Multitasking and the Brain: The Science of Task Switching

A switch can take less than a second and still leave a fingerprint. The old task does not vanish when the new one arrives; for a moment, both are competing backstage.

You stop writing, answer a message, and return to the document. The movement looks almost trivial. Yet the brain has had to preserve an unfinished goal, interpret a new demand, activate a different rule, choose a response, and then reconstruct where the first task was going. The visible interruption may be brief. The internal changeover is not.

This is the core of task switching: the brain is not merely moving attention from one object to another. It is changing the set of instructions that determines what matters and what response should follow. That reconfiguration is one reason rapid multitasking can feel smooth while producing slower work, more errors, or a strangely expensive return to the original task.

The same distinction matters across Cognitive Train’s brain training tools. Reaction speed, working memory, inhibition, and cognitive flexibility may cooperate during one ordinary interruption, but they are not the same ability.

The Brain Carries a “Task Set”

Before you can perform a task, the brain needs an active working model of it. Researchers often call this a task set. It may contain the current goal, the feature to watch, the rule that converts a stimulus into a response, and the information that must stay available until the task is finished.

Imagine sorting cards first by colour and then by shape. The cards themselves do not change. What changes is the instruction governing them. Under one task set, a blue triangle belongs with blue items. Under the other, the same card belongs with triangles. A switch therefore asks the brain to alter the meaning of an otherwise familiar stimulus.

There Is No Single “Multitasking Centre”

Task switching does not belong to one isolated patch of brain tissue. It recruits a distributed control system, especially frontal and parietal regions that help maintain goals, represent rules, direct attention, and select actions.

A meta-analysis of brain-imaging studies by Kim and colleagues found a broad frontoparietal switching network. Some areas, including the inferior frontal junction and posterior parietal cortex, appeared across different kinds of switches, while other regions were more involved depending on whether the change concerned perception, context, or the required response.

That pattern is important. The brain does not appear to press a universal “switch” button. It reorganises a network according to what is changing. Moving from reading to calculating is not identical to changing which hand should respond, even though both can create a measurable switch cost.

Earlier imaging work also showed that different control problems can recruit partly distinct machinery. In an event-related fMRI study by Dreher and Berman, lateral prefrontal activity was especially sensitive when people switched back to a task performed recently, whereas the anterior cingulate was recruited when a new task sequence began. The study did not reveal a tidy command centre; it separated pieces of control that everyday language tends to bundle together.

The Previous Task Leaves Residue

One reason switching is costly is that the old task set can remain active after it has stopped being useful. This is often described as task-set inertia. The previous rule has momentum, so the new task must be established while irrelevant settings are still exerting influence.

That lingering activity is more than a metaphor. Evans and colleagues used event-related brain potentials while participants alternated between memory and perceptual judgments. On the first trial after a switch, neural activity associated with the now-irrelevant memory process was still detectable, and stronger residual activity was related to a larger behavioural switch cost.

Residual interference is also why two familiar tasks can produce different costs depending on their order. A strongly activated rule may be harder to leave. A recently suppressed task may be slower to retrieve. The path between tasks matters, not only the tasks themselves.

Preparation Helps—but the Stimulus Still Has a Say

When a switch is predictable, the brain can begin preparing before the next item appears. It can orient toward the new rule and reduce some of the coming delay. Preparation is useful, but it rarely completes the whole transition in advance.

Using high-density electrical recordings, Wylie, Javitt, and Foxe traced the timing of an anticipated task switch. They found early switching-related activity over posterior parietal areas, followed later by frontal differences. Related work from the same group found that some elements of switching occurred only after the new stimulus arrived. In other words, the brain can set the stage, but part of the changeover depends on seeing what it must act upon.

This is why an advance warning can make an interruption less disruptive without making it free. “In five minutes, move to the next section” is easier than an unexpected message that demands an immediate answer. Yet even the expected switch still requires the new rule to meet a real stimulus and produce the correct action.

Watch the Switch Cost Appear

The test below makes the task set visible. A cue tells you whether colour or shape is currently relevant. On repeat trials, the same rule remains active. On switch trials, the stimulus may look familiar, but the rule used to judge it changes. The results compare those two conditions.

Your switch cost is the difference between performance on repeat and switch trials within this particular task. It is not a permanent measure of how “flexible” your brain is. Familiarity with the rules, speed–accuracy trade-offs, fatigue, the device used, and the number of valid trials can all affect the result.

It is most useful as a concrete demonstration: changing the instruction adds work even when the visual stimulus is simple. For a broader profile, Cognitive Train’s four-part Multitasking Test compares switching with dual-task cost, visual-audio divided attention, and parallel monitoring rather than treating them as one skill.

Why Everyday Switching Feels More Expensive

Laboratory switches are deliberately clean. The rules are short, the possible responses are known, and the abandoned task does not contain a half-written argument or an emotionally charged message. Real work is messier.

A meaningful task has context: what has already been decided, which possibility was rejected, what the next sentence was meant to accomplish. Working memory holds only part of that structure. When attention moves elsewhere, some of it must be stored externally, reconstructed later, or lost. That restart burden is one reason frequent interruptions can cost more than the split-second switch measured in a simple experiment.

The wider collection of cognitive and brain tests separates several ingredients involved in that return—working memory, processing speed, inhibition, and attention. A difficult restart does not automatically mean slow switching alone.

The Brain Switches by Rebuilding Control

Task switching is not attention teleporting from one place to another. It is a brief act of cognitive reconstruction. The brain must decide which rule is active, keep the new goal available, limit interference from the old goal, and connect the current stimulus with the response that now counts.

That is why the cost can survive preparation, why the order of tasks matters, and why the previous activity sometimes seems to follow you into the next one. The change is fast enough to feel seamless but complicated enough to leave evidence in reaction time, accuracy, and neural activity.

Continue with what the “multitasking is a myth” claim gets right and wrong, or explore the full collection of multitasking tests and training tools to compare switching with other forms of competing attention.