Originally Posted by urapnes1
When you wrote what I quoted below, I interpreted it to mean that a bias controlled comparison is required in order to hear an audible difference.
First of all, I never said a DBT was 'required' in order to 'hear' an (actual) audible difference. I wrote about standards of proof. But is one required to verify every possible
reported audible difference, including complete absence of output versus normal listening output level
? No, of course not, that is absurd. I didn't really think this needed explaining. Clearly a mistake on my part.
You provided a plausible reason why a huge audible difference would be perceived in a normal listening situation (i.e., not a lab test where you are measuring 'just noticeable difference' versus silence) : in one state, there was NO OUTPUT. The poster I was originally addressing said the difference he perceived was 'huge'...but he surely didn't say it was output vs no output. And he too was listening at his normal levels. The cause he cited was that he had upgraded a DAC or an amp. Faced with such a claim , " A caused B", you have a choice to believe
1. A is most likely to cause B
2. Something else is equally or more likely to cause B
Cognitive bias is not likely to cause a person to think one DUT has no output whatsoever, and the other has a normal-level output. You, btw, have not determined what the actual cause of the system failure was..i.e., whether it was the 'fault' of the Pioneer amp, or whether you simply badly mismatched an amp with a pair of difficult-to-drive loudspeakers (in which case, the fault is yours).
But psychological biases are
*very* likely to lead a person to believe that an 'upgrade' improved output. Even if there wasn't really a change (a so-called 'phantom switch' experiment). An analogous experiment with taste includes a by now famous study of people given the same wine labelled with different prices. They tended to report that the pricier wine tasted better
Please note that, as with all real audible difference, you could have 'verified' your difference independently of listening if you wanted to, with measurements. You could verify that your system is *broken*, without doing any listening to it at all...just by watching it shut down.