How fMRI Scans Work: From Brain Activity to Applications

Functional magnetic resonance imaging, or fMRI, is a specialized non-invasive technique that allows scientists and clinicians to see which parts of the brain are active during specific tasks. Unlike structural MRI, fMRI provides functional information by creating a dynamic map of neural activity. This process tracks the metabolic changes that accompany neural firing, relying on monitoring minute changes in blood flow as a proxy for localized brain function.

How fMRI Measures Brain Activity

The ability of fMRI to detect brain activity rests on a phenomenon known as the Blood-Oxygen-Level Dependent (BOLD) contrast. Neural activity requires energy, and when a population of neurons begins firing, it consumes oxygen and glucose delivered by local blood vessels. This consumption leads to a temporary, slight increase in the concentration of deoxygenated hemoglobin in that specific brain region.

The body overcompensates for this localized oxygen usage by rapidly increasing the flow of oxygenated blood to the active area. This influx of fresh blood is significantly greater than what the active neurons consume, resulting in a surplus of oxygen-rich blood. This hemodynamic response, the change in blood flow following neural activity, peaks about two to six seconds after the neurons first fire.

The key to the BOLD signal is the different magnetic properties of hemoglobin depending on its oxygenation status. Oxygenated hemoglobin is diamagnetic, meaning it has little effect on the magnetic field generated by the MRI scanner. Deoxygenated hemoglobin, however, is paramagnetic due to the presence of iron, which makes it act like a tiny contrast agent.

When the concentration of deoxygenated blood is high, the paramagnetic nature of the iron distorts the scanner’s magnetic field, causing the MR signal to drop slightly. Conversely, the arrival of the oxygenated blood surplus “washes away” this deoxygenated hemoglobin, reducing the magnetic field distortion and causing the MR signal to increase. The fMRI scanner detects this localized signal increase and maps it to the corresponding brain location.

The Subject Experience of an fMRI Scan

The fMRI procedure takes place within the same large, tube-shaped machine used for conventional MRI scans. The participant lies on a narrow table that slides into the bore of the magnet, and they are asked to remain as still as possible throughout the entire session. Even slight head movements, which can be less than a millimeter, can introduce artifacts that corrupt the subtle BOLD signal data.

One of the most noticeable aspects of the scan is the extremely loud noise produced by the gradient coils during image acquisition. This sound requires that earplugs and noise-canceling headphones are always provided to attenuate the sound to a safer level. The headphones also allow the technologist to communicate instructions and task cues to the participant.

During the scan, the participant typically performs specific tasks designed to activate particular brain regions. Stimuli can be delivered through specialized equipment, such as visual displays projected onto a screen inside the bore or auditory cues delivered through the headphones. Sessions can last from 30 to 90 minutes, depending on the number and complexity of the tasks being performed.

Key Uses in Research and Clinical Settings

The ability of fMRI to map functional activity makes it valuable in both neuroscience research and clinical medicine. In research settings, fMRI is used extensively to map the neural correlates of complex cognitive processes. Researchers use the technique to identify the brain networks involved in memory formation, emotional regulation, decision-making, and language comprehension.

Researchers can pinpoint areas that show increased blood flow, such as the amygdala for emotion or the prefrontal cortex for planning, by having participants view emotionally charged images or solve complex puzzles while in the scanner. Another approach, resting-state fMRI, analyzes spontaneous fluctuations in the BOLD signal to reveal how different brain regions are functionally connected even when the person is not performing a specific task. This helps to characterize the brain’s default organizational networks.

In clinical practice, the most common application of fMRI is in pre-surgical planning for patients with brain tumors or epilepsy. Neurosurgeons use fMRI to precisely locate “eloquent” cortex areas responsible for motor function or language before an operation. This preoperative mapping helps the surgical team maximize the removal of diseased tissue while minimizing the risk of damaging functional areas, which can preserve the patient’s abilities.

fMRI data must be interpreted with an understanding of its inherent limitations. Because the technique measures the slow hemodynamic response, not the rapid electrical firing of neurons, it has a low temporal resolution and cannot measure neural activity directly. The technique shows a strong correlation between a task and a location of increased blood flow, but it does not establish causation or explain how the underlying neural circuitry interacts.