Stimulating While Listening - When the Brain Is No Longer Merely Observed

For a long time, an important part of neuroscience was organized around a relatively comfortable situation: first we observe the brain, and then we try to understand what that activity means.

We record EEG.

We present a stimulus.

We ask a person to imagine a movement.

We compare conditions.

We look for differences.

But what happens when we decide to do something much more difficult?

What if we stimulate the brain while trying to keep listening to what it is doing?

This is the methodological and conceptual question that makes the work of Mareike Vermehren, Niels Peekhaus, Annalisa Colucci, Marian Wiskow, David Haslacher, Gabriel Curio, and Surjo R. Soekadar particularly interesting. Published in 2026 in the IEEE Transactions on Neural Systems and Rehabilitation Engineering, their study, Robust Brain–Computer Interface (BCI) Control During Frequency-Tuned Transcranial Alternating Current Stimulation (tACS), confronted a problem that at first seems almost paradoxical: using an EEG-based brain–computer interface while an external electrical current is being applied to the brain precisely within the frequency range one wishes to investigate.

This is not only an engineering difficulty.

The question touches on a central issue in contemporary neuroscience:

how can we causally test the role of a brain dynamic without destroying, through the intervention itself, our ability to observe it?

A question that requires perturbing the system

A large part of what we find in EEG remains correlational.

We observe an oscillation.

We observe a behavior.

We find a relationship between the two.

But correlation does not necessarily mean that the oscillation is causally producing, modulating, or required for that behavior.

One way to go further is to deliberately interfere with neural dynamics.

Transcranial alternating current stimulation, or tACS, is particularly interesting because it allows oscillating electrical currents to be applied at specific frequencies. The broader hypothesis is that these currents can interact with endogenous neural oscillations and alter their dynamics. Recent reviews discuss the potential of tACS to modulate oscillations, excitability, plasticity, and functional networks, while also emphasizing important uncertainties regarding mechanism, dose, brain state, and clinical effects.

The problem appears immediately.

If we use EEG to track a brain oscillation while simultaneously applying an external oscillating current, the stimulation signal can overlap with the neural signal we want to measure.

It is like trying to hear someone speaking quietly while placing a loudspeaker next to the microphone.

The easiest solution would be to switch off the loudspeaker.

But then we would lose exactly what Vermehren and colleagues wanted to investigate:

the brain while it is being perturbed.

The elegance of the design lies in not interrupting the event

The researchers worked with 14 healthy participants and used a BCI based on motor imagery.

In motor imagery, a person imagines performing a movement without necessarily executing it physically. This process produces detectable changes in sensorimotor EEG rhythms, allowing a BCI to classify different states and translate that activity into commands.

The authors compared two signal-processing approaches.

The first used a Laplacian filter, a conventional method.

The second combined spatio-spectral decomposition — SSD — with beamforming, creating a real-time spatial pipeline specifically designed to better separate neural activity of interest from stimulation-related artifacts.

The BCI was tested both without stimulation and during amplitude-modulated, frequency-tuned transcranial alternating current stimulation — AM-tACS.

The hypothesis was clear:

without stimulation, both methods should perform similarly.

During stimulation, however, only the SSD-beamforming strategy should preserve robust BCI control.

This creates a particularly elegant experimental design because the new method is not simply tested in isolation.

It is placed precisely in the condition in which it must prove its value.

When there was no stimulation, little separated the methods

Without AM-tACS, the two pipelines performed similarly.

SSD-beamforming achieved approximately 73% accuracy, while the Laplacian filter achieved approximately 72%.

The difference was not significant.

This result is methodologically important.

If the new method had been superior under every condition, it would have been difficult to determine whether its main advantage truly lay in its capacity to cope with stimulation artifacts.

But the difference appeared precisely when the experimental environment changed.

During AM-tACS, BCI performance using Laplacian filtering fell to approximately 58%, close to chance level for that classification.

With SSD-beamforming, however, performance remained robust, at approximately 76%.

In other words:

the stimulation nearly destroyed the conventional pipeline’s ability to interpret the EEG correctly.

But the method developed by the authors preserved decoding.

The merit is not only in the 76%

It would be easy to turn this paper into a technological headline:

“New algorithm keeps a BCI working during brain stimulation.”

That is true.

But the scientifically more interesting aspect is different.

The authors created a situation in which it now becomes possible to stimulate and decode simultaneously.

That means we can begin asking different questions.

Instead of:

“does this oscillation appear when the participant performs a certain task?”

we can ask:

“what happens to the task if we interfere precisely with this oscillation while it is participating in the process?”

That is a major shift.

We move from a predominantly observational neuroscience toward a more causal experimental possibility.

And there is a third step.

If we can observe the activity, identify a state, and modify stimulation according to that state, we begin to approach bidirectional and adaptive systems.

The brain produces a signal.

The machine interprets that signal.

The machine intervenes.

The brain changes.

The machine reads again.

And the cycle continues.

This is why the authors themselves present the work as a step toward bidirectional brain–computer interfaces and more direct tests of the causal contribution of brain oscillations.

The brain we measure is already happening under some condition

This is where the article begins to speak to our concepts of Body-Territory and 5D Consciousness.

Not because the study investigated consciousness.

It did not.

And not because it demonstrated that a particular oscillation is “consciousness.”

It did not.

Its importance for our reflection lies elsewhere.

The study forces us to recognize something that often becomes invisible in the way experiments are described:

every brain we measure is already in a condition.

It is sitting or moving.

Waiting or responding.

Imagining a movement.

Receiving sensory input.

Being electrically stimulated.

Fatigued, attentive, anxious, familiarized, or surprised.

The recording does not capture a neutral entity called “the brain” and then add conditions afterward.

It captures a dynamic already unfolding in a situated organism.

In this experiment, that becomes impossible to ignore because the researchers deliberately change a material condition of the system while continuing to interpret it.

Intervening does not mean writing directly onto the brain

There is also an important caution.

The expression “brain stimulation” may create the impression that we select a frequency, apply a current, and then directly determine what the brain will do.

Reality is much more complex.

Recent tACS literature emphasizes that its effects depend on the existing neural state, applied frequency, individual anatomy, intensity, electric-field distribution, and plasticity mechanisms that may vary across participants and conditions.

This is particularly interesting for our concept of Body-Territory.

An intervention does not arrive in an empty organism.

It encounters a system that already has a dynamic.

Therefore, perhaps it is more appropriate to imagine stimulation not as a command entering the brain, but as a new condition that becomes part of the possibilities of that system.

The outcome depends on the encounter between perturbation and state.

Frequency is not identity

There is another conceptual trap.

When we find a certain oscillatory band associated with a behavior, we may be tempted to transform that frequency into a functional entity:

“this is the frequency of movement.”

“this is the frequency of attention.”

“this is the frequency of consciousness.”

The very logic of this work suggests caution.

Oscillations matter, but their meaning depends on location, network, phase, task, state, and interaction with other dynamics.

If a frequency can be stimulated and its causal contribution investigated, that does not mean that the frequency is the function being investigated.

It may participate in a configuration.

This distinction is central to 5D Consciousness.

The perception of Being does not need to be located in a single frequency, area, or isolated moment.

We can think of multiple processes participating simultaneously in the configuration of the Body-Territory, some gaining greater functional weight at a given moment, others remaining active without dominating the attentional channel.

Read, perturb, read again

Perhaps we can represent the contribution of this paper in three movements:

Read.

EEG detects a dynamic related to motor imagery.

Perturb.

AM-tACS alters the oscillatory conditions of the system.

Read again.

The algorithm continues trying to identify what emerges after — and during — that perturbation.

But there is something even more interesting:

the third moment never returns exactly to the first.

After the intervention, we are facing a new configuration.

This logic approaches what we have been discussing about consciousness as movement.

The organism does not need to be understood as a sequence of independent snapshots.

It may be understood as a trajectory.

One state participates in the next.

A perturbation changes possibilities.

A response changes what may happen afterward.

From brain–computer interface to Body-Territory interface?

BCIs emerged, to a large extent, from attempts to establish a communication pathway between brain activity and a machine.

But as these systems become more complex, perhaps the very name “brain–computer interface” begins to conceal part of the phenomenon.

A person imagines a movement.

That imagination depends on learning, intention, bodily experience, proprioception, memory, and context.

EEG records only part of that configuration.

An algorithm selects particular features.

The machine responds.

That response returns to the organism as sensory information.

The organism perceives the result.

That perception modifies strategy.

And then a new signal is produced.

We are no longer looking only at a brain sending commands to a computer.

We are looking at a circuit involving organism, technology, and environment.

A small artificial territory has been produced.

And that territory begins to participate in what the organism can do.

A technology that can also produce new questions

The clinical applications are evident.

BCIs have been studied extensively in neurorehabilitation, especially after stroke, and recent reviews discuss their potential for motor recovery and the induction of plasticity.

tACS has likewise been explored as a tool capable of modulating oscillations and potentially supporting plastic processes, although clinical use still requires caution, personalization, and stronger mechanistic evidence.

Combining these two technologies opens a fascinating possibility:

not only detecting the neural state associated with an attempted movement, but potentially adjusting the intervention according to the state being detected.

We still do not know how far this can go.

And technological enthusiasm should never replace evidence.

But Vermehren and colleagues solved a necessary condition for questions of this kind to be investigated far more rigorously.

That is a considerable achievement.

Perhaps observation was never only observation

In the previous text of this series, we encountered two people moving together through an experience of exclusion.

Now we encounter something different:

an organism, a machine, an algorithm, and an electrical current temporarily forming a new experimental system.

In both cases, the comfortable boundary of the isolated brain becomes less simple.

Vermehren and colleagues show that neural activity can continue to be decoded even while the conditions under which that activity is being produced are deliberately altered.

But their work allows an even larger question.

If what we call brain activity changes as we intervene in the system; if the effects of intervention depend on the prior state; and if the response itself produces new conditions for the next moment, then perhaps we should not ask only:

“What does this oscillation represent?”

We can ask:

“What can this oscillation do within this configuration?”

And perhaps, later:

“What new possibilities of Being emerge when a Body-Territory learns to perceive and transform its own dynamics through a technology that has itself become part of its territory?”

At that point, the interface is no longer merely a window into the brain.

It becomes part of the event.

References

Vermehren, M., Peekhaus, N., Colucci, A., Wiskow, M., Haslacher, D., Curio, G., & Soekadar, S. R. (2026). Robust Brain–Computer Interface (BCI) Control During Frequency-Tuned Transcranial Alternating Current Stimulation (tACS). IEEE Transactions on Neural Systems and Rehabilitation Engineering, 34, 3983–3992. https://doi.org/10.1109/TNSRE.2026.3732552

Agboada, D., Zhao, Z., & Wischnewski, M. (2025). Neuroplastic effects of transcranial alternating current stimulation (tACS): from mechanisms to clinical trials. Frontiers in Human Neuroscience, 19, 1548478. https://doi.org/10.3389/fnhum.2025.1548478

Li, D., Li, R., Song, Y., Qin, W., Sun, G., Liu, Y., Bao, Y., Liu, L., et al. (2025). Effects of brain-computer interface based training on post-stroke upper-limb rehabilitation: a meta-analysis. Journal of NeuroEngineering and Rehabilitation. https://doi.org/10.1186/s12984-025-01588-x

Zhang, Y., Gao, Y., Zhou, J., Zhang, Z., Feng, M., & Liu, Y. (2025). Advances in brain-computer interface controlled functional electrical stimulation for upper limb recovery after stroke. Brain Research Bulletin, 226, 111354. https://doi.org/10.1016/j.brainresbull.2025.111354

Application and research progress of different frequency tACS in stroke rehabilitation: A systematic review. (2025). Brain Research, 1852, 149521. https://doi.org/10.1016/j.brainres.2025.149521