Nidhi Seethapathi
@nidhise.bsky.social
assistant professor at MIT building computational models to understand human movement
The IMSI team was proud to participate in Dynamic Walking Pacific Northwest, contributing presentations spanning biomechanics, motor control, robotics, and assistive technologies while connecting with researchers across the movement sciences. #IMSI #MovementScience #Robotics
Natural behavior unfolds as a continuous stream of actions. Because these actions often occur in rapid succession, the brain must prepare multiple future actions while the current action is being executed. Our TICS piece explores how this works. www.sciencedirect.com/science/arti...
sciencedirect.com
A systematic investigation reveals dissociable effects of ageing on implicit and explicit components of sensorimotor learning
A systematic investigation reveals dissociable effects of ageing on implicit and explicit components of sensorimotor learning
Nature Human Behaviour, Published online: 19 May 2026; doi:10.1038/s41562-026-02468-7Ageing impairs strategic re-aiming but enhances implicit recalibration during sensorimotor learning. The strategic deficit probably reflects reduced memory caching, whereas the implicit enhancement is linked to altered multisensory integration.
dlvr.it
Thrilled to have this preprint out, from the amazing Zekun Sun! "Motor abstraction training generalizes to the refinement of specific movement patterns" In brief, across 8 experiments we found evidence of a causal role for abstract representations in motor learning www.biorxiv.org/content/10.6...
A new paper on guinea fowl gait biomechanics for National Biomechanics Day! How do bipedal animals adjust movement to avoid falls in slippery terrain? We found that guinea fowl slow down, take shorter steps and adjust posture to reduce fall risk in slippery terrain, just like humans. 1/3
If you're interested in emerging ideas in neural interfaces, I humbly suggest my lab's latest: www.nature.com/articles/s42... Neural interfaces create dynamic interactions between the brain & devices. This means mean we need new engineering approaches beyond typical ML to "decode" a static brain
Computational framework to predict and shape human–machine interactions in closed-loop, co-adaptive neural interfaces - Nature Machine Intelligence
Madduri et al. introduce a computational framework grounded in control and game theory to model co-adaptation between users and decoders in neural interfaces. This framework enables a principled desig...
nature.com
"The challenge ... is identifying the level of description that generalizes over the scientifically relevant domain" Totally agree with Xaq on this point! The default assumption that mechanism is the correct level is problematic: Marr 1 may be the most generalizable!
The Transmitter @thetransmitter.bsky.social · 5mo agoNeuroscience has become increasingly concerned with prediction, and machine learning with causal explanation, with each field adopting methods from the other, writes @gershbrain.bsky.social. Will this bring us closer to understanding neural systems? www.thetransmitter.org/the-big-pict...
How does perceived risk shape adaptation and learning? Our new work reveals that locomotor adaptation proactively navigates a "fall risk landscape" , modulating learning parameters that dictate optimality to prioritize safety. (work with Inseung Kang and Kanishka Mitra) doi.org/10.64898/202...
Fall risk-aware adaptation explains suboptimal locomotor performance
Human locomotion requires balancing multiple biological objectives, such as metabolic energy efficiency, stability, and symmetry. While models based on optimization successfully predict how humans walk in familiar settings, they fail to explain why individuals adopt inefficient movement patterns in novel environments, even after extensive practice. Here, we show that such suboptimality in a novel environment arises from a fundamental prioritization of safety. We find that individuals do not simply fail to reach an optimal solution; instead, they navigate an environment-dependent risk landscape by mitigating the statistical probability of falling. We find that this risk-averse strategy is explained by adjusting internal learning parameters: specifically, the learning rate and the tradeoff between metabolic cost and symmetry, in a manner that lowers fall risk. To quantify this process, we developed an ‘inverse adaptation’ modeling framework; this approach works backwards from locomotor performance data to mathematically infer the underlying internal learning parameters and how they vary with fall risk. Our analysis reveals that the observed motor performance is explained by a global probabilistic fall risk rather than a local step-based measure of instability. Ultimately, these findings reveal that fall risk-aware adaptation explains suboptimal locomotor behavior, providing a new data-driven framework to understand the drivers of motor performance. ### Competing Interest Statement The authors have declared no competing interest.
doi.org
Paper is out by @harikalidindi.bsky.social : feedback control model of neural population activity www.cell.com/cell-reports...
Feedback control of random networks as a model of flexible motor cortical dynamics across tasks
Kalidindi and Crevecoeur develop a computational framework linking feedback-controlled networks to limb dynamics. They demonstrate that optimal control of fixed network reproduces key motor cortical d...
cell.com
Meet Surabhi Simha, a Postdoctoral Fellow at Georgia Tech and Emory University, whose research investigates neural control of human movement with a focus on mechanistic muscle spindle models and energy-efficient gait adaptation. Part of our Women at IMSI spotlight series. #WomenAtIMSI #Biomechanics
I hope this new-ish paper will challenge people to think about the interpretations of our models STRIATUM SUPPORTS REINFORCEMENT AND NOT ACTION SELECTION (!!!) tinyurl.com/HodgeAndYttri @cmuscience.bsky.social @cmu-neuroscience.bsky.social #neuroskyence
Striatal modulation supports context-specific reinforcement and not action selection
Hodge et al. used closed-loop optogenetic stimulation and naturalistic behaviors to determine that striatal activity does not select the action to perform but rather biases performance through reinfor...
tinyurl.com
Yang ICoN researchers are revealing the shared rules of balance across species. 🧠🚶♀️🐭🪰 Humans, mice, and flies all use the same error-correction strategy to stay upright, thanks to new work led by ICoN Center’s @nidhise.bsky.social & ICoN Fellow @antoinecomite.bsky.social.
Staying stable
Scientists at MIT’s McGovern Institute have determined that animals with very different bodies likely use a shared strategy to balance themselves when they walk.
news.mit.edu
New preprint; Shared feedback control principles across standing and walking. With Yang Geng and @jvandieen.bsky.social. Feedback welcome! www.biorxiv.org/content/10.1...
Shared feedback control principles across standing and walking
Feedback control is required to maintain stability in both walking and standing, whereas the control parameters may differ due to the inherently different characteristics of these tasks. How feedback ...
biorxiv.org
Let's compare our world models. I find that different people seem to have rather distinct internal world models. E.g. I personally have neither visual imagination nor an inner voice, found it weird others do. Here is a quick google forms to check idea: docs.google.com/forms/d/e/1F...
World-models in your head
Talking with a lot of people, they have rather shocking different kinds of world-models. I believe that people have somewhat specialized simulators. Let me list some and then give you the chance to ad...
docs.google.com
The best place in the world to be at the cutting edge of muscle models!
Kiisa Nishikawa@kiisa.bsky.social · 9mo agoThe Nishikawa lab at Northern Arizona University and the NSF-funded Integrative Movement Sciences Institute are hiring a postdoc to help develop multiscale muscle models - please apply and share widely! careers.nau.edu/cw/en-us/job...
I think almost all scientific projects should be planned carefully. And I think an app can dramatically improve that. So I wrote an app for that (free for now, if you can fund this let me know). I tested it quite a bit (>8000 users in beta so far). try it: planyourscience.com
My research group has an open position for a postdoc! Interested in investigating the postural transition towards mammalian gait using movement simulations? We might have the right position for you! More details and application info here: www.asm.tf.fau.de/en/2025/11/1...
Postdoctoral researcher for DFG-funded project “FossilGaitSim”
The Biomechanical Motion Analysis and Creation (BioMAC) group at the Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) invites applications for a postdoctoral position with the goal to…
asm.tf.fau.de
On a similar note, I feel like we ought to stop writing papers as flowing text and instead just have the bullet points you would have used as a prompt. I'm actually not joking.
MLMC2025 is tomorrow! For those joining remotely, here is the link for the meeting: harvard.zoom.us/j/9544921964... Reminder, the conference will start promptly at 9am PST. Please visit the website for additional information, schedule, registration, etc! motor-conference.org/openconf.php
Join our Cloud HD Video Meeting
Zoom is the leader in modern enterprise cloud communications.
harvard.zoom.us
Come to my poster at SfN on Wednesday morning! I'll present my work on how the cerebellum uses internal models to control neocortical dynamics in a non-motor learning task. I'm biased, but I think it's pretty exciting! www.abstractsonline.com/pp8/#!/21171...
Program Planner
abstractsonline.com
India won the Women’s Cricket World Cup for the first time with a 52-run victory over South Africa on Sunday.
India wins Women’s Cricket World Cup for first time with 52-run victory over South Africa
India has won the Women’s Cricket World Cup for the first time with a 52-run victory over South Africa on Sunday.
bit.ly
1/N How do neural dynamics in motor cortex interact with those in subcortical networks to flexibly control movement? I’m beyond thrilled to share our work on this problem, led by Eric Kirk @eric-kirk.bsky.social with help from Kangjia Cai! www.biorxiv.org/content/10.1...
We are excited to share our new paper that dives into how ongoing decision deliberation reflects ongoing movements: (jneurosci.org/content/45/3... A tour-de-force by the incredible Jan Calalo.
The cerebellum isn’t just about coordinating movement. It’s implicated in nearly every domain of cognition—from language to social behavior. But how exactly does the cerebellum contribute to action and cognition? 🧵 Check out our new paper w/ Rich Ivry. arxiv.org/abs/2509.09818
Cerebellar Contributions to Action and Cognition: Prediction, Timescale, and Continuity
The cerebellum is implicated in nearly every domain of human cognition, yet our understanding of how this subcortical structure contributes to cognition remains elusive. Efforts on this front have ten...
arxiv.org
Stoked to see our study out in final form! Big kudos to @jonathanamichaels.bsky.social for driving this along for the past 5 years.
Jonathan A. Michaels@jonathanamichaels.bsky.social · 10mo agoThrilled that our paper is out today in Nature! www.nature.com/articles/s4...
New Pre-Print: www.biorxiv.org/cgi/content/... We’re all familiar with having to practice a new skill to get better at it, but what really happens during practice? The answer, I propose, is reinforcement learning - specifically policy-gradient reinforcement learning. Overview 🧵 below...
Policy-Gradient Reinforcement Learning as a General Theory of Practice-Based Motor Skill Learning
Mastering any new skill requires extensive practice, but the computational principles underlying this learning are not clearly understood. Existing theories of motor learning can explain short-term ad...
biorxiv.org
Our 2024 paper showed that policy gradient RL (with performance-based memory updates) predicts long-horizon motor learning. Now, @adrianhaith.bsky.social shows that policy-gradient RL also explains learning in other shorter horizon tasks. Exciting! www.biorxiv.org/content/10.1...
Policy-Gradient Reinforcement Learning as a General Theory of Practice-Based Motor Skill Learning
Mastering any new skill requires extensive practice, but the computational principles underlying this learning are not clearly understood. Existing theories of motor learning can explain short-term ad...
biorxiv.org
Important work by @nidhise.bsky.social discovering signatures of control for stable locomotion across species. Excited to see the models applied to diverse species as part of IMSI @movementscience.bsky.social
Nidhi Seethapathi@nidhise.bsky.social · 10mo agoExcited that this work discovering cross-species signatures of stabilizing foot placement control is now out in PNAS! pnas.org/doi/10.1073/... @antoinecomite.bsky.social
Check out this #incredible work on #stability and #control during movement across #humans, #mice, and #flies. Congratulations @nidhise.bsky.social and @antoinecomite.bsky.social 👏🏽⭐
Nidhi Seethapathi@nidhise.bsky.social · 10mo agoExcited that this work discovering cross-species signatures of stabilizing foot placement control is now out in PNAS! pnas.org/doi/10.1073/... @antoinecomite.bsky.social
Excited that this work discovering cross-species signatures of stabilizing foot placement control is now out in PNAS! pnas.org/doi/10.1073/... @antoinecomite.bsky.social
Foot placement control underlies stable locomotion across species | PNAS
Animals navigate their environment stably without inefficient course corrections despite unavoidable errors. In humans, this stability is achieved ...
pnas.org
Nidhi Seethapathi@nidhise.bsky.social · 2y agoHow do animals walk stably in the presence of intrinsic noise or external perturbations? biorxiv.org/content/10.1... We lack a unified answer to this question across species. In this preprint, we find that legged species use a shared strategy for stable locomotion: foot placement control! (1/6)