The main purpose of this study was to analyze the effects of computer-assisted practice on reading and spelling in children with learning disabilities (LD). We compared three practice conditions, one with reading and two with spelling, in order to test whether computer-based reading and spelling practice has an influence on the development of reading and spelling ability in children with LD. A sample was selected of 85 children with LD, with age range between 8 years and 10 years (age, M=111.02, SD=9.6), whose spelling performance was two years below grade level. The participants were randomly assigned to one of four groups: 1) Copy the target word from the computer screen (n=22), 2) Memorize the target word and write it from memory (n=21), 3) Word reading (n=21), and 4) the untrained control group (n=21). We administered measures of pseudoword reading, phonological awareness, phonological word decoding and orthographical word decoding tasks. We examined the learning effects and transfer effects on words classified as a function of length, consistency, and complexity of syllable structure. Overall, the results showed that reading training did not improve spelling; however, the children who participated in the copy training condition improved their spelling skills.