetone is a very volatile compound, mainly eliminated through respiration, thus can be measured non-invasively in breath. A reduction of acetone in breath has been found to highly correlate with BHB in blood and be associated with symptom severity in schizophrenia (Jiang et al. 2022). Using MR spectroscopy, A08 aims to (1) identify whether acetone and other volatile organic compounds in breath are associated with aggression and acute threat processing in mental disorders and (2) to examine whether these breath markers are associated with direct metabolic brain correlates (like BHB, glutamate) and with the brain-derived neurotrophic factor (BDNF) levels in plasma. In a translational approach, (3) we will test if supplementation of BHB reduces aggressive behavior in mice.

TDCS will be used as an interventional tool to decrease aggression. Using a simultaneous tDCS – fMRI approach, the project aims to enhance cognitive control by repeated prefrontal brain stimulation, investigating its effect on aggression. In addition to gauging tDCS responsivity, identifying the role of individual factors such as genetic profiles in aggression will be a particular focus of this project. By examining brain activity at multiple time points (e.g., before, during multiple stimulation sessions and after tDCS), it will add to the understanding of mechanisms underlying neural tDCS effects and help to identify individual factors that predict responsiveness to the stimulation. To determine the therapeutic potential, we will include psychiatric patients with substance use problems, a group of criminal, violent offenders, and healthy matched controls.

This project aims to identify cognitive and emotion control deficits in the context of negative valence and threat interference and their association with ACE in young offenders. Complementary to other projects, this project will focus on a group of young people defined by their propensity to aggression showing at the same time more severe psychopathologies. In a series of studies using multimodal imaging (EEG-fMRI, EEG-sMRI) in combination with naturalistic longitudinal follow-up (ecological momentary assessment (EMA)) B02 will identify the neural mechanisms and predictors of self-regulation deficits as a putative common developmental pathway for both, aggressive behavior, and psychopathology. Additionally, B02 will seek to causally confirm neural network mechanisms of inhibitory control and emotion regulation deficits as the basis of aggressive behavior and associated psychopathology by real-time EEG-triggered TMS-stimulation in young offenders.

Probe the self-regulation of CS networks in adults and adolescents diagnosed with mental disorders related to frequent stress-associated affective outbursts and aggressive symptoms in posttraumatic stress disorder (PTSD), and BPD. The patients will subsequently be trained to regulate the frontal control network to varying acute threat in a double-blind, randomized, controlled design. An immersive, virtual brain- computer-interface (BCI) will allow for a culture- and age-sensitive, personalized training approach. The aim of the present investigation is to assess feasibility of the approach according to four clinical markers: Reduction of perceived threat and aggressive behavior in daily life, improved control in the face of unfair provocation, and neurofeedback-specific modulation of the neural networks.

Focus on the neural correlates of characterizing cognitive control deficits during conflict situations. The project will investigate patients with varying levels of cognitive control along with their close partners (sibling or intimate partner) to identify the dynamics of self-regulation and co-regulation in provoked conflict situations in patients with control deficits. To identify the precursors and dynamics of conflict escalation, the project will apply measures of behavioral reactions, skin conductance, simulated or real conflict, fMRI and fMRI-hyperscanning techniques and physiological measures. Neuroimaging data will also be combined with information on stress, control and conflicts in real-life via EMA.

Focus on the investigation of a lack of cognitive control in bullies and victims that contributes to the risk of developing mental health problems. Therefore, the project will assess bullies and their victims in real-life and digital social interactions to investigate how aberrant cognitive and affective prefrontal control and sensitivity to peer rejection with accompanied alterations in autonomic arousal may increase externalizing and internalizing behavior. To this end, a unique combination of ambulatory assessments of (cyber-)bullying, functional neuroimaging (emotion regulation, inhibition, social exclusion), physiological assessments (heart rate variability) and clinical trait-related questionnaires will be applied. Decoding dynamic

Sex-dependent effects and gene-environment interactions will be investigated by applying escalating aggression paradigms. Specifically, the project will investigate the effects of early life stress on aggression in response to threat and hyperactivity as well as social decision-making in 32 BXD mouse strains, the progenitor strains (C057Bl/6J and DBA/2J), and the F1 BXD cross. The project aims to identify the quantitative trait loci (QTL) and putative candidate genes contained within the QTL and associate them with specific behavioral responses of stressed and unstressed cohorts of mice. The publicly available database GeneNetwork (www.genenetwork.org) will be used to validate the findings which include measurements of mRNA and protein expression, and methylation patterns in mouse brains

Develop virtual scenarios to assess decision strategies in cartoon-like and naturalistic contexts. The core question is how healthy individuals and patients make (mal-)adaptive aggressive decisions in social conflicts given their threat sensitivity, cognitive functions, and learning experience. We plan to present mathematically well-defined aggressive decision scenarios to healthy participants as well as patients across diagnostic categories with high scores of aggressive behavior, threat sensitivity, and inference of hostile intent in others. Computational models that accurately explain behavioral choices and neural responses (tested using fMRI and pupillometry) will be developed to identify the aggressive decision strategies humans employ in approach-avoidance conflicts of increasing complexity and ecological realism. The purpose will be to determine if patients use overly aggressive strategies that are not warranted by the necessary defense of self-threats and underlying neural circuits.

Investigate context-dependent aggression triggered by frustrative non-reward or acute social threats. Using newly developed approaches, multiple behavioral domains will be assessed in a semi-naturalistic, autonomous mouse habitat. Specifically, the habitat assesses the inter-individual dynamics of social interactions, aggressions, and hierarchy and the individual reward learning and impulsivity through different integrated modules. Intermittent challenges comprise intruder aggression and frustrative non-rewards. Within this LCD, circuit mechanisms are dissected through chemogenetic interventions, in vivo recordings, and functional MRI in awake mice during task performance. This approach in the first funding period will enable us to disentangle the specific functions of candidate entry points in prefrontal to ventral striatum pathways with respect to their modulation of aggression and dominance for potential interventions.

Address sex-specific NVS (reactive aggression) and CS (different dimensions of psychopathy, proactive aggression) associated risk factors, and risk factor-based biosignatures in young people. Considering the interacting genetic, environmental, and hormonal factors related to these specific aggressive behavior dimensions, C04 will identify specific and shared factors and mechanisms related to NVS and CS in female and male youth with and without pathological aggression. Implementing deep-learning algorithms, sex-specific, data-driven subgroups in relation to dimensions of aggressive behavior will be described and probed against the NVS and CS. Group-level risk factors of aggressive behavior dimensions, and individual risk factor-based subgrouping will be the basis of developing a biologically informed stratification strategy for tailored treatment. Models and classifiers will be established cross-sectionally in available data and replicated in the prospectively collected cross-sectional data (Q01). In addition, C04 will test the models and classifiers for predictive validity in the longitudinal data of the TRR Q01 cohort.

Link questionnaire measures of aggression to specific neural substrates using structural MRI. The resulting patterns of aggression-related neuroanatomical variability will be co- registered with the Allen Human Brain Atlas providing gene-expression data, to highlight genes with a spatial pattern of expression that matches the neuroimaging findings. Utilizing the neurotypical control data, a normative model of neuroanatomical diversity within the NVS and CS will be established to quantify neuroanatomical abnormalities within these systems in individual cases

Identify specific neuronal mechanisms related to the NVS and CS in female and male clinical samples with a history of early-life maltreatment (ELM) who exhibit externalizing, aggressive psychopathologies as opposed to internalizing, non-aggressive psychopathologies. We will therefore explore the interaction of the NVS and CS as well as the modulating effects of theory-of-mind (ToM) on the NVS and CS using a series of fMRI and behavioral tasks. Furthermore, we will investigate the role of hormonal stress responses and will use EMA to assess anger and aggression in everyday life. Thus, we will be able to combine behavioral phenotyping in natural conditions of everyday life and neurobiological correlates of psychopathology in order to detect clinically relevant biosignatures for AMD.

Test the interaction of the CS and frustrative non-reward as part of the NVS. It will investigate the electrophysiological correlates of frustrative feedback in aggression-prone patients. In the aftermath of induced stress, an EEG task-battery including frustrative feedback will be applied for extraction of error-related negativity (ERN) and contingent negative variation to monitor electro-physiologic signaling of the relevant learning and frustration processes. In half of the participants, tDCS over the prefrontal cortex will be applied to enhance cognitive control, with participants being put into a stress context inducing frustration.

The central recruitment platform for collecting and curating a longitudinal dataset for studying individual aggression dynamics related to the neural, cognitive-emotional, neurobiological, psychopathological and environmental factors in patient groups.

Data management and training platform. A decentralized data management infrastructure will help focus on developmental and therapeutic longitudinal data, training all participating researchers in the necessary skills for future use. This strategy will lay the foundations for further data-driven computational modelling projects in the next funding period.

Bundle all training and mentoring activities in a research training group (RTG) for the young researchers participating in this collaborative research center.

Governance, and communication among participating researchers and the public.