We understand the mechanisms of our brain better every day, but the link between our physical brain and our mind largely remains a mystery. Leading experts from the Vrije Universiteit Amsterdam (VUA) and the University of Amsterdam (UVA) will join forces to tackle this issue in the Amsterdam Brain & Mind Project

Our research aims to understand the higher cognitive functions and its underlying neural basis. To this end, we integrate research across different levels of analysis, i.e. genetic, molecular, cellular, synaptic, circuit, network analyses, and behavioural analyses. The Amsterdam Brain and Mind Project focuses on five core and interrelated Research Themes.

Theme 1: Emotion, stress and affective regulation

Theme 2: Attention, awareness, and action

Theme 3: Decision making and executive control

Theme 4: Higher cognitive functions, development & modelling

Theme 5: Brain mechanisms in health & disease

Central within ABMP are the projects of ten young investigators, each supervised by two principal investigators from VUA and UvA respectively and across the five research themes – © 2018

Theme 1

Project 1 – Dissection of prediction error and reconsolidation circuits of aversive-emotional memory

Team: Priyanka Rao-Ruiz, Sabine Spijker (VUA) and Merel Kindt (UvA)

Rationale: Personal experiences shape our memories that are reflected by changes in connections in our brain. When these experiences are salient, it can result in maladaptive memories that impinge on our daily life. By elucidating the neurobiology of fear, we gain insight into the psychological construct of emotional memory. Rodent behavioral paradigms (Spijker) that closely simulate those used in human cognitive behavioral therapy (Kindt) will be used to understand the molecular and cellular mechanisms of human fear behavior, which should lead to improved or adapted forms of treatment.

Summary: Aberrant memory processing lies at the root of a plethora of fear, anxiety and addiction disorders. Understanding how memories are processed at the molecular and cellular level enables us to obtain important biological clues of what goes awry in the brain in these disorders. In pre-clinical human studies, the reconsolidation of the emotional component of a memory can be disrupted by administration of the β-adrenergic receptor blocker, propranolol, once the memory is retrieved and destabilized. However, this form of pharmaceutical intervention does not always work and the underlying mechanisms remain unknown. Here, we will initiate true translational research, in which human studies (Kindt) instruct the development of behavioral protocols for memory-modifying resiliency in mice (Spijker). Using a multilevel approach, we aim to 1) causally identify mediating mechanisms of how propranolol disrupts the restabilization of aversive emotional memories after their retrieval and 2) identify brain region(s) generating aversive Prediction Error (PE) signals that are necessary for memory destabilization in neurons encoding aversive emotional memories.

© 2018

Theme 1

Project 2 – Advanced in vitro and in vivo modeling to study the role of astrocytes in MDD

Team: Claudia Sestito, Vivi Heine (VUA) and Aniko Korosi (UvA)

Rationale: The use of iPSC for this project represents an unique approach to identify defects in brain cells (astrocytes) in patients affected by a complex mental illness such as MDD. The study represents the first step toward the screening for MDD disease mechanisms and the identification of novel target genes, which will ultimately leads to the development of more effective treatments. In addition, the generation of iPSC lines from MDD opens new opportunity for investigating the role of other neural cells (e.g. neurons or oligodendrocytes) in MDD pathology. Finally we will also gain further information on the possible role of astrocytes in conveying the early-life stress induced vulnerability to develop MDD

Summary: Major Depressive Disorder (MDD) is among the most common psychiatric disorder characterized (among others) by hyperactive neuroendocrine hypothalamus pituitary adrenal (HPA) stress axis and therefore, high cortisol levels. Current treatments are not effective and the risk factors and mechanism underlying the pathology still remain elusive although a role for astrocytes is emerging. By combining advanced induced pluripotent stem cell (iPSC) technology and the resources from the Netherlands Study for Anxiety and Depression (NESDA), the project aims to investigate defects in astrocytes under basic and stress conditions (treatment with dexamethasone, a cortisol analog). In addition, because early-life stress has been proposed as risk factor for development of MDD, we study how early-life stress exposure affects astrocytes through life.

© 2018

Theme 2

Project 3 – Untangling the elusive influence of prediction on attentional capture and conscious awareness

Team: Matthew Weaver, Simon van Gaal (UvA) and Artem Belopolsky (VUA)

Rationale: This collaborative project between the UvA and VU Amsterdam has provided a unique opportunity to bring together complementary expertise to explore the influence of predictions and task relevance on attentional control and conscious awareness. By uncovering how this influence is implemented at the neural level, we are bridging the gap between the brain and the mind. Such fundamental questions speak to what it means to be human as we perceive and attend to the world around us

Summary: It has been proposed that generation of predictions may be a universal principle of neural computation, which can explain the majority of the operations of the brain and strongly influence our perception and cognition. However, the exact mechanisms of prediction are still unknown and strongly debated. We conducted three studies using state-of-the-art neuroimaging methods, combined with sophisticated eye-tracking measures, to uncover how expectation and task relevance impact attentional selection and subjective conscious awareness.

© 2018

Theme 2

Project 4 – Perception is more than meets the eye: flexibility and control in sensory processing

Team: Martijn Barendregt, Tomas Knapen (VUA) and Heleen Slagter (UvA)

Rationale:  7T fMRI is applied to study 1. Cortical visual information processing, and 2. Activations in subcortical brain structures that mediate pupil-linked arousal. We will be able to more closely link the knowledge bases at VU and UvA, through a collaboration with the Spinoza Centre, in which both our universities participate. This will further allow us to combine our skills and expertise (VU: pupil dilation, 7T fMRI, population receptive field mapping, visual cortex function) and the UvA (theories of attention and predictive coding, convolutional neural networks).

Summary: How do internal states, such as predictions and attention influence the processing of stimuli by our brains? Surprisingly little is known about how our predictions, independently from and in combination with attention, shape our perceptual experience. Especially the instances in which our predictions are violated, so called prediction errors (PE), are informative regarding the state of the world. We have conducted an experiment in which we investigated the influence of attended and unattended PEs on behaviour and changes in pupil dilation. Pupil dilations are thought to index of arousal and correlate with changes in cortical state. Here, we investigated whether these signals can be used as a proxy for the impact of PEs on perceptual decision-making. Using detailed mathematical modelling of behaviour, we showed that only attended PEs change the speed of evidence accumulation in the decision-mechanism itself, whereas unattended PEs do not. We also found a strong link between PE-induced pupil dilations and decision-making, particularly when observers make a mistake (Figure). This linkage between pupil and behavior was so strong, that we are able to “decode” from the pupil’s dilations whether an observer was going to commit an error, far before their manual response. Intriguingly, the ability to predict mistakes was exclusive to situations in which the behavioural error was prompted by an attended PE (and not an unattended PE). These findings indicate that the interaction between attention and prior expectations is implemented at a very low level in sensory cortex. Furthermore, our findings imply that these interactions are strongly linked to arousal, changes in neuromodulatory circuits and global brain state.

© 2018

Theme 3

Project 5 – Processing multisensory evidence for decision-making: control by frontal versus sensory cortex

Team: Jean Pie, Cyriel Pennartz (UvA)and Christiaan de Kock (VUA)

Rationale: Our project addresses the neural substrates (“brain”) of prototypical “mind” processes – integrated perception and decision-making. What makes our ABMP project truly “bridging” is that we study brain-mind processes at multiple levels of organization: from single neurons, via ensembles in local brain areas and via multi-area interactions, to cognition and behavior.  The ultimate target is to conceptually understand neural activity (“read”) and manipulate activity patterns (“write”) in a key process in brain-mind research: perception.

Summary: Perception, decision-making, and all our conscious experiences are inherently multimodal. One of the deepest questions in mind-brain research is how our brain manages to forge a coherent representation from multisensory inputs to reach quick and efficient decisions. Using a binary decision task in which mice report whether and where left vs. right) near-threshold visual and/or haptic stimuli appear, we propose to test two contrasting, non-exclusive hypotheses on the contribution of lower sensory versus higher cortical areas to multimodal decision-making:

  • The decision to report the location of stimuli may depend on the feedforward combination of visual and tactile inputs in association areas PPC (posterior parietal cortex) and/or RL (rostro-lateral).
  • This decision involves a higher-order cortical area (cingulate cortex Cg1), which first receives information from both modalities and then orchestrates behavioural output and sensory processing via top-down control.

We are combining our complementary expertise in ensemble and juxtacellular single-neuron recordings and optogenetics to tackle this essential aspect of mind-brain research.

© 2018

Theme 3

Project 6 – Reframing the addiction-occupied mind by disrupting reconsolidation of alcohol memories through working memory interference

Team: Anne-Marije Kaag, Reinout Wiers (UvA) and Taco de Vries (VUA)

Rationale: Over the past two years, the VUmc, AMC and UvA have been working together intensively in this project, and will continue working together in new projects, to optimize and develop novel interventions in substance use disorder. The translational character of this collaboration allowed us to take the first step in bridging the gap between fundamental and clinical researchers, which we believe is of crucial importance in modern neuroscience and thus within Amsterdam Neuroscience.

Summary: In this proposal, experts in the field of experimental psychology, clinical neuroscience and preclinical behavioral neuroscience join forces to test a novel hypothesis: high working memory load following the retrieval of alcohol-associated memories will disrupt the reconsolidation of those memories and reduce alcohol cue-reactivity. This hypothesis is based on a dual process account and will be tested in rats and humans. Besides a cross-species analysis at the behavioral and cognitive level, we also aim to investigate the neurobiological substrate of this memory retrieval – working memory interference. Innovative methods are being employed to address the research hypotheses: In humans, a randomized controlled study was performed to test the effectiveness of inducing a high working memory load following the retrieval  of alcohol related memories on alcohol craving and alcohol intake. In addition, fMRI measures were taken to assess intervention-related changes in neural cue-reactivity. Parallel to the human study, two animal studies were/are performed. In the first animal study we investigated whether we could chemically target working memory related neural pathways using DREAD technology. In the second animal study, that is currently ongoing we assessed the effectiveness of inducing a high working memory load during the retrieval  of alcohol related memories on cue-induced reinstatement of alcohol-seeking. It will be additionally assessed how chemically inhibiting working-memory related neural pathways during reinstatement influences cue-induced reinstatement.

© 2018

Theme 4

Project 7 – The interplay between genetic profiles and environment in higher cognitive function

Team: Cesar Vroom, Danielle Posthuma (VUA) and Lourens Waldorp (UvA)

Rationale: Cognitive  disorders  are  a  major  economic,  societal  and  personal  burden.  Current  treatments  are  seriously  hampered by a lack of insight into underlying cognitive functions and neurobiological mechanisms. Major scientific efforts have focused on identifying causal genetic variants for cognitive (dys)function. After a century of limited etiological progress, the past decade has seen unprecedented advances in our understanding of the fundamental genetic architectures of cognitive traits, and numerous robust and replicable genetic variants are now available.

Summary:  To benefit from these exciting findings and use them to gain insight into the higher cognitive functions involved in  cognitive  disorders,  we  need  to  translate  these  findings  to  phenotypic  differences  in  higher  cognitive  function and the brain. We here propose to capitalize on the genetic results by relating the most recent genetic findings  for  higher  cognitive  traits  to  IQ  and  MRI‐based  measures  of  the  brain.  We  will  do  so  by  calculating  polygenic risk scores in the ID-1000 cohort and using them together with measured environmental effects to predict IQ and a variety of brain measures. The main goal is to elucidate the link between genetic risk profiles for cognitive traits, environmental moderators, higher cognitive function (IQ) and brain structure and function. Identification  of  brain  measures  relevant  to  higher  cognitive  function  will  provide  further  insight  into  the  underlying route from genetic risk to cognitive (dys)function.

© 2018

Theme 4

Project 8 – Characterization of functional brain network organization in dyslexia and development

Team: Gorka Fraga Gonzalez, Maurits van der Molen (UvA) and Eco de Geus (VUA)

Rationale: The current study details dysfunctional EEG network organization and parasympathetic sensitivity in dyslexia and, thus, relates a high cognitive ability such reading to brain network organization and autonomic responding.

Summary: The project aims at delineating the topology of functional brain networks in dyslexic children and adults observed during resting-state and task-related conditions using network analysis on EEG data. Behavioral and electrophysiological data have been obtained from both adults and children, dyslexic as well as typical readers. The data set includes two baseline (resting-state) recordings and a recording during performance of a letter-speech sound-mapping task, tapping a fundamental deficit in developmental dyslexia. The analysis of adult EEG revealed differences in resting-state networks between typical readers and dyslexics. Subsequent analyses focused on how network metrics change between resting-state and task performance. In addition, the analysis of electrophysiological manifestations associated with task performance indicated age-group differences in the efficiency of feedback processing. Ongoing analyses will focus primarily on age-age group differences in EEG network metrics during resting state and task performance.

© 2018

Theme 5

Project 9 – Disambiguating similar experiences: influence of the perirhinal cortex on hippocampal pattern separation

Team: Charlotte Oomen, Guus Smit (VUA) and Carien Lansink (UvA)

Rationale: A critical feature of episodic memory is the ability to discriminate between similar, but distinct, experiences. Even day-to-day problems as “Where did I park my car this morning” can only be solved by recognizing that the car was parked in a different lot today compared to yesterday and last week. Storing similar memory representations in a distinct, non-overlapping fashion is thought to depend on a pattern separation (PS) process that is attributed to the hippocampal dentate gyrus (DG). PS is not unique to the DG but also occurs in the perirhinal cortex (PRC), which is located in the parahippocampal region and projects to the hippocampal DG and CA subregions via the entorhinal cortex. The influence of the PRC on the decorrelation of neuronal patterns in DG, however, is yet unknown.

Summary: In this project, we aim to elucidate the cellular and molecular mechanisms in the hippocampus and PRC causally related to PS. To this end, we use chemogenetic technology to transiently silence principal cells in the PRC and DG. In conjunction, we perform simultaneous large-scale electrophysiological recordings of neuronal ensembles in the PRC, the DG, and its projection areas CA3 and CA1. We plan to further combine this behavioral/chemogenetic setup with a novel cell type-specific proteomics approach. We hypothesize that suppression of DG activity will lead to an impaired ability to discriminate between two similar stimuli and that the degree of overlap between firing patterns of neuronal populations in the DG and CA3/CA1 representing these two stimuli will be higher as a result of DG inactivation. Furthermore, we postulate that the suppression of principle neurons in DG will distort temporal firing relations between PRC and DG and that hippocampal (DG/CA3/CA1) ensembles representing the two stimuli will have a differential molecular profile, enabling stimuli discrimination over time. Together, these experiments will shed light on how our memory system retrieves and disambiguates representations of similar experiences; fundamental knowledge that is critical for understanding conditions characterized by impaired memory processing.

© 2018

Theme 5

Project 10 – Oxytocin as novel treatment of Prader Willi Syndrome

Team: Rocio Díez Arazola, Ruud Toonen (VUA) and Marco Hoekman (UvA)

Rationale: Prader Willi Syndrome (PWS) is a genetic disorder affecting >10.000 people in Europe. PWS is characterized by feeding problems associated with cognitive and social impairments similar to autism spectrum disorder (ASD). PWS patients depend on lifetime care but effective therapies for PWS are lacking. Recent evidence points towards the neuropeptide oxytocin as promising treatment of PWS but progress is hampered by lack of mechanistic insight. By combining unique assets at the VU and UvA including disease models (mice and human induced neurons) and expertise in analysis of oxytocin-vesicle dynamics we aim to elucidate the mechanisms underlying rescue of PWS alterations by oxytocin. This knowledge will aid the development of next-generation approaches to target the oxytocin system to improve social function in PWS and ASD.

Summary: Oxytocin is difficult to measure reliably in the brain. In addition to the well-known release of oxytocin in the bloodstream numerous axonal connections are made between oxytocin producing neurons in PVN and SON and several brain regions. However, it is currently unknown where, when and how oxytocin is released. To address this, we have generated fluorescently-labeled oxytocin for use in live cell imaging experiments. Mouse primary neurons infected with lentiviral particles expressing oxytocin-EGFP show a distinctive punctate pattern indicative of secretory organelles. These puncta fully overlap with canonical cargo of neuropeptide-containing vesicles (NPY-mCherry) and dynamically traffic through axons and dendrites. Stimulation of central neurons with trains of action potentials induces the release of these vesicles. Currently, we are testing oxytocin-vesicle transport and release in human iPSC-derived neurons and in mouse brain slices in order to come to a full description of the transport and release mechanisms of these vesicles. In addition, we screen for molecules involved in oxytocin secretion and have identified two essential members of the oxytocin-release machinery in cultured neurons. Inducible null mutant mice of these proteins will now be used to test in vivo whether oxytocin release is impaired via injection of cre-recombinase and measurements of oxytocin in cerebrospinal fluid and the blood. These mice offer a unique opportunity to assess the importance of oxytocin release on postsynaptic neurons and will be used in addition to oxytocin application to assess (epi-)genetic effects.

© 2018