Sitting at the computer this morning. Running the QM simulation of an electron hitting a potential barrier. Real and imaginary components of the non-existent wave function shimmer and spike, the reflected front-parts interfering with the still-oncoming rear-parts. They settle down into two diminished sets of waves - one moving onwards, the other retracing the previous path. Neither set of waves is the electron.
I tick the Ψ2 box and watch the two wave packets: the probabilities of finding the electron in that position with that momentum in an experiment. Practical Magic.
In fact, take a look.