Functional magnetic resonance imaging (fMRI) is one of the most powerful tools in modern neuroscience. It allows scientists to observe how different parts of the brain become active while a person performs tasks such as moving, speaking, or thinking. You have probably seen brain images covered with coloured patches that appear to highlight “active” areas. These images are widely used in research papers, news articles, and documentaries about the brain. But what do those coloured areas actually represent? Understanding this requires a basic grasp of how MRI works, how brain activity relates to blood flow, and how scientists analyse the data. In this article, I will break down the core ideas behind functional MRI in a clear and accessible way.
From MRI Scans to Functional Brain Imaging
How MRI creates images of the brain
Magnetic resonance imaging (MRI) uses powerful magnets and radiofrequency pulses to produce detailed pictures of structures inside the body. The technique relies on hydrogen atoms found in water molecules throughout the body.
When a person lies inside the scanner, the magnetic field causes many of these hydrogen atoms to align in the same direction. The scanner then sends radiofrequency pulses that briefly disturb this alignment.
As the atoms return to their original state, they emit small signals that the scanner detects. Computers convert these signals into detailed images of the brain.
Structural scans vs functional scans
MRI machines can operate in different modes depending on the type of data needed.
Structural MRI produces very detailed images of the brain’s anatomy. These scans are often used in medicine to detect tumours, bleeding, or other abnormalities. A single full-brain image can take several minutes to collect.
Functional MRI, on the other hand, captures images much more rapidly. Instead of one detailed snapshot, the scanner repeatedly measures the brain over time.
Typically, a full brain image is recorded every few seconds. This creates a sequence of images showing how signals in the brain change moment by moment.
Understanding voxels
MRI data is divided into tiny three-dimensional units called voxels. A voxel is similar to a pixel in a digital photograph, but it represents volume rather than a flat point. Each voxel contains a single numerical signal value measured by the scanner. In many studies, a voxel might represent a cube around three millimetres wide on each side. Even within this small space, there are millions of neurons. Because of this, fMRI does not measure individual neurons. Instead, it captures the combined activity of very large populations of cells.
How Brain Activity Changes Blood Flow
Neurons and energy demand
The brain is an energy-hungry organ. Neurons constantly require oxygen and nutrients to send electrical signals and communicate with one another.
When a group of neurons becomes more active, their demand for oxygen increases. The body responds by sending more oxygen-rich blood to that region.
Neurovascular coupling
This link between neural activity and blood flow is known as neurovascular coupling.
In simple terms, when neurons work harder, nearby blood vessels expand slightly. This allows more blood to reach the area and supply the extra oxygen needed.
This relationship is crucial because fMRI does not measure electrical signals directly. Instead, it detects changes related to blood flow and oxygen levels.
The haemodynamic response
The increase in blood supply following neural activity is called the haemodynamic response.
This response does not occur instantly. Instead, it unfolds gradually over several seconds. First, blood flow rises as vessels widen. Then the response slowly returns to its baseline level once the increased demand has passed.
Because of this delay, the signals measured by fMRI reflect brain activity indirectly rather than in real time.
The BOLD signal
Functional MRI relies on something called the BOLD signal, which stands for Blood Oxygen Level Dependent signal.
The key idea is that oxygenated and deoxygenated blood behave slightly differently in a magnetic field. Oxygen-rich blood disturbs the magnetic field less than oxygen-poor blood.
When neural activity increases, extra oxygenated blood flows to the area. This changes the magnetic properties detected by the scanner.
The fMRI system measures these subtle changes and records them as variations in signal intensity across the brain.
Identification of Active Brain Regions
Designing tasks for participants
To understand which brain areas support a particular behaviour, researchers ask participants to perform carefully designed tasks inside the scanner.
These tasks can be simple or complex. Examples include:
-
tapping fingers
-
viewing images
-
listening to sounds
-
solving mathematical problems
-
remembering words
The experiment alternates between periods of activity and periods of rest. This creates a predictable pattern of when brain regions should become more active.
Building a model of expected activity
Scientists then build a mathematical model representing when brain activity should increase during the task.
Because the haemodynamic response is delayed, the model also includes the expected rise and fall of the blood-flow signal following neural activity.
This predicted signal acts as a reference pattern for analysis.
Comparing the model to real data
During the experiment, the scanner collects brain images every few seconds for many minutes. Each voxel therefore has a signal that changes across time. Researchers compare the time-series signal of every voxel to the predicted model. If the two patterns match closely, it suggests that region of the brain is involved in the task. This comparison is performed using statistical methods commonly known as a general linear model (GLM).
Creating the famous coloured maps
Finally, voxels that strongly match the predicted pattern are highlighted on the brain image. Different colours often represent different levels of statistical strength. Areas with the strongest evidence appear as bright clusters. These clusters are the coloured patches people often see in neuroscience images. They indicate regions where the signal closely followed the expected pattern during the task. In other words, the coloured regions mark brain areas that were likely engaged while the participant performed the activity.
Conclusion
Functional MRI provides a remarkable window into how the human brain works. Rather than measuring electrical signals directly, it tracks changes in blood oxygen levels linked to neural activity. The process begins with rapid MRI scans collected over time. Each scan records signals from thousands of tiny three-dimensional voxels across the brain. When neurons in a region become active, increased blood flow changes the magnetic signal detected by the scanner. These changes produce the BOLD signal that forms the basis of fMRI analysis. Researchers then use carefully designed tasks and statistical models to identify which brain regions respond during specific behaviours. The familiar coloured clusters on brain images are simply voxels whose signals closely match the predicted activity pattern.
Although the underlying physics and mathematics can be complex, the core principle is surprisingly intuitive: when neurons work harder, they require more oxygen, and that change in blood flow can be detected by MRI.
Thanks for reading and please comment on YouTube with any questions! Click here to see the other blogs and associated videos I have about neuroscience! 🙂
If you want to support this evidence based neuro-explainer content then feel free to buy me a coffee or become a patreon! 😊
