Visual Search Test

You can now see whether you are keeping pace, if you want to

Every timed question test has a bar: a number of correct answers in a number of minutes. Until today that number was only ever shown to you afterwards, on the results screen, and the run itself gave you a clock counting down and a question number counting up without ever saying whether the two were in step.

The briefing now does the division for you. Our bar on Directions and Distances is 48 answers in 16 minutes, which is about 20 seconds each, and that sentence is on the page before you start. There is also a tickbox: turn it on and the run tells you "on pace" or "behind pace" next to the question number, with how many you should have answered by now.

That 48 is our own training benchmark, set deliberately above a competent standard so clearing it here leaves you margin on the day. It is not a published pass mark, and we have no way of knowing the real one.

It is off unless you ask for it, and it is remembered once you do. The real test shows you nothing of the kind, and a rehearsal you have to switch back to being faithful is not much of a rehearsal.

Why it was built: looking at everybody with four or more runs of the same test, people do get faster with practice on almost every test here. On Directions and Distances they barely do, and some candidates are answering at about a third of the speed the target asks for while getting most of what they attempt right. They were not making mistakes. They were running out of time, and nothing on the screen had told them.

One honest limit. The guide counts answers against a bar that counts correct ones, so it is telling you the least you can get away with rather than the figure to aim for. If you are getting four out of five right you need to attempt rather more than the target to reach it.

The two Visual Search tests had their bar set far too low

Both were marked against a target of 24 correct answers in three minutes, a number picked before anybody had sat them. Candidates actually average 58. The result was that almost everyone finished on the maximum score of 150%, which tells a strong candidate and a struggling one exactly the same thing.

The target is now 50 on both, taken from what people genuinely score rather than from a guess. A typical run now reads around 100%, a weaker one around 89% and the best on record 125%, so the number means something again.

This moved scores. Your Visual Search scores will read lower than before, around 100% instead of 150%. Nobody got worse. The old runs are kept separately rather than compared with new ones, because they were measured against a bar that was wrong.