Search
Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Search Tips
Register for SRCD21
Personal Schedule
Change Preferences / Time Zone
Sign In
X (Twitter)
Introduction. The purpose of the present study is to investigate the timeline of the acquisition of verbal grammatical categories in Russian-speaking children during their first three years of life. For that we analysed two corpora of longitudinal speech recordings: of a boy Y and of a girl T. We focused on the acquisition of four grammatical categories: aspect, tense, person, and number. In line with previous studies (Bar-Shalom, 2002; Gvozdev, 2008; Gagarina, 2008), we expected to find that imperfective forms would be acquired earlier than perfective, present tense — earlier than past and future, the order of acquisition of person would be 3→1→2, and singular forms would be acquired earlier than plural. Importantly, we analysed parents’ speech in both corpora and compared the usage of grammatical categories in parents’ and children’s speech. We hypothesized that grammatical properties of the children’s speech will be similar to those of the child-directed speech.
Method. In 2016-2019, child-parent everyday interactions were video-recorded and transcribed according to the guidelines of the CHILDES database (MacWhinney, 2000). The overall size of the corpora are 57184 tokens for Y (recorded from 1;5 till 3;0) and 37252 tokens for T (recorded from 0;10 till 3;0). All the verbs were morphologically annotated using the PyMystem3. Morphological homonymy was disambiguated by an annotator. To investigate the timeline of acquisition of grammatical categories we divided both corpora into three periods with the comparable amount of verbs produced by children in each of the periods.
Results. First, we found that T produced more imperfective than perfective verbs during all the three periods. However, Y produced more perfective than imperfective verbs during the first period (1;5 - 1;10), and vice versa during the second and the third periods. For both children, imperfective verbs prevailed in the input during all the three periods. Second, we showed that for Y, past tense was more frequent at the early stage of language development but from 2 years of age, he produced present tense more often than past tense. T produced present tense more often than past tense during all the three periods. For both children, future tense was least frequent during all the periods. In the input, both children heard present tense more often than past and future tense. Third, both children acquired 2nd person verb forms last. However, 1st person was acquired earlier than 3rd person that contradicted previous observations for Russian. In the input, both children heard 2nd person more often than 3rd and 1st person forms. Fourth, our results indicated that verb forms in singular were acquired earlier than in plural which is in line with the input during all the periods.
Discussion. Our results showed individual differences in the acquisition of tense and aspect and partly contradicted previous findings. The order of acquisition of person differs from that in (Gagarina, 2008). We also found that statistical properties of grammatical categories in child-directed speech did not change as children grew older.