Neurotech notes

fMRI for BCI Task Design: From Voxels to BOLD Models

A technical note for connecting fMRI measurements to experimental design: start with voxels, follow blood oxygenation, model the hemodynamic delay, and only then construct BCI training tuples.

Aman BhandulaMay 26, 2026BCI, fMRI, task design
Figure 0From task to signal to model
Experimental pipeline from stimulus timing to fMRI model fittingfMRI task analysis pipelineThe measured signal is delayed relative to the neural event, so the task design must be modeled before voxel activity is interpreted.ATask timingtimestimulusBNeural activityactive regionCHemodynamic responseeventpeakDVoxel model fitmeasuredpredictedtask design + hemodynamic response = predicted fMRI regressor
fMRI analysis links an experimental timeline to delayed blood-oxygen-dependent measurements, then tests whether voxel time series match the predicted response.

A common mistake is to treat fMRI as a live camera pointed at neurons. It is not. It is a slow, spatially detailed measurement of how blood oxygenation changes around active neural tissue. That distinction matters for every downstream decision in BCI task design.

The useful sequence is simple: structural MRI gives anatomy; fMRI adds time by repeatedly sampling brain volumes; neural activity changes local blood flow; blood oxygenation changes the MRI signal; and the task designer has to align those measurements with what the participant was seeing, hearing, or doing.

This is a simplified working model, not a clinical explanation and not a claim that fMRI directly reads thoughts.

1. First, a brain image is a stack of voxels

In normal images, we talk about pixels. In MRI, the practical unit is a voxel: a tiny 3D cube. A practical approximation is a cube around 3 x 3 x 3 mm, though actual acquisition depends on the scanner and protocol. Each voxel stores a signal value. Put enough voxels together and you get a 3D brain volume.

Figure 1Voxels and repeated volumes
Voxel grid and repeated fMRI volumesVoxel sampling in fMRIEach acquired 3D volume is decomposed into voxels; each voxel contributes one value to a time series.A3D voxel gridhighlighted voxelBRepeated volumest0t0 + TRt0 + 2TRCVoxel time seriessampled signal over time
Structural MRI focuses on anatomy. fMRI repeatedly samples volumes over time, often on the order of every 2-4 seconds.

This is why functional MRI should not be treated as a static anatomical image. It is a time series of brain volumes. The scanner might capture one volume at t0, another at t0 + 2.4s, another at t0 + 4.8s, and so on. Every voxel gets a little signal trace across time.

The important move is to stop looking for a thought location and start asking which measured voxel signals vary with the task I designed.

2. The scanner does not see neurons directly

Neurons need energy. When a region becomes more active, the local vascular system sends more blood toward that region. The phrase for this is neurovascular coupling: neural activity and local blood delivery are coupled, not identical.

Figure 2Neurovascular coupling
Mechanism of neurovascular coupling and BOLD contrastNeurovascular couplingBOLD contrast is driven by vascular and oxygenation changes following local neural activity.ANeural activitymetabolic demand risesBLocal blood-flow responseoxygenated blood delivery increasesregional signal reflects vascular response, not direct spikesCMRI contrastBOLD signaldeoxy-Hb
Active neurons increase local metabolic demand. Nearby vessels respond by changing blood flow and oxygenation.

MRI is sensitive to magnetic properties. Deoxygenated hemoglobin distorts the local magnetic field more than oxygenated hemoglobin. When fresh oxygenated blood floods an active area, the measured local signal changes. This is the basis of the BOLD signal: blood-oxygen-level dependent contrast.

fMRI measures neural activity indirectly through blood oxygenation, not through the electrical activity itself.

3. BOLD is delayed, and the delay must be modeled

The hemodynamic response is slow. If a person taps a finger right now, the neural activity is immediate. The BOLD response rises later, peaks a few seconds after the event, and then returns toward baseline. This means an fMRI frame at time t is not a clean label for what the brain was doing at time t.

Figure 3The BOLD response curve
Canonical hemodynamic response functionCanonical BOLD responseNeural activity is fast; the blood-oxygen-dependent response rises and peaks several seconds later.04812162024seconds after neural eventrelative BOLD signalbrief neural inputpeak lag, about 4-6 sundershoot
A short task event produces a delayed BOLD response. The peak can arrive several seconds after the neural event.

The BOLD curve is the central modeling constraint. The brain event, the blood response, and the scanner sample are three different clocks. Good fMRI analysis starts by respecting those clocks.

4. Task design creates the signal you can later model

This is why simple task/rest designs are useful. If you ask someone to perform a task for one minute, rest for one minute, repeat the block, and sample the brain every few seconds, you get a clean timing structure. Then you can predict what the BOLD response should look like if a voxel is related to that task.

Figure 4Task blocks become an expected BOLD model
Task design convolved with hemodynamic responseFrom block design to model regressorA task sequence is convolved with an HRF kernel to create the predicted fMRI response used in voxel-wise modeling.ATask designtaskrestBHRF kernelconvolutionCPredicted BOLD regressorcompare predicted regressor against each measured voxel time series
The task timeline is convolved with a canonical hemodynamic response model, then compared against measured voxel time series.

The shorthand is to convolve the task design with a canonical hemodynamic response function. In plain English: take the task timeline, smear it by the expected hemodynamic delay, and compare that predicted curve with each measured voxel curve. Voxels that match the model are candidates for regions responding to the task.

  • Separate the stimulus timeline from the measured brain signal.
  • Remember that fMRI is spatially rich but temporally slow.
  • Treat the BOLD curve as a delayed proxy for neural activity, not the activity itself.
  • Use simple task/rest blocks before trying naturalistic tasks.
  • Store images, text, audio, labels, EEG, fMRI frames, and timestamps together.

5. Why this matters for our BCI work

The BCI angle is not to use fMRI as an undifferentiated model input. The useful angle is dataset design. If we want a model to learn from brain activity, we need aligned tuples: what was on the screen, what text or audio was present, what the person was instructed to do, what EEG was doing in real time, what fMRI volumes were sampled, and which delayed BOLD target belongs to which earlier task event.

Figure 5A BCI dataset is a timing object
Alignment of multimodal BCI training dataBCI data alignmentTraining examples are indexed by task time, not by raw scan time.AShared experimental clock0s4s8s12s16s20sstimuluslabelEEGfMRIBOLD targetimagetextaudiomotorlanguagerestfastTRhemodynamic lagBTraining tupletask inputbehavior labelEEG windowfMRI volumeBOLD target
A usable BCI training example needs a shared clock across stimuli, task labels, EEG, fMRI volumes, and delayed BOLD targets.

In that setup, EEG and fMRI are complementary. EEG gives fast timing and poor spatial certainty. fMRI gives rich spatial localization and slow timing. For task design, the job is to create situations where both streams can be useful without pretending they have the same resolution.

So the practical rule is: design the task first, timestamp everything, model the delay, and only then train. If the task is messy, the data tuple will be messy. If the tuple is messy, the BCI model is learning from a misaligned supervision signal.

The operational summary

fMRI is a repeated 3D measurement of brain-related blood oxygen changes. Neurons work first, blood responds after, the scanner samples periodically, and the analyst models the delay. For BCI, this turns into a design discipline: clean tasks, clean rest periods, clean timestamps, and explicit alignment between stimulus, action, EEG, fMRI, and labels.

VoxelNeurovascular couplingHemodynamic responseBOLD signalTask/rest designBCI alignment