This study was designed to examine the effects of auditory memory, visual memory, and vocabulary knowledge on intramodal and intermodal delayed matching-to-sample tasks. A secondary objective was to assess the value of task performance and trait measures in predicting reading achievement. A total of 52 first graders were randomly selected as participants and their cognitive traits were measured individually with the visual-sequential memory subtest from the ITPA, the digit span subtest from the WISC-R, the picture vocabulary from the Stanford-Binet, and the vocabulary subtest of the WISC-R. Each child then participated in four delayed matching-to-sample tasks: (1) matching a visual stimulus with a previously seen visual stimulus, (2) matching an auditory stimulus with a previously heard auditory stimulus; (3) matching an auditory stimulus with a previously seen visual stimulus and (4) matching a visual stimulus with a previously heard auditory stimulus. The stimuli were ten pronounceable nonsense trigrams and response choices were formed from systematic alterations of these trigrams. A few weeks after the trait measures and matching-task data were collected, all first graders took the SRA as part of the district testing program. The data were analyzed with a series of multiple regressions and the results are presented and discussed in detail. (JMB)