The Multitasking Myth: What the Cognitive Science Actually Shows

The claim that multitasking is a myth has itself become something of a cliche. Productivity writers repeat it as though it settles the question. The research underneath is more specific and more interesting than the summary suggests, and understanding the mechanism tells you something useful about how to structure your work that the slogan alone does not.

What happens when you “multitask”

When you attempt to do two cognitively demanding things at once, such as writing an email while participating in a conference call, the brain does not process both simultaneously. It switches between them. Each switch involves two operations: disengaging from the current task (goal shifting) and reconfiguring cognitive resources for the new one (rule activation) [1]. Both operations take time, and both introduce errors.

Goal shifting is the process of updating your internal representation of what you are trying to do. When you switch from writing an email to listening to a speaker, you have to suppress the writing goal and activate the listening goal. Rule activation is the reconfiguration of the cognitive rules governing each task: writing uses grammatical rules, word selection, and sentence construction, while following a conversation uses parsing, context tracking, and inference. Loading a new set of rules takes time, and the old rules do not fully deactivate, which is why you sometimes type a word someone just said or find yourself composing a sentence in your head while trying to listen.

Rubinstein and colleagues measured these costs directly and found that switching between tasks increased completion time by 25 to 40 percent compared to doing the same tasks sequentially [2]. The cost increased with task complexity: switching between two simple arithmetic problems was fast, while switching between two tasks requiring different rules was substantially slower. The rule-activation component was the primary driver of the cost, which means that tasks with more complex or distinct rule sets (writing and coding, for instance) produce larger switching penalties than tasks with overlapping rules (two similar analysis tasks).

This is not a controversial finding. The psychological refractory period, the delay that occurs when a second stimulus arrives while the brain is still processing the first, has been documented across hundreds of experiments [5]. Pashler’s review established that the brain has a serial bottleneck in central executive processing that cannot be trained away. The bottleneck is structural, not a matter of insufficient practice, and it appears across all populations tested.

What the research shows

The cost scales with similarity

Tasks that use similar cognitive resources interfere more than tasks that use different ones. Writing an email while listening to a podcast (both language-heavy) creates more interference than writing while listening to instrumental music. Reading a report while someone speaks to you is particularly costly because both tasks require the language-processing system [1].

This explains why some “multitasking” feels easy (walking while talking, folding laundry while listening to a podcast) while other combinations feel impossible (writing while listening to a conversation). The difference is whether the tasks compete for the same processing resources. The brain has multiple processing channels, including separate systems for visual and auditory input, for language and spatial reasoning, and for motor control and cognition. Tasks that draw on different channels can proceed in partial parallel; tasks that draw on the same channel create a bottleneck.

The practical implication is that not all multitasking is equally costly. Listening to instrumental music while coding uses different channels (auditory processing for music, language and logic for code) and produces minimal interference for most people. Listening to a podcast while coding uses the same language-processing channel for both, and the interference is substantial. Knowing which channel each task uses helps you predict which combinations will work and which will not.

Heavy multitaskers perform worse, not better

Ophir and colleagues compared self-described heavy media multitaskers (people who routinely use multiple media simultaneously) with light multitaskers on a battery of cognitive tasks [3]. The heavy multitaskers performed worse on task-switching tests, not better. They were more susceptible to distraction from irrelevant stimuli and less effective at filtering information. On a test that required them to ignore irrelevant visual elements, the heavy multitaskers were significantly slower and less accurate.

This is a correlation, not a proven causal relationship: it could be that people who are more distractible gravitate toward heavy multitasking rather than that multitasking makes them more distractible. But the finding challenges the intuition that practice at multitasking improves the ability. If anything, the evidence suggests that habitual multitasking may train the brain to attend to everything rather than to filter effectively, which is the opposite of what focused work requires.

The study also challenges the self-assessment of multitaskers. The heavy multitaskers in the sample believed they were good at managing multiple inputs simultaneously. Their objective performance told a different story. This gap between subjective confidence and measured ability is a recurring theme in the attention literature: people are generally poor judges of their own attentional capacity.

Real-world interruption costs

In the workplace, the cost of task-switching manifests as interruption recovery time. Mark and colleagues observed information workers and found that interrupted tasks took substantially longer to return to, with an average of 23 minutes to re-engage at the same level of concentration [4]. The interrupted workers also reported higher stress and frustration, even when they completed the task successfully.

The 23-minute figure includes the full reorientation process: remembering where you were, rereading context, re-establishing the mental model. For complex tasks, this recovery is the hidden cost of every notification, email check, and “quick question.” The cost is not symmetrical: a two-second glance at a notification can cost minutes of recovery, because the glance is enough to shift the goal-state representation even if you do not act on the notification.

Mark and colleagues also found that interrupted workers attempted to compensate by working faster after the interruption. This compensation came at a cost: higher error rates and higher stress. The work got done, but it was done worse, and the worker felt worse doing it. This pattern, speed-accuracy tradeoff under interruption pressure, is consistent with laboratory findings on task-switching and suggests that the workplace cost of interruptions is underestimated by completion-time measures alone.

When multitasking isn’t the problem

True parallel processing does work when one task is automatic. Walking, eating, and other well-practiced motor activities can proceed alongside cognitive work without significant interference, because they do not require central executive resources. The boundary is clear: when a motor task becomes cognitively demanding (walking on an unfamiliar trail, eating while following a complex recipe), the interference reappears.

Some work legitimately requires rapid switching between inputs: a help desk, a newsroom, an emergency department. In these contexts, the switching cost is the price of the job, and the goal is to minimize it through environmental design (clear triage, reduced unnecessary interruptions) rather than to eliminate switching entirely. These roles are designed around the switching, with systems, checklists, and protocols that reduce the cognitive cost of each transition.

The “multitasking myth” framing can also overstate the problem. Checking email briefly between work sessions is not the same as trying to write and listen to a conversation simultaneously. The cost of switching exists on a spectrum, and brief, deliberate switches between tasks during natural break points carry much less cost than continuous interleaving. The key variable is whether the switch is self-initiated at a natural stopping point or externally imposed in the middle of a cognitive operation. The former involves a planned transition with a saved context; the latter involves an involuntary interruption that displaces the current mental state.

Practical considerations

The actionable implication is structural, not motivational. Telling yourself to stop multitasking is about as useful as telling yourself to stop being hungry. The question is what changes reduce the number of involuntary switches.

Turn off notifications during focused work. Each notification is an interruption, and each interruption carries a switching cost whether or not you act on it [4]. The ping itself is enough to shift attention, and the cost applies even when you decide not to respond. The most effective notification strategy is not “respond only to important ones” but “receive none during focused blocks.”

Batch similar tasks. If you must switch, switching between similar tasks (two writing tasks, two analysis tasks) is less costly than switching between dissimilar ones, because the reconfiguration step is smaller [2]. Grouping email responses, administrative tasks, or review work into dedicated blocks reduces the total switching cost compared to interleaving them with deep work throughout the day.

Protect transition points. When you do need to switch tasks, take ten seconds to note where you are in the current task and what the next step would be. This external save-state reduces the recovery cost when you return, because the reorientation process begins with reading your note rather than reconstructing your mental state from scratch.

Accept the serial bottleneck. Your brain does one demanding thing at a time. Designing your work around this constraint, rather than trying to override it, is the practical takeaway from several decades of cognitive research. The goal is not to become better at multitasking but to reduce the number of times your work forces you to do it.

If difficulty concentrating is new, persistent, or significantly affecting your life, it’s worth speaking with a doctor. Changes in focus can sometimes reflect sleep problems, mood disorders, or health conditions that are worth addressing directly. This site covers technique, not diagnosis.