In previous research, we found that English-acquiring children produce medial wh-phrases more frequently when the question is a subject question, like in (1), than when the question is an object or adjunct question, like in (2) and (3), respectively.

  1. Who do you think who kissed the boy?
  2. Who do you think who the boy saw?
  3. How do you think how the boys cut the rope?

If there is a production planning window that includes spec,CP and spec,TP to the exclusion of VP (and if children make these errors because of their developing executive function, not their grammar), then this difference is expected because there are two wh copies that the child must inhibit the pronunciation of in the hypothesized planning window for the subject question, rather than just the one wh copy that would be in the hypothesized planning window for the object and adjunct questions.

We're designing an experiment to test this planning window hypothesis with adults, and we would appreciate your feedback on the experimental design.