I admit I may be wrong so I asked you to show me an example, since you talk with such certainty.
In fact this is an open question so nobody knows, even though most of the researchers believe human computation is equal as far as I know.
Backwards jumps are equally decidable for turing machines as for humans. Compiler and static code linters can trace through very complicated code and know that it will or will not halt, and optimize or warn based on the result.
I believe there’s nothing special about human computation abilities. Anything that is decidable by a human will be also decidable by a computer.
Please prove me wrong by showing a counter example.
Most of the “undecidable” are only undecidable for a subset of the problem instances, while a vast number of instances can be even trivially decidable. For example in the undecidable halting problem, both you and a computer can trivially deduce that while(true) will not halt. In the same way a computer can deduce that many instances of two pieces of code are semantically equivalent.
I’d like to see an instance of the problem where a human could decide it and the computer could not.
I just learned that people who are allergic to shrimp are also allergic to cockroaches.