Interestingly this work contradicts not just previous theory, but also previous experimental result. I would be interested to see if they have a critique for why the previous runs of the experiment provided a different experimental result.
They discuss a few possibilities. The first is cross-cueing, which they describe as " ... behavioural tricks, such as touching the left hand with the right hand" [1]. However, they also discuss a number of reasons why they think cross-cueing did not play a role. The second (and imho quite interesting) explanation is that the split-brain phenomenon could be transient: their subjects have had between 10 and 20 years to recover since their callosectomy. It may be that the brain learns to use other pathways to communicate inter-hemispherically.
I think your second explanation is probably right. We keep discovering that brains are more adaptable than we previously thought. I wouldn't be surprised if their brains had learned to 'route around' the damage in some way over the years.
Very interesting. I'm also wondering whether we might just be approaching this with too much b/w thinking - it might simply differ widely between individual patients. The operation was done by only a small number of doctors, so it might well be that each set of patients (regionally somewhat separate) has slightly different results. Or even that each individual differs - they weren't operated with modern precision and even with modern techniques you likely still would get slightly different operation results on each attempt.
For the second suggestion to be tested they would have to identify someone who was tested 10/20/30/40 years ago and try to repeat the experiment.
One last thought: that information flows between brain halves doesn't exclude the possibility of there still being some disconnect and/or even two 'personas'.
That's what leapt out at me. These brains have had years and maybe decades to learn to reunify themselves. Split brain experiments have always shown a strong drive for presenting a unified personality even as the experiments showed a non-unified experience and response.
So even if neuroplasticity in the cranium can be ruled out, the "bandwidth" and subtlety of cross-cueing, or some unknown means of communication, might have grown.
That was where my mind went as well. We can't ignore the incredible feats of neuroplasticity that undoubtedly happen in a split brain patient over a decade.That would make an interesting body of work in and of itself.