Vol. 11:2 (2020) ► pp.163–195
The processing of passive sentences in German
Evidence from an eye-tracking study with seven- and ten-year-olds and adults
This study examines the processing and interpretation of passive sentences in German-speaking seven-year-olds, ten-year-olds, and adults. This structure is often assumed to be particularly difficult to understand, and not yet fully mastered in primary school (Kemp, Bredel, & Reich, 2008), i.e. in children aged between six and eleven. Few studies provide empirical data concerning this age range; it is therefore unknown whether this assumption is warranted. Against this background, we tested whether the three age groups differed in their off-line comprehension of passive sentences. In addition, we employed Visual World eye-tracking to measure processing difficulties that may differ between age groups and may not be reflected in the final interpretations. Previous studies on adult language processing in German and English have documented a preference to interpret sentences according to an agent-first strategy. Our results show that all three groups make use of this strategy, and that all of them are able to revise this interpretation once the first cue indicating a passive sentence is encountered (the auxiliary verb form wurde). We conclude that at least from age seven on, children have the linguistic and cognitive prerequisites to process the passive morphosyntax of German and to revise initial sentence misinterpretations.
Article outline
- 1.Introduction
- 1.1The agent-first strategy: Theoretical accounts
- 1.2Acquisition of passives: Evidence from off-line comprehension studies
- 1.3On-line processing of passive sentences in children
- 1.4Off-line and on-line comprehension of passive sentences in adults
- 2.The current study
- 2.1Research questions and predictions
- 2.2Participants
- 2.3Materials
- 2.4Procedure
- 2.5Analyses of the eye-tracking data
- 2.6Statistical analysis
- Off-line data
- On-line data
- 3.Results
- 3.1Accuracy
- 3.2Eye movement data
- 4.General discussion
- Implications for a language processing model
- 5.Conclusion
- Acknowledgments
- Notes
-
References