> why can't the results be explained by "Some babies dislike certain toys and prefer new toys over those they dislike."
It actually is explained this way in the paper: "If infants do exhibit choice-induced preference change, they should prefer the novel toy and avoid the previously unchosen toy."
The thing is, they are not only testing for this "obvious" conclusion. Each subsequent experiment tries to invalidate the previous one's conclusions. They thought of 4 experiments, and maybe even more could be done.
> It's swap methodology is also unconvincing. Why couldn't the baby simply be avoiding a toy that was effectively taken away?
This is a good reasoning — if the infants are aware of the swap, their thinking could be "I chose that and didn't get it, now I don't like it / don't want it anymore". Even then, it still points to the same direction: the infants are changing their preferences based on their first choice, now rejecting something they wanted but didn't get.
They raise that possibility when discussing Experiment 3:
> Recall that in the induction trial, approximately equal numbers of infants chose each block—there was no systematic preference for any particular block. But what if all infants had their own idiosyncratic but consistent object preferences? ...
Then they sort of test that hypothesis in Experiment 4, by making the infants choose a block in the induction trial but then giving them the block they had not chosen – and even in this case the infants, in the next step of the trial, tend to reject the block they didn't get in the induction trial, which in this case was the block they had initially chosen. This needs to be combined with the previous findings, such as those from Experiment 2, in which the experimenters gave the infants a block, and Experiment 3, in which the infants' choice is blind (they din't know what was inside each box) — in these cases, the infants didn't show the same rejection tendency for the block they didn't get as in Experiments 1 and 4.
Anecdotically, QWERTY also seems to be sort of arranged in alphabetical order but with certain keys moved here and there — for exemple, the middle row is:
A .. D . F G H . J K L
(dots indicating the missing letters.)
Then it goes down for the M N from right to left, up for O P, then the top row is:
https://www.mesen.ca/docs/tools.html#netplay
Also RetroArch (which has cores for many systems, including Genesis):
https://www.retroarch.com/index.php?page=netplay