Go / No-Go Practice
Build a frequent Go response, then withhold it on rare No-Go trials. Results separate commissions, omissions, correct rejections, Go reaction time and timing consistency instead of hiding the run inside one score.
Start Go / No-Go
Choose Easy, Standard or Challenge and a pace. Press Space or GO for frequent Go stimuli; withhold for the visible No-Go rule.
What Go / No-Go asks you to do
A Go / No-Go block deliberately makes one action common. Most stimuli are Go trials, so pressing quickly becomes the default response. A smaller set are No-Go trials, where the correct action is to withhold that already-prepared response. DailyBrainArc keeps that structure explicit: the rule is visible before play, the Go button and Space key are the only response, and a No-Go trial is correct only when you do not press during its response window.
Why commission and omission errors are separate
Pressing on a No-Go trial is a commission error. Failing to press on a Go trial is an omission. Those mistakes point in different directions, so the specialist never merges them into one vague error count. A fast player can have very few omissions but too many commissions; a cautious player can suppress No-Go well but miss Go trials. The result shows both patterns directly beside overall and No-Go accuracy.
How Easy, Standard and Challenge differ
Easy uses 48 trials with 36 Go and 12 No-Go trials. The No-Go rule is one shape and the base response window is 1,050 milliseconds. Standard uses 64 trials with 52 Go and 12 No-Go trials, making Go more prepotent; the No-Go rule is one color and the base window is 800 milliseconds. Challenge uses 80 trials with 68 Go and 12 No-Go trials, a 620 millisecond base window, and a shape-plus-color conjunction so one-feature lures remain Go.
Why Challenge includes lures
A Challenge No-Go can be, for example, only one particular color and one particular shape together. Other trials deliberately share just the shape or just the color. Those are still Go trials. The generated source contract requires substantial counts of both one-feature lure types so the block cannot be solved by watching only one dimension.
The source bank is generated locally
The committed V0.139 bank contains 90 project-generated sources split 30/30/30 across Easy, Standard and Challenge. It does not use a copied clinical item set, scraped competitor stimuli or a remote API. The generator constructs exact Go and No-Go counts, rejects consecutive No-Go trials, limits excessively long Go runs, and keeps an initial Go-only region so the response tendency is actually established before the first inhibition event.
Exact response authority
Every abstract trial stores shape and color attributes. The active rule defines whether shape, color, or their conjunction is the No-Go condition. When a seeded Practice block is materialized, shape and color labels are deterministically permuted. The runtime then recomputes No-Go from the transformed attributes and transformed rule. The independent verifier repeats the rule test separately instead of trusting a stored no-go answer flag.
Why reaction-time consistency is included
Median Go reaction time describes the typical correct Go response without letting one interruption dominate the result. RT MAD is the median absolute deviation of correct Go reaction times from that median. It is a transparent within-block consistency measure: a lower MAD means correct Go responses clustered more tightly around the block median. It is not converted into a population score.
What post-error change means
The trainer also compares correct Go reaction times after an error with correct Go reaction times not following an error. The displayed post-error change is the difference between those medians. A positive value means the sampled post-error Go responses were slower in that run. Short blocks can have too few qualifying trials, so the metric is shown as unavailable when there is not enough data rather than inventing a value.
Clean and Assisted practice
The No-Go rule itself is always visible because the task is about withholding a response under a known rule, not guessing the instructions. Training highlight is optional assistance: when enabled, a visible STOP cue appears on No-Go trials. That makes the discrimination easier, so the entire run is marked Assisted. Assisted runs remain in recent history but do not replace Clean bests.
Adaptive pace changes timing, not truth
Relaxed, normal and fast modes change the response window using a pace multiplier. Adaptive starts from a browser-local pace for the selected level. After a Clean Adaptive run, very high overall and No-Go accuracy can move the pace up slightly; poor accuracy can move it down. The underlying source sequence and Go/No-Go identity never change because of pace.
Deterministic Practice and replay
Opening Practice without a seed inserts a stable seed into the URL. That seed selects a frozen source and deterministic shape/color permutations. Reloading the same URL reconstructs the same materialized block. Fresh creates a different seed. This makes regressions and personal comparisons reproducible while keeping generation local.
Why this differs from Stroop
Stroop creates interference between a relevant and irrelevant attribute while still requiring a response on each trial. Go / No-Go instead creates a frequent motor response and then occasionally requires withholding that response altogether. Both involve control, but the defining error here is a commission on a rare No-Go trial rather than choosing the wrong response category.
Why this differs from Task Switching
Task Switching asks which rule is active and then which of two responses matches that rule. The defining event is a rule change. Go / No-Go holds one inhibition rule stable for the whole block. The challenge comes from frequent Go responses making action habitual while rare No-Go events require stopping the action.
What the results do not claim
The specialist does not diagnose ADHD, impulse-control disorders or any other condition. Browser timing depends on hardware, input method, background load, familiarity and interruptions. The metrics are transparent practice measurements from one local block. They are not IQ scores, clinical norms or proof that training transfers to unrelated real-world behavior.
A useful practice routine
Start at Standard 1.0× and prioritize low commission errors before chasing speed. If commissions are high, use Training highlight for one Assisted run to study the No-Go feature, then return to Clean. If omissions dominate, a relaxed pace may be more informative. When both error types are low and Go timing is stable, try fast or Adaptive rather than shortening your own decision rule.