Paired associate learning: form 1 and form 2 of the Wechsler Memory Scale

Despite its popularity as a clinical measure, the paired Associate Learning (PAL) subtest of Wechsler's Memory Scale (WMS) still remains relatively undocumented for its psychometric properties. This article reports on the performance of 500 (form 1) and 600 (form 2) general medical patients hospitalized for other than neuropsychiatric conditions across five (20-69) and six (20-79) ten-year age bands respectively. In both alternate forms of PAL, strong age effects emerged while sex differences were more marked in form 2. Temporal reliability, assessed in separate student samples, proved adequate for either form. Parallel form correlations also indicated a moderate relationship between them. When easy and hard associates were analysed as separate measures with MANOVA the two forms were found to differ. Some possible reasons for these results are explored with respect to current information processing concepts postulated to affect verbal learning performance, and their utility in helping distinguish between pseudodementia and mild Alzheimer type dementia. Separate factor analyses also showed the two equivalent forms to diverge somewhat in their structural pattern with form 1 demonstrating clear clusterings of the pairs along a continuum ranging from very easy (overlearned) and mediate (superordinate) to hard associates.