Sandhi Interface · Independent EEG Research · Puebla, México

The brain acts. You find out after. Half a second before the impulse — what's there?

Studying the Readiness Potential as a neural marker of pre-impulse states — where neuroscience meets clinical relevance.


Why this matters

The gap in clinical intervention

Impulse control disorders — eating disorders, anxiety, addictions, compulsive behavior, self-harm — share a common mechanism. The brain prepares an action before conscious awareness of intent. Clinical intervention almost always arrives too late.

1 in 4

People worldwide will experience an impulse-related disorder in their lifetime

Kessler et al., NCS-R 2005

1–2 s

The neural window before conscious intent — the Readiness Potential window we study

Kornhuber & Deecke, 1965

7–10 s

Predictive decoding of decisions before conscious awareness — via fMRI

Haynes et al., 2008

Scientific background

The neuroscience of the pre-impulse window

Over sixty years of research show that voluntary actions are preceded by distinct neural signatures — measurable before the person is consciously aware of their intention to act.

Impulsive states manifest as detectable patterns of motor preparation in the brain — measurable via EEG — that differ significantly from controlled or inhibited responses. This "pre-impulse window" may eventually enable early intervention.

1965

The Bereitschaftspotential

Kornhuber & Deecke describe a slow negative cortical potential beginning up to 2 seconds before self-initiated voluntary movement — originating in the supplementary motor cortex. The canonical neural signature of preparation for action.

1983

Libet and conscious intent

Libet et al. demonstrate that the readiness potential precedes the felt "urge to move" by several hundred milliseconds — revealing that the brain initiates actions before conscious awareness of the intention. This finding opened the neuroscience of volition.

2008

Predictive decoding

Haynes et al. demonstrate via fMRI that the outcome of a freely chosen decision can be decoded from brain activity 7–10 seconds before the subject is consciously aware of their choice — extending the pre-conscious window far beyond what EEG studies had shown.

now

Sandhi Interface

Can this pre-impulse window be characterized using low-cost, accessible EEG hardware — and does it look different in the seconds before an unwanted compulsive action? That is our question.


Experimental protocol

How we work

Designed to be reproducible, transparent, and low-cost. Every step is documented in the open research log.

FPz ref AF7 AF8 TP9 TP10
Current hardware

Muse 2

4-channel wireless EEG (AF7, AF8, TP9, TP10) · 256 Hz sampling rate · Bluetooth LE. Data streamed via Lab Streaming Layer (LSL) to LabRecorder for timestamped acquisition. Electrode coverage spans anterior frontal and temporal-parietal regions — sufficient for readiness potential detection and synchronization validation.

Next hardware: Unicorn Hybrid Black (g.tec) — arriving June 2026. 8 active dry EEG channels, purpose-built for RP research. Funded in part by our €1,000 award at g.tec BR4IN.IO Spring School 2026.

Muse 2 LSL MNE-Python 256 Hz Bluetooth LE Dry electrodes
Experimental phases
I

Spontaneous RP characterization

Voluntary free-movement paradigm. Baseline acquisition and full pipeline validation. Confirming that the readiness potential is detectable with our hardware configuration before advancing to task conditions.

II

Go / No-Go inhibitory control

Comparison of pre-motor neural activity between executed and successfully inhibited responses. Mapping what "stopping an impulse" looks like at the cortical level.

III

Impulsivity EEG markers

Differentiation of neural patterns preceding controlled versus impulsive actions. The core clinical application phase — characterizing the pre-impulse window in ecologically valid conditions.

Signal analysis strategy
Step 1

Temporal segmentation

Pre-action windows extracted around event markers — isolating the 1–2 s window before movement onset for epoch-based analysis.

Step 2

Spectral power analysis

Band power across delta, theta, alpha, beta, and gamma ranges. Identifying frequency-domain signatures of preparation states.

Step 3

Slow cortical potential extraction

RP extraction via averaging across trials — isolating the slow negative deflection that precedes voluntary action from ongoing neural noise.

Step 4

Cross-condition comparison

Statistical comparison of pre-motor patterns across conditions — controlled vs. impulsive, executed vs. inhibited — to identify discriminative neural features.

Research milestones · June 2026

From proof of concept to validated infrastructure

Phase 01 is complete. Sandhi has moved from internal prototyping to a validated pipeline with external participants. Here is what we confirmed, refined, and documented openly.

Completado Fase 01 · Pilot Trial

End-to-end pipeline validated

Validated the full data acquisition chain: Muse 2 → LSL → LabRecorder. Temporal data integrity confirmed across all pipeline stages with external participants — the project's first milestone beyond solo prototyping.

Validado Logro técnico

Atomic sync: jitter < 10 ms

Confirmed sub-10 ms jitter between visual stimulus onset and EEG marker — meeting gold-standard thresholds for wireless BCIs. Stimulus-to-marker synchronization is production-ready for RP research.

Refined Refinamiento científico

The preparation biomarker

Following clinical advisory, we refined our working definition: the Readiness Potential (RP) and beta desynchronization are studied jointly as a precognitive motor preparation biomarker — mapping the neural transition from impulse to action.

Documentado Transparencia en la señal

Signal quality: real-world variables

Openly documented: hair density, eyeglass use, and cranial morphology are critical modulators of HSI signal quality in uncontrolled environments. These are now tracked variables in our acquisition protocol, not unacknowledged confounds.

Field documentation
Sandhi Interface setup at Expo Ingenierías
Beta participant during Fase 01 EEG session
Photo · coming soon

The people behind the research

Team

Portrait of Biniza Veronica Vazquez Moreno
Biniza Verónica Vázquez Moreno
Principal Investigator · Sandhi Interface

Robotics and Systems Engineering student at Tecnológico de Monterrey Campus Puebla. NASA and MIT-adjacent experience. Leading Sandhi Interface as an independent research initiative with the conviction that meaningful neuroscience doesn't require institutional scale — only rigor, curiosity, and an honest research log.

Premio Mujer Tec 2026 g.tec BR4IN.IO Spring School ITESM Puebla
Contributors
Linda Michelle Silva Ramos

Research Interface Specialist
Mechatronics Engineering · 8th semester
Physical & UI interface for RP measurement

Gianluca de la Rosa Bandini

Research Interface Specialist
Mechatronics Engineering · 8th semester
Physical & UI interface for RP measurement

Sofía Navarro Rebolledo

Human Factors Research Assistant
Future neuroscientist at UBC
Human factors & participant experience

Advisors
Dr. Ana Luisa Santaolalla

Harvard Neuroscience · Hult MBA
Scientific direction

Yolanda Fajardo

Clinical Psychologist
Founder, Serena Mente Puebla
Clinical framing


Get involved

Support & collaborate

This research is self-funded and independently conducted. If this question matters to you — as a researcher, a clinician, someone with lived experience, or simply someone who believes this science is worth doing — there are real ways to help.

Fund the next hardware

We won €1,000 at g.tec BR4IN.IO Spring School 2026 — a real start. Our crowdfunding goal is the Unicorn Hybrid Black, a g.tec headset purpose-built for RP research with active dry electrodes and higher channel density. Every contribution gets us closer.

Support on GoFundMe

Collaborate

We're looking for researchers in neuroscience and clinical psychology, engineers in signal processing and robotics, and anyone interested in replication or co-authorship.

Open an issue on GitHub
Literature

References

  1. Kornhuber, H.H., & Deecke, L. (1965). Hirnpotentialänderungen bei Willkürbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pflügers Archiv, 284, 1–17.
  2. Libet, B., Gleason, C.A., Wright, E.W., & Pearl, D.K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness potential): the unconscious initiation of a freely voluntary act. Brain, 106(3), 623–642.
  3. Haynes, J.D., Sakai, K., Rees, G., Gilbert, S., Frith, C., & Passingham, R.E. (2008). Reading hidden intentions in the human brain. Current Biology, 17(4), 323–328.
  4. Kessler, R.C., et al. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593–602.
  5. Uhlhaas, P.J., & Singer, W. (2006). Neural synchrony in brain disorders: relevance for cognitive dysfunctions and pathophysiology. Neuron, 52(1), 155–168.
  6. Gramfort, A., et al. (2013). MEG and EEG data analysis with MNE-Python. Frontiers in Neuroscience, 7, 267.