Objective: HIV-associated neuropsychological impairment is marked by poor cognitive flexibility and strategy use, due in part to the virus's affinity for impairing the frontal-striatal network. Verbal fluency research has shown impaired performance is associated with poor clustering and switching strategies, but little research has assessed if nonverbal fluency impairment is driven by a similar mechanism in HIV. This study examined figural cluster strategies as a mediator of the relationship between the CNS penetration index of antiretroviral medication (CPE index) and total figures produced on the Ruff Figural Fluency Test (RFFT). Method: 20 HIV-positive adults (85% male) recruited from an outpatient primary care clinic completed a neuropsychological batteryincluding the RFFT. Nadir CD4+ count, medication history, and years since diagnosis were derived from chart review of medical records. Results: After controlling for age and Average Cluster Size (ACS), the relationship between the CPE Index and Total Score on the RFFT was fully mediated by ACS (β = 3.542, t(16) = 0.904, p = 0.380). The unstandardized regression coefficients between CPE Index to ACS and ACS to Total Score were statistically significant (p < 0.05). The unstandardized indirect effect was 6.907, which was statistically different from zero (p = 0.01). Number of Clusters (NOC) failed to mediate CPE Index to Total Score. Conclusion(s): Lower CNS penetration levels are associated with poor figural fluency performance, which is mediated by smaller figural cluster strategies. This result is consistent with research on verbal fluency tasks, suggesting similar executive deficits may be driving impaired performance on both verbal and nonverbal modalities in HIV-positive individuals.