My apologies if this has already been covered elsewhere. I work in the field of nutritional science and research. My wife is involved in medical sciences and by nature we’re engrossed in a wonderful world of double-blind, placebo controlled studies on a daily basis; this experimental model is the gold standard in determining whether variable X has any affect on variable Y. Where possible, objective measurements (e.g. blood cholesterol, NK cell activity) are used, however subjective or self-reported changes are often used concurrently. Presently, this model is the most effective means of determining which drugs or chemicals have an impact (positive or negative) on a specific health outcome (such as pain, mobility, mood etc). By using a control group, and / or a placebo group, researchers can account for the confounding variable of what the test subject can “expect” when he or she is administered the independent variable (drug, chemical). It should be noted that in the majority of cases, if a subject does experience a positive result, it is because the subject expects the medicine to improve their health. Whether the subject was administered the drug, or a fake look alike (placebo) the majority of the benefit is experienced by both groups (sometimes test groups number in the 1000’s of people). A drug or chemical is only said to be effective if it produces an effect ABOVE AND BEYOND the placebo. The moral of the story here is that if people expect to achieve benefit, they will, regardless of whether any change actually occurred. If I give 100 people a bottle of snake oil and tell them that it will improve their mood, I can predict with statistical certainty that almost half of them will respond with improved mood. …..for a period of time. Immediately, anecdotal evidence and testimonials are called into question. What does this have to do with audio? Audio engineers or objectivists have been at odds with audio subjectivists for decades. A golden eared audiophile who claims that they can hear a difference by altering audio variable “x” is in essence no different than the statistical group who were duped by my snake oil. Maybe he or she is golden eared, or maybe they aren’t, but we are assumed to take their word for it by default. If they buy a more expensive cable they are expecting an improvement, and statistically they will – if they are using their human sense organs to interpret the data. And yet the audiophile’s golden ears insist that they trump all findings obtained in controlled audio experiments when it comes to cables, pebbles or CD markers. And hence, this often falsely perceived difference is why we rely on empirical data and measurements in field of science. Science uses controlled experiments and objective data for a reason – it’s trust worthy. Human interpretation of data from stimuli are not. Sometimes our observations are easily backed by solid measurable data (e.g. I think we can all agree that the DT770 is for bassheads, and the FQ graphs are in support of that observation). But when our senses defy all known sound physics / sound science, we have to wonder if our interpretations should be brought into question. I bring this up as a point of discussion, because as a scientist I have to be open to this logic (er…scientific mehod). I’m not trying to start an argument, but rather suggesting that all of us golden ears must admit that we are human and thus are subject to the placebo effect like anyone else. If there is a reason that we all agree to disagree on what we hear, I can’t think of a better reason that this. Perhaps I’m offering up this discussion to help reconcile my thoughts on the absolutely massive variance in opinions here at head-fi! Please don’t kill me.