cundare2 – Good points on the mechanical side — the Klippel papers are legitimate, and you’re right that loudspeaker transducers measurably change with use. The "no substantial evidence for mechanical changes" claim is worth pushing back on. And people in this thread already have. I thought that aspect of the article was settled.
Keep in mind, that claim is the article’s, not mine. I was summarizing what the authors concluded, not endorsing it. Your quarrel there is with them. As I pointed out fairly early in this thread, I find this type of research interesting and was asking for more. Most on this thread have just taken to beating up on this article, crying "foul," defending their experiences, etc. Ok, well, this is just an audio forum and most people are not interested in searching out more literature. Some are, but most would rather offer their view, or think through the logic or illogic in the original post.
Coming back to your argument, where I’d push back is in where you take the engineering papers. They establish that mechanical changes occur and are measurable. They don’t establish that those changes fall within the audibility threshold, or that they’re what you were actually responding to in the LCD-X case. That’s a further inference the papers don’t provide. But thanks for citing – I am glad to have the references.
The personal experience example is vivid, but "like dialing a tone control from 10 to 3" is exactly the kind of dramatic, unblinded, cross-session comparison that psychological accommodation explains most parsimoniously. I don’t doubt the experience was real. The question is what it’s evidence of. Strong confidence in the absence of confirmation bias isn’t itself evidence of its absence — that’s rather the point of controlling for it experimentally.
The asymmetry you invoke at the end — positive reports constitute evidence, negative reports don’t — is worth thinking through. If every positive report counts as evidence for burn-in but no negative report can count as evidence against it, then there’s no possible observation that could challenge the claim, which means it isn’t being held to any evidential standard at all.
The "you can’t prove a negative" move is also slightly off-target: nobody needs to prove burn-in never happens. The question is what standard of evidence we require before accepting that it did happen in a specific case. And under controlled conditions the asymmetry actually runs the other way — a well-controlled negative result (blind, level matched, same session) is more informative than an uncontrolled positive one, precisely because it rules out the competing explanations the positive report leaves open. Unblinded, cross-session personal experience doesn’t clear that bar — not because the experience wasn’t real, but because it’s compatible with too many alternative explanations.
None of this closes the case against burn-in. It just means the case isn’t closed by personal experience either, however compelling it felt in the moment.
