Clean Signal, Real Body, Trustworthy Collective

Carbon wire loops, EEG inside and outside MRI, and the methodological foundation of a serious science of collectives

There are moments when we become so fascinated with the idea of the collective that we risk forgetting something basic: without robust method, relational neuroscience turns into loose metaphor. We speak about synchrony, shared presence, coordination between brains and bodies, but if the signal is dirty, if artifacts are poorly controlled, and if the design does not allow true comparison across contexts, what looks like discovery may be nothing more than well-narrated noise.

That is exactly why this blog matters. It reminds us that a “we” also needs engineering, calibration, and critique.

Tsutsumi and colleagues provide a very strong methodological anchor here. They released a dataset with EEG and fMRI from 39 healthy adults, collected during rest, visual oddball, and N-back tasks, both inside and outside the MRI scanner. Part of the sample was also recorded in two different scanners, which makes traveling-subject analyses possible. For us, the key point is that the study does not treat method as a secondary detail. It starts from the assumption that, if we want to speak seriously about brains in relation to context, we need data that support real comparison rather than mere fascination with simultaneity. (Nature)

That becomes even more important because the paper deals directly with one of the biggest problems in simultaneous EEG-fMRI: noise. The MRI environment introduces gradient artifacts, body- and head-motion contamination, and the ballistocardiogram, which is especially difficult because its amplitude and spectral profile can overlap with important EEG bands. Tsutsumi and colleagues explain that carbon wire loops were used to independently measure motion-related and ballistocardiogram artifacts so they could later be removed, improving signal quality. In other words, before we interpret relation, we need to measure what is brain and what is machine, what is coordination and what is contamination. (Nature)

This is where the text meets the I-mode / We-mode problem directly. If we want to distinguish instrumental coordination from shared agency, measuring performance alone is not enough. It is not enough to know that two participants pressed keys with a small temporal difference. In the task structure we are building, We-mode means both participants win or lose together, while I-mode preserves interdependence but individualizes reward, creating coordination without genuine “we-intentionality.” Paulo also points to something decisive: even in We-mode, residual egocentric strategies may remain, and those should show up in self-reports and interbrain-coupling measures.

That changes everything. It forces us to abandon a very common naivety: the idea that declaring a shared goal is enough to make the collective real. It is not. We can give a joint instruction and still capture only disguised competition. We can generate an appearance of coordination and still be measuring tense bodies, artifact-dominated brains, and reports that reveal a persistent private logic. The collective, then, cannot be romanticized. It has to be tested under conditions where signal, body, and context are truly comparable.

That is why this blog proposes a methodological ethics for a decolonial neuroscience of collectives. We do not want to deny the importance of interbrain synchrony, cooperation, or physiological coordination. We want to strengthen them. And here, strengthening means measuring better. It means accepting that a “we” never appears in a pure state. It appears mixed with instrumental noise, physiological noise, defensive strategies, respiratory tension, competitive vigilance, differences between recording environments, and the limits of the experimental design itself. A trustworthy collective, in science, is not born from enthusiasm. It is born from comparability.

In that sense, the Tsutsumi paper offers more than a dataset. It offers a stance. By combining EEG and fMRI in cognitive tasks and resting state, inside and outside MRI, in a public and machine-readable format, the study creates a basis for other groups to compare, replicate, and test denoising and multimodal-integration pipelines. That matters a great deal for us, because relational neuroscience still runs the risk of producing beautiful images and too little comparable evidence. What this work suggests is the opposite: the beauty of the collective has to pass through the rigor of the signal first. (Nature)

From a BrainLatam2026 perspective, this speaks directly to the Damasian Mind. If mind is a living body in situation, then measuring the body badly means thinking the mind badly. If consciousness emerges through the articulation of interoception, proprioception, and situated action, then data quality is not a peripheral technical issue. It belongs to the very ontology of what we are studying. The same holds for Triple-Aspect Monism: if the physiological, the experiential, and the informational-social cannot really be separated into clean isolated layers, then method has to become even more rigorous. The more complex the articulation between body, brain, and bond, the more careful we must be not to call artifact “relation.”

Read through Jiwasa, the question becomes both beautiful and demanding: how do we build a scientific “we” that is not seduced by its own narrative? A “we” that recognizes that the common itself can become a methodological illusion if the design is not strong enough. Producing shared agency with the reader here does not mean asking the reader to believe with us. It means asking the reader to measure with us, to doubt with us, to clean the signal with us. Science enters We-mode when we share not only enthusiasm, but criterion.

That has practical consequences for the I-mode / We-mode experiment we are shaping. Beyond temporal performance and interbrain measures, it becomes increasingly clear that autonomic and muscular signals should also be followed, along with serious artifact tracking. Respiration, ECG with HRV/RMSSD, EMG from masseter and trapezius, and perhaps GSR can help show whether the observed coordination came with bodily stabilization or with tense vigilance. But that only gains value if the methodological design remains comparable across contexts and if denoising pipelines are transparent. The Tsutsumi paper does not solve the “we” problem by itself, but it helps build the ground without which that problem remains suspended in abstraction. (Nature)

In the end, this blog wants us to feel something simple: there is no serious neuroscience of collectives without a radical commitment to signal quality. The collective is too important to be treated loosely. If we want to distinguish instrumental coordination from shared agency, it is not enough to measure that two people matched their timing. We need to know what, in that record, was brain; what was body; what was noise; what was artifact; and what, perhaps, had really begun to become relation.

Maybe that is exactly where a decolonial neuroscience of collectives starts to mature: when we understand that the “we” cannot only be celebrated. It has to be earned by the method.

Reference

Tsutsumi, M., Kishi, T., Ogawa, T., Kuroda, T., Kobler, R. J., & Kawanabe, M. (2026). An EEG dataset with carbon wire loops in cognitive tasks and resting state inside and outside MR scanners. Scientific Data, 13(1), 351. doi:10.1038/s41597-026-06734-1. (Nature)