Predicting Response to Neuropsychological Intervention in Developmental Dyslexia: A Retrospective Study

Abstract

Identifying the patients who are likely to be non-responders to a certain treatment may allow clinicians to provide alternative strategies and avoid frustration and unrealistic expectations for the patients and their families. A retrospective study on 145 children treated with visual hemisphere-specific stimulation examined the specific profiles (reading, writing, metaphonology, memory, callosal functions) of non-responders, and identified predictors of response to intervention (reading, reading and writing) through linear regression models. The effects of additional variables such as rapid automatized naming (RAN) and Visual Search were investigated in a subsample of 48 participants. Subgroups related to gender and dyslexia subtype were considered in the analyses. The results highlight an Intervention Differential Effect (IDE) not depending on regression to the mean and mathematical coupling effects. The characteristics of non-responders for reading seem to correspond children with mild reading and severe writing impairments; non-responders for reading and writing are those with impaired callosal transfer. Predictors of overall response to intervention were pre-test reading and writing scores; phoneme blending, accuracy in visual search and speed in rapid automatized naming contributed to explaining response variance. Specific predictors for female vs. male participants and dyslexia subtypes were identified.