SymSeqBench: a unified framework for the generation and analysis of rule-based symbolic sequences and datasets
arXiv · doi
About
I study how the brain computes the mind. My research combines mathematical modeling, biophysics, and dynamical systems theory to build mechanistic models of cortical circuits—from molecular and synaptic processes through dendritic integration and network dynamics to cognitive function. I argue for an integrative approach where biological constraints at every level guide computational models, establishing causal links between biophysical mechanisms and neural computation. I lead the Computational Neuroscience & Neural Computation group at the Center for Neuroscience and Cell Biology, University of Coimbra.
I design and implement large-scale spiking neural network simulations, develop custom analysis and benchmarking tools, and build research software for the computational neuroscience community. I work daily with Python, PyTorch, JAX, and the NEST simulator, and have led computing projects on high-performance infrastructure at the Jülich Supercomputing Centre and, more recently, at the Portuguese National Advanced Computing Infrastructure. My technical work spans biophysical modeling, machine learning and AI, dynamical systems analysis, and scientific software engineering.
Before establishing my group in Coimbra, I held a tenured Assistant Professorship in Cognitive Artificial Intelligence at Radboud University, worked as a postdoctoral fellow at the Forschungszentrum Jülich, and completed a joint PhD across the University of Freiburg and the University of Edinburgh through the Erasmus Mundus EuroSPIN programme (summa cum laude). I write about neuroscience and AI-augmented research at Grounded Neuro.
Research
My research programme bridges cognition and biophysics through computational modeling, unified by dynamical systems theory and a commitment to biological plausibility.
Sequential Computation & Cognition
How do cortical circuits acquire, represent, and generalize sequential rules? I investigate the neural substrates of temporal structure processing, working memory, and compositional generalization—from sensorimotor coordination to language. This line uses spiking neural networks, reservoir computing, and formal language tasks to study domain-general cortical computation.
Biophysical Mechanisms & Microcircuits
Learning emerges from coupled electrochemical processes spanning multiple spatial and temporal scales. I build models of heterosynaptic plasticity, dendritic nonlinearities, and canonical cortical microcircuits—grounded in molecular biology and electrophysiology—to explain how local synaptic and cellular mechanisms give rise to circuit-level computation and adaptive behavior.
Dynamical Systems & NeuroAI
Transient dynamics, metastability, and attractor landscapes are the computational primitives of recurrent neural circuits. I use dynamical systems analysis, reservoir computing theory, and biologically-constrained machine learning to characterize neural computation and to build brain-inspired AI architectures—bridging neuroscience and modern deep learning.
Career
2024 – present
Principal Investigator & Assistant Researcher
Center for Neuroscience and Cell Biology (CNC-UC), University of Coimbra
Leading the Computational Neuroscience & Neural Computation group. PI on FCT-funded research on heterosynaptic plasticity and compartmentalized learning.
2021 – 2023
Assistant Professor of Cognitive AI tenured
Department of Artificial Intelligence, Radboud University, Nijmegen
Designed and taught Neural Computation (MSc) and Reinforcement Learning (BSc, 250+ students). Tenure awarded 1.5 years ahead of schedule. Educational Innovation Award.
2018 – 2021
Postdoctoral Researcher
Institute of Neuroscience and Medicine (INM-6), Forschungszentrum Jülich
Led research projects on functional neural architectures and synaptic dynamics. JARA-HPC computing allocation (50M+ core-hours). Guest Researcher at the Max Planck Institute for Psycholinguistics.
2011 – 2018
Doctoral Researcher summa cum laude
Erasmus Mundus EuroSPIN — Albert-Ludwigs-Universität Freiburg & University of Edinburgh
Joint PhD programme across Freiburg, Edinburgh, Bochum, and Jülich. Thesis: “State-dependent processing in Spiking Neural Networks.”
2009 – 2011
MSc Cognitive Neuroscience & BSc Pharmaceutical Sciences
University of Algarve (MSc, Distinction) · University of Coimbra (BSc)
Skills
Simulation & Biophysical Modeling
Machine Learning & NeuroAI
Scientific Computing & HPC
Dynamical Systems & Theory
Data Analysis & Signal Processing
Software Engineering
Open Source Software
Research tools and contributions to community infrastructure.
FNA
activeFunctional Neural Architectures — a Python framework for benchmarking neural circuits across abstraction levels, from rate networks to biophysically detailed spiking models.
NMSAT
archivedNeural Microcircuit Simulation and Analysis Toolkit — a high-level wrapper for reproducible simulation and analysis of spiking neural microcircuits at scale.
MEA-Flow
activeA Python library for multi-electrode array data analysis with a focus on neural population dynamics, manifold learning, and latent state-space analysis.
NEST Simulator
contributorCore contributions to the NEST simulator ecosystem, including neuron and synapse model implementations, and co-authorship of NESTML code generation tools.
More projects at github.com/rcfduarte and github.com/CNNC-Lab.
Teaching
Course Design & Coordination
Neural Computation
9.5/10 student evaluations
Reinforcement Learning
250+ students · Educational Innovation Award
Systems & Computational Neuroscience
5-day intensive with international guest lecturers
Introduction to Scientific Programming with Python
Guest Lectures & Schools
Synapse Assembly, Function and Plasticity
Neuronal Circuits & Behavior
Spring/Fall School in Computational Neuroscience
Summer School in Computational Biology
Selected Publications
For the full list and live citation metrics, see Google Scholar.
Nonlinear dendritic integration supports Up-Down states in single neurons
The Journal of Neuroscience · doi
A layered microcircuit model of Somatosensory cortex with three interneuron types and cell-type-specific short-term plasticity
Cerebral Cortex · doi
Phenomenological modeling of diverse and heterogeneous synaptic dynamics at natural density
New Aspects in Analyzing the Synaptic Organization of the Brain · doi
A refined information processing capacity metric allows an in-depth analysis of memory and nonlinearity trade-offs in neurocomputational systems
Scientific Reports · doi
Signal denoising through topographic modularity of neural circuits
eLife · doi
The Tripod neuron: a minimal structural reduction of the dendritic tree
The Journal of Physiology · doi
Unsupervised learning and clustered connectivity enhance reinforcement learning in Spiking Neural Networks
Frontiers in Computational Neuroscience · doi
Neuronal spike-rate adaptation supports working memory in language processing
Proceedings of the National Academy of Sciences (PNAS) · doi
Leveraging heterogeneity for neural computation with fading memory in layer 2/3 cortical microcircuits
PLoS Computational Biology · doi
Synaptic patterning and the timescales of cortical dynamics
Current Opinion in Neurobiology · doi
Dynamic stability of sequential stimulus representations in adapting neuronal networks
Frontiers in Computational Neuroscience · doi
All Publications (25)
- Zajzon, B., Bouhadjar, Y., Fabre, M., Schmidt, F., Ostendorf, N., Neftci, E., Morrison, A., & Duarte, R. (2025). SymSeqBench: A unified framework for the generation and analysis of rule-based symbolic sequences and datasets (No. arXiv:2512.24977; Version 1). arXiv. https://doi.org/10.48550/arXiv.2512.24977 doi
- Quaresima, A., Fitz, H., Hagoort, P., & Duarte, R. (2025). Nonlinear dendritic integration supports up-down states in single neurons. Journal of Neuroscience. https://doi.org/10.1523/JNEUROSCI.1701-24.2025 doi
- Jiang, H.-J., Qi, G., Duarte, R., Feldmeyer, D., & van Albada, S. J. (2024). A layered microcircuit model of somatosensory cortex with three interneuron types and cell-type-specific short-term plasticity. Cerebral Cortex, 34(9), bhae378. https://doi.org/10.1093/cercor/bhae378 doi
- Korcsak-Gorzo, A., Linssen, C., Albers, J., Dasbach, S., Duarte, R., Kunkel, S., Morrison, A., Senk, J., Stapmanns, J., Tetzlaff, T., Diesmann, M., & van Albada, S. J. (2024). Phenomenological modeling of diverse and heterogeneous synaptic dynamics at natural density. In J. H. R. Lübke & A. Rollenhagen (Eds.), New Aspects in Analyzing the Synaptic Organization of the Brain (pp. 277–321). Springer US. https://doi.org/10.1007/978-1-0716-4019-7_15 doi
- Schulte to Brinke, T., Dick, M., Duarte, R., & Morrison, A. (2023). A refined information processing capacity metric allows an in-depth analysis of memory and nonlinearity trade-offs in neurocomputational systems. Scientific Reports, 13, 10517. https://doi.org/10.1038/s41598-023-37604-0 doi
- Zajzon, B., Duarte, R., & Morrison, A. (2023). Toward reproducible models of sequence learning: Replication and analysis of a modular spiking network with reward-based learning. Frontiers in Integrative Neuroscience, 17. https://doi.org/10.3389/fnint.2023.935177 doi
- Zajzon, B., Dahmen, D., Morrison, A., & Duarte, R. (2023). Signal denoising through topographic modularity of neural circuits. eLife, 12, e77009. https://doi.org/10.7554/eLife.77009 doi
- Schulte to Brinke, T., Duarte, R., & Morrison, A. (2022). Characteristic columnar connectivity caters to cortical computation: Replication, simulation, and evaluation of a microcircuit model. Frontiers in Integrative Neuroscience, 16. https://doi.org/10.3389/fnint.2022.923468 doi
- Quaresima, A., Fitz, H., Duarte, R., Broek, D. van den, Hagoort, P., & Petersson, K. M. (2023). The Tripod neuron: A minimal structural reduction of the dendritic tree. The Journal of Physiology, 601(15), 3265–3295. https://doi.org/10.1113/JP283399 doi
- Weidel, P., Duarte, R., & Morrison, A. (2021). Unsupervised Learning and Clustered Connectivity Enhance Reinforcement Learning in Spiking Neural Networks. Frontiers in Computational Neuroscience, 15. https://doi.org/10.3389/fncom.2021.543872 doi
- Fitz, H., Uhlmann, M., Van Den Broek, D., Duarte, R., Hagoort, P., & Petersson, K. M. (2020). Neuronal spike-rate adaptation supports working memory in language processing. Proceedings of the National Academy of Sciences of the United States of America, 117(34), 20881–20889. https://doi.org/10.1073/pnas.2000222117 doi
- Zajzon, B., Duarte, R., Mahmoudian, S., Morrison, A., & Duarte, R. (2019). Passing the Message: Representation Transfer in Modular Balanced Networks. Frontiers in Computational Neuroscience, 13(December), 79. https://doi.org/10.3389/fncom.2019.00079 doi
- Bachmann, C., Tetzlaff, T., Duarte, R., & Morrison, A. (2020). Firing rate homeostasis counteracts changes in stability of recurrent neural networks caused by synapse loss in Alzheimer’s disease. PLOS Computational Biology, 16(8), e1007790. https://doi.org/10.1371/journal.pcbi.1007790 doi
- Duarte, R., & Morrison, A. (2019). Leveraging heterogeneity for neural computation with fading memory in layer 2/3 cortical microcircuits. PLOS Computational Biology, 15(4), e1006781. https://doi.org/10.1371/journal.pcbi.1006781 doi
- Duarte, R., Uhlmann, M., van den Broeck, D., Fitz, H., Petersson, K. M., & Morrison, A. (2018). Encoding symbolic sequences with spiking neural reservoirs. Proceedings of the International Joint Conference on Neural Networks, 2018-July, 1–8. https://doi.org/10.1109/IJCNN.2018.8489114 doi
- Zajzon, B., Duarte, R., & Morrison, A. (2018). Transferring State Representations in Hierarchical Spiking Neural Networks. Proceedings of the International Joint Conference on Neural Networks, 2018-July, 1–9. https://doi.org/10.1109/IJCNN.2018.8489135 doi
- Duarte, R. (2018). State-dependent processing with Spiking Neural Networks [Doctoral dissertation]. Albert-Ludvigs Universitat Freiburg. http://hdl.handle.net/2128/19337
- The best spike filter kernel is a neuron https://ccneuro.org/2017/abstracts/abstract_3000204.pdf
- Duarte, R., Seeholzer, A., Zilles, K., & Morrison, A. (2017). Synaptic patterning and the timescales of cortical dynamics. Current Opinion in Neurobiology, 43, 156–165. https://doi.org/10.1016/j.conb.2017.02.007 doi
- Weidel, P., Djurfeldt, M., Duarte, R., & Morrison, A. (2016). Closed loop interactions between spiking neural network and robotic simulators based on MUSIC and ROS. Frontiers in Neuroinformatics, 10(31), 1–19. https://doi.org/10.3389/fninf.2016.00031 doi
- Duarte, R. (2015). Expansion and State-Dependent Variability along Sensory Processing Streams. The Journal of Neuroscience, 35(19), 7315–7316. https://doi.org/10.1523/JNEUROSCI.0874-15.2015 doi
- Toledo-Suárez, C., Duarte, R., & Morrison, A. (2014). Liquid computing on and off the edge of chaos with a striatal microcircuit. Frontiers in Computational Neuroscience, 8(November), 130. https://doi.org/10.3389/fncom.2014.00130 doi
- Duarte, R. C. F., & Morrison, A. (2014). Dynamic stability of sequential stimulus representations in adapting neuronal networks. Frontiers in Computational Neuroscience, 8(October), 124. https://doi.org/10.3389/fncom.2014.00124 doi
- Duarte, R., Seriès, P., & Morrison, A. (2014). Self-Organized Artificial Grammar Learning in Spiking Neural Networks. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), Proceedings of the Annual Meeting of the Cognitive Science Society (Vol. 36, Issue 36, pp. 427–432). Cognitive Science Society. https://escholarship.org/uc/item/75j1b2t0 link
- Duarte, R. (2011). Self-organized sequence processing in recurrent neural networks with multiple interacting plasticity mechanisms [University of Algarve]. http://sapientia.ualg.pt/handle/10400.1/3786 link