Background: Studies of the neurocognitive effects of long-term benzodiazepine use have been confounded by the presence of neurocognitive deficits characterizing the clinical conditions for which these medications are taken. Similarly, studies of the neurocognitive effects of anxiety disorders have been confounded by the inclusion of chronically benzodiazepine-medicated patients. This study was designed to tease apart the potentially confounding effects of long-term benzodiazepine use and panic disorder (PD) on memory and visuoconstructive abilities. Methods: Twenty chronically benzodiazepine-medicated and 20 benzodiazepine-free patients with PD with agoraphobia were compared with a group of 20 normal control participants, group-matched for age, education, and gender on a battery of neuropsychological tests assessing short-term, episodic long-term, and semantic memory, as well as visuoconstructive abilities. Results: Results indicated that benzodiazepine-free panic patients were relatively impaired in nonverbal short-term and nonverbal episodic long-term memory and visuoconstructive abilities, whereas verbal short-term and verbal episodic memory and semantic memory were preserved. Only limited evidence was found for more pronounced impairments in chronically benzodiazepine-medicated PD patients. Conclusions: This study provides evidence that patients with PD are characterized by relative impairments in nonverbal memory and visuoconstructive abilities, independent of benzodiazepine use. Nonetheless, we found evidence that chronic treatment with benzodiazepines is associated with intensification of select relative impairments in this realm. Documentation of these deficits raises questions about the broader etiology of neurocognitive impairment in PD as well as its impact on daily functioning. Depression and Anxiety 28:999-1007, 2011.