Usefulness of a Short Form of the Pyramids and Palm Trees Test To Differentiate Variants of Primary Progressive Aphasia (P02.044)

Abstract

Objective: To evaluate a short form (15 item version) of a picture association test of object semantics, to distinguish semantic variant primary progressive aphasia (svPPA) from nonfluent/ agrammatic variant (nfvPPA) and logopenic variant (lvPPA), and healthy controls. Background The Pyramids and Palm Trees Test (Howard and Patterson, 1992, Pearson, Inc) is widely used as a test of semantics. However, several items depend more on cultural knowledge (in some cases specific to the UK) rather than knowledge of the defining features of objects (e.g. Are acorns associated with pigs or goats?). Published norms may not be applicable in the US and may depend on education. However, a test of object semantics would be useful in distinguishing patients with svPPA (who often have impaired object semantics) from nfvPPA and lvPPA (who have spared object semantics). Design/Methods: We selected 15 items from this test that could be answered correctly on the basis of knowing the defining features of the objects. We administered this short form to 20 controls and 47 PPA patients: 17 with svPPA, 16 with lvPPA, and 14 with nfvPPA. We used independent t-tests to compare mean accuracy scores across groups. We also evaluated whether age or education influenced scores on the short form of the test in either patients or controls using linear regression. Results: Mean accuracy for patients with svPPA was significantly lower (66.6±27.4) than for patients with nfvPPA (99.9±0.5) (t=4.5; df29; p Conclusions: The 15-item version of this test appears to be useful for assessment of object semantics and distinguishing svPPA from other variants of PPA and healthy controls. Disclosure: Dr. Lala has nothing to disclose. Dr. Race has nothing to disclose. Dr. Tsapkini has nothing to disclose. Dr. Hillis has nothing to disclose.