This project grew out of a collaboration on the kiddraw project (https://github.com/brialorelle/kiddraw). A core question underlying that project is: What developmental changes explain age-related variation in children’s ability to produce and comprehend graphical representations (e.g., drawings of object categories)?

Question

Here we ask: How does the ability to decompose complex visual shapes into parts and recompose these parts into composite shapes develop throughout childhood?

Approach

To investigate this phenomenon, we plan to take the following strategy:

1. Measure age-related variation in the ability to copy composite shapes by drawing. Prediction: younger children will have more difficulty in producing high-fidelity reconstructions of these composite shapes.
2. Measure age-related variation in the ability to copy simple shapes by drawing. Prediction: performance will be relatively high and more similar across age groups than in the composite shapes condition.
3. Test whether a “curriculum” that provides experience with simple shapes and some simple combinations of simple shapes facilitates transfer to a task requiring the reconstruction of a composite shape.

This could be accomplished by training in the “bottom-up” strategy of starting with practice reproducing simple shapes and simple combinations of them, and/or by the “top-down” strategy of directing children to the compositional structure by explicitly segmenting the composite shape into simpler parts (e.g., by color coding).