NeuroMat - Research Center for Neuromathematics

22/12/2025

Take a word like "champagne". Our brain does interesting things in order to store it in our mental lexicon — that is, our memory of words. "Champagne" is formed by several phonemes and, still, it comprises a single morpheme, a single unit. Hence, our linguistic apparatus can treat it as such when making sense of sentences, for example.

Using Dehaene and colleagues’ definition, let’s consider “chunk" a “group of contiguous items that frequently recurs as a whole and that are therefore usefully encoded as a single group by the nervous system” (2015; see Ref. 2). As they explain, a sequence of events might well be grouped together as a “chunk” by the brain and stored for further processing.

One interesting question, then, would be what cues the brain uses to segment continuous sequences into smaller units. Thinking of spoken sentences, for example, we listen to a continuous acoustic signal and — effortlessly — segment it into words. In written texts, periods help us identify when the sentence ends. Likewise, blank spaces cue our eyes to word boundaries. In both cases, we find our units — sentences and words, respectively.

Now, let’s put language aside and consider more general experiences. How does the brain look for patterns in any auditory stimuli?
- - -
Electrophysiological data opens a window through which neural processes can be assessed in several aspects. Thanks to this method, Najman and colleagues (2025; see Ref. 1) addressed the issue of which classes of probabilistic models the brain uses to encode sequences of events.

Participants were exposed to a samba-like rhythm made of hand-claps, while some small portion of the sounds were omitted. The sequences of auditory stimuli were random. The innovation proposed was how to tackle the question, given that samba has a pattern of strong beats, weak beats and silent units.

A novel procedure was introduced to investigate whether and how the brain forms clusters of data based on probabilistic features of the stimuli. The paper unveils “hidden features” that could not be retrieved by a context tree — a subject of a previous post (see Ref. 3). The results show that the brain uses “the occurrence of the strong beat to identify the structure embedded in the sequence of stimuli”.

The key statistical role of strong beats, specifically, is highlighted.

It turns out that sequences of auditory stimuli, just like words in sentences, may also contain their own cues, markers for boundaries — and they can be seen in EEG data registered from the pre-frontal cortex. Moreover, the researchers’ new clustering procedure allowed them to group sets of EEG data by intrinsic law. Once again, interesting results come along with methodological innovation that can benefit the whole field.
- - -
How each chunk is treated, being a single unit, depends on the phenomenon being observed; the neural or mental process. As usual, there is much to be elucidated. We know that, in some cases, units can be merged in a higher level to form another chunk (another unit), meaning the brain can process information in a hierarchical fashion — as it does with language when building sentences, to recall our initial example.
- - -
Now, broadly speaking, several probabilistic models can be recruited to parse random sequences like. Mathematical lens, they help the brain calibrate its “vision” accordingly. Hidden patterns come to light; claps get more predictable. Auditory sequences become samba.

That is it for this week. Don’t forget to check the paper — links below!
- - -
Chunking | The “Black Box of Science” series
References:
1. Najman et al. (2025) - The ‘design features’ of language revisited. https://doi.org/10.1371/journal.pcbi.1012765
2. Dehaene et al. (2025) - The Neural Representation of Sequences: From Transition Probabilities to Algebraic Patterns and Linguistic Trees. https://doi.org/10.1016/j.neuron.2015.09.019
3. NeuroMat (2025) - "Waltz in the Dark | The “Black-Box of Science” Series", Facebook post. https://www.facebook.com/share/p/1GPnV1JvoQ/

Author summary A classical conjecture is that the brain is constantly estimating regularities from sequences of events to be able to properly act upon the environment. We assume that, by doing statistics, the brain chooses a model from a class of possible models. Which class of models is used by the...

01/12/2025

Parkinson’s Disease (PD) has typical motor symptoms that are critical for a qualified diagnosis. They also influence public understanding about the characteristics of Parkinson’s and the experience of people with the disease. Research effort mirrors this fact: most works focus on motor symptoms.

However, that is not the whole picture. Importantly, non-motor symptoms are worth the attention of those in search of a better understanding of different aspects of life with Parkinson’s. Emotional and social variables, for example, have meaningful influence on these peoples’ lives.
- - -
Katia Nóbrega and colleagues (2025; see Ref. 1) point out that Sexual Health is one fundamental aspect of this discussion and had not received much attention until recently. It would not be surprising, though, to presume motor symptoms are decisive for evaluating this dimension of people with PD. Let’s say we can find a correlation here.

The study conducted by NeuroMat members shows, however, that correlation does not necessarily mean prediction — hence, we can’t properly comprehend the subject without a proper account of emotional and social factors. Here’s a turning point that many wouldn’t expect.
- - -
“Sexual Health is influenced by a complex interplay of biological, psychological, and social factors”. That applies to typical, healthy adults, and also to people with PD. In all cases, it deeply impacts quality of life.

Nóbrega and colleagues conducted a cross-sectional study with 100 women with PD who were part of the AMPARO Network (see Ref. 2) and accepted the invitation. Using remote interviews, they collected data on several aspects: demographic, clinical, cognitive capacity, motor and non-motor experiences, fatigue, self-esteem, sleep disorders, etc.

Most interestingly, “while several motor, non-motor, and social factors were correlated with short-term and long-term sexual health, only two factors—couple’s relationship quality and sleep quality—were significant predictors of
both short-term and long-term sexual health outcomes”. Regarding our previous point, let’s explore the black-box of statistical terms.

A correlation between factors — for example, motor factors and sexual health — means that changes in one tends to occur when the other changes. However, that does not mean causality.

Finding a predictive relationship, on the other hand, means going one step further. In this case, sleep quality and couple relationship were found to significantly predict both short- and long-term sexual health. It is worth mentioning this analysis was possible thanks to multiple regression models.
- - -
The study reinforces the importance of tailored therapeutic approaches for people with PD, which once again recall us of the value of multidisciplinary effort — in both, Scientific research and Clinical work.

Check the article and learn more about the AMPARO Network in the links below.
- - -
Prediction | The “Black Box of Science” series

References:
1. Nóbrega et al. (2025) - The impact of motor, non-motor, and social aspects on the sexual health of women living with Parkinson’s disease. DOI: 10.1177/1877718X251315375
2. AMPARO Network -

The Research, Innovation and Dissemination Center for Neuromathematics (NeuroMat), hosted by the University of São Paulo (USP), Brazil, and funded by the São Paulo Research Foundation (FAPESP), is offering two post-doctoral fellowships for recent PhDs with outstanding research potential. The fel...

17/11/2025

The game is set on the computer screen. The participant of the experiment is given a task: assuming the role of the goalkeeper in a sequence of penalty trials, choose where to jump to save the kick by pressing the arrows on the keyboard: ‘left’, ‘center’ (down) or ‘right’. In each trial, the penalty kick takes place only after the participant has conveyed his decision by pressing a button.

Response times are recorded at each trial.
- - -
“More than a century ago, Helmholtz conjectured that the human brain is able to detect statistical regularities in a sequence of events” (Cabral-Passos and colleagues, 2024; Ref. 1). This sort of conjecture often pushes science ahead for centuries. We have seen this before in this series, when discussing Alan Turing’s wonderings on machines, intelligence and brains (Ref. 2). Here, the renowned Hermann von Helmholtz represents a time reference for the long history of research on a particular topic (Ref. 3).

“Since then, psychophysiological measurements have been employed to study this conjecture”, add Paulo Cabral-Passos, Antonio Galves, Jesus Garcia and Claudia Vargas in the paper. The authors’ bring a new conjecture, one emerged from the centuries-long history of scientific work. They adopt the tradition of applying a new probabilistic framework to assess our cognitive system — one of the defining points of NeuroMat’s trajectory and legacy.

The researchers revisited the topic through an experimental paradigm. Following other studies that paved the way with EEG (see the post 1 of our series, “Waltz in the Dark”; Ref 4), they conjectured that the same pattern — i.e., a probability distribution — would also occur at a behavioral level.

The measurement? Response time: the interval between a stimulus onset and the participant’s reaction in a given cognitive task.

- - -

“The participant is asked to predict the successive choices of the penalty taker.” Conjecture: “the probability distribution of response times is a function of the specific sequence of past choices governing the algorithm used by the penalty taker”. Empirical evidence was indeed reported by the authors.

Most interestingly, the success or failure of the previous prediction also affected the distribution of response times — participants were slower after incorrect predictions. Furthermore, the dependence propagates up to two steps forward.

“To the best of our knowledge, this is the first study in which the structure of random stimuli is retrieved from the participant’s response times”, they add to the discussion.

- - -
Among behavioral methods, response time is perhaps the most known and adopted. From Psychology all the way to Physiology and Neuroscience, it is widely instrumentalized in experiments to investigate mechanisms underlying perception, attention, language and decision-making. In many cases, it depends on cheaper tools to extract data from the brain circuitry and human’s cognition, compared to a fMRI or EEG set up.

It is worth mentioning that a study like this is part of a wider net of works and projects that often branch from conjectures. In favor of science construction, questions and methods can be arranged in many elucidative ways.

Do you have favorite methods, tools or techniques? Let us know in the comments!
And check out the references below!
- - -
Post: Response Times | The “Black Box of Science” series

References:
1. Cabral-Passos et al. (2024) - Response times are affected by mispredictions in a stochastic game. https://doi.org/10.1038/s41598-024-58203-7

2. NeuroMat (2025) - "Sandpiles and Neuronal Networks | The “Black-Box of Science” Series", Facebook post. https://www.facebook.com/share/p/1Zjto2wC2c/

3. Helmholtz, H. V. (1867) - Handbuch der Physiologischen Optik.

4. NeuroMat (2025) - "Waltz in the Dark | The “Black-Box of Science” Series", Facebook post. https://www.facebook.com/share/p/1GPnV1JvoQ/

Acting as a goalkeeper in a video-game, a participant is asked to predict the successive choices of the penalty taker. The sequence of choices of the penalty taker is generated by a stochastic chain with memory of variable length. It has been conjectured that the probability distribution of the resp...

03/11/2025

The 4th season of NeuroMat’s podcast, "A Matemática do Cérebro" is on the way.
Felipe Parlato, from our Science Dissemination Team, is preparing five new episodes, each one featuring a paper published by NeuroMat researchers. The episodes explore studies on topics such as the Goalkeeper’s Game, a research tool developed by NeuroMat, and measurements of brain responses to random rhythmic auditory stimuli.
Photo: Felipe with investigators Cláudia Vargas, Fernando Najman, and Paulo Cabral-Passos, after the recording sessions.

31/10/2025

Sandpiles and Neuronal Networks | The “Black-Box of Science” Series

Alan Turing, in his memorable paper Computing Machinery and Intelligence, was probably the first one to draw an analogy between brain activity and critical processes (See Refs 1. and 2). A critical point, in short, is where a phase transition takes place — for example, the temperature at which liquid water turns into ice. In other words, qualitative change caused by quantitative change.

It turns out we can observe such a pattern all around us in many scales, meaning complexity can emerge in natural settings. Sandpiles have been widely used to illustrate the idea. Look at them for long enough, and perhaps you will see a transition: a single grain on top may appropriately find its place and compose the whole; plausibly, the same small particle can heavily disturb the pile.

Rapid return to stability and a sudden avalanche are both caused by the same undermost event.
— — —
Back to 1950, Turing wondered how a thinking system would fit that scene:
“Each such neutron will cause a certain disturbance which eventually dies away. If,
however, the size of the pile is sufficiently increased, tire disturbance caused by such an incoming neutron will very likely go on and on increasing until the whole pile is destroyed. Is there a corresponding phenomenon for minds, and is there one for machines?”
— — —
Ever since, plenty of theoretical works were built upon the sandpile analogy, even though researchers have never been successful in experimentally reproducing the model in this specific setting. Nevertheless, remarkable progress have been made in many aspects, and experiments actually did found evidence for phase transition in neuronal networks.

Now, Self-Organized Criticality refers to systems able to maintain themselves relatively close to a critical point, without external pre-programming. One main implication would be that, independently of scale, very small changes could lead whether to stability or phase transition.

The concept seems to accurately describe a wide range of phenomena — e.g. forest fires, earthquakes, and market fluctuations, each of which along with its own mathematical model. In fact, its appearance in neuronal networks takes us further into the discussion.

In the scope of NeuroMat, the paper published by Kinouchi, Pazzini and Copelli in 2020 (Ref 1) pleasantly tell us a lot in the very title: Mechanisms of Self-Organized Quasicriticality in Neuronal Network Models. The review of several proposed biological mechanisms highlights the quasi aspect of these models.

“Conservative sandpile models should not be used to model neuronal avalanches because neurons are not conservative”. Neuronal systems “hover around, but do not exactly sit on the critical point”. Hence, they would fit the description of SOqC models.

That raises questions such as: having phase transitions, what would be the advantages of the system behaving like this?
— — —
Theoretical neuroscience is one of the most active areas that employ such ideas to improve our understanding of the brain. Kinouchi, Pazzini and Copelli’s paper underscores that “the study of criticality in neuronal networks developed itself as a research paradigm, with a large literature”.

Some physicists eventually stepped in the field, searching for solutions to this sort of challenge. This transdisciplinary movement is partly due to a wide set of mathematical theories developed during the 20th century. Tools like those behind SOqC models.

Importantly, disciplines converge or interact not just because a field offers different ways of seeking answers for others’ inquiries. Rather, theoretical frameworks provide us with mechanisms for identifying, describing, even reimagining phenomena across the universe; in this case, within the brain.

— — —

“The “skin-of-an-onion” analogy is also helpful”, wrote Turing. “In considering the functions of the mind or the brain we find certain operations which we can explain in purely mechanical terms. This we say does not correspond to the real mind: it is a sort of skin which we must strip off if we are to find the real mind. But then in what remains we find a further skin to be stripped off, and so on. Proceeding in this way do we ever come to the "real" mind, or do we eventually come to the skin which has nothing in it? In the latter case the whole mind is mechanical.”
— — —
That’s it for this week. Check out the references below!

1. Kinouchi, Pazzini, Copelli (2020) - Mechanisms of Self-Organized Quasicriticality in Neuronal Network Models. Frontiers in Physics. https://doi.org/10.3389/fphy.2020.583213
2. Turing, A. M. (1950) - Computing Machinery and Intelligence. Mind LIX, p. 433. https://doi.org/10.1093/mind/LIX.236.433

15/10/2025

𝐈𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐨𝐧 𝐯𝐢𝐚 𝐬𝐢𝐦𝐩𝐥𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧 | 𝐓𝐡𝐞 “𝐁𝐥𝐚𝐜𝐤-𝐁𝐨𝐱 𝐨𝐟 𝐒𝐜𝐢𝐞𝐧𝐜𝐞” 𝐬𝐞𝐫𝐢𝐞𝐬

People often associate innovation and scientific progress with increased technological complexity. When it comes to health care, for example, most of us can list sophisticated devices that might be found in a well-equipped clinic or hospital: such as an fMRI or EEG. Curiously — and fortunately —, science also provides ways of improving life via simplification.

Consider a hypothetical exercise. You are invited to help discuss ideas to improve medical treatment of, say, people with Parkinson’s Disease. You are taught about how postural instability is a debilitating cardinal symptom and how early detection can help prevent falls and their negative consequences. Then you read a list of tests and devices currently used for clinical evaluation of postural control and requested to freely raise ideas on how to improve them.
Your turn. Brainstorming time. What would you say?

Adding new tasks? Makes sense. Changing existing metrics? Alright. Perhaps adding new equipment. A lot can be done — as it is in universities and other research centers around the world. However, how likely it would be that you suggested simplifying a process in order to get better results?

This award-worthy idea certainly looks challenging. But Venas and colleagues, in 2023 (see Ref. 1) elaborated a strategy, took on the task, and made it happen. The study’s goal was investigating “the effectiveness of a two-dimensional balance assessment to identify the decline in postural control associated with Parkinson’s Disease progression”. The great contribution, the Postural Instability Index, was successful in differentiating the first three stages of the disease evolution, which means a “hopeful prospect of employing a clinical tool to detect subtle alterations in the postural control of individuals with PD”.

Besides, as the paper’s title tells us, it is non-expensive. While “the use of sophisticated equipment (...) has limited clinical utility because they are expensive and demand a highly trained team”, here’s innovation built upon already existing tools.
- - -
Laboratories that host this sort of research often need those “sophisticated equipment”. Venas et al. (2023) report quite a list of tests and tools: TUG test, Pull test, Timed Up and Go, Balance evaluation, cameras, computers, a movement analysis system. One huge benefit of science is that we can often concentrate the technological infrastructure to the lab and progressively optimize the work done in hospitals and rehabilitation clinics. Once such results become public, the society foresees transformation — which means, in this case, that less expensive devices to achieve an accurate, reliable assessment.
- - -
In fact, simplicity is widely assumed as a valuable concept for scientific theories, as well. It is worth mentioning that defining what is ‘simple’ in terms of science is far from an easy task. Again, fortunately, things are more transparent when we deal with the cost of technological assets. Furthermore, for a public health system such as the one in Brazil, whose population is above 210 million, “simplifying” a complex assessment also means a huge impact in terms of improving life-quality and democratizing the access to health service.
- - -
We can say that a black-box of science may be full of new methods, practices, and tools. We can’t ignore that some of them plausibly contain the ones we already know, organized differently.

- - -
References:
1. Venas et al. (2023). A non-expensive bidimensional kinematic balance assessment can detect early postural instability in people with Parkinson’s disease. Frontiers in Neurology, v. 14, art. 1243445. Link below!

01/10/2025

LASCON 2026 – X Latin American School on Computational Neuroscience

January 5–30, 2026 | University of São Paulo, Brazil

https://sisne.org/en/lascon-x/

APPLICATION DEADLINE: October 11, 2025 (until midnight, US Pacific Time)

The 10th edition of the Latin American School on Computational Neuroscience (LASCON 2026), which will be held at the University of São Paulo, Brazil, from January 5 to 30, 2026.

LASCON 2026 offers an intensive four-week training program designed for graduate students, advanced undergraduates, and early-career researchers interested in applying computational and mathematical methods to the study of neurons and neural networks. The curriculum spans multiple scales of brain modeling—from biophysically detailed neurons to large-scale networks and theoretical frameworks of brain function and dysfunction—and covers topics such as extracellular field modeling, synaptic and structural plasticity, brain disease and brain state modeling, artificial intelligence, criticality in brain dynamics, consciousness, and brain–machine interfaces.

The program combines morning lectures on theory, afternoon tutorials using leading simulation tools (NEURON, NEST, NetPyNE), and evening group project work, culminating in student project presentations.

This 10th edition marks a special milestone: 20 years of LASCON. Since its first edition in 2006, the school has become a recognized international initiative in computational neuroscience training, attracting students from Latin America and beyond. Over the years, LASCON has welcomed participants from all continents, and its alumni (278 to date) have gone on to successful careers in academia and industry worldwide.

To celebrate this anniversary, LASCON 2026 will feature the largest faculty in the school’s history, with lectures and tutorials covering a wide spectrum of subfields in computational neuroscience. Confirmed faculty include:

Alain Destexhe (Paris-Saclay Inst. of Neuroscience, France) · Alessandro Treves (SISSA, Italy) · Aline Duarte (USP, Brazil) · Arnd Roth (UCL, UK) · Cláudia Vargas (UFRJ, Brazil) · Daniel Takahashi (UFRN, Brazil) · Demian Battaglia (Univ. of Strasbourg, France) · Enzo Tagliazucchi (Univ. of Buenos Aires, Argentina) · Fernanda Matias (UFAL, Brazil) · Florencia Iacaruso (The Francis Crick Institute, UK) · Gaute Einevoll (NMBU, Norway) · Gustavo Rohenkohl (USP, Brazil) · Hans Ekkehard Plesser (NMBU, Norway) · Harel Shouval (Univ. of Texas Medical School at Houston, USA) · Horacio Rotstein (NJIT & Rutgers, USA) · Junji Ito (Jülich, Germany) · Leonardo Elias (Unicamp, Brazil) · Lyle Muller (Western Univ., Canada) · Marja-Leena Linne (Tampere Univ., Finland) · Markus Diesmann (Jülich, Germany) · Maurício Girardi-Schappo (UFSC, Brazil) · Mauro Copelli (UFPE, Brazil) · Netta Cohen (Univ. of Leeds, UK) · Oswaldo Baffa (USP, Brazil) · Patricia Reynaud-Bouret (Côte d’Azur Univ., France) · Rodrigo Pavão (UFABC, Brazil) · Rodrigo Pena (Florida Atlantic Univ., USA) · Sonja Grün (Jülich, Germany) · Sophia Sanborn (Stanford Univ., USA) · Viktor Jirsa (Aix-Marseille Univ., France) · Valeriy Bragin (SUNY, USA) · William Bialek (Princeton University., USA) · William Lytton (SUNY, USA).

Applications: The number of students is limited to 40. Applications must be submitted via the school’s website (link) and include a CV (in English) and two recommendation letters. The application deadline is October 11, 2025 (midnight, US Pacific time).

Costs and support: There is no registration fee. Subject to available funding, accommodation may be provided for participants coming from outside Greater São Paulo. Travel costs must be covered by participants.

Selection: Priority will be given to Latin American students, but applications from all regions are warmly encouraged.

LASCON 2026 is organized as an activity of the FAPESP Research, Innovation and Dissemination Center for Neuromathematics (NeuroMat).

We look forward to welcoming you in São Paulo.

LASCON XLASCON ScopeLASCON X ProgramLASCON X ParticipantsLASCON X VenueLASCON X LodgingLASCON X TransportationLASCON X SupportPrevious editions of LASCONLASCON Alumni X Latin American School on Computational Neuroscience – LASCON 2026 January 5-30, 2026 NeuroMat (Antonio Galves building) Institute...

30/09/2025

𝐓𝐡𝐞 𝐁𝐫𝐚𝐢𝐧’𝐬 𝐃𝐚𝐫𝐤 𝐌𝐚𝐭𝐭𝐞𝐫 | 𝐓𝐡𝐞 “𝐁𝐥𝐚𝐜𝐤-𝐁𝐨𝐱 𝐨𝐟 𝐒𝐜𝐢𝐞𝐧𝐜𝐞” 𝐒𝐞𝐫𝐢𝐞𝐬

Once upon a time, astrophysicists from around the world found themselves troubled by the same puzzle: why did there seem to be invisible bodies in outer space? Did general relativity and other theories really suggest the existence of “ghosts” in the cosmos? Matter that does not interact with light?

More than 100 years have passed, and Dark Matter remains hypothetical, captivating, and plausible — to the point of becoming a metaphor adopted by neuroscientists.
- - -
Astrocytes are star-shaped cells that populate the brain and spinal cord. By no means invisible, they’re all around: a reliable neuroscience handbook would tell you they’re almost half of the number of brain cells (see ref. 2). So why don’t we usually hear about these mysterious stellar entities? Do they puzzle neuroscientists, too?

Our central nervous system is often pictured as a dense universe full of interconnected galaxies that interact with each other via complex electrical utterances — i.e. synapses. Something galaxies and neural populations do have in common is their own chemical language under the surface.

If shining stars usually get more attention, let’s not condemn the barely invisible ones to a black-box on the shelf.
- - -
As dark matter is claimed to serve as gravitational bedrock for cosmic entities, astrocytes provide brain structures with neuronal support: potassium regulation, metabolic maintenance, neurotransmitter recycling and so on. It turns out it can be much more than that.

They seem to act as “active regulators of neural activity, participating in information processing, encoding, and even influencing cognitive functions and behavior” (see ref. 1). While astrophysicists would point their own devices to the sky and look for answers, monitoring isolated astrocytes and their complex chemical life poses a challenge for researchers. The good news, as explained by Bezerra and Roque, is that simulations can take us on step ahead in comprehending the cosmic web inside our heads.

Astrocytes have bidirectional communication with neurons and are affected by neurotransmitters generated by neurons. Meaning, neurons activated somewhere plausibly provoke astrocytes; in turn, they release their own molecules that affect neurons’ activities. Two types of neurotransmitters able to do that are glutamate and dopamine, usually resulting from local and distant “calls”, respectively.

Besides crafting a simplified model that simulates these interactions at lower computational cost, Bezerra and Roque explored the influence of the astrocyte’s morphology on its potential behavior. The results suggest both neurotransmitters play specific roles, and the extent of propagation does depend on the cell’s star-like shape.

- - -

Gravity and solar systems put aside, perhaps we are witnessing neural elements once again sharing the spotlight and becoming co-protagonists of a broad research effort. Thanks to computational and mathematical modeling, we are able to look with fresh eyes at familiar cells and observe something new.

Plausibly, some shadows are projected by dark — but not invisible — stars.

References:
1. Bezerra, T. O., Roque, A. C., 2024. Dopamine facilitates the response to glutamatergic inputs in astrocyte cell models. PLoS Computational Biology, v. 20, n. 12, e1012688. DOI: 10.1371/journal.pcbi.1012688

2. Kandel et al., 2013. Principles of neural science. 5. ed. New York: McGraw-Hill. ISBN 978-0-07-139011-8.

23/09/2025

Images of the Workshop Transcranial Magnetic Stimulation: Breakthroughs in Instrumentation and Neuromodulation

15/09/2025

𝐍𝐞𝐢𝐠𝐡𝐛𝐨𝐫𝐢𝐧𝐠 𝐯𝐨𝐱𝐞𝐥𝐬 | 𝐓𝐡𝐞 “𝐁𝐥𝐚𝐜𝐤-𝐛𝐨𝐱 𝐨𝐟 𝐒𝐜𝐢𝐞𝐧𝐜𝐞” 𝐬𝐞𝐫𝐢𝐞𝐬

“Brain plasticity consists in the ability of the central nervous system to modify in response to changes in behavior, as a consequence of skill acquisition or following central/peripheral injury”. The link between clinical questions and brain mechanisms may not be much of a surprise. It’s not that obvious, however, that neuroscience is currently way beyond the classic approach of locating single functions in discrete brain areas.

The excerpt that opens this post is the very first sentence of the paper (see ref. 1) and is built upon previous works as well (see refs. 2, 3, and 4). It highlights one of the reasons that the human brain is often referred to as the most sophisticated system or machine we know. A circuit able to “rewire” itself.

Fraiman and colleagues’ work touches on critical questions that shed light on the issue.
- - -
“Severe traumatic brachial plexus lesions with root avulsion leads to motor and sensory function loss of the arm”. Fortunately, “nerve transfer can be performed to regain function” by rerouting nerves. At first, post-surgery patients can contract their biceps via respiratory effort only. Interestingly, with time, they may regain voluntary control of the muscle.
- - -
In the human brain, as the paper exemplifies, specific areas are correlated to specific tasks, such as motor control. Nevertheless, other functions — especially higher-order cognitive processes — involve wider networks and sometimes more indirect communication pathways. So, as we learn from this paper, scientists frequently keep an eye — and an fMRI — on the dialogue between neurons, populations, and even complex dynamics between whole areas.

The voxel abstraction, as we explained in a previous post, captures a bunch of neurons working together. As Fraiman and colleagues question, it may be possible that brain regions have their ability to work together rewired due to contextual needs — after a surgery, for instance. It’s plausible, then, that neighboring voxels have their interactions influenced by post-injury accommodation.

- - -
Some black-boxes may remain invisible to those looking exclusively for discrete, single voxels driving processes alone — which is not the case here. Even though controlled experiments most often demand isolating objects and phenomena, processes may rely on interactions rather than in single, specific cells.

So, when it comes to neuroscience, answers often lie not just in single pieces, but in how the whole puzzle comes together.
- - -

References:

1. Fraiman et al., 2016. Reduced functional connectivity within the primary motor cortex of patients with brachial plexus injury. Neuroimage: Clinical, 12, 277-284. https://doi.org/10.1016/j.nicl.2016.07.008

2. Buonomano, D.V., Merzenich, M.M., 1998. Cortical plasticity: from synapses to maps. Annu. Rev. Neurosci. 21, 149–186.

3. Kaas, J.H., 1991. Plasticity of sensory and motor maps in adult mammals. Annu. Rev. Neurosci. 14, 137–167.

4. Garraghty, P.E., Kaas, J.H., 1992. Dynamic features of sensory and motor maps. Curr. Opin. Neurobiol. 2 (4), 522–527.