The 'Double Blind' Listening Test

I hope I don't unintentionally offend anyone but I am thinking that as a number of you guys are relatively newcomers to hifi, perhaps you guys have not heard of this before.

The background to me starting this discussion is that I have noticed recently that as all good hifi enthusiasts do, eventually everyone start talking about whether and how one can determine, to a degree of certainty, that a change to the system has changed the sound or not (setting aside the question of whether it is for the better or for the worse). Certainly there was a danger recently of a big flame-war in the KLAV report thread where there were contentions whether it is possible to hear the effects of reversed (power supply) polarity on equipment. What about those what I called 'exoticas', super expensive cables (power cords, interconnects), or even more, the 'black art' things like 'Shun Mook' tuning pucks and so forth ? Are some people really gifted with 'golden ears'. Have you sat silently and uncomfortably amongst a group of enthusiasts who are shouting about how great an improvement the latest power cord have made to the sound while you yourself did not hear anything. But of course you did not want to say it out since you worry about spoiling the party or worse still being labelled 'wooden ears'.

HOW do we settle the question, once and for all, whether something really have made a difference ? The 'Double Blind' Listening test is one such method. Note however that like all scientific test, this DBL test is not the total answer to everything and there are valid arguements why it does not work. But let me explain the test to those of you who have not heard about it before.

The DBL works by having a group of people listen to a 'control system', a hifi system made up of sufficiently high quality components (to the level that the listeners are working at). I.e. DBL works for all kinds of systems, from entry level beginners type hifi, way through to top-of-the-range reference grade systems. We just have to make sure the system if at the level we are testing at.

There are various ways to run the test, some more extreme than others. My favourite is the 'sighted' test where the listener has full view of the system (some testers even put a curtain between the listeners and the system). Now we start with the listeners subjected to a range of carefully chosen music.

Next, the 'change' is introduced, whatever is being tested. If say a fancy new Sunyata power cord is being tested, then that is swapped into the system. HOWEVER, the difference here is that some means is made to hide this from the listener. I.e. while the listeners knows they are supposed to listen for the effect of a change in power cord, from say the manufacturer supplied one to a Sunyata cord, they are not told when or whether the change has been made.


What ???

Well, what happens is that the listening session stops and the tester goes to the system and fiddles with it. He MAY or MAY NOT change the power cord. But he won't say. He just fiddles (or pretends to fiddle) with the system. Now he walks back to the listening group and plays the same set of music selection again. The listeners will now WRITE DOWN the differences they hear on paper and when the session has ended, hand it to the tester.

Now, the test IS REPEATED AGAIN. It is repeated several times and the listeners hears the same selection of music and are expected to hear any difference and commit their reputation (or their 'golden ear') by writing their findings down. Of course the tester also documents down the actual change in the system, i.e. which session was with the control equipment (the manufactuer supplied power cord) and which session was with the equopment being tested (the sunyata power cord). The test can also be adapted to test several alternatives at the same time, with the tester not announcing which power cord is used at any session and the 'control', the manufacturer power cord being swapped in for a few sessions.

The name 'double blind' refers to the fact that the listeners DO NOT KNOW which equipment is being used. I think (not sure) if the listeners can see the equipment, it is simply called 'blind test' but when they can't even see the equipment, it is called 'double blind'.

The theory is that if the listeners can really hear the change in sound (this is a test of the equipment or for some wicked testers, it is a test of how good the ears of the listeners are, or how honest they are), then regardless of whether they know it or not, they should be able to consistently identify the improvement or rather, change in the sound of the system whenever the equipment being tested is put into the system.

In theory, this DBL test is THE authoritative way to conclusively determine if something really delivers what it is supposed to. It is based on solid scientific procedure. So for once, we can conclusively test to see if fancy power cords improves the system sound or not (actually, I have found that they do).

Now, back to the statement that there are valid arguments AGAINST this. Some people admits that seeing or knowing that a fancy expensive piece of equipment may lead one to PHYCOLOGICALLY THINK that the system's sound has improved but it may not. However, the biologists argues that this is exactly the point. Listening is a PERCEPTION. There may not be a real absolute reference because the ear picks up the sound waves but it is the brain which interprets that sound waves. This is why some people like heavy bass and some don't. And some like rock while others like classical. In the end, it all boils down to perception and preference. So the argument is SO WHAT if any improvement perceived is all in the mind. If the owner really -thinks- he can hear an improvement, then that is the most important thing of all.

There are other arguements against DBL. E.g. the human ear/brain is very adaptive and we quickly get used to the improvements. Stated another way, our standards changes as we 'upgrade' our experience to better and better equipment. So over the course of the DBL, the listeners notes can become inconsistent as towards the end of the test, he becomes accustomed to the better sound so his notes now changes as he makes reference to the better sound and not to the control equipment which the test wants it to be.

Nevertheless, the DBL test is one of the most accepted means of conclusively determining if an equipment really changes the sound or if a listener really hears what he claims to be able to hear.

Sorry for the double posting. I have deleted the duplicate thread.

I don't intend to start a flame war - but below are some subjects that hifi buyers should be aware of:

http://www.audioholics.com/education/cables/the-truth-about-interconnects-and-cables

http://2eyespy.tripod.com/id3.html

BTW -
"In a double-blind experiment, neither the individuals nor the
researchers know who belongs to the control group and the experimental
group. Only after all the data have been recorded (and in some cases,
analyzed) do the researchers learn which individuals are which." http://en.wikipedia.org/wiki/Blind_experiment

For myself, I will go 'on record' and say that I have heard differences in interconnects, speaker cables and power cords. And in some cases, it was an unplanned type blind listening session. But only if the cable in question is very good.

On the other hand, I am usually quite wary of 'the truth about' type articles (no offence meant yah). I feel the writer usually have some sort of hidden agenda, and something to prove.

To those of you who sometimes get together for music sessions (I understand a few of you do), do try out this test and tell us your experience. It would be an interesting discussion.

CN Yee wrote:I don't intend to start a flame war - but below are some subjects that hifi buyers should be aware of:

http://www.audioholics.com/education/cables/the-truth-about-interconnects-and-cables

http://2eyespy.tripod.com/id3.html

BTW -
"In a double-blind experiment, neither the individuals nor the
researchers know who belongs to the control group and the experimental
group. Only after all the data have been recorded (and in some cases,
analyzed) do the researchers learn which individuals are which." http://en.wikipedia.org/wiki/Blind_experiment

The problem with the audioholics post is that they do not qualify their claims of exotic hifi cables being a hoax.

Its the same argument that many many years back when Stereo Review wrote that "all CD Players sound the same" - Julian Hirsh.


Here's a scenario to consider:

If you are using "Edifier" or "Logitech" type of computer desktop speakers, would you expect to hear a difference in sound between the standard OEM chicken noodle wire vs a length of vandenHul or Kimber cable ?

CN Yee wrote:I don't intend to start a flame war - but below are some subjects that hifi buyers should be aware of:

http://www.audioholics.com/education/cables/the-truth-about-interconnects-and-cables

http://2eyespy.tripod.com/id3.html

BTW -
"In a double-blind experiment, neither the individuals nor the
researchers know who belongs to the control group and the experimental
group. Only after all the data have been recorded (and in some cases,
analyzed) do the researchers learn which individuals are which." http://en.wikipedia.org/wiki/Blind_experiment

Also just to add some words of wisdom; There is an old Japanese saying:
If you believe in everything you read, then it is better not to read at all.