B – 136 Examining the Effects of Word-Level Characteristics on Short Term Recall in an Adapted US-English Verbal Memory Task

Abstract Objective This study examined how word-level features such as frequency, syllables, serial position, and familiarity affect recall on a Verbal Memory Test, as part of adapting the Word List subtest of the Believe battery, originally developed in Castilian Spanish, for US English. Method Three versions of the Word List Test were developed: a direct translation, a partial adaptation, and a full adaptation based on AERA and ITC guidelines. 51 English-monolingual participants were recruited from the University of North Carolina Wilmington (UNCW) using convenience and snowball sampling and were randomly assigned one of the three versions. The versions differed in word selection based on frequency, number of syllables, tangibility, and appropriateness across geographic and socioeconomic backgrounds. Version 3 also included a category change to uphold the word criteria described. Participants also completed a word familiarity task to assess their level of familiarity with each word within their assigned list. Results Spearman’s correlations revealed positive but non-significant trends for frequency (ρ = .27, p = .063) and familiarity (ρ = .28, p = .055), and a negative trend for syllables (ρ = –.19, p = .206). Semantic category did not affect recall, F(4, 46) = 1.44, p = .236. Word position significantly influenced recall, F(15, 32) = 3.36, p = .002; and words at exactly the halfway point (position 8) had significantly lower recall than several other positions. Conclusion The present study showed that word position significantly impacted recall. Although other word-level characteristics showed limited effects, trends suggest that differences may be found with more power.