A cryptography-based approach for movement decoding

Dyer, Eva L.; Azar, Mohammad Gheshlaghi; Perich, Matthew G.; Fernandes, Hugo L.; Naufel, Stephanie; Miller, Lee E.; Körding, Konrad P.

doi:10.1038/s41551-017-0169-7

Cited by 58 publications

(44 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, the map into the latent manifold could be nonlinear, e.g. using Isomap [7,41]. There also exist several nonlinear alignment procedures, such as kernel CCA or Distance Covariance Analysis , although such extensions have yet to translate into BCI practice [42].…”

Section: Across-animal Decoding In the Rodent Olfactory Bulbmentioning

confidence: 99%

Across-animal odor decoding by probabilistic manifold alignment

Herrero-Vidal

Rinberg

Savin

2021

Preprint

View full text Add to dashboard Cite

Identifying the common structure of neural dynamics across subjects is key for extracting unifying principles of brain computation and for many brain machine interface applications. Here, we propose a novel probabilistic approach for aligning stimulus-evoked responses from multiple animals in a common low dimensional manifold and use hierarchical inference to identify which stimulus drives neural activity in any given trial. Our probabilistic decoder is robust to a range of features of the neural responses and significantly outperforms existing neural alignment procedures. When applied to recordings from the mouse olfactory bulb, our approach reveals low-dimensional population dynamics that are odor specific and have consistent structure across animals. Thus, our decoder can be used for increasing the robustness and scalability of neural-based chemical detection.

show abstract

Section: Across-animal Decoding In the Rodent Olfactory Bulbmentioning

confidence: 99%

Across-animal odor decoding by probabilistic manifold alignment

Herrero-Vidal

Rinberg

Savin

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The stable latent dynamics reported here offer an intriguing alternative to existing approaches: BCIs based on latent dynamics as opposed to recorded neural activity could be periodically aligned through a linear procedure such as CCA, thereby achieving stable performance through months or even years of neural turnover. Recent work has shown that approaches based on latent dynamics can be used to improve decoding stability 70,71,38 , adjust for changes in neural inputs 72 , and enable unsupervised decoding 68 .…”

Section: Practical Implications For Braincomputer Interfacesmentioning

confidence: 99%

“…Due to neural turnover, the neural signals that provide inputs to these decoders typically change after a few days, rapidly leading to degraded BCI performance 65,66 . Many groups have been working to develop computational techniques to continually recalibrate these decoders, and thereby restore some of their degraded function 67,68 . Other groups have suggested that the use of input signals such as multiunit threshold crossings 65 or local field potentials 69 could reduce the magnitude of these changes, at the risk of reducing the amount of available information.…”

Section: Practical Implications For Braincomputer Interfacesmentioning

confidence: 99%

A stable, long-term cortical signature underlying consistent behavior

Gallego

Chowdhury²,

Solla³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Animals readily execute learned motor behaviors in a consistent manner over long periods of time, yet similarly stable neural correlates remained elusive up to now. How does the cortex achieve this stable control? Using the sensorimotor system as a model of cortical processing, we investigated the hypothesis that the dynamics of neural latent activity, which capture the dominant co-variation patterns within the neural population, are preserved across time. We recorded from populations of neurons in premotor, primary motor, and somatosensory cortices for up to two years as monkeys performed a reaching task. Intriguingly, despite steady turnover in the recorded neurons, the low-dimensional latent dynamics remained stable. Such stability allowed reliable decoding of behavioral features for the entire timespan, while fixed decoders based on the recorded neural activity degraded substantially. We posit that latent cortical dynamics within the manifold are the fundamental and stable building blocks underlying consistent behavioral execution.

show abstract

“…In addition, we could interpret the improvement in the decoding performance gained with the encoding model from the perspective of statistics of movements. Kording, et al 42 proposed a fundamentally new approach, alignment decoding (DAD), leveraging the statistics of movements. The understanding of the statistics of movements can help us in many situations where obtaining simultaneous recordings of both neural activity and kinematics is challenging, expensive, or impossible.…”

Section: Discussionmentioning

confidence: 99%

Interactions of encoding and decoding problems to understand motor control

Wen

Yin

Zheng

et al. 2019

Preprint

View full text Add to dashboard Cite

Learning a map from movement to neural data (Encoding Problem) and vice versa (Decoding Problem) are crucial to understanding motor control. A principled encoding model that understands underlying neural dynamics can help better solve the decoding problem. Here, we develop a new generative encoding model leveraging deep learning that autonomously captures neural dynamics. After training, the model can synthesize spike trains given any observed kinematics, under the guidance of the learned neural dynamics. When neural data from other sessions or subjects are limited, synthesized spike trains can improve cross-session and cross-subject decoding performance of a Brain Computer Interface decoder. For cross-subject, even with ample data for both subjects, neural dynamics learned from a previous subject can transfer useful knowledge that improves the best achievable decoding performance for the new subject. The approach is general and fully data-driven, and hence could apply to neuroscience encoding and decoding problems beyond motor control.1 Current deep generative models can only generate samples from the distribution they have been trained on. For example, in machine vision, a generative model can only generate images of dogs and cats if it has only been trained on images of cats and dogs. We do not expect it can generalize to images of birds. Here, because each distinct kinematics is a new category such as birds, we do not expect our model to generalize well to novel kinematics in terms of neural dynamics. However, we show that, after fine-tuning with a small amount of neural data from another session or subject, our encoding model can generalize to novel situations improving cross-session and cross-subject decoding performance.

show abstract

A cryptography-based approach for movement decoding

Cited by 58 publications

References 40 publications

Across-animal odor decoding by probabilistic manifold alignment

Across-animal odor decoding by probabilistic manifold alignment

A stable, long-term cortical signature underlying consistent behavior

Interactions of encoding and decoding problems to understand motor control

Contact Info

Product

Resources

About