Working-Memory Load in Microlearning Bursts

Microlearning Stream

Chunk size (items/chunk) 3

Working-memory capacity K 4

Burst pace (s between chunks) 1.8s

Interleaved review strength 40%

Buffer State

Working-memory load0 / 4

Consolidated chunks

Raw units learned

Overloaded / forgotten

Avg. activation

—

How it works

Microlearning works because it respects two limits of working memory, not because "short" is inherently magic:

Capacity (Cowan, 2001): K ≈ 4 ± 1 chunks held at once
Activation decay:        dA/dt = -A / τ   (τ ≈ 6s here)
Consolidation rule:      ∫A dt ≥ θ  →  chunk moves to long-term store

Each incoming chunk occupies one working-memory slot with activation A = 1. Left alone, activation decays exponentially. Every time a new chunk arrives, prior chunks get an interleaved review boost (a quick recap before new content — a real microlearning design pattern), which fights the decay. If a chunk's accumulated exposure (∫A dt) crosses a threshold before it decays away, it is consolidated into long-term memory (outer shell) and frees its slot.

If a new chunk arrives while all K slots are full, the weakest (lowest-activation) occupant is displaced and lost — this is cognitive overload. Chunking raises how much raw content each slot carries (bigger chunk size = more items learned per consolidated chunk) but also makes each chunk slightly harder to hold, so it decays a bit faster — the classic chunk-size ↔ cognitive-load trade-off from Sweller's Cognitive Load Theory.

Chunk size — raw items grouped per chunk. Bigger chunks cover more content per slot but decay faster.
Capacity K — how many chunks working memory can hold at once (try 4 vs. the old "7±2" folk number).
Burst pace — how fast new chunks arrive; too fast overloads the buffer before consolidation can happen.
Review strength — how much each new arrival "recaps" what's already in the buffer.
Take a Break — pauses new arrivals so the buffer can consolidate or quietly decay, showing why spacing bursts matters.

Working-memory slot Long-term memory (consolidated) Forgotten (overloaded out)

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.