Chattering to babies may be even better than reading to them

Mimicry comes first, so an infant can learn that when they see Dora they should say "Dora". When they put something down or finish eating, everyone seems to say "all done" and clap, so they start saying "all done" and clapping. There's a whole range of sophisticated seeming behaviours that can be trained easily by simple mimicry.

So, wrt. to reading. Things that infants can learn easily are picking up / putting down / fetching books. The nice feel of sitting with a parent and a book (which will hopefully lead to favourable thoughts towards reading, later). Also, mimicking sounds that parents make on specific pages. E.g. "BOOM, BOOM, BOOM, Mister Brown makes thunder" might trigger "BOOO BOOO BOOO".

Think about it this way -- imagine reading with an infant. You could substitute / make up all of the words but go through the motions and they would imitate your words instead of what's in the book. I.e. mimic conversation in context. They really have no idea of the content of the book, or even that it's significant, compared to all the actions / conversation around the activity of reading a book.

One of the most interesting findings from Scarborough & Dobrich, which looks like it was replicated by the study you linked to, is that the number of books in a child's home is a much greater predictor of later literacy ability than the amount parents actually read to their kids. A couple things to note though:

- Intrinsic motivation plays a big role in this. E.g. if a kid enjoys have books read to them, it will probably be beneficial. But if a kid would rather be doing something else, then reading to them may be actively harmful, in that it may actually make them read less on their own later even though otherwise they would have eventually come to enjoy reading.

- The way you read to your kid matters. E.g. asking questions about the book as you're reading is more beneficial than just reading, and even the way you phrase the questions makes a difference. That Hart & Risley book I linked to above has some data about how the cognitive benefits of asking questions depends on their grammatical structure.

Also, for what it's worth this may be related to the audio/video gap. I.e. talking to kids has a huge benefit for them, but playing recordings or videos of people talking has essentially no benefit.

The Slate article you linked doesn't deal with nature vs. nurture at all. It simply assumes that nature doesn't matter, approvingly citing the views of one researcher who makes the astonishing claim that "Children aren’t born smart. They’re made smart by their parents talking to them." This view is definitively disproven both by studies of adoptive children and by studies of twins reared apart (not to mention common sense—cognitive ability must be at least partly heritable, else there would be no way for ever smarter animals to have evolved). For more details, I recommend The Blank Slate by Stephen Pinker.