What switch cost means
RT switch cost is the median correct RT on switch trials minus repeat trials in this exact run. Accuracy cost compares the same two groups. Neither is a clinical score or population percentile.
Read the cue, answer the active rule, and see exactly how repeat and switch trials differ. No opaque flexibility grade.
Read the cue first. Switch and repeat trials are scored separately.
PARITY: F = Odd · J = Even
MAGNITUDE: F = Low 1–4 · J = High 6–9
Keyboard: F / ← = left · J / → = right
24 scored · 50% switches · 0 ms cue · 1800 ms deadline
Finish a run to inspect every scored trial.
Accuracy first. Read the cue before the digit. A small or negative switch-cost number from one short run is not a universal grade; inspect accuracy, usable-trial counts and repeat/switch medians together.
Benchmark is deliberately longer. It preserves single-rule baselines so mixing cost can be estimated separately from local switch cost. Practice, Daily and Sprint trade that depth for repeat use.
Measure pure-rule baselines separately from mixed repeat and switch trials. Cue changes, three-rule load and response validity stay visible instead of collapsing into one opaque flexibility score.
This is browser-local practice and self-tracking. Switch/mixing/cue costs are descriptive difference scores, not a diagnosis, IQ score, standardized norm or population percentile.
RT switch cost is the median correct RT on switch trials minus repeat trials in this exact run. Accuracy cost compares the same two groups. Neither is a clinical score or population percentile.
Benchmark preserves the inherited pure-rule baselines and mixed block so it can also estimate mixing cost. Practice, Daily and Sprint are shorter repeat-use modes.
Your run history stays in this browser. Daily and shared Practice/Sprint links preserve the exact task stream so you can replay the same sequence.