Coloring Book: A new method for testing language comprehension

Abstract

We present a new method for investigating children's language comprehension and argue that it has the potential to mitigate known task-related biases and expose children's grammatical and lexical knowledge in a more natural and ecologically valid manner. The new method consists of filling in a digital coloring page, according to sentence stimuli (e.g., The green monkey is being scratched by the blue monkey; The rabbit is red.). Through the playful act of coloring in the page, children reveal their interpretations of grammatical constructions and lexical items. We argue that this method gives more accurate results than existing methods, in which children are asked to choose among several pictures representing a number of possible interpretations. We present two experimental studies: one with Dutch-speaking children, tested on four types of grammatical constructions, and a second study with children learning Dutch as a second language, tested on their vocabulary knowledge. In both studies, the new method was compared with one of the most widely used methods: the picture selection task. In the first study where children's performance is said to be underestimated, the new method reveals better performance whereas in the second study where children's performance is assumed to be overestimated, the new method reveals lower performance. The results suggest therefore that the new task indeed decreases external task-related effects and offers a more reliable measurement of children's linguistic knowledge.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.