The study sought evidence consistent with the hypothesis that phonological recoding of printed words is important during reading acquisition. Children at the end of their Kindergarten year were given a test of nonsense word reading (as a measure of phonological recoding skill) as well as tests of sight word reading and verbal intelligence. Two groups of 28 children were matched on sex, school attended, sight word reading, and verbal intelligence, but differed on phonological recoding skill. If phonological recoding was important in reading acquisition, the children with greater skill in this area should make greater gains in reading achievement over the following years. When reading achievement was tested at the end of Grades 1 and 2, these children were found to be significantly ahead.

The tendency to translate from a weaker to a stronger language was investigated as a function of proficiency in the nondominant language. Three groups of native Hebrew speakers with varying degrees of proficiency in English read passages of alternating English and Hebrew sentences. Subjects identified a following recognition sentence as either “identical” to a sentence read in the passage, a “translation” of a sentence read in the passage, or “new,” and then rated their confidence. The results showed no evidence for translation processes; with respect to “identical” and “translation” pairs, none of the groups was more accurate in recognizing sentences originally presented in their dominant language. These findings, which parallel those of an earlier study in which nonfluent bilinguals were tested auditorally, reflect similar psycholinguistic mechanisms underlying nondominant language processing in speakers of different levels of proficiency. The results were related to the distinction between automatic and controlled processes involved in dominant versus nondominant language processing.

Using a random-block design, 52 profoundly retarded nonverbal children were matched on relevant variables and randomly assigned to one of three treatment groups or a no-treatment group. In Phase One, two groups were trained in either the Premack or NON-SLIP graphic systems. Children in the Premack group learned an average of 2.3 symbols; no child in the NON-SLIP group reached the symbol stage. In Phase Two, these two groups and a third group were taught manual signs from an American Sign Language lexicon. There were significant differences in IQ and language age from pre- to posttest, and children in the two groups exposed to both graphic symbols and signs learned significantly more signs than graphic symbols. Signing appears to be a demonstrably superior intervention for profoundly retarded nonverbal populations.