Babies hate touchscreens

One of those pieces of self-serving anecdotal evidence I see a lot is how much some designer's kids love their iPhone, iPad, etc. I am now even seeing tweets and posts about kids raised with touch finding keyboards weird. Any day now someone is going to start calling them "Touch Natives" or "Gesture Children" or something else stupid. This sort of stuff annoys me because... It's totally anecdotal. Not only is it just the kids of designers, I feel it's a lot like my overview of devices I see in the wild. If you are inclined to notice iOS devices, or touch screen use, that's what you see. These kids don't use remote controls for their TV? They don't use game consoles, or have a Nintendo DS? You /really/ think they do not use buttons? I don't believe you. It's got serious selection bias. These are folks who love, love, love their touch devices, and only reluctantly even test their work on Android. They think keyboards are for suckers, because Apple says so I guess, so their kids have relatively less exposure to hardware selection mechanisms or hardware keyboards. It's not my experience. I have a lot of devices. And when I say a lot, I mean a lot of variation. The big one there is a Tablet PC. It's basically indestructible, so children (including random foster kids who come through) get to use it. I have tried some experimentation with this, such as showing them it takes touch input, then walking away. And they are shown by other kids, or themselves discover the pen. So far (of about a dozen kids from 4 years old onward) prefer the pen interface; they happily sit in the corner poking the device with the pen, in the dark, for hours. A lot of them have (I guess) observed keyboard-based computers, and ask to use a "real computer" (at home? on tv?) when given a post-PC tablet (e.g. iPad, Touchpad) and they get tired of typing searches, etc.
But what would be really fun is to get as far away as possible from preconceptions and learned behaviors. What if we could see how babies deal with things like this? I don't have the time or money, but I can throw out my own anecdotes as well as anyone else. Like this: What I see (and you can only get so much from the video clip) is the 12-month-old child:

• Grabbing items on the screen as though they are real objects, with physical presence.
• Trying to move, or perform drag actions on the items (to the point it works when supported; the program shown however does not).
• Rotating the device, to see the dimensionality of the item, see around the object, get behind it, or try to approach grabbing it from another angle.
• Interacting with the hardware, such as becoming involved with the bezel (as a grabbable object) when trying to grasp the items on the screen.
• Tasting the items, or chewing on the device.

There is no clear instance I have observed of a deliberate touch action (incidental is not deliberate). Which makes perfect sense when you look at what skills he's acquiring with the objects he encounters in the real world. He grasps, feels, pushes, pulls, picks up and puts down, stacks, turns and so on. This points me to some of the precepts of NUI. Touch in the sense of tapping an object on a flat screen isn't very natural, but how about dragging, or grabbing actions? I'd like to see dimensional behaviors; at least simulated 3D but maybe even multi-sided devices, so flipping has a real consequence, and shows different data. Hell, why not? I am not so sure about chewing interactions. That may be less useful to the general population. We'll have to do more research on that one.