Task Switching Test — Cognitive Flexibility
Judge digits under two alternating rules — blue asks 'greater than 5?', orange asks 'is it even?' — with the rule switching every 2 trials. Your score is your switch cost in milliseconds.
What this test measures
Cognitive flexibility — the ability to shift rapidly between mental rule sets, known in the literature as set shifting. Each trial shows a digit from 1-9 (never 5) on a colored background: blue asks 'is it greater than 5?', orange asks 'is it even?'. The rule switches every 2 trials (the alternating-runs design of Rogers & Monsell 1995), so half the trials repeat the previous rule and half demand a switch. You answer Yes or No by button or keyboard across 12 unscored practice trials and 33 scored trials; the first scored trial is a warm-up that counts for accuracy but never enters the timing averages, leaving 16 switch and 16 repeat trials measured. Your score is the switch cost: median response time on switch trials minus median response time on repeat trials, computed from correct answers only (responses faster than 150ms are discarded as anticipations). Lower is better; 0ms would mean switching costs you nothing, and a slightly negative cost occasionally happens through measurement noise — it is shown as measured but scored like a 0ms cost. Runs below 75% accuracy, or with too few correct trials in either condition, are not scored at all.
The science behind it
Task switching is one of the oldest paradigms in cognitive psychology, dating back to Jersild's 1927 work on 'mental set and shift'. The modern form — predictable rule alternation over digit judgments in runs of two, as used here — descends from Rogers & Monsell 1995 (Journal of Experimental Psychology: General), who showed the cost shrinks with preparation time but never fully disappears (the 'residual switch cost'). Monsell's 2003 review in Trends in Cognitive Sciences puts typical switch costs at roughly 100-200ms for untrained adults, and Kiesel et al. 2010 (Psychological Bulletin) attribute the cost to task-set reconfiguration plus interference from the just-abandoned rule. Set shifting is one of the three core executive functions in Miyake et al. 2000 (Cognitive Psychology), and Dale & Green 2017 report task-switching advantages in action video game players.
How to improve your cognitive flexibility
Switch cost shrinks with practice — but only so far. Rogers & Monsell 1995 showed that even with full preparation time a residual cost remains, and Monsell 2003 notes well-practiced participants still pay a measurable penalty. Task-switching training does reduce switch costs and shows some transfer to similar untrained switching tasks (Karbach & Kray 2009), but claims of broad 'brain training' benefits are contested (Simons et al. 2016). Within a session, exploit the predictable structure: the rule changes every 2 trials, so prepare the next rule during the feedback pause; verbally rehearse the active rule; and prioritize accuracy — only correct responses count toward your averages, and a run below 75% accuracy is not scored at all. Sleep and an undistracted environment measurably help executive tasks.
Frequently Asked Questions
What is a good task switching score?
Against the switch-cost norms used here (mean 150ms, SD 95ms — the SD is widened to include single-run measurement noise), a cost under 90ms puts you roughly in the top 26%, under 30ms is around the top 10%, and near 150ms is average. Typical untrained adults land in the 100-200ms range (Monsell 2003); costs of 250-300ms — common in older or unpracticed adults — land around the bottom 15% to bottom 6%. These norms are an initial literature-based estimate pending recalibration on live data.
What is switch cost in a task switching test?
Switch cost is the reaction-time penalty you pay on trials where the rule just changed, compared with trials that repeat the previous rule. Here it is your median correct response time on switch trials minus repeat trials — the scored phase measures 16 switch and 16 repeat trials (plus one warm-up trial right after practice that is shown but not timed). Typical values are 100-200ms (Monsell 2003); it approaches zero in well-practiced switchers but rarely disappears entirely. A slightly negative cost — repeat trials slower than switch trials — occasionally happens through measurement noise; it is displayed as measured and scored like a 0ms cost.
Is task switching the same as multitasking?
Related but not identical. Task switching measures set shifting — sequentially replacing one mental rule with another — which Miyake et al. 2000 identify as one of three core executive functions. What people call multitasking usually involves overlapping demands, and switch costs are a major reason it is inefficient: every shift of attention carries a time penalty. Try our multitasking test to measure the overlapping side.
Can you improve cognitive flexibility with practice?
Partly. Practice reliably lowers switch cost within a task, and training studies show gains transfer to similar untrained switching tasks (Karbach & Kray 2009). But a residual cost remains even with full preparation time (Rogers & Monsell 1995), and evidence for broad 'brain training' transfer to everyday life is weak (Simons et al. 2016). A realistic goal is to shrink your cost, not eliminate it.
Why do wrong answers not count toward my switch cost?
It is the standard convention in switch-cost research: error-trial response times are contaminated — fast guesses shorten them and confusion inflates them — so they do not reflect true switching efficiency. This test uses only correct responses slower than 150ms (anything faster is an anticipation, not a decision), and the 12 practice trials never count. Rushing therefore backfires twice over: errors remove trials from your medians, and if your accuracy drops below 75% — or too few correct trials remain in either condition — the run is not scored at all, so button-mashing cannot produce a score.
More Cognitive Tests
Track a bouncing dot and click it the moment it turns red — while classifying digits as odd or even. 16 dual-task rounds measure how much accuracy you keep when both tasks compete.
Watch cells light up on a 3×3 grid and press Match whenever the position repeats from exactly 2 steps back. 30 fixed-pace trials (28 scored, exactly 9 true matches) measure your working-memory balanced accuracy.
Color words flash in sometimes-mismatched ink — respond to the INK color, not the word. 40 scored trials measure your Stroop effect: the interference cost in milliseconds.