(~20 minutes)

Semantic Noun Matching

This test is similar to the written version of the Pyramid and Palm Trees Test (Howard & Patterson, 1992). Unlike that test and the similar Camel and Cactus Test (Bozeat et al., 2000), which combine trials requiring similarity judgments with trials requiring thematic associations, our version uses only similarity judgments. Our version also evaluates knowledge about abstract concepts that cannot be easily depicted, and thus allows a comparison between concrete object and abstract concept knowledge. Behavioral and neurophysiological studies indicate substantial differences in the processing of abstract and concrete concepts (Binder et al., 2005; Kounios & Holcomb, 1994; Paivio, 1971; Sabsevitz et al., 2005; Wise et al., 2000), and neuropsychological evidence for sensitivity to this factor in aphasia (Coltheart et al., 1980; Goodglass et al., 1969; Katz & Goodglass, 1990; Roeltgen et al., 1983). In addition to overall accuracy, category specificity is measured by comparing performance on various superordinate categories, and task “difficulty” is manipulated using similarity distance ratings obtained from healthy controls (Sabsevitz et al., 2005).

Materials:

• 60 concrete noun triads; half easy and half hard; half artifacts and half living things
• 60 abstract noun triads; half easy and half hard; matched to concrete trials on average word frequency

Procedure:

• sample word and two choices presented visually in pyramid formation (see figure)
• patient manually selects the alternative that is “most similar in meaning” to the sample item
• The 120 trials are divided into two balanced 60-trial blocks to avoid fatigue.

Data analysis:

• %correct overall and each condition; chi-square tests for category effects; mean RT by condition