Switch-Access Scanning Simulator

Grid Size

Scan Pattern

Scan rate 800 ms/step

Switch acceptance delay 250 ms

Current Trial

Row scan — step 1 of 5

Performance

Last selection time

—

Success rate

—

Theoretical avg. time (this setup)

—

How it works

Switch-access scanning is a real access method for augmentative & alternative communication (AAC) devices used by learners who cannot reliably point or type — a single switch (button, puff-sensor, eye-blink sensor) is the only required input. A cursor automatically sweeps through the on-screen options; the user presses the switch when it reaches the item they want.

Row-column scanning (R rows, C cols):
  1. Cursor sweeps rows, one row per "step"
  2. Press → that row is selected
  3. Cursor sweeps cells within the row
  4. Press → that cell is chosen

E[steps] ≈ (R+1)/2 + (C+1)/2
E[time]  ≈ E[steps] × scan_rate + dwell

Linear scanning (all R×C cells, one pass):
E[steps] ≈ (R·C+1)/2
E[time]  ≈ E[steps] × scan_rate + dwell

Scan rate — how long the cursor dwells on each row/cell before advancing. Faster rates mean quicker selection but demand faster reaction time.
Acceptance delay (dwell) — real switches accept the press only after a short debounce window, filtering accidental muscle spasms. Here it shifts your effective selection forward by round(delay ÷ scan_rate) steps — press too early relative to the delay and you may land on the next item instead.
Row-column vs linear — row-column needs two decisions but far fewer average steps for larger grids; linear is cognitively simpler (only ever "is this the one?") but scales linearly with grid size. This is the real trade-off clinicians weigh when configuring an AAC device for a specific learner.

Purple = scanning cursor · Gold ring = your target · Green/Red flash = result

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