The signaller memory grid — “the dots”

Applicants call this one "the dots", and those who get it consistently name it the hardest part of the assessment. A sequence of positions lights up on a grid. You reproduce it in order. In between, a second task interrupts you and demands an answer.

Who gets it

Here the accounts genuinely conflict, and it would be dishonest to smooth that over. Applicants describe getting the memory grid on Grade 8 and Grade 9 applications. At least one describes not getting it on a Grade 2 trainee application. Others say it does appear on lower grades too.

Our reading: if you are applying for a higher grade, expect it. If you are applying for a trainee post, treat it as possible rather than certain — worth an hour of practice, not worth a week of it. We would rather tell you that than sell you a section by implying it is guaranteed to appear.

Why the interruption is the whole thing

Holding four positions in mind for two seconds is easy. Almost everyone can do it. The exercise is not really testing that.

What you are doing, when you hold a sequence, is rehearsing it — running it round in your head. The interrupting task competes for exactly the same mechanism. You cannot answer "38 + 17" without using the loop you were using to keep the sequence alive, so the sequence has to be either dropped or converted into something that survives without rehearsal.

That is why people's scores collapse when the interference switches on, and why practising sequence recall without interference improves almost nothing that transfers. Our version therefore lets you set the interruption to none, light or full, and the useful measurement is the gap between your span at "none" and your span at "full". That gap is what practice actually shrinks.

The technique that works

Stop trying to remember positions. Convert the sequence into a shape.

Four dots in a grid are four separate things to hold, and four is roughly where most people's capacity runs out under interference. But the same four dots traced as a path — down the left, across, up to the corner — is one thing. A shape survives an interruption in a way that a list does not, because you are not rehearsing it, you are holding a picture.

Concretely, while the sequence is playing:

• Watch the movement between positions rather than the positions themselves. Think "down two, right three, up one" — the path, not the coordinates.
• If the path makes a recognisable shape — an L, a zigzag, a diagonal — name it. A name is one item.
• Break longer sequences at a natural point: two shapes of three beats one list of six.

This is not a trick and it is not cheating. It is what people who are good at this task are already doing, usually without being able to describe it.

During the interruption

Answer it properly and quickly, then immediately re-picture the shape before you touch the grid. The instinct is to hold the sequence tightly through the interruption, half-answering the sum while trying not to let go — and that reliably loses both. Put it down deliberately, deal with the interruption, pick it up again.

Being willing to put something down and pick it back up is, not coincidentally, one of the eight non-technical skills Network Rail assesses, and it is a large part of what a signaller does all shift.

A realistic target

Most people land somewhere between four and six positions once the interruption is on, and one or two higher without it. If your "full interference" span is four and your "no interference" span is seven, the problem is not your memory — it is the handover between holding and doing. That is the trainable part, and it usually moves within a few sessions.

Practise just above what is comfortable. Our version adapts: get one right and the sequence grows, get one wrong and it shrinks, so after a few trials it settles on your actual ceiling rather than testing a fixed length you have already beaten.