Sign in

Danila Shelepenkov

@shpen.bsky.social
151 followers 349 following 35 posts

Cognitive neuroscience, stupid jokes and philosophy

PostsRepliesMedia
Reposted by Danila Shelepenkov
Decision, Action, and Neural Computation (DANC) lab @danclab.bsky.social · 30/09/2026
Our multilayer MEG paper is out today in Nature Communications: depth-resolved laminar inference in humans 🧠#neuroskyence www.nature.com/articles/s41...
nature.com
Multilayer MEG source modelling enables depth-resolved laminar inference in humans - Nature Communications
Researchers developed a multilayer MEG framework that estimates activity across cortical depth non-invasively. Simulations and three human datasets show when reliable laminar inference is possible, en...
18243
Reposted by Danila Shelepenkov
Markus Butz @markusbutz.bsky.social · 24/08/2026
PS: And this is Matteo: @mattgmasp.bsky.social
052
Reposted by Danila Shelepenkov
Viktor Timokhov @vutyan.bsky.social · 08/04/2026
And here it is! The U.S. is way ahead. Next come Germany and the UK, along with the Netherlands—and based on my impression after reviewing all the data, there often seemed to be some collaboration between them. (4/n)
121
Reposted by Danila Shelepenkov
CoPhy team @cophy.bsky.social · 07/02/2026
Thrilled to share this work, spearheaded by the Danc lab (led by James Bonaiuto) @danclab.bsky.social, in collaboration with us, especially Matteo Maspoli @mattgmasp.bsky.social and Danila Shelepenkov @shpen.bsky.social, and other groups who share the belief that MEG still has untapped potential!
biorxiv.org
098
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
15/ I don't like all the modern elegant (more like tiaras or EEG) systems in the movies, so we'll finish here with one MRI like, TOTAL RECALL!
000
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
14/ This Canadian Robocop (Vindicator) is really scary, besides, I don't think you can just put your brain in plastic
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
13/ Perhaps a couple more invasive ones, for example, how can you sleep with this in your head? And how to wear a hat?
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
12/ Looks brutal(Johnny mnemonic 1994). This and the matrix are my favorite loaded designs (lots of wires, hardware and other stuff)
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
11/ This is opposite, good style, you can see right away that only the frontal lobes are needed (Surrogates 2009) I also think this is a reference to Schwarzenegger's BCI from the movie 6th day
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
10/ Want to dive into VR and become a genius in your garage? This is Lawnmower Man movie, pure trash
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
9/ Back to the early OPM era, the bulb lights inside make everything cooler pls add them to real MEG
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
8/ Brainstorm (1983) - this according to the plot records experiences and memories, cool and stylish but I have a question why two staples on the sides?
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
7/ It's a head drying or maybe brainwashing, who knows, but it looks cool for 50s
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
6/ Old and clean. They were able to transfer consciousness in Metropolis using only 3 electrodes and this was in 1927, impressive
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
5/ My top 1, legend from Ghostbusters, you can still make pasta using this
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
4/ According to the description from neurafutures.com, most often such devices are used in science fiction to control memory or transplant consciousness. Well, I don't think that anyone is working on this now, but what is definitely cool in these films is the design of the devices. Below is my top
neurafutures.com
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
3/ As a database I used bciwiki.org/index.php/Br... add some info from the excellent project www.neurafutures.com and also posts on Reddit and from my own memory (cool device photo starting from 5/ )
bciwiki.org
Brain Computer Interfaces In Fiction - Brain Computer Interface Wiki
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
2/ Well, quite a few movies use BCI or something similar, and I didn't include magical objects like the sorting hat from Harry Potter
100
Danila Shelepenkov @shpen.bsky.social · 07/02/2025
1/ There is a private cinema museum in Lyon, the owner collects various artifacts from movies. Among them is a part of Johnny Depp's (fake) head with electrodes from the Transcendence. I wondered how often BCI devices like this are used in sci-fi films and series (and my own top of them below)
110
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
20/ More important that, predictive coding was able to elegantly combine perception, learning, and attention (and sensory attenuation) into one model. Also while it did change the idea of ​​what a neural signal is, we're still 25 years later trying to figure out how exactly error signals are encoded
000
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
19/ So what about the strange inversion? From my point of view It wasn't really a real inversion from stimulus–cognition–response towards prediction. Although the concept of predictive error was new to psychology and neuroscience
110
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
18/ Finally, there were no studies pointing to suppression in the visual cortex, a key element in pioneer predictive coding papers. For example see, Predictive coding: A fresh view on inhibition in the retina (1982) and Rao & Ballard's work, both mention end-inhibition neurons
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
17/ Additionally, the mathematical foundation were not on the table (like in 90s). Ideas related to optimal information processing, developed within cybernetics and information theory, only emerged in the 60s-70s (when a New Look approache was drowned in criticism).
110
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
16/ Also because many early cognitive psychologists in the 50s-60s were reluctant to work with "set," which was seen as a garbage term tied to psychoanalysis and Gestalt psychology. New Look failed to formalize this correctly
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
15/ I think this happened because it was the simplest interpretation of experimental results (e.g., faster reactions to expected stimuli and slower to unexpected ones) Occam's razor slaughtered the theory's growth
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
14/ These concepts assumed “directive factors, given a stimulus input of certain characteristics, operate to organize the perceptual field in such a way as to maximize percepts relevant and expectations and to minimize percepts inimical to such needs and expectations”
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
13/ What's even stranger is that the opposite happened. New Look proponents developed concepts like Perceptual Defense and perceptual vigilance.
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
12/ Even experiments from the 20th century, like Bruner & Postman's card experiment, closely resembled the oddball paradigm, a classic in predictive coding?
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
11/ So, describing the 20th century solely as stimulus–cognition–response is inaccurate. But why didn’t we get a fully developed precursor theory of predictive coding 50 years earlier? After all, the "set" also includes prediction
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
10/ And apparently for quite a long time, Gibson's 1941 classic review of the "set" concept revealed that, with its almost infinite subdivisions (mental set, motor set, neural set, etc.), including ideas like "neurotic anxiety," "incubation process in reasoning," "time error in psychophysics."
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
9/ It also pointed out that the concept of “set” was a “traditional wastebasket construct until recently” (mid-70s article).
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
8/ article A new look at the new look: Perceptual defense... 1974 shows : “It is not particularly instructive... to review the various studies showing that expectations about input do indeed affect thresholds... for this would be merely to document the existence of a well-accepted phenomenon.”
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
7/ They also clearly stated: “its perception represents a violation of expectation” and “when well-established expectancies fail of confirmation, the organism may face a task of perceptual reorganization.”
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
6/ For instance, Bruner & Postman in their 1949 paper On the perception of incongruity: A paradigm wrote: “It is a truism worth repeating that the perceptual effect of a stimulus is necessarily dependent upon the set or expectancy of the organism.”
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
5/ There were also works in psychology within the New Look framework that followed a set-stimulus–cognition–response model, much closer to modern predictive coding.
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
4/ For example, F. Attneave's 1954 paper Some informational aspects of visual perception where he demonstrated redundancy of sensory information. He also concluded: “predictions are not possible on the basis of the present visual field, but depend also upon previous fields”
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
3/ In reality, precursor ideas to predictive coding were "in the air" throughout the 20th century in psychology, waiting for a unifying theory
110
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
2/ From his point of view this paper fundamentally altered the dominant 20th-century stimulus–cognition–response framework. While Friston's essay is excellent, it simplifies the real picture and shifts focus away from the contributions of Rao & Ballard
100
Danila Shelepenkov @shpen.bsky.social · 31/01/2025
1/ In various sciences, what Dennett called a "strange inversion" occasionally occurs. In his 2018 essay Does predictive coding have a future? Karl Friston argued that the same thing happened in neuroscience in 1999 after the publication of Rao & Ballard's Predictive coding in the visual cortex
120