The current study aimed at promoting receptive vocabulary found in expository texts (ET) dealing with animals. While playing a board game, the participants were exposed to information about each animal. During six playing sessions, they gathered pieces of information about the target animal by scanning QR codes and watching videos of a speaking avatar (SA) of the animal delivering information about its characteristics, or by listening to a trained human mediator (HM) giving the same information. Twenty-three preschool boys (Mage = 6.12 years; SD = .48) at risk for Specific Learning Disorder (R-SLD) participated in this study. They were randomly assigned into the above-mentioned intervention groups: 12 and 11 participants, respectively. The receptive vocabulary of the ETs (i.e., comprehension of 'academic vocabulary') was tested using a word–picture match test at pre-intervention (T1), one-week post-intervention (T2), and three weeks post-intervention (T3). The results were analysed using nonparametric statistics because of the small number of participants in each group. According to the results, the SA intervention group outperformed the HM intervention group at T2 and T3 in receptive vocabulary scores. While the HM intervention group scores decreased significantly at T3 in comparison to T2, the same comparison for the SA intervention group did not reach significance. These preliminary results add to the accumulative evidence pointing to the advantages of using 'agents' for supporting children's language development and comprehension of ET genre. It also suggests a practical implementation of using state-of-the-art digital technology for vocabulary enrichment among preschool children with special needs.

Keywords: Preschool, risk for Specific Learning Disorder, expository text, avatar.