September 2021 - Comp-neuro - lists.cnsorg.org

Biologically plausible models of path planning in humans and rodents
by Jeffrey Krichmar Sept. 10, 2021

Sept. 10, 2021

Dear Colleagues, We recently published two papers on biologically plausible models of path planning that simulate and analyze navigation strategies in rodents and humans. Some of the ideas presented in these papers may be applicable to autonomous systems. If you are interested, please see: Krichmar, J.L., and He, C. (2021). Importance of Path Planning Variability: A Simulation Study. Topics in Cognitive Science. https://onlinelibrary.wiley.com/doi/10.1111/tops.12568 Individuals vary in the way they navigate through space. Some take novel shortcuts, while others rely on known routes to find their way around. We wondered how and why there is so much variation in the population. To address this, we first compared the trajectories of 368 human subjects navigating a virtual maze with simulated trajectories. The simulated trajectories were generated by strategy-based path planning algorithms from robotics. Based on the similarities between human trajectories and different strategy-based simulated trajectories, we found that there is a variation in the type of strategy individuals apply to navigate space, as well as variation within individuals on a trial-by-trial basis. Moreover, we observed variation within a trial when subjects occasionally switched the navigation strategies halfway through a trajectory. In these cases, subjects started with a route strategy, in which they followed a familiar path, and then switched to a survey strategy, in which they took shortcuts by considering the layout of the environment. Then we simulated a second set of trajectories using five different but comparable artificial maps. These trajectories produced the similar pattern of strategy variation within and between trials. Furthermore, we varied the relative cost, that is, the assumed mental effort or required timesteps to choose a learned route over alternative paths. When the learned route was relatively costly, the simulated agents tended to take shortcuts. Conversely, when the learned route was less costly, the simulated agents showed preference toward a route strategy. We suggest that cost or assumed mental effort may be the reason why in previous studies, subjects used survey knowledge when instructed to take the shortest path. We suggest that this variation we observe in humans may be beneficial for robotic swarms or collections of autonomous agents during information gathering. Krichmar, J.L., Ketz, N.A., Pilly, P.K., and Soltoggio, A. (2021). Flexible Path Planning through Vicarious Trial and Error. bioRxiv, 2021.2009.2008.459317. https://www.biorxiv.org/node/2148912.abstract Flexible planning is necessary for reaching goals and adapting when conditions change. We introduce a biologically plausible path planning model that learns its environment, rapidly adapts to change, and plans efficient routes to goals. Unlike prior models of hippocamapl replay, our model addresses the decision-making process when faced with uncertainty. We tested the model in simulations of human and rodent navigation in mazes. Like the human and rat, the model was able to generate novel shortcuts, and take detours when familiar routes were blocked. Similar to rodent hippocampus recordings, the neural activity of the model resembles neural correlates of Vicarious Trial and Error (VTE) during early learning or during uncertain conditions. Similar to rodent studies, after learning, the neural activity resembles forward replay or preplay predicting a future route, and VTE activity decreases. We suggest that VTE, in addition to weighing possible outcomes, is a way in which an organism may gather information for future use. Best regards, Jeff Krichmar Department of Cognitive Sciences 2328 Social & Behavioral Sciences Gateway University of California, Irvine Irvine, CA 92697-5100 jkrichma(a)uci.edu http://www.socsci.uci.edu/~jkrichma

1 0

Postdoctoral position available
by Sara A. Solla Sept. 10, 2021

Sept. 10, 2021

*Integrating Human and Machine Learning for Enabling Co-Adaptive * *Body-Machine Interfaces* * Sandro Mussa-Ivaldi and Sara A. Solla PI’s * *Northwestern University* We are seeking postdoctoral applicants with interests and experience in computational methods, machine learning, and human-machine interactions to participate in an NSF-funded project. The goal of the project is customizing human-machine interfaces through a collaborative interaction between human and machine learning. This has relevance to the recovery of function after injury to the spinal cord. The same goals apply also to non-clinical domains. Common to the different domains is the necessity to learn novel movement skills. The project will consider how this challenge is met through the interaction with an adaptive interface, the body-machine interface, that maps body motions onto commands to external devices. The working hypotheses are H1) that through a dynamical process of learning the human operators form an internal model of the interface and connected device, and H2) that the human machine interface can be adapted to its users by an update rule that tracks the learning process of the human operator. The success of the research project will lead to a new family of intelligent human machine interfaces capable of customizing themselves to the evolving ability of their users. Applications, including a brief statement, a bio sketch, and the names of two references, should be sent to bomiproject2021(a)gmail.com

1 0

Postdoc position on neural mechanisms of flexible behavior
by Luca Mazzucato Sept. 5, 2021

Sept. 5, 2021

Dear colleagues, We are looking for a computational neuroscience postdoc to lead a project on the neural mechanisms of flexible behavior. The theory position will be co-supervised by Luca Mazzucato and David McCormick and funded by the NIH BRAIN Initiative collaboration between the McCormick, Niell, Jaramillo and Mazzucato labs at University of Oregon. The project aims at elucidating the whole-brain dynamics enabling flexible modality switching in an audiovisual decision-making task. A successful candidate will use state-space models and multi-area recurrent network models to elucidate the activity and functional connectivity of cortical circuits from large datasets in behaving mice (2-photon mesoscope with ~10k neurons; multiple neuropixel probes). Experience with deep learning is recommended for integrating behavioral analyses with task performance and neural activity. The postdoc will be part of the vibrant neuroscience and machine learning community at the University of Oregon including the Institute of Neuroscience <https://ion.uoregon.edu/> and the Knight Campus <https://accelerate.uoregon.edu/> for Accelerating Science. The University is located in beautiful Eugene, the country's leader in the "20-minute friendly <https://tinyurl.com/y8jd8pmr>" bike commute movement, at close range between the Cascades ski areas and the Pacific Ocean. Successful candidates will have a PhD in Physics, Neuroscience, Mathematics, Computer Science, Statistics or a related discipline and, preferably, experience in theoretical neuroscience and statistical analysis of large neuroscience datasets. For inquiries please contact Luca Mazzucato or David McCormick.

1 0

8th Baltic-Nordic Summer School on Neuroscience and Neuroinformatics (BNNI 2021)
by Pawel Herman Sept. 5, 2021

Sept. 5, 2021

Dear Fellow Neuroscientists and Students We would like to invite you to the 8th Baltic-Nordic Summer School on Neuroscience and Neuroinformatics, which will be organised as a virtual event on 21-25 September, 2021. The 8th Baltic-Nordic Summer School on Neuroscience and Neuroinformatics 2021 “Learning in the Brain and NeuroRobots – from Molecules to Behaviour” offers interdisciplinary course and covers modelling at different levels of organization of the brain, from single neurons to microcircuits, neural networks and neurorobotics. The course offers lectures on the latest achievements in understanding learning, neural and network dynamics and function in health and disease, neurorobotic theory and applications, and hands-on tutorials on the EBRAINS services and tools. The summer school targets advanced master students, doctoral students and postdoctoral researchers in biomedical and technology sciences, ranging from medicine, biology, psychology, to mathematics, informatics, information technology, physics and chemistry, who would like to get an introduction to neuroinformatics and computational neuroscience and especially the EBRAINS Infrastructure<http://ebrains.eu/> (ebrains.eu<http://ebrains.eu>). Further information and registration: https://www.humanbrainproject.eu/en/education/BNNI2021/ Best regards Pawel Herman on behalf of the organising committee ------------------------------------ Paweł Herman Associate Professor, PhD, Docent KTH Royal Institute of Technology School of Electrical Engineering and Computer Science (EECS) Division of Computational Science and Technology (CST) Lindstedtsvägen 5 114 28 Stockholm, Sweden room 4442 (4th floor, D-building) tel. +46 8 790 6513

1 0

DEADLINE EXTENDED: Summer School on Network and Control Sciences for Psychiatry
by Jasper Bischofberger Sept. 2, 2021

Sept. 2, 2021

Dear all, We are happy to announce a Summer School on “Current and Future Applications of Network and Control Sciences for Psychiatry” Host Institution: *University of Tübingen, Germany* Location: *Online event* Dates: *September 29th – October 1st 2021* Abstract deadline: *September 10th, 2021* Registration fees: *free* *Focus* This Summer School gives an opportunity to students and young researchers to get acquainted with and exchange within a range of current topics in network and control sciences and its applications in Psychiatry. ____ *Keynotes* Danielle S. Bassett (University of Pennsylvania, USA) Peter Dayan (MPI Tübingen, Germany) John Medaglia (Drexel University, USA) For the full list of speakers visit: https://www.medizin.uni-tuebingen.de/de/summer-school-controlpsychiatry/spe… ____ *Registration* The summer school is free of charge for all participants, but to get all important information please register online via: https://www.medizin.uni-tuebingen.de/de/summer-school-controlpsychiatry ____ *Poster Submission* Participants can present their latest research results as a poster. Some of the most relevant submissions will have the possibility to present their work in a dedicated oral session and receive an official certificate. Please submit your abstract (limited to 250 words) latest by September 10th to: controlpsychiatry(a)med.uni-tuebingen.de ____ Questions? Please contact: controlpsychiatry(a)med.uni-tuebingen.de We are excited to see you soon! Best wishes, The organizing committee

1 0