Impaired planning in patients with obsessive-compulsive disorder and unaffected first-degree relatives: Evidence for a cognitive endophenotype.

Patients with obsessive-compulsive disorder (OCD) show deficient planning capacity in the Tower of London (TOL) problem solving task. Preliminary evidence for similar deficits in unaffected first-degree relatives suggests that impaired planning may constitute an endophenotype of OCD. However, results on this issue are inconsistent, possibly owing to small sample sizes and variability in problem structure across TOL tasks. Here, we adopted a computerized version of the TOL task featuring a 2 × 2 factorial design (high/low search depth × full/partial tower goal state) and examined a well-characterized sample of n = 72 OCD patients, n = 76 unaffected first-degree relatives and n = 102 healthy comparison subjects. Both OCD patients and relatives exhibited significantly less accurate problem solving than controls. Search depth, goal hierarchy, or the number of minimum moves did not moderate these group differences. Medication, OCD symptoms, and depressive comorbidity did not affect TOL performance in patients, suggesting a state-independent effect. In conclusion, we found that OCD patients as well as unaffected first-degree relatives show deficient TOL performance across a range of task conditions, strongly supporting the role of impaired planning as an endophenotype of OCD, and contributing to the growing evidence for fronto-striatal dysfunctions in OCD.