Move colored balls between pegs to match a goal - using as few moves as possible. Classic test of executive planning.
Tower of London (Shallice, 1982) is the standard executive planning task in clinical neuropsychology.
Three pegs hold colored balls. You must reach a goal configuration in a minimum number of moves.
Solving optimally requires looking ahead 3+ moves before acting - pure planning load.
Performance differentiates frontal-lobe injury from healthy controls more reliably than almost any other task.
The task loads the dorsolateral prefrontal cortex strongly. fMRI studies show DLPFC activity scales with planning depth.
Tower of London is more sensitive to subtle planning deficits than the older Tower of Hanoi.
How far ahead you can plan a sequence of moves before acting.
Tower of London isolates look-ahead planning: holding a goal configuration in mind, mentally simulating a chain of moves, and choosing the path that reaches it in the fewest steps. Solving in the minimum moves reflects how well you can sequence sub-goals internally rather than acting impulsively and correcting as you go.
Unlike a pure memory or speed task, the difficulty here is combinatorial: each move opens new possibilities, so success depends on inhibiting the first tempting move and evaluating consequences. Shallice introduced it precisely to probe frontal-lobe planning, and the same skill underlies tasks like routing errands efficiently or staging a multi-step repair.
Plan the whole sequence in your head before you touch the first ball.
Start with three-ball problems and work toward longer goal states only when you regularly hit the minimum-move solution. Before moving, trace the full path mentally and identify which ball must move first to avoid blocking others. Treat extra moves, not time, as the metric to drive down.
The classic mistake is acting on the most obvious move and then untangling the mess, which inflates your move count. Watch for goal-blocking: placing a ball in its final spot too early can trap one beneath it. Pausing a few seconds to plan almost always beats fast trial-and-error here.
Read moves-over-minimum first, solution time second.
The cleanest measure is how close you come to the minimum number of moves, not raw speed; solving a five-move problem in five moves is a strong result even if you paused to plan. Longer planning time with fewer excess moves usually signals better strategy, so track excess moves trending toward zero across sessions.
Performance is state-dependent and sensitive to fatigue and impatience, so a rushed session inflates move counts. A good score reflects planning on this specific puzzle format; it does not certify real-world organisation or decision-making, where time pressure, emotion, and incomplete information change the problem entirely.