This is the standard change-detection paradigm used in visual working-memory research (Luck & Vogel, 1997; Cowan, 2001). Each trial: a study array of N colored items appears briefly, disappears for a blank retention delay, then a test array appears — on half of trials one item's color has changed. You judge "Same" or "Different".
Working memory holds only a handful of discrete items at once, not a fixed amount of visual detail. From your hit rate (correctly spotting a real change) and false-alarm rate (wrongly reporting a change that didn't happen), Cowan's formula estimates how many items you actually held in memory, independent of guessing:
K = N × (H − F)
H = hits / (hits + misses)
F = false alarms / (false alarms + correct rejections)
- K — estimated number of items truly retained. Typical adult visual working-memory capacity is K ≈ 3–4, regardless of how large the array is — this is the empirical basis for the popular "magic number 4" (a refinement of Miller's earlier "7 ± 2" for immediate verbal recall).
- Set size N — above your true capacity, K should plateau even as N keeps rising: you simply guess on the extra items.
- Study time / delay — real experiments show capacity is largely insensitive to a longer glance or a longer delay once items are encoded — memory failures here are capacity-limited, not time-limited.
Real-world relevance: visual working-memory capacity predicts fluid intelligence scores, is reduced in ADHD and schizophrenia, and is the mechanism radiologists and air-traffic controllers push against when scanning multi-item displays.