I think change for the sake of a "fresh look" which is often advocated for by middle management and marketing yahoos with too much power over design would count as "just for fun".

>It is very unlikely that people are changing things just for fun.

Why is that very unlikely? Many of these changes are not objectively better: they're just different and sometimes arguably worse. If you and your team decide you don't have anything to do for the next release, guess what happens next.

Here's how I think about it. Let's take the example of changing how to move the cursor from having the magnifier to dragging the cursor directly & using the spacebar long press trackpad.

Scenario 1 is that Apple has enough money, time, and manpower that they are constantly doing user studies, in-person and via survey, to see how people currently use and learn to adopt the product.

They find out that moving the cursor is something that people want to do often, but that the current method has some problems. It is hard for new users to discover, and the tap and hold takes a long time if you're doing it frequently, like in a text revising workflow. The magnifier was useful for the old low-res screens, but less helpful now that everything is retina.

They implement a few different approaches, and measure them against the old method. In their previous tests, they've seen that one thing new users try is dragging the cursor directly, so they try making that work, which is more discoverable. It's faster because you don't have to wait for the long press. They add some feedbacks so you can see both where it's tracking your finger, and where the cursor will drop in the text. And if those feedbacks aren't precise enough, you can use the spacebar trackpad method. In their benchmarks, the combination of the methods is easier to discover, and a big speed improvement.

They know that some users will hate being forced to change, because nobody likes being forced to change, but they honestly think it's worth the improvements. They take their findings to management, who personally try out the new methods and approve the change.

Scenario 2 is that Apple is such a large company that nobody really knows what other people or teams are doing, even management. There's nothing pressing for the UX teams to work on - I mean, aside from the AR HMD and the car. So, they decide to change up one of the core interactions of the company's most important product. They don't need to justify the change to upper management (all of whom use the product heavily), so it just sort of slips under the radar.

If you've got a more plausible scenario 2 I'd be interested to hear it, but yes, I think scenario 1 is more likely.