Task Switching
Background
Task switching examines how people shift between different activities and the mental costs of doing so. In everyday life, we often say we are multitasking. Most of the time, we are not doing two demanding things at the same time. We are switching our attention back and forth very quickly.
Think about checking a message while writing an email, glancing at a navigation app while talking with a passenger, or toggling between a spreadsheet and a chat window at work. Each time you look away and come back, your mind has to reconfigure what to pay attention to and which rules to apply. That reconfiguration takes time and can cause mistakes.
The roots of this work trace back to Arthur Jersild (1927), who showed that people are slower when alternating between tasks than when repeating the same task. This insight suggested that switching requires extra mental operations beyond performing either task alone.
Since the mid-1990s, task switching has become a central paradigm for studying cognitive control. In a typical experiment, participants rapidly alternate between two simple tasks, for example, classifying numbers as odd or even, or letters as consonant or vowel. Researchers quantify two core performance costs:
- Switch cost: Slower and often less accurate performance on trials that switch tasks compared with immediately repeating the same task within mixed-task blocks.
- Mixing cost: Slower performance on task repeats in mixed blocks compared with pure blocks, which reflects the overhead of maintaining and selecting between multiple task sets.
A widely used design is the "alternating runs" paradigm (Rogers and Monsell, 1995), where participants perform two trials of Task A, then two trials of Task B (A A, B B, ...). This predictable pattern cleanly separates repeat and switch trials within mixed blocks and supports precise estimation of switch and mixing costs.
In the figure below, reaction times are shown for task-repeat versus task-switch trials in mixed blocks. On average, participants are slower on switch trials than on repeat trials.
Giving people time to prepare for an upcoming task, for example by lengthening the cue to stimulus interval, reduces switch costs but usually does not remove them. This residual switch cost remains even after extensive practice. These findings suggest limits in how quickly we can reconfigure task sets and overcome interference from the previous task. The exact mechanisms are still an active topic of research and debate.
Further Reading
(2023). Principles of cognitive control over task focus and task switching. Nature Reviews Psychology, 2(11), 702-714.
(1927). Mental set and shift. Archives of Psychology, 89, 5-82.
(2003). Task switching. Trends in Cognitive Sciences, 7(3), 134-140.
(2007). Extensive practice does not eliminate human switch costs. Cognitive, Affective, and Behavioral Neuroscience, 7, 192-197.
About This Demonstration
Estimated Time to Complete: 15 minutes
You will complete two kinds of blocks:
- Pure blocks: Perform only one task (numbers: odd or even, or letters: consonant or vowel).
- Mixed blocks: Alternate between the two tasks using an alternating runs schedule.
As you switch between classifying numbers and letters, notice your mental experience. Many people report a brief gear shift on switch trials. Your performance illustrates how cognitive control supports flexible adaptation to changing task demands.
At the end, you will see:
- Your switch cost: mixed-block switch trials compared with mixed-block repeat trials.
- Your mixing cost: mixed-block repeat trials compared with pure-block repeat trials.
Click the button below to open the experiment in a new tab. This page will remain open.