The vigilance test is ten minutes of almost nothing happening. A grid of sixteen values updates continuously, all of them routine, until one of them quietly is not. You have about two seconds to catch it. The test is engineered around a well documented fact of human attention: detection performance decays with time on task, and selection wants to see your decay curve. This guide covers the format, the false alarm economy, and how to train detection that lasts.
What is the Vigilance Test?
The test measures sustained monitoring of low-frequency events, the purest laboratory version of watching a quiet radar picture. The professional reality is blunt: most of a monitoring shift is uneventful, and the entire value of the human in the loop is what happens in the uneventful parts. An overnight sector with three aircraft is this test with better graphics.
The design detail that matters: every grid update pulses identically, whether the new value is routine or anomalous. Motion cannot help you. Only reading the numbers separates a 480 from a 7400, so the test forces genuine semantic checking, not reflex to change. It is the long-form cousin of the sustained attention test, stretched from ninety seconds of letters to ten minutes of numbers.
Which Exams Include the Vigilance Test?
Long-duration monitoring tasks with rare targets are standard in controller screening. Candidates commonly report this test type in:
- FEAST and related ATC selection batteries, where vigilance maps directly to the job profile
- DLR screenings, in monitoring and attention stages
- ADAPT, within its sustained attention tasks
- Aon / cut-e batteries, in monitoring scales
- COMPASS and Advanced COMPASS by EPST, in attention sections
Providers change their content without notice and publish no official syllabus, so treat this as candidate-reported guidance.
How the Test Works: Format and Rules
The rules are sparse, which is the point:
- A 4x4 grid of values updates continuously for 10 minutes. Routine values live between 100 and 500.
- Occasionally one cell jumps far out of range, into the thousands.
- Tap the anomalous cell within about 2 seconds; miss the window and it resets as a missed anomaly.
- Tapping a routine cell is a false alarm and costs you.
- Every update pulses the same way; only the number itself tells you anything.
Suppose the grid ticks over and a cell now reads 4870 among neighbours in the 100s and 400s. That is the target; tap it now. This is exactly how it looks in the test:
The window and the discipline in two images:
The tenth minute is the test. The first nine exist to soften you up for it.
Scoring: What a Good Result Looks Like
Four readings: hits, misses, false alarms, and the trend across time, your vigilance decrement. The decrement is computed by comparing early blocks against late ones, and it is the number assessors care about most. A strong profile holds a stable hit rate from first block to last with false alarms near zero. Fast but trigger-happy candidates reveal themselves in the false alarm column; steady readers reveal themselves in the flat trend line.
How to Prepare: Strategies That Actually Work
- Scan in a fixed loop. Sweep the grid in a consistent serpentine path, roughly one full pass every few seconds. Structure outlasts motivation.
- Read magnitudes, not digits. You do not need the exact value, only its order: three digits routine, four digits target. Train that one-glance judgment.
- Refuse the pulse reflex. Updates flash identically everywhere; teach your hand that motion alone is never a reason to tap.
- Re-arm yourself on a cadence. Attention sags in waves; a small internal reset every thirty seconds, straighten posture, one deliberate full scan, flattens the decrement.
- Train full length, always. Five-minute practice sessions never reach the zone where the test is actually decided. The boredom is the training load.
- Sharpen the short game separately. The sustained attention test drills the same readiness at sprint distance, and the two transfer both ways.
Common Mistakes to Avoid
- Fixating the cell that last alerted while the next anomaly appears elsewhere.
- Tapping on peripheral motion without reading the value.
- Letting the scan loop dissolve into random glancing around minute six.
- Treating a quiet stretch as safe; anomaly spacing is irregular by design, and clusters happen.
Practice This Skill
The decrement flattens with time-on-task training and with nothing else. Build it with the Vigilance module, which reproduces the 4x4 grid, the identical pulses, the two second windows and the block-by-block trend report exactly as described here. Level 1 is free, and the guided tutorial walks you through the format before your first attempt.