Scientific Activity

Journal Club

We conduct our BrainWorlds Journal Club to keep ourselves updated on the state-of-the-art literature on internal world models and align our thoughts on what we will contribute to the field. On Tuesdays at 12:30 PM, our investigators and early career researchers meet to present and discuss a paper relevant for the topics of BrainWorlds.

Upcoming

Session Nr. Date Topic Presenter
13 Feb 6, 2024 Transformers Can Do Bayesian Inference. doi: 10.48550/arXiv.2112.10510 Frank Hutter
14 Feb 13, 2024 Volitional activation of remote place representations with a hippocampal brain-machine interface. doi: 10.1126/science.adh5206 Marlene Bartos
15 Feb 20, 2024 Adaptation in Auditory Processing. doi: 10.1152/physrev.00011.2022 Nicole Roßkothen-Kuhl
16 Feb 27, 2024 Johannes Letzkus
17 March 5, 2024 Andreas Vlachos
18 March 12, 2024 Claus Normann
19 April 16, 2024 Andreas Schulze-Bonhage

Archive

Session Nr. Date Topic Presenter
12 Jan 30, 2024 How language can shape learning of world models. doi: 10.48550/arXiv.2308.01399 Abhinav Valada
11 Jan 23, 2024 The bittersweet lesson: data-rich models narrow the behavioural gap to human vision. doi: 10.1167/jov.22.14.3273 Adam Kortylewski
10 Jan 16, 2024 Human-like systematic generalization through a meta-learning neural network.
doi: 10.1038/s41586-023-06668-3
Joschka Bödecker
9 Jan 9, 2024 Data Distributional Properties Drive Emergent In-Context Learning in Transformers.
doi: 10.48550/arXiv.2205.05055
Thomas Brox
8 Dec 5, 2023 Contextual inference underlies the learning of sensorimotor repertoires. doi: 10.1038/s41586-021-04129-3 Carsten Mehring
7 Nov 28, 2023 How deep is the brain? The shallow brain hypothesis. doi: 10.1038/s41583-023-00756-z Ilka Diester
6 Nov 21, 2023 An Anatomically Constrained Model for Path Integration in the Bee Brain. doi: 10.1016/j.cub.2017.08.052 Andrew Straw
5 Nov 7, 2023 Prior information differentially affects discrimination decisions and subjective confidence reports.
doi: 10.1038/s41467-023-41112-0
Natalie Mrachacz- Kersting
4 Oct 31, 2023 Representation of Real-World Event Schemas during Narrative Perception. doi: 10.1523/JNEUROSCI.0251-18.2018 Monika Schönauer
3 Oct 24, 2023 Studying the neural representations of uncertainty. doi: 10.1038/s41593-023-01444-y Florian Steebergen
2 Oct 17, 2023 A cerebellar internal model calibrates a feedback controller involved in sensorimotor control. doi: 10.1038/s41467-021-26988-0 Christian Leibold
1 Oct 10, 2023 Learning task-state representations. doi: 10.1038/s41593-019-0470-8 Lisa Graf

Lecture Series

Upcoming: April 17th, 2024 – Professor Karl Friston

Upcoming: March 13th, 2024 – Professor Cyriel Pennartz

Consciousness, predictive processing and multisensory integration

Cyriel Pennartz is Professor of Cognitive and Systems Neuroscience at the Swammerdam Institute for Life Sciences and the University of Amsterdam, the Netherlands. He studied biology and philosophy at the same institution. His current research is focused on perception, consciousness and memory. During his career he worked, amongst others, at the California Institute of Technology, University of Tennessee (Memphis) and University of Arizona (Tucson). At the University of Amsterdam, he leads a department of about 25 researchers, including 5 faculty. His group uses a multidisciplinary combination of techniques to understand the relationships between distributed neural activity and cognition, including in vivo electrophysiology and optical imaging, animal behavior, optogenetics and computational modelling.

September 12th, 2023 – Anil Ananthaswamy

Altered Selves and Self Models

Could changes in how we experience ourselves—in conditions such as Xenomelia and Autoscopy – be due to changes in how our brains model our bodies and the environment? And could these altered selves give us clues to the nature of our sense of self? In this talk, Anil Ananthaswamy explores the idea that our sense of self is a construction, a process that is active moment-by-moment. A construction can come apart, as it does in conditions such as Alzheimer’s disease, schizophrenia or even during out-of-body experiences; in each case, certain aspects of our sense of self are disrupted, even destroyed. Examining the ways in which the self is altered helps us understand how the self is put together in the first place, by the brain and body acting in concert.

Anil Ananthaswamy is an award-winning science writer and former staff writer and deputy news editor for New Scientist magazine and a 2019-20 MIT Knight Science Journalism fellow. He’s currently the journalist-in-residence at the Heidelberg Institute for Theoretical Studies. Anil writes regularly for Quanta, Scientific American, New Scientist and Nature, among others. He is the author of three books: The Edge of Physics, The Man Who Wasn’t There and Through Two Doors at Once. Anil’s next book, Why Machines Learn: The Elegant Math Behind Modern AI, will be published by Dutton, Penguin Random House, in May 2024.

Workshop

Freiburg-Oxford Workshop “Internal World Models in Animals, Humans, and AI”

November 8-9, 2023

The workshop was designed as a unique opportunity to think beyond your own discipline and to look for synergies with other research fields. Internal world models are a requirement for biological and artificial intelligence to enable flexible behaviour.

Participants discussed the underlying principles of internal world models with cutting-edge experts from neuroscience und AI. It was an event of great scientific value for the BrainWorlds Research initiative and the partnership between Freiburg and Oxford Universities.

Over 100 participants including several guests from our partner, the University of Oxford, gathered at IMBIT to discuss internal world models in animals, humans and artificial intelligence. ©Jürgen Gocke
Professor Ilka Diester, the BrainWorlds spokesperson in chief and chair of the workshop organizing committee, welcomes her colleagues from Freiburg and Oxford, as well as the younger generation of scientists, to the NEXUS Lab Venue. ©Jürgen Gocke
A good mixture of 9 presentations from Oxford and 15 presentations from Freiburg explored and discussed the concept of internal worlds models from the points of view of systems and computational neuroscience, AI and robot learning, psychology and human psychiatry. ©Jürgen Gocke
Junior Professor Monika Schönauer explains how prior knowledge about the structure of our environment supports new learning and the formation of stable neocortical representations.
On the second day, we were privileged to receive a special visit from the Rector of the University of Freiburg, Professor Kerstin Krieglstein, a neuroscientist and anatomist, who addressed the participants and expressed her support for the BrainWorlds initiative both administratively and conceptually. ©Jürgen Gocke
Ethical implications are crucial for BrainWorlds: History-of-science expert Professor Veronika Lipphardt (pictured, left) prompted us to consider public perception of our scientific endeavours. ©Jürgen Gocke
The workshop included a special session addressing the needs of early career researchers. Dr. Nicole Rosskothen-Kuhl und Dr. Julia Veit discussed how BrainWorlds can support the next generation of scientists. ©Jürgen Gocke
Lively discussions and networking continued during the lunch and coffee breaks.
In the evening, FreiDOG demonstrated its fine motor skills together with “Akaishi Daiko”, a traditional Japanese drums ensemble. The robot was choreographed to dance to the drum beats by the Robot Learning Lab of Professor Abhinav Valada, Faculty of Engineering, University of Freiburg.

Chairs

Prof. Dr.

Thomas Brox

Department of Computer Science

University of Freiburg

Prof. Dr.

Ilka Diester

Head

Optophysiology - Optogenetics and Neurophysiology

University of Freiburg

Prof.

David Dupret

MRC Brain Network Dynamics Unit

University of Oxford

Prof. Dr. med.

Andreas Vlachos

Institute for Anatomy and Cell Biology

University of Freiburg

Skip to content