I spent most of the time observing the 4-year-olds and the Kindergarten class. The four-year-olds had more displays of Autonomy and Initiative behavior, whereas the Kindergartners had displays of all three in pretty equal proportions.

For both age groups, the things that caught their attention of the bat was dynamic colors and things that they could clearly see had multiple components. This was put off slightly if there was a line for being able to try or do something, waiting for the action was not okay. Another drawback was if it had explicit instruction or if they were being talked at, they no longer had any interest in the thing. Both age groups like having the object or thing light up or make noise as a way to draw them in, however, without consistent time at the machine or lack of interaction the attention dissipated rather quickly. If the object did not directly interact with their hands or their eyes or ears, most wanted nothing to do with it.

I noticed while watching the Kindergartners, that when a child made a guess or claim that something would happen, another child would immediately try to put it to the test. For example, one child explained their reasoning for laying out the track the way he did for the train-set and two other children tested his layout to see if his way of thinking made sense based on what he had said. When one of the two things he had said did not turn out the intended way, all three went to see how they could fix the problem. Although the problem did not get solved while I was watching, based on some of the other instances I saw, the child instigating child would probably have been content with their work once it did get solved.

So based on what I observed, the ideal setup would involve noise or music, at least one way that the setup would have dynamic user input and finally, it could quickly give goal for the user to work towards so they can be proud of their efforts.