Tower of London Task

The Tower of London (ToL) task is a neuropsychological test of planning and problem-solving ability developed by Tim Shallice in 1982, based on the mathematical Tower of Hanoi puzzle. The task presents the participant with three vertical pegs of different heights and three colored discs (or balls). The discs begin in a start configuration, and the participant must rearrange them to match a goal configuration in the minimum number of moves, following simple rules: only one disc can be moved at a time, and a disc can only be placed on a peg that has room for it. Problems range from 1-move (trivial) to 5+ move configurations that require complex planning, subgoal management, and the ability to make counterintuitive "backward" moves that temporarily move away from the goal state in order to ultimately reach it.

Task Structure and Administration

• Materials — The original Shallice version uses three pegs of descending height (holding 3, 2, and 1 disc respectively) and three colored balls. The Drexel University version (TOL-DX, Culbertson & Zillmer) uses a standardized board with normative data. Computerized versions allow precise measurement of planning time, execution time, and move sequences.
• Problem complexity — Problems are classified by the minimum number of moves required (1-move through 7+ move). Complexity increases nonlinearly: each additional move roughly doubles the number of possible sequences, and problems requiring counterintuitive moves are disproportionately difficult. The search space for a 5-move problem includes hundreds of possible move sequences, only a few of which reach the goal in the minimum moves.
• Key measures — Planning time (latency before the first move), execution time (total time to completion), total moves (accuracy — how close to the minimum), rule violations (illegal moves), and first-move accuracy (whether the first move is on the optimal path) provide a comprehensive profile of planning ability.
• Think-ahead requirement — The critical cognitive demand is "look-ahead planning" — mentally simulating the consequences of moves before executing them. Effective performance requires the participant to visualize the state of the pegs after each potential move, identify sequences that lead toward the goal, and select the optimal path — all in working memory before touching a disc.

Cognitive Processes Involved

• Planning — The primary construct the ToL assesses: formulating a sequence of actions to achieve a goal. Effective planning requires identifying the goal state, analyzing the current state, determining the sequence of intermediate states needed, and constructing an action plan to traverse them. Planning on the ToL correlates with planning ability in daily life, educational settings, and occupational functioning.
• Working memory — Holding the current state, the goal state, and the planned sequence of intermediate states simultaneously in working memory places heavy demands on visuospatial working memory (for representing disc positions) and central executive function (for managing and updating the plan). Working memory capacity is a strong predictor of ToL performance.
• Inhibition — Some ToL problems require counterintuitive moves — moving a disc away from its goal position in order to create space for another disc. Making these "backward" moves requires inhibiting the prepotent tendency to move each disc directly toward its target. This inhibitory demand makes certain problems disproportionately difficult for individuals with inhibitory control difficulties.
• Cognitive flexibility — When an initial plan fails or proves suboptimal, the ability to abandon the current approach and generate an alternative plan is essential. Perseveration — continuing with a failed strategy — is a common error pattern on the ToL and is associated with frontal lobe dysfunction.
• Subgoal management — Complex ToL problems require decomposing the overall goal into subgoals (intermediate configurations) and managing the ordering and dependencies between subgoals. This hierarchical planning ability is considered a hallmark of prefrontal cortex function.

Neural Basis

• Dorsolateral prefrontal cortex (DLPFC) — The primary brain region associated with ToL performance. Neuroimaging studies consistently show DLPFC activation during ToL planning, with activation intensity increasing with problem complexity. Patients with DLPFC lesions show the most severe ToL impairments, confirming the causal role of this region in complex planning.
• Anterior cingulate cortex — Active during conflict monitoring and error detection on the ToL. The ACC detects when the current plan is going wrong and signals the need for plan revision, contributing to the flexibility required for optimal performance.
• Parietal cortex — The visuospatial demands of mentally simulating disc movements engage the parietal cortex, particularly the intraparietal sulcus and superior parietal lobule. Parietal activation reflects the spatial working memory load of representing disc positions and movement trajectories.
• Basal ganglia — The striatum contributes to sequencing actions and selecting among competing response options. Conditions affecting the basal ganglia (Parkinson's disease, Huntington's disease) produce ToL deficits that reflect difficulty selecting and sequencing motor plans rather than generating the overall strategy.
• Cerebellum — Cerebellar activation during ToL performance reflects the role of the cerebellum in sequencing, timing, and coordinating cognitive operations, consistent with its broader role in cognitive, not just motor, coordination.

Clinical Applications

• Autism spectrum disorder — ToL performance is often impaired in autism, reflecting the planning and executive function difficulties that characterize the condition. Autistic individuals typically show longer planning times, more moves to solution, and particular difficulty with problems requiring counterintuitive moves. However, performance varies significantly: some autistic individuals with strong visuospatial abilities perform at or above neurotypical levels, reflecting the uneven cognitive profile of ASD.
• ADHD — Individuals with ADHD often show reduced planning time (acting before thinking) and more moves to solution, reflecting impulsive responding rather than impaired planning capacity. When forced to wait before moving (imposed delay), ADHD performance may improve, suggesting that the difficulty is in deploying planning ability rather than in the ability itself.
• Frontal lobe lesions — The ToL was originally designed to assess planning deficits in patients with frontal lobe damage. Patients with dorsolateral prefrontal lesions show the most severe impairments, while patients with orbital or medial frontal lesions may show relatively preserved performance.
• Schizophrenia — Planning deficits on the ToL are a consistent finding in schizophrenia, reflecting the executive function impairments that contribute to functional disability in the condition.
• Neurodegenerative diseases — Progressive ToL decline is observed in Alzheimer's disease, frontotemporal dementia, Parkinson's disease, and Huntington's disease, reflecting the progressive executive function deterioration in these conditions.

Relationship to Tower of Hanoi

• Tower of Hanoi — The mathematical puzzle that inspired the ToL: three pegs with discs of graduated sizes, where larger discs cannot be placed on smaller ones. The Tower of Hanoi has a recursive solution structure and places heavy demands on working memory and planning. It is more difficult than the ToL and is used in research on complex problem-solving and means-end analysis.
• Key differences — The ToL uses discs of equal size (no size constraint) and pegs of different heights (capacity constraint), making problems less mathematically elegant but more clinically useful because difficulty can be precisely calibrated. The ToL also provides better discrimination at lower ability levels.

Psychometric Considerations

• Standardized versions — The TOL-DX (Culbertson & Zillmer, 2001) provides age-based norms for children and adults, standardized administration procedures, and validated scoring criteria. It is the most widely used clinical version.
• Reliability concerns — Test-retest reliability for the ToL is moderate, which is a limitation for repeated assessment. Practice effects can be significant, as individuals may remember solution strategies from previous administrations.
• Ecological validity — The ToL has demonstrated correlations with real-world planning ability (cooking, route planning, project management), supporting its ecological validity as a measure of planning, though the relationship is modest, reflecting the multifaceted nature of real-world planning.