Background: What can be done at school to reduce the number of children having difficulties in learning to read, especially those considered at risk for academic under-achievement, such as those who live in poverty ? This research seeks to identify early predictors of reading in French children from low-income families, and to develop and test the efficacy of 3 different reading training programs to 5 years old children. What are their direct and indirect effects on later reading ? Are they limited to what have been trained or have they a more general effect ? Do their benefits maintain over a one year period of reading instruction, when children finish grade one ?