#brain machine interface

Standardized and reproducible measurement of decision-making in mice. Aguillon-Rodriguez, V., Angelaki, D., Bayer, H., Bonacchi, N., Carandini, M., Cazettes, F., … Zador, A. M. (2021). eLife, 10, e63711.

Neural integration underlying naturalistic prediction flexibly adapts to varying sensory input rate. Baumgarten, T. J., Maniscalco, B., Lee, J. L., Flounders, M. W., Abry, P., & He, B. J. (2021). Nature Communications, 12(1), 2643.

Predictive Visual Motion Extrapolation Emerges Spontaneously and without Supervision at Each Layer of a Hierarchical Neural Network with Spike-Timing-Dependent Plasticity. Burkitt, A. N., & Hogendoorn, H. (2021). Journal of Neuroscience, 41(20), 4428–4438.

Integration of predictions and afferent signals in body ownership. Chancel, M., Hasenack, B., & Ehrsson, H. H. (2021). Cognition, 212, 104722.

A brain-computer interface that evokes tactile sensations improves robotic arm control. Flesher, S. N., Downey, J. E., Weiss, J. M., Hughes, C. L., Herrera, A. J., Tyler-Kabara, E. C., … Gaunt, R. A. (2021). Science, 372(6544), 831–836.

Response-based outcome predictions and confidence regulate feedback processing and learning. Frömer, R., Nassar, M. R., Bruckner, R., Stürmer, B., Sommer, W., & Yeung, N. (2021). eLife, 10, e62825.

Multidimensional population activity in an electrically coupled inhibitory circuit in the cerebellar cortex. Gurnani, H., & Silver, R. A. (2021). Neuron, 109(10), 1739-1753.e8.

Dorsal CA1 hippocampal place cells form a multi-scale representation of megaspace. Harland, B., Contreras, M., Souder, M., & Fellous, J.-M. (2021). Current Biology, 31(10), 2178-2190.e6.

How do non-human primates represent others’ awareness of where objects are hidden? Horschler, D. J., Santos, L. R., & MacLean, E. L. (2021). Cognition, 212, 104658.

Robotic hand augmentation drives changes in neural body representation. Kieliba, P., Clode, D., Maimon-Mor, R. O., & Makin, T. R. (2021). Science Robotics, 6(54).

Trading mental effort for confidence in the metacognitive control of value-based decision-making. Lee, D. G., & Daunizeau, J. (2021). eLife, 10, e63282. https://doi.org/10.7554/eLife.63282

Experience replay is associated with efficient nonlocal learning. Liu, Y., Mattar, M. G., Behrens, T. E. J., Daw, N. D., & Dolan, R. J. (2021). Science, 372(6544).

A Hierarchy of Functional States in Working Memory. Muhle-Karbe, P. S., Myers, N. E., & Stokes, M. G. (2021). Journal of Neuroscience, 41(20), 4461–4475.

Mouse prefrontal cortex represents learned rules for categorization. Reinert, S., Hübener, M., Bonhoeffer, T., & Goltstein, P. M. (2021). Nature, 593(7859), 411–417.

Neurons including hippocampal spatial view cells, and navigation in primates including humans. Rolls, E. T. (2021). Hippocampus, 31(6), 593–611.

Nonlinear spatial integration in retinal bipolar cells shapes the encoding of artificial and natural stimuli. Schreyer, H. M., & Gollisch, T. (2021). Neuron, 109(10), 1692-1706.e8.

Transsynaptic modulation of presynaptic short-term plasticity in hippocampal mossy fiber synapses. Vandael, D., Okamoto, Y., & Jonas, P. (2021). Nature Communications, 12(1), 2912.

Evidence for a single mechanism gating perceptual and long-term memory information into working memory. Verschooren, S., Kessler, Y., & Egner, T. (2021). Cognition, 212, 104668.

The hippocampal‐entorhinal system represents nested hierarchical relations between words during concept learning. Viganò, S., & Piazza, M. (2021). Hippocampus, 31(6), 557–568.

New study allows brain and artificial neurons to link up over the web

Neuralink: Beyond the hype

The sewing machine like robot that is the linchpin for neuralink.

Brain machine interfacing, as someone who does research in the field and is getting a PhD in a brain machine interface lab, I think I’m qualified to comment on the progress neuralink. There’s a lot of hype out there, curing disease, ending paralysis, a world where we are part of the machine and the machine is part of us. Is it…

Ez az első értelmes videó, amit Elon Musk agybaültetős brain-machine-interface-éről láttam, ami ha úgy alakul, akkor az okostelefonok ÉS az internet ÉS a televízió ÉS a számítógépes játékok ÉS a hadviselés stb jövője is lehet. 

The probe—which does not require a power supply—directly records neural voltage changes by being able to interface with all regions of the brain from the level of single neuron through circuits and networks, in which the mesh recording electrode is connected by passivated metal lines (that is, having a protective coating applied to its surface) to input/output pads located at the opposite end of the mesh structure. These I/O pads, in turn, are then connected to Flat Flexible Cables (FFC) and plugged in external system for recording. The researchers also conducted systematic post-implantation studies, finding minimal or absent neural immune responses, and moreover that brain tissue had penetrated and merged with the mesh probe. The scientists note that the mesh implant may never require removal—but if it does, doing so would be a straightforward if not issue-free procedure. They conclude that most areas of fundamental neuroscience research could benefit from mesh electronics providing long-term stability and single-neuron resolution—unique capabilities not found in conventional neuroprosthetics—and state in their paper that ultraflexible open mesh electronics probes could in the future enable a wide range of opportunities for in vivo chronic recording and modulation of brain activity.

"The horizontal and sagittal brain slices—which contain cross-sections of implanted mesh probes, and nearly the entire implanted mesh probe, respectively—were stained with antibodies that can target neuron somata, axons, astrocytes and microglia," Lieber continues, Moreover, he points out that these studies demonstrated that unlike conventional probes, neuron somata and axons around mesh probes were not harmed, and led to natural tissue levels at the mesh probe surface. "Similarly, the markers for immune response inflammation highlighting astrocytes and microglia showed that these species became background level after only a couple of weeks in the mesh electronics—but they proliferated and accumulated at the interfaces of conventional probes." The scientists also found that the ability of allowing the neurons and mesh to interpenetrate is universal for all meshes they injected and imaged at 6~12 weeks post injection, prompting them to initiate additional experiments to find out how the size of mesh electronics structural elements and other parameters might be tuned to enhance the capability of neuron interpenetration.

It should be noted, Lieber tells Phys.org, that the researchers are being conservative when in their paper they write a minimal immune response pending more detailed marker analyses. "In fact, we believe there is no immune response from the mesh because our results show that any initial enhancement in astrocyte and microglia return to background with no measureable difference proximate or distal to the probe at 12 weeks—and, as shown in our 2016 Nature Methods paper1—up to at least one year. We thus believe that the response is due to acute damage that occurs when inserting the needle (or for that matter, any probe) into the brain—but given the lack of immune response from the mesh probe, this acute damage all heals over time, as opposed to worsening, as is the case with conventional probes."

However, Lieber adds, should the mesh need to be removed, it can be directly extracted with minimum force and damage to the brain. "While this could cause a small amount of damage due to seamless integration with neural tissue, we believe the unprecedented stability and absence of chronic immune response of our neural tissue-like mesh electronics will lead to a paradigm change where the probe is a lifelong implantation that does not require removal."

Moving forward, Lieber says that they are conducting ongoing studies of new mesh designs having high large numbers of electrodes and multisite injections. "Moreover, our next steps include implantations of mesh electronics into tissues and organs other than the brain—for example, in the eye for in vivo recording of single retinal ganglion cells, in the spinal cord, in the muscle for studying signal propagation at the neuromuscular junction, and so on. We're also beginning studies exploiting the unprecedented stability and absence of chronic immune response of the mesh electronics in Alzheimer's and Parkinson's disease models, and are working on implantation of mesh electronics in non-human primates subjects and human patients."

Recent Advances in Electrical Neural Interface Engineering: Minimal Invasiveness, Longevity, and Scalability. Luan, L., Robinson, J. T., Aazhang, B., Chi, T., Yang, K., Li, X., … Xie, C. (2020). Neuron, 108(2), 302–321.

Placing language in an integrated understanding system: Next steps toward human-level performance in neural language models. McClelland, J. L., Hill, F., Rudolph, M., Baldridge, J., & Schütze, H. (2020). Proceedings of the National Academy of Sciences of the United States of America, 117(42), 25966–25974.

Artificial Intelligence and the Common Sense of Animals. Shanahan, M., Crosby, M., Beyret, B., & Cheke, L. (2020). Trends in Cognitive Sciences, 24(11), 862–872.