- Joined
- Jan 27, 2019
- Messages
- 7,337
- Likes
- 12,303
Yeesh. Ok....
With respect: You actually have no idea whether the scenario I described was due to the Barnum effect.
Yet, in a very unscientific-leap-of-inference, you've attributed the sonic descriptions to the barnum effect, with no way of ruling out alternative explanations like: the speakers really DID have the sonic character that produced the similar impressions in both listeners.
That really, already, is the nail in the coffin. You can't go any further in justifying your claim it was the Barnum effect, let alone that that Barnum Effect is responsible for many or most such convergences in people reporting hearing similar sound characteristics.
But just to further indicate how dubious your suggestion is, and how far it can go when one is showing any degree of caution:
As a life-long devotee of the skeptic movements (e.g. Skeptical Inquirer, Skeptic Magazine and plenty of other skeptical inquiries) I'm well aware of the Barnum effect. It's a classic reference in skepticism of astrology, psychic readings, etc.
What I described is not suitably analogous to the Barnum effect. As you are in a forum that prizes scientific rigor, you should remember just how cautious scientists are when making inferences from any particular experiment to what can be explained. Leap outside the constraints of any specific experiment at your peril! The Barnum effect is a very specific effect which has to do with describing human personalities. "meaning general characterisations attributed to an individual are perceived to be true of them, even though the statements are such generalizations, they could apply to almost anyone."
Note in the link you gave, that even in the specific use-cases, there are variables involved in achieving the effect, and a fine line has to be walked. For instance: include too many negative traits in your description and the effect fails, the personality description is rejected as inaccurate.
Now, you are trying to make the leap from a specific research in humans evaluating descriptions of THEMSELVES to claiming the effect explains
the type of descriptions people arrive at concerning speakers, including the one I gave. That's quite a leap. Remember, we aren't talking about, say, a priming effect where a huckster tells you what you should expect to hear before you listen. My examples is listening first, forming an impression, and then later identifying other people's descriptions as being similar to your own.
But before we take on your Barnum explanation, we already have another plausible explanation: The speakers have an audible sonic character that produced the impression I had when I auditioned them, and which also explains why they reviewer reported hearing that same character.
We know many speakers deviate audibly from one another due to different designs - audible differences in dispersion, frequency response, driver integration, room interaction, distortion, audible resonances etc. So if someone says "I heard two quite different speaker designs today and they sounded different" that is not just plausible....that is likely! And with some practice, one can start to identify and describe the changes in sound.
If you took a neutral speaker and an EQ, and created an exaggerated "smiley" EQ, and switched back and forth between flat and the exaggerated EQ, you think you'd need a blind test to have any justification in reporting or describing the difference? No, you don't, not in any practical sense.
Why not? Because while it's certainly *possible* some level of bias is involved, the fact is different EQ curves like that are well established as being audible. This justifies our not having to blind test every damned time we touch our EQ. Look at Amir's speaker reviews. Often in his listening tests he ends up altering the sound with EQ, evening out some dips or bumps that were bothering him. Then he reports "now, it sounds better." Strictly speaking, it COULD be that because Amir knows he dialed some EQ knobs that it's all an expectation effect and he isn't REALLY perceiving any change in the sound. But as a practical matter, we can accept that, yeah, he's reporting a change in the sound because he's making changes that are in the known audible range, and the effects of the changes are also generally fairly well known.
Speakers in a sense often come with different "EQ settings" - all the different design choices can alter the sound profile. And the sound can be described. So if I say "This speaker sounded bright and steely/irritating in the high frequencies, recessed in the midrange, and bloated in the bass region" and another person who carefully listens, with whom I've had no contact, reports the same thing, we have plausible grounds for having identified the actual sonic profile of this speaker. Could be wrong! Yes! Not perfectly reliable. But the hypothesis has a lot behind it making it plausible.
WHEREAS:
You want to propose the Barnum effect as the alternative explanation.
Do you realize that to take that seriously, as an analog to the Barnum personality test effect, you'd have to be able to produce a speaker description that is sufficiently vague-yet-compelling that anyone would think it applies to ANY speaker they own or have heard?
Remember: the Barnum Effect posits that you can take a room of say 50 or 100 people, give them all the same sufficiently vague personality profile, and the vast majority will say "Yes, that's an accurate description of ME." In other words, the single description will be seen to fit many different people. It does this by being vague, and by playing to known human biases about how people think of themselves.
The analog of this is that you can produce a sonic description of a speaker that will be seen to fit many different speaker designs!
You'd have to produce a description of the sonic character of a speaker where you could take a hundred very different speakers, and yet someone (or a group of people) would rate the description accurate for almost every speaker!
It posits that, for instance, I heard a speaker, formed a distinct impression, but a reviewer's description had no relationship to the actual sound of the speaker, but it was sufficiently vague that I could have heard ANY speaker and his description would have "fit what I heard."
I would like to see you actually demonstrate the plausibility of this claim, such that it becomes as plausible or more plausible than the hypothesis that our impressions came from actually hearing the sonic character of a speaker.
For instance, if you wrote a review of a pair of speakers I own, say the Thiel 2.7s, and you described them as "bright/steely upper frequencies, recessed midrange, bloated bass" there's no way in hell I'd read that as fitting my own impression of how those speakers sound in my home. They sound if anything precisely the opposite in every respect. Those sonic characters I described are not vague at all; they are damning deficits.
If you really think The Barnum Effect is the more likely explanation in play here, then you should be able to produce a Barnum-like sonic description of a loudspeaker's sonic characteristics such that I would agree it could fit ANY speaker I own or have owned. (For that matter, every ASR member should read it and say "Yeah, that describes just how my speakers sound...in fact, almost every speaker I've owned!).
And note also: in order for it to actually be analogous to the type of sonic description I gave, and which are often found in subjective reviews, your description can't just be something like "It produced sound! There, see how that can fit the sonics of any speaker?" You actually have to, like a subjective reviewer, get in to some detail - bass quality, mids, highs, imaging, etc, and produce a description that will be accepted as describing practically any speaker someone has heard.
Ready to try creating this Barnum Effect covers-all-speakers description?
Go!...
With respect: You actually have no idea whether the scenario I described was due to the Barnum effect.
Yet, in a very unscientific-leap-of-inference, you've attributed the sonic descriptions to the barnum effect, with no way of ruling out alternative explanations like: the speakers really DID have the sonic character that produced the similar impressions in both listeners.
That really, already, is the nail in the coffin. You can't go any further in justifying your claim it was the Barnum effect, let alone that that Barnum Effect is responsible for many or most such convergences in people reporting hearing similar sound characteristics.
But just to further indicate how dubious your suggestion is, and how far it can go when one is showing any degree of caution:
As a life-long devotee of the skeptic movements (e.g. Skeptical Inquirer, Skeptic Magazine and plenty of other skeptical inquiries) I'm well aware of the Barnum effect. It's a classic reference in skepticism of astrology, psychic readings, etc.
What I described is not suitably analogous to the Barnum effect. As you are in a forum that prizes scientific rigor, you should remember just how cautious scientists are when making inferences from any particular experiment to what can be explained. Leap outside the constraints of any specific experiment at your peril! The Barnum effect is a very specific effect which has to do with describing human personalities. "meaning general characterisations attributed to an individual are perceived to be true of them, even though the statements are such generalizations, they could apply to almost anyone."
Note in the link you gave, that even in the specific use-cases, there are variables involved in achieving the effect, and a fine line has to be walked. For instance: include too many negative traits in your description and the effect fails, the personality description is rejected as inaccurate.
Now, you are trying to make the leap from a specific research in humans evaluating descriptions of THEMSELVES to claiming the effect explains
the type of descriptions people arrive at concerning speakers, including the one I gave. That's quite a leap. Remember, we aren't talking about, say, a priming effect where a huckster tells you what you should expect to hear before you listen. My examples is listening first, forming an impression, and then later identifying other people's descriptions as being similar to your own.
But before we take on your Barnum explanation, we already have another plausible explanation: The speakers have an audible sonic character that produced the impression I had when I auditioned them, and which also explains why they reviewer reported hearing that same character.
We know many speakers deviate audibly from one another due to different designs - audible differences in dispersion, frequency response, driver integration, room interaction, distortion, audible resonances etc. So if someone says "I heard two quite different speaker designs today and they sounded different" that is not just plausible....that is likely! And with some practice, one can start to identify and describe the changes in sound.
If you took a neutral speaker and an EQ, and created an exaggerated "smiley" EQ, and switched back and forth between flat and the exaggerated EQ, you think you'd need a blind test to have any justification in reporting or describing the difference? No, you don't, not in any practical sense.
Why not? Because while it's certainly *possible* some level of bias is involved, the fact is different EQ curves like that are well established as being audible. This justifies our not having to blind test every damned time we touch our EQ. Look at Amir's speaker reviews. Often in his listening tests he ends up altering the sound with EQ, evening out some dips or bumps that were bothering him. Then he reports "now, it sounds better." Strictly speaking, it COULD be that because Amir knows he dialed some EQ knobs that it's all an expectation effect and he isn't REALLY perceiving any change in the sound. But as a practical matter, we can accept that, yeah, he's reporting a change in the sound because he's making changes that are in the known audible range, and the effects of the changes are also generally fairly well known.
Speakers in a sense often come with different "EQ settings" - all the different design choices can alter the sound profile. And the sound can be described. So if I say "This speaker sounded bright and steely/irritating in the high frequencies, recessed in the midrange, and bloated in the bass region" and another person who carefully listens, with whom I've had no contact, reports the same thing, we have plausible grounds for having identified the actual sonic profile of this speaker. Could be wrong! Yes! Not perfectly reliable. But the hypothesis has a lot behind it making it plausible.
WHEREAS:
You want to propose the Barnum effect as the alternative explanation.
Do you realize that to take that seriously, as an analog to the Barnum personality test effect, you'd have to be able to produce a speaker description that is sufficiently vague-yet-compelling that anyone would think it applies to ANY speaker they own or have heard?
Remember: the Barnum Effect posits that you can take a room of say 50 or 100 people, give them all the same sufficiently vague personality profile, and the vast majority will say "Yes, that's an accurate description of ME." In other words, the single description will be seen to fit many different people. It does this by being vague, and by playing to known human biases about how people think of themselves.
The analog of this is that you can produce a sonic description of a speaker that will be seen to fit many different speaker designs!
You'd have to produce a description of the sonic character of a speaker where you could take a hundred very different speakers, and yet someone (or a group of people) would rate the description accurate for almost every speaker!
It posits that, for instance, I heard a speaker, formed a distinct impression, but a reviewer's description had no relationship to the actual sound of the speaker, but it was sufficiently vague that I could have heard ANY speaker and his description would have "fit what I heard."
I would like to see you actually demonstrate the plausibility of this claim, such that it becomes as plausible or more plausible than the hypothesis that our impressions came from actually hearing the sonic character of a speaker.
For instance, if you wrote a review of a pair of speakers I own, say the Thiel 2.7s, and you described them as "bright/steely upper frequencies, recessed midrange, bloated bass" there's no way in hell I'd read that as fitting my own impression of how those speakers sound in my home. They sound if anything precisely the opposite in every respect. Those sonic characters I described are not vague at all; they are damning deficits.
If you really think The Barnum Effect is the more likely explanation in play here, then you should be able to produce a Barnum-like sonic description of a loudspeaker's sonic characteristics such that I would agree it could fit ANY speaker I own or have owned. (For that matter, every ASR member should read it and say "Yeah, that describes just how my speakers sound...in fact, almost every speaker I've owned!).
And note also: in order for it to actually be analogous to the type of sonic description I gave, and which are often found in subjective reviews, your description can't just be something like "It produced sound! There, see how that can fit the sonics of any speaker?" You actually have to, like a subjective reviewer, get in to some detail - bass quality, mids, highs, imaging, etc, and produce a description that will be accepted as describing practically any speaker someone has heard.
Ready to try creating this Barnum Effect covers-all-speakers description?
Go!...
