Home▸Rehabilitation Outcome Digital Assessment▸3D Cognitive Rehabilitation Gamified Task Difficulty Adaptive

3D Cognitive Rehabilitation Gamified Task Difficulty Adaptive

Orbit a genuine Three.js attention-grid landscape driven by the same 2-down-1-up staircase rule and challenge-vs-skill flow-channel index as the 2D simulator — now a real 3D scene you can rotate, with a live rolling-accuracy chart and flow-channel plane instead of a flat dashboard.

Rehabilitation Outcome Digital Assessment3DModerate60 FPS📱 Mobile-adapted⇄ 2D version
3d-cognitive-rehab-gamification ↗ Open standalone

This is the 3D companion to the Cognitive Rehabilitation Gamified Task Difficulty Adaptive simulator: the same model — success probability p = clamp(0.5 + (skill − level)·0.075, 0.04, 0.97), the 2-down-1-up staircase rule (two correct in a row raises difficulty by step·0.5, any single miss drops it by step·0.5), and the flow index that scores how close the rolling 8-trial accuracy sits to the 70–85% target band — now drives a genuine, freely-orbitable Three.js scene instead of a flat dashboard. An N×N grid of attention tiles (N = clamp(round(2 + level·0.42), 3, 7), exactly as in the 2D version) flashes a target tile on every trial; a green particle burst marks a correct response, a red cross marks a miss. A live 3D bar strip behind the grid replays the rolling-accuracy history against the same 70–85% band, and a tilted flow-channel plane to the side plots skill against difficulty level in real time, with a glowing marker tracking whether the pairing sits inside the diagonal flow corridor.

⚙ Under the hood

The same closed-form staircase and flow-channel model as the 2D simulator drives a real Three.js attention-grid landscape: a resizing N×N tile grid with flash/burst feedback, a rolling-accuracy bar chart against the 70–85% target band, and a skill-vs-difficulty flow-channel plane, all recomputed on every trial and every control change.

CanvasBiomedicineThree.js

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)