Tower of London Test

What this measures

How often you can finish a rearrangement in the fewest possible moves, in this task — planning, as the Tower of London measures it. Three pegs of different heights hold three discs; you are shown where they have to end up and you move one at a time. A problem counts only when you finish it in the minimum, because solving it the long way is a different thing from solving it. The number is a count of problems, not a time: the clock beside it is shown because it is interesting, never because it scores you.

A commit step keeps your thinking time and your moving time apart — play the balanced twelve, climb the Ascent until the spare moves run out, or sprint the clock, then challenge a friend.

How to get better

You mostly get better at this task — reading the goal before you touch anything, noticing which disc has to move last and therefore which has to move first, and resisting the pull to make the obvious move and see what happens. That is a real, honest thing to practise on a screen. It does not make you a better planner anywhere else, and we would rather say so than sell you the other thing.

Questions

What is the Tower of London task?

Three pegs of different heights hold three discs. You are shown a goal arrangement and you move one disc at a time — only the top disc of a peg, and never onto a peg that is already full — until the discs match the goal. Tim Shallice introduced it in 1982 as a way to look at planning specifically, rather than at speed or memory.

Why does a problem only count when I solve it in the minimum?

Because solving it the long way is a different behaviour from solving it the short way. Finishing in the fewest possible moves means you worked the sequence out first; finishing in more means you searched by doing. Averaging the two would produce a number that describes neither, so there is no partial credit and there are no stars.

Why is there a Plan button?

It is the only way to measure planning in a browser. The pegs stay locked until you press it, so the time before the press is thinking time and the time after it is moving time. Both are recorded and both are shown — and neither of them is your score in the Survey, because a planning task that rewards speed teaches you to stop planning.

Is this the Tower of Hanoi?

No. In Hanoi a larger disc may never rest on a smaller one. Here any disc may rest on any disc, and the only constraint is how many discs a peg can hold — three, two and one. The disc widths are there so you can tell them apart, not to restrict where they go.

Why these twelve problems?

Because an unbalanced set measures which problems you happened to get rather than how you plan. Kaller, Unterrainer and Stahl showed that in 2012, and the Survey's twelve are balanced across the three parameters they identified: how many moves the problem takes at minimum, how many moves must be planned before the first one can be chosen, and whether the goal fixes the order the discs go in. The seed changes which problems you get; it never changes the mix.

Does my device or my speed affect the number?

Not in the Survey or the Ascent. Both are counted in moves, so the number is the same on a phone as on a laptop. The 30-second sprint is the one mode that reads a clock, and its records are kept per device class and never compared across them.

How stable is this number?

Kaller and colleagues reported split-half reliability of .72 and internal consistency of .69 for a balanced set of this kind, which is respectable for an executive measure and is still a long way from precise. Read one run as a rough reading, and your own runs over time as the thing worth watching.