"While we cannot say with full assurance that infants at this age cannot learn printed words, our results make clear they did not learn printed words from the baby media product that was tested," said Susan Neuman, a professor in New York University Steinhardt's Department of Teaching and Learning and the study's senior author.
In exit interviews, there was the belief among parents that their babies were learning to read and that their children had benefited from the programme in some areas of vocabulary development.
"It's clear that parents have great confidence in the impact of these products on their children. However, our study indicates this sentiment is misplaced," Neuman said.
Researchers examined 117 infants, aged nine to 18 months, who were randomly assigned to treatment and control groups.
Children in the treatment condition received a baby media product, which included DVDs, word and picture flashcards, and flip books to be used daily over a seven-month period; kids in the control condition did not receive these materials.
Over the course of seven months, the researchers conducted a home visit, four laboratory visits, and monthly assessments of language development.
To test children's emerging skills in the laboratory, the researchers examined the capacity to recognise letter names, letter sounds, vocabulary, words identified on sight, and comprehension.
Using a state-of-the art eye-tracking technology, which follows even the slightest eye movements, the researchers were able to closely monitor how the infants distributed their attention and how they shifted their gaze from one location to another when shown specific words and phrases.
The results, which included criterion and standardised measures of emergent and early reading skills, showed no differences between the infants exposed to baby media and the control group on 13 of the 14 assessments.
The only assessment that showed a difference was parents' beliefs that their child was learning new words despite countervailing evidence from a standardised measure indicating no differences between groups.
The study appears in the Journal of Educational Psychology.