Reading Intervention to Children With Developmental Language Disorder

The main purpose of this study is to investigate if the internet-based computer-game Grapholearn (GL) is an effective tool for early and intensive intervention of the reading decoding skills in Swedish children with DLD, as compared to two control groups of children with DLD (I) playing a math computer game, and (II) attending usual schooling. Hypothesis: Five weeks of 20 sessions of GL will improve the accuracy of word and nonword decoding skills in children with DLD from baseline to T2 and T3. The GL intervention group will perform higher number of accuratly decoded words and nonwords than the two control groups at T2, and at T3. The second aim is to elucidate how the parents' language and reading skills, prevalence of language-related diagnoses, and socio-economic-status, and the children's self-reported self-esteem is related to their reading skills. Hypothesis 1: Children with DLD who's parents show poor language and/or reading skills will perform lower on the word/nonword decoding tests. Hypothesis 2: Children who show no significant improvement of word/nonword decoding skills will to a higher degree have parents with reading difficulties. Hypothesis 3: Parents highest level of education will not be associated with the participating childrens language and reading skills. Hypothesis 4: Results from the self-reported self-esteem will be lower than normdata from the test manual, and will not be associated with the children's language and reading skills.

(I) The investigators will collect the following cognitive measures from each participant with DLD: Decoding of words and nonwords, and words occurring in the Grapholearn computer game; naming of the letters in the Swedish alphabet; non-verbal IQ; articulation; grammar production; nonword repetition; verbal working memory; and general language ability. The tasks assessing reading decoding skills are administered at T1, T2, and T3, while all other tests are administered only at baseline. The test administrators are blinded for group (intervention or control groups). (II) Anamnestic interviews with the participants' parents aiming at collecting information about the child (diagnoses or developmental challenges); reading activities in the family; family aggregation of language-related diagnoses and problems in the parents and siblings, and highest level of education of each parent, as well as their current occupation. (III) Language and reading tests of parents, same reading tests as in (I); and tests of nonword repetition, grammatical production of past tense; and sentence repetition. (IV) Questionnaire to the schools covering type and dose of formal reading instruction, and who is providing this instruction (e.g., teacher or special teacher) and in what kind of context (individual, whole class, and/or in small groups of children). The investigators will perform analysis of primary and secondary outcomes and influence of background factors in a mixed linear model with time- and groupinteraction effects. Background factors will be presented with relevant descriptive statistics, such as mean and standard deviations.

Children with DLD who attend a special school for children with DLD. This group will play the computer game GL in school for 5 weeks, 20 sessions, 15-30 min per session. Supervised by a teacher.

Children with DLD who attend a special school for children with DLD. This group will play a computer game focusing on math in school for 5 weeks, 20 sessions, 15-30 min per session. Supervised by a teacher.

Children with DLD who attend a special school for children with DLD. This children will have normal schooling as usual without playing any of the above computer games.

GL is a computer game based on the phonics-method (i.e., graphem-phoneme correspondence). It was developed at Jyväskylä University in Finland, in several languages, among others also for Swedish.

The rate (number) of accurate decoding of words and nonwords measured with lists of real words (n=20), grapholearn-words (n=20) and nonword (n=20) consisting of two-, three-, four-, and five-letter words distributed in three word lists, short words with simple orthographic structure first and long words with complex orthographic structure last.

The letter naming task is used to measure the rate of correct naming of 26 lower case letters. The task is presented with a chart of 4 by 6 rows of letters with the Times New Roman font in size 48 pt.

Inclusion Criteria: DLD diagnoses attending a special school for children with DLD attending Swedish schoolgrades 0 to 4 Exclusion Criteria: - non-Swedish speaking