The subprojects
1. Extinction learning in the 4th dimension. We plan to study the neural fundaments of extinction of appetitive discrimination in pigeons in a design that permits to analyze the neural events governing acquisition, extinction and renewal along the full time-frame (the 4th dimension).
2. The cellular and the systems level. By using single cell recordings in different forebrain areas and by employing imaging in task performing pigeons at high magnetic field strengths, we are able to track neural changes with different time frames during extinction learning at the cellular and the systems level.
3. Recording neurons during extinction learning. We will record single cells from the pigeon’s visual forebrain during extinction learning. Single unit properties as well as the resulting population code will be the focus of the analyses to reveal the underlying coding schemes.
4. Optogenetics – Causal interventions. By using optogenetic tools, we will be able to intervene into specific system components. Thus, we aim to reveal causal interactions within and between different key areas of the extinction network.
5. Pigeons working in the scanner. Pigeons will extinguish their learned responses to specific cues in a 7T-scanner under context-specific conditions. We can thus visualize long-term learning related changes at the systems level by analyzing BOLD-responses as well as changes of functional connectivity.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Maximizing the ecological validity of human fear and extinction learning research. We will establish a novel conditioned trauma-film paradigm that allows investigating the acquisition and extinction of intrusive memories in an MRI scanner.
2. Assessing the role of cue and context generalization for the formation and extinction of intrusive memories.
3. Characterizing the specific representational features of intrusive memories and differentiating bottom-up and top-down interactions by laminar 7T recordings.
4. Revealing how active avoidance impairs the extinction of intrusive memories. We will embed active avoidance techniques in the conditioned trauma-film paradigm and elucidate their impact on the representational structure of intrusive memories, predictive cues, and trauma contexts.
5. Our overall goal: Understanding the representational structures of extinction learning. Across the SFB funding periods, A02 aims to dissect neural representations and the mechanisms that are responsible for their transformation from the level of single cells, oscillations, and cell layers to large-scale brain networks.
1. Predicting inter-individual learning differences using 7T fMRI and DTI recordings. We will examine if interindividual differences in fear and extinction learning can be predicted by different measures of brain microstructure and functional network connectivity at an unprecedented spatial resolution.
2. Imaging feedforward and feedback interactions within the human extinction network. Successful extinction learning is highly dependent on the interaction between core regions of the extinction network. We will use laminar 7T recordings in order to disentangle feedforward and feedback connectivity.
3. Electrophysiological basis of extinction network interactions. All experiments will be conducted via simultaneous EEG and 7T fMRI recordings in order to unravel the electrophysiological origins of fMRI connectivity. We will relate EEG oscillations at different frequencies to layer-resolved fMRI activity and connectivity.
4. Investigating the physiological foundations of avoidance behavior. Avoidance after fear acquisition may impair subsequent extinction, but the neural mechanisms underpinning this phenomenon are largely unknown in the human brain. By conducting simultaneous EEG and 7T fMRI recordings during an integrated fear acquisition, avoidance, and extinction paradigm, we will investigate how the neural correlates of avoidance behavior influence the neurocognitive processes during subsequent extinction learning.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Neuroanatomical substrates: We shall explore the extent to which brain regions and neuronal subcompartments, that engage in acquisition, EL and retrieval, overlap. We shall investigate if short intervals between acquisition and EL promote generalization of learned experience.
2. Connectivity: Using fMRI we shall identify the functional connectome of appetitive EL. We shall clarify to what extent this differs from brain connectivity during renewal or acquisition.
3. Attention: We will explore to what extent attention determines efficacy of EL and renewal. We shall establish if trial-by-trial learning during acquisition can predict the subsequent efficacy of EL.
4. Role of consolidation: We shall explore to what extent prolonged consolidation (>24h) and long intervals (>weeks) between acquisition, EL and renewal impacts on encoding and retrieval. We shall determine if resiliencecc against EL depends on the robustness of consolidation.
5. Our methods: Fluorescence in situ hybridisation of experience-dependent immediate early gene expression (FISH), rodent functional magnetic resonance imaging (fMRI), wide-field calcium imaging in freely behaving rodents, behavioural paradigms, in vivo electrophysiology (local field potentials, single- units), optogenetics, DREADDS and transgenic rodents.
1. The cerebellum: a frequently ignored co-player in extinction. The likely contribution of the cerebellum to extinction of learned fear has only rarely been studied. This is surprising, despite the well-known contribution of the cerebellum to associative learning. The main aim of our project is to provide further experimental evidence that the cerebellum has to be included as part of the neural circuitry underlying extinction of conditioned fear responses.
2. The cerebellum, reward prediction error and extinction. Our most important finding in the first funding period suggests that the cerebellum is involved in processing reward prediction error, which is assumed to be the central driver of extinction learning. In the second funding period, four 7T fMRI experiments are planned, including studies in patients with cerebellar disease and studies using dopaminergic drugs, to substantiate our initial observation.
3. The cerebellum – ventral tegmental area (VTA) connection in extinction learning. Dopaminergic neurons in the VTA have been shown to signal reward prediction error in the initial fear extinction learning trials in mice. We predict that the contribution of the cerebellum to reward prediction error driving extinction learning is conveyed via the cerebellum’s recently described anatomical connection with the VTA.
4. The cerebellum, aversive and appetitive conditioning. To further test the hypothesis that the cerebellum is involved in processing of reward prediction error, we will compare cerebellar activations in reversal learning paradigms of differential fear and appetitive conditioning in the 7T MR scanner.
5. New MAGNETOM Terra 7-Tesla (7T) MRI system (Siemens Healthcare GmbH, Erlangen) at the Erwin L. Hahn Institute (ELH). Our project will strongly benefit from the new system and its stronger and faster gradients.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Even before the SFB 1280 started, the subprojects presented their research plans to a number of experts in a short video.
The renewal effect describes the return of a previously unlearned reaction. This reinstatement occurs when a situation or a context is different from that in which extinction happened. Renewal occurs in very different learning situations, in fear conditioned behaviour as well as in learning through reward or punishment. In our studies without fear components only about half to two thirds of participant exhibit a renewal effect and related activation pattern in the hippocampus and the ventromedial prefrontal cortex. Additionally subjects with renewal show hippocampal activation already during the initial learning, not only during extinction. Here we want to gain a deeper understanding of which internal and external factors influence the occurrence of renewal. We manipulate context parameters such as serotonin levels or with the help of noninvasive brain stimulation, and look at how subjects perform a predictive learning task in an MRI scanner to analyse the resulting manifestation of the renewal effect.
1. Cell-type specific optogenetic control of 5-HT2A and 5-HT2C receptor signaling pathways in the amygdala. By using light-activated G protein-coupled receptors (GPCRs), we will optogenetically control 5-HT2A/2C receptor signaling pathways during fear memory formation and extinction learning in different amygdala cell-types.
2. Visualization of Gq/11 signals in GPCR-specific microdomains. We plan to visualize and control 5-HT2A and 5-HT2C receptors signaling dynamics in awake mice using our customized optogenetic tools in combination with miniature microscopy imaging technique. We will perform longitudinal analysis of individual cells during fear acquisition, extinction, renewal and reinstatement.
3. Visualization of stress-induced 5-HT2A/2C receptor signaling in awakemice. We will use mLoCal-XR to visualize and monitor endogenous Ca2+ signal dynamics of 5-HT2A and 5-HT2C receptors in the amygdala of awake mice in combination with miniature microscopy imaging. The study will explore the neural mechanisms underlying 5-HT2A/2C receptor mediated contribution to the potentiation of fear responses of stressed mice.
4. 5-HT 2C receptor knockout mice – a mouse model for fear-related behavior and extinction learning. We will explore the neural mechanisms underlying 5-HT2C-R-mediated facilitation of fear extinction with special focus on the bed nucleus of the stria terminalis (BNST). We will monitor Ca2+ dynamics in the dorsal raphe nucleus (DRN) and the BNST. They will allow us to refer changes of neuronal activity to a specific stimulus and behavioral phenotype.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Stress hormone effects on contextualized extinction memories. We will investigate the time-dependent influence of stress hormones on extinction learning, extinction retrieval, and its interaction with the context in healthy human participants.
2. Extinction retrieval in a novel context. Stress and context appear to interact in determining the return of fear. We will test this prediction by using not only the initial acquisition context but also a completely novel retrieval context (ABA and ABC renewal) in all four proposed studies.
3. Physical exercise versus psychosocial stress. In two studies, we will test if physical exercise compared to psychosocial stress can similarly boost extinction learning in a time-dependent fashion.
4. Rapid versus delayed effects of cortisol on extinction retrieval and multiple extinction contexts. Cortisol can have rapid non-genomic but also delayed genomic effects on the brain. In a third study, we will systematically characterize these distinct cortisol effects on extinction retrieval and if extinction in multiple contexts can reduce the cortisol-associated return of fear.
5. The fate of emotional versus cognitive fear memories. Using categorical fear conditioning combined with virtual reality a fourth study will test the neural correlates of cortisol on emotional and more implicit compared to episodic and more cognitive extinction retrieval.
1. Extinction learning across threat modalities. We plan to implement clinically relevant aversive interoceptive and exteroceptive stimuli to elucidate mechanisms underlying extinction of conditioned fear and
safety in the context of the gut-brain axis.
2. Effects of the stress hormone cortisol on extinction efficacy. Effects of acute stress on pain-related extinction learning remain unclear. We will test the effects of hydrocortisone versus placebo on memory retrieval and reinstatement in a conditioning model with nociceptive and non-nociceptive threats.
3. Context-dependent interoceptive conditioning. We will elucidate the contribution of the central fear network to the context-dependent generation, extinction, and resurgence of conditioned fear responses to benign interoceptive stimuli as predictors of pain.
4. Resurgence of fear-inducing “gut feelings”. Resistance to extinction and preferential return of interoceptive pain-related fear may pave the way to understand and ultimately treat or even prevent persisting interoceptive hypervigilance.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Pain-related aversive and appetitive conditioning in chronic back pain. Using a novel capsaicin-induced tonic heat pain paradigm, we will investigate whether behavioral mechanisms and temporal dynamics of acquisition, extinction and reinstatement differ between appetitive learning and aversive learning and between chronic back pain patients and healthy controls.
2. Characterizing neural mechanisms of appetitive and aversive pain-related learning. Using rsfMRI and DTI, we will test whether functional or structural predictors are related to an individual’s acquisition, extinction and reinstatement behavior.
3. Instructed vs. uninstructed extinction training. Previously acquired aversive US-CS associations seem to be more resistant to extinction than their appetitive counterparts in healthy subjects. We will test, whether prior knowledge about US-CS contingencies alters extinction and conditioned responses differentially for appetitive and aversive US-CS associations.
4. The role of state and trait variables. We aim to gain insights into individual vulnerability factors related to maladaptive acquisition and extinction behavior.
5. Contribution to joint analyses of the SFB. By acquiring physiological (SCR, pupil dilation) and brain imaging (rsfMRI, DTI) data in healthy participants and patients, we will contribute to the joint analysis on neural predictors of acquisition and extinction learning (F02).
1. Fear extinction in the light of chronic inflammation: a translational approach. We plan to study the behavioral and neural effects of chronic inflammation on fear learning and extinction in patients and in a preclinical mouse model of inflammatory bowel disease (IBD).
2. Impaired fear learning and extinction in IBD patients. We will include IBD patients with active and inactive disease to elucidate immune mechanisms and the role of disease activity in altered fear learning and extinction.
3. Multiple hits: the role of recurrent inflammation and disease progression. We will experimentally address in the dextran sulfate sodium (DSS) mouse model if impaired fear extinction in chronic IBD is the result of multiple inflammatory hits and disease progression.
4. Inflammation-induced changes in the fear extinction network. Using brain imaging in IBD patients and molecular techniques in the IBD mouse model, we will analyze inflammation-induced structural and functional changes in the fear extinction network.
5. Mechanistic insights and potential therapeutic targets. We will test in the mouse model if in vivo neutralization of TNF-α as a candidate cytokine can prevent or ameliorate inflammation-induced effects on fear extinction in chronic IBD.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Verbal instructions (VI) about CS-US contingencies and fear extinction in patients with anxiety disorders and healthy controls. We plan to study effects of VI about CS-US contingencies over the course of fear extinction training and retrieval in both healthy controls and patients with various anxiety disorders.
2. Time-dependent effects of VI. VI will be administered either prior to and/or after fear extinction training to study changes in fear extinction learning and retrieval (including fear renewal and fear reinstatement).
3. Assessment and modification of contingency biases (CBs) prior to exposure therapy in acrophobia. Patients with AD tend to overestimate the contingency between fear-relevant stimuli and aversive outcomes (CBs). We aim to assess CBs and to develop a training experimentally modifying CBs, both in the context of acrophobia.
4. Boosting the patients’ coping ability, therapy outcome, and
generalization of therapeutic effects. Acrophobic CBs will be modified prior to a virtual reality exposure for acrophobia to promote the patients’ coping ability during exposure therapy, but also to increase therapy outcome and generalization of therapeutic effects.
1. Why does it help to use multiple exposure contexts in exposure-based therapy? We will use computational modeling to study whether exposure-based therapy in multiple contexts reduces renewal because the extinction training generalizes across contexts and thereby becomes context-independent, or, alternatively, because using multiple contexts increases the chance that one of the exposure contexts matches the unknown acquisition context.
2. Neural correlates of two different types of context-dependent representations. Our previous modeling suggests that higher network layers learn distinct and task-dependent contextual representations, whereas lower layers show sensory representations, which differ between contexts because of their different sensory features and are independent of learning. We will look for evidence for the activity patterns predicted by the model in neural activity (fMRI and single-cell) recorded by collaborating projects.
3. Unifying classical and operant conditioning. Usually, classical and operantconditioning are viewed as conceptually distinct and are studied in isolation. We will explore how aspects of both learning paradigms can be integrated into a single learning paradigm and what the differential effects of their interplay are, using contextual fear conditioning as an example.
4. A better understanding of anxiety disorders. While clinical studies have emphasized that anxiety disorders are characterized by a vicious, reinforcing cycle of avoidance behavior and the stimulus’ threatening valence, this process has not been studied using computational models. We will use a reinforcement learning model that integrates, and can describe, both avoidance and threatening valence to better understand how this vicious cycle works and how to disrupt it.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Identification of interconnected brain regions in extinction of taste-immune associative learning. The insular cortex is considered to be the one mediator in extinction/persistence of taste-immune associative learning. We will perform anterograde tracing methods starting from this brain region to label connected anatomical brain structures additionally involved in taste- immune associative learning.
2. Analyzing projections and neural circuits modulating extinction and reconsolidation in taste-immune associative learning. By using
chemogenetic methods, the next experimental step will identify downstream projections from the insular cortex to relevant structures that are functionally involved in taste-immune associative learning and extinction.
3. Abrogating extinction of learned immune responses in a neuromuscular disease model. To analyze the therapeutic effectiveness of reconsolidation-like processes on learned immune responses, we will employ a rat model of myasthenia gravis, an autoimmune disease leading to muscle weakness.
4. Extinction and reconsolidation of learned immunopharmacological
effects with FTY720. We will investigate whether and to what extent taste-immune associative learning applies for fingolimod (FTY720), a drug used for the treatment of multiple sclerosis. Employment of this drug allows to continuously analyse transient neuroendocrine and immunological changes during extinction of taste-immune associative learning.
1. What is a context? Here we will translate modelling simulations of A14 into behavioral experiments in pigeons. We will explore if physical stimulus properties or learning alone decide in how far a stimulus serves as context.
2. Can we cause a stimulus to become context? Using optogenetic stimulation in VTA we will determine if any stimulus can become a context solely based on principles of reinforcement learning. We will further test the involvement of the avian hippocampus using optogenetic inhibition.
3. What are the neural dynamics of context generalization in a spatial context? Here we will perform wireless neurophysiology in crows. We will address if context-dependency of extinction is caused by attention or in generalization.
4. Do rule-learning or associative learning underlie extinction-renewal across species? Together with F01 and A09 we will explore in practical classes and in graduate thesis if the same model applies to crows and to humans to increase the visibility of SFB1280 among students.
5. Technical aspects: We will establish a novel setup for ethologically more plausible, yet controlled behavior in birds. To study cognitive processes, we will use high-density and wireless neurophysiology, wireless optogenetics and we will develop standards for data-management and analysis within and outside SFB1280 (with INF, A01, A07).
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Cerebellar Contribution to Extinction Learning. The main goal of this project is to investigate the cerebellar contribution and its intrinsic
mechanisms to fear extinction in mice.
2. Identifying the Cerebellar Neural Circuitry in Extinction. We will identify the specific regions and cell types in the cerebellum which are active during the distinct phases of extinction.
3. Optogenetic Control of Extinction from the Cerebellum. To functionally implement regions and specific cell types in the cerebellum that are key circuits in extinction, we will use optogenetics to delay or accelerate fear extinction in mice.
4. Investigating Intrinsic Cerebellar Learning Mechanisms underlying Extinction. We will gain more insights on the fundamental learning mechanisms important for extinguishing fear memories and predicting errors in the cerebellum by measuring changes in calcium levels in behaving mice.
5. Contribution of Cerebellar Disorders to Extinction Deficits. Individuals suffering from P/Q type calcium channel specific diseases such as episodic ataxia type 2 (EA2) and spinocerebellar ataxia type 6 (SCA6) have been reported to display attention deficit, personality and anxiety disorders. We will utilize our cerebellar specific EA2 and SCA6 mouse models to identify and rescue deficits in fear conditioning and extinction.
1. Phenomenological learning rates in acquisition and extinction learning. We will study the variation of phenomenological learning rates of acquisition and extinction in ABA renewal experiments across different groups of human participants and conditions.
2. Trans-species comparison of repeated ABA renewal sessions. We will study whether other species (corvids, rats, and humans) exhibit a similar diversity and whether their learning curves can also be accounted for by associative learning.
3. Neural correlates of the reward prediction error. A key variable that drives learning in reinforcement learning (RL) models is the reward prediction error. We will look for neural correlates of the reward prediction error extracted from a RL model.
4. Automated behavioral analysis. By necessity almost all experimental studies extract simple measures of the behavior of a subject at predefined time points, but behavior is far more complex than this approach acknowledges. We will develop methods that automatically extract information about the subject’s behavior from position, trajectory and video data in less biased ways.
5. Neural correlates of uninstructed behaviors. We have found that extinction learning drives the emergence of alternative behaviors. We will study the neural correlates of uninstructed behaviors in extinction learning.
Our research in a nutshell, find out more about the projects with the short introductory videos.
Our research in a nutshell, find out more about the projects with the short introductory videos.
1. Integrating neuroimaging data across subprojects. F02 streamlines data acquisition and analysis pipelines and helps establishing advanced data-analysis tools across the SFB.
2. Developing novel pipelines for resting-state connectivity. We implement new parcellation and connectivity approaches for resting-state fMRI and DTI data.
3. Setting new standards for neuroimaging meta-analyses. We will apply machine-learning tools, deep neural networks and predictive modeling to analyze the extinction learning network. Results will be validated by 7T MRI data, including laminar recordings.
4. Integrating genetic information. We will use advanced bioinformatics tools to identify novel genetic pathways associated with structure and function of the extinction network.
5. Towards a comprehensive understanding of extinction learning from genes to brain networks. Analyses of the relationship between genes, resting-state connectivity patterns and extinction behavior will provide a comprehensive understanding of the extinction network.
Research data has become one of the most important resources in science and requires a particularly high level of attention to ensure good scientific practice. Therefore, sustainable research data management (RDM) is the basis for present and future research. It is a main aim of SFB 1280 to ensure that research data is findable, accessible, interoperable and reusable according to the FAIR principles.
Due to the strong interdisciplinary structure of SFB 1280, a large amount of heterogeneous research data is produced in the individual subprojects by a variety of different measurement techniques: From magnetic resonance imaging, electroencephalography, measurements of skin conductivity, single cell recording technologies, eyetracking and pupillometry, medical data like histology and hormone data to questionnaires for psychological studies. In addition to the variety of data, especially single cell recordings and magnetic resonance imaging techniques generate a large amount of data and thus impose high demands on an efficient storage infrastructure.
Find out more on the project page >>
Christian Merz (A09), Marcella Woud (A13) and Rainer Holl talking for the Podcast “Kannste Vergessen – Der Podcast vom Lernen, Errinnern und Vergessen.”
The Ö Project focuses on communicating the scientific discoveries of the SFB 1280 to the public in an adequate and interesting way. This is achieved through three core principles:
- Local Outreach
- National and International Dissemination
- Transgenerational Dissemination
Our research is financed by taxpayer’s money. Thus, it is the obligation of us scientists to tell in in an intelligible way what we have discovered. This is especially true for our immediate environment, our city, our region, we summarize these efforts under “Local Outreach”. As important an exhibition or a public talk in the local theater is, it is just for this one evening. So, we established a second line of outreach that is permanent and can be reached from everywhere. Our activities with this intention are summarized under “National and International Dissemination”. Finally, we scientists should spark enthusiasm for our science in the generation that comes after us. activities of the SFB 1280 with this goal fall under “Transgenerational Dissemination”.
More information on the alumni projects from the first funding period can be found here
