Swingley & Aslin tidbit

[mzettersten]

There are a couple of potential tweaks that could be made to the Swingley // Aslin dataset.

ONe idea is to emphasize the usefulness for teaching a little bot more (Annissa is working on that).

Another question is whether we might use a different dataset than Swingley & Aslin, because (a) the condition manipulation has multiple levels and doesn't lend itself as neatly to a simple story (when comparing the paper to what we generate in the tidbit) and (b) the original paper doesn't actually include the kind of time course plot we end up creating.

[mzettersten]

Possible alternative datasets (2-level condition manipulations, time course plots to compare in paper):

• perry_cowpig
• garrison_bergelson_2020
• byers-heinlein_2017

[mzettersten]

I think we will stick with swingley aslin.