Join 3,438 readers in helping fund MetaFilter (Hide)


You probably don't even know what you like
August 26, 2010 9:28 AM   Subscribe

Choice blindness occurs when subjects are unaware that the choice they made is opposite their previously stated preferences. In this recent paper, subject preferences were reversed between tastes of jam and scents of tea. Overall, only a third of all the manipulated trials were detected by subjects whose preferences had been switched by the experimenters.

Abstract:
We set up a tasting venue at a local supermarket and invited passerby shoppers to sample two different varieties of jam and tea, and to decide which alternative in each pair they preferred the most. Immediately after the participants had made their choice, we asked them to again sample the chosen alternative, and to verbally explain why they chose the way they did. At this point we secretly switched the contents of the sample containers, so that the outcome of the choice became the opposite of what the participants intended. In total, no more than a third of the manipulated trials were detected. Even for remarkably different tastes like Cinnamon-Apple and bitter Grapefruit, or the smell of Mango and Pernod was no more than half of all trials detected, thus demonstrating considerable levels of choice blindness for the taste and smell of two different consumer goods.
posted by scrutiny (31 comments total) 25 users marked this as a favorite

 
The two-at-a-time thing is messing me up, but luckily there's this:
...showed male subjects a pair of female faces. The subjects were asked to choose the face that they found more attractive. Then, the mischievous scientists used a “card trick” to reverse the outcome of the choice. Here’s where the results get a little sad: Less than 30 percent of subjects noticed that their choice had been changed. Our eyes might have preferences, but this doesn’t mean our mind can remember them.
That's not what this shows. If you just couldn't remember, it would be 50%. But the subjects are showing an actual aversion towards their original choice. So maybe they are responding to novelty more than forgetting what they liked.
posted by DU at 9:34 AM on August 26, 2010 [3 favorites]


Chances are, to most of these people it is completely unimportant which jam or tea or woman's face they choose. But ask them if they prefer (for example) the Democratic Party of the Republican Party, and you would not find that their choices would be so easily reversible.
posted by grizzled at 9:38 AM on August 26, 2010 [3 favorites]


responding to novelty more than forgetting what they liked.

That's definitely a possibility. I was thinking that the preferences of 2/3 of the subjects just weren't that strong to begin with so they defended whatever their choice happened to be, assuming it was the one they had made. It just really surprises me (it probably shouldn't) that people can be so inconsistent in their choice.
posted by scrutiny at 9:41 AM on August 26, 2010


This FPP is nowhere near as good as this one is.
posted by Eideteker at 9:43 AM on August 26, 2010 [17 favorites]


This is certainly true of editors. I routinely only make about 50% of requested changes. On the next revision, they almost never notice this, but they will find other things they now want changed that they thought were fine the previous time around or, yes, reversing an earlier decision.

Repeat as needed until one side is exhausted or deadline looms.

So yes, novelty, and a desire to appear decisive.
posted by Herodios at 9:49 AM on August 26, 2010 [7 favorites]


But the subjects are showing an actual aversion towards their original choice.

No, they're acquiescing to the researchers' deceptive indication of what their original choice was.

This FPP is nowhere near as good as this one is.

That's what I've been saying the whole time!
posted by Horace Rumpole at 9:51 AM on August 26, 2010 [2 favorites]


This is certainly true of editors. I routinely only make about 50% of requested changes. On the next revision, they almost never notice this, but they will find other things they now want changed that they thought were fine the previous time around or, yes, reversing an earlier decision.

As an editor, I can say this is definitely true. Of course, that's because English is a funny language, and depending on your state of mind, different solutions to the same problem can seem better at any given point in time. Also, it's impossible to notice every single passage that could be improved, and even if you do, you're usually just trying to hit the things that seem most in need of fixing on any given revision pass.

And as a writer, and having known a lot of writers in my time, if you're claiming you don't do the same thing when doing your own revisions, you're the only one.
posted by Caduceus at 10:06 AM on August 26, 2010


So maybe they are responding to novelty more than forgetting what they liked.

I suspect it's less about this than about the power of narrative to shape perception/experience.

The people in question have made a choice, have been asked to create a narrative about their choice. The human tendency is probably to let the story shape the experience. There's a lot of power in that, actually, because it can help you reconfigure negative feedback or ignore quick gratification impulses into the service of a greater goal. It also, however, lets you create a rigid value system where new inputs are attenuated, which can mean you don't adapt quickly to new feedback.

It's probably also worth pointing out that in these experiments, it's not only self-reinforcement going on. The researchers are also more or less encouraging them to apply their constructed narrative over the supposed preference. That's an interesting experiment in itself, but it's worth considering that the results are probably pushed towards fewer people recognizing what's going on because of this.

I was thinking that the preferences of 2/3 of the subjects just weren't that strong to begin with so they defended whatever their choice happened to be

I think this is also a good explanation, and dovetails with my own.

they will find other things they now want changed that they thought were fine the previous time around or, yes, reversing an earlier decision.

I do this all the time with my own work... trying something, seeing if I like it, changing it back when I don't, realizing after starting at something for a long time that something I thought was OK is actually a problem. Isn't this the normal refining process?
posted by weston at 10:06 AM on August 26, 2010 [3 favorites]


Hey wait, this isn't the FPP I favorited!
posted by Kabanos at 10:16 AM on August 26, 2010


In some way, I've noticed that with myself. I have a set of "favorites" and, honestly, sometimes I'm not even sure why they've made the cut.

I like them, I like them enough to go back to them, and it's an easy answer when people ask or when I need to make a decision. Otherwise, I'd probably just want to try everything before I make an informed choice.

Because of that, I could definitely see myself falling into a trap like that and realize why others do, too. I think it's just a matter of making a decision rather than actually having a strong preference.
posted by inmediasres at 10:18 AM on August 26, 2010


grizzled: Chances are, to most of these people it is completely unimportant which jam or tea or woman's face they choose. But ask them if they prefer (for example) the Democratic Party of the Republican Party, and you would not find that their choices would be so easily reversible.

Choice blindness can be a huge factor in politics, actually. If you oppose deficit spending and corporate bailouts but you want your kids to go to a good school then voting Republican is pretty much the antithesis of your previously stated preferences, and yet…
posted by paisley henosis at 10:42 AM on August 26, 2010 [8 favorites]


"This FPP is nowhere near as good as this one is."

What a moronic comment! What kind of moron made such a moronic comment?

Oh.
posted by Eideteker at 11:02 AM on August 26, 2010


isn't it possible that in the taste tests, the flavor combinations are made up of individual elements that can each occur in both formulas? That is, I thought at first that if you gave me two different teas, one tasting of mango and the other tasting of pernod, I could damn sure tell the difference and remember it.

But OTOH, there are like a hundred different kinds of mango, each one with a slightly different flavor. And when you're analyzing two items for comparison, you tend to try to pick out distinctive notes or ingredients in each. And maybe, on being given the deceptive second tasting, I might think to myself, "Gee, it's funny how much these two different things really taste kind of alike, or how there's that unexpected pernod element contributing to this mango flavor."

You know, like when you actually read the marketing label on the wine bottle and then try to actually identify all the elements (smoky, currant, peach etc) while you're drinking?

So I don't know if it's a case of not knowing what you like as it is a case of being asked to examine something so closely that it's hard to see what it really is.
posted by toodleydoodley at 11:08 AM on August 26, 2010


What a great experiment. I bet people get tripped up because the testers separate out the choosing part from the explaining part. Part one, try these jams and choose your favorite. Mhm, it's good jam, isn't it? Lovely day day for jam. Part two, tell me why you like this here jam, which I'm now giving you some more of.

At that point, people aren't considering what jam they're tasting, just trying to think of reasons why they like it because they have a pressing question to answer and, hey, this is your favorite jam, remember?
posted by katerschluck at 11:14 AM on August 26, 2010 [3 favorites]


But ask them if they prefer (for example) the Democratic Party of the Republican Party, and you would not find that their choices would be so easily reversible.

You might think that, but there was a study I'm unfortunately having trouble finding that shows many people who identify as Republicans also choose Democratic positions as their own on a majority of issues. I don't think the study specifically tested your hypothesis but it seems likely that given a fair, informed choice many Republicans would switch affiliation. The root problem seems to be that the Republican Party has a knack for political deception & uses it to bolster their rolls.
posted by scalefree at 11:47 AM on August 26, 2010 [1 favorite]


I've also seen the experiment where they will swap out the person they are talking to with someone else and the subject doesn't notice. You don't expect the jam to have changed so you discount that and carry on as normal.
posted by zeoslap at 11:59 AM on August 26, 2010


Research is predicated on the idea that people are or should be predictable.

That's the religious belief at the core of sciences having to do with people.

People have a right to be unpredictable!

Researchers also seem to believe that researching people is a neutral activity that in no way impacts or helps to construct the interaction that takes place during the research.
posted by vitabellosi at 12:14 PM on August 26, 2010


A foolish consistency is the something of something something.
posted by Astro Zombie at 12:18 PM on August 26, 2010 [1 favorite]


(Stares at mashed potatoes) DAMMIT! THIS MEANS SOMETHING!
posted by Trochanter at 12:32 PM on August 26, 2010 [2 favorites]


Researchers also seem to believe that researching people is a neutral activity that in no way impacts or helps to construct the interaction that takes place during the research.

This gives me a great idea for an experiment about locking people in boxes filled with poison gas.
posted by Copronymus at 12:39 PM on August 26, 2010


I've also seen the experiment where they will swap out the person they are talking to with someone else and the subject doesn't notice. You don't expect the jam to have changed so you discount that and carry on as normal.

The former is called "change blindness" and I think you've hit a good point there. Is what this experiment uncovering really choice blindness? Or is it just change blindness using the taste sense rather than the visual sense?
posted by storybored at 1:19 PM on August 26, 2010


zeoslap: I've also seen the experiment where they will swap out the person they are talking to with someone else and the subject doesn't notice. You don't expect the jam to have changed so you discount that and carry on as normal.

Professor Daniel J. Simons has the original video of the "door" experiment up on the website for his book, the Invisible Gorilla (scroll down for the "door" video, but the others are also worth watching). Here's another Simons video on Change Blindness.
posted by dilettanti at 4:09 PM on August 26, 2010


I once had a boss who argued with me whether a text I was showing him was in font size 14 or not (it certainly was, because he always wanted it that way and I had set it as default on my computer). After a short argument (polite on my side, very shout-y and angry on his side), I went back to my room and printed the same document again without changing the font size. This time, he accepted it quietly.

If he hadn't marked a big red cross on the first copy, I wouldn't have bothered printing it again.
posted by vidur at 5:25 PM on August 26, 2010 [1 favorite]


One of the admins should make the link on the main page go to a totally different discussion instead.
posted by brenton at 5:26 PM on August 26, 2010


I'm curious to know the sorts of explanations and comparisons the subjects gave. I imagine a preference would be based on things like comparative sweetness, sourness, or identifiable flavors. I wonder if the tricked subjects described their initial choice accurately but failed to notice the difference, described the current product accurately instead, or if they all just gave non-distinct responses.
posted by luftmensch at 6:08 PM on August 26, 2010


People often fail to understand the consequences of the choices that they make, which is why a person who opposes deficit spending might still vote Republican; it's not exactly the same thing as choice blindness, although it does deserve to be called blindness of some kind, perhaps consequence blindness. But the point remains, if a person has a strong preference for something, such as for the Republican Party in preference to the Democratic Party, they are not going to carelessly reverse that preference without even realizing what they are doing. It would take great soul-searching before such allegiance is reversed.
posted by grizzled at 6:01 AM on August 27, 2010


(Stares at mashed potatoes) DAMMIT! THIS MEANS SOMETHING!

Yeah, it means someone has swapped out your plate of beans without you noticing.
posted by logopetria at 6:03 AM on August 27, 2010 [1 favorite]


If the parameters of the experiment were changed to make the cost of a choice very dear, perhaps more subjects would notice the switch.

Spend 2000USD on a vacation package and get sent to the opposite side of the globe. See if you notice.

otoh, make a reservation for a flight and you might not notice if the airline is swapped out when your itinerary shows up in your inbox.

I wonder if a surgeon would have choice blindness?

a programmer who calls for a code review and has the code switched?

someone on a dating site sending a message to a prospect and getting a reply back from a different account?

ideas:
vary the amount of cognitive engagement to make the choice more memorable. alter the experiment to introduce interventions to improve perfomance on recognition tasks. vary the novelty of the situation or the choices (something a person does everyday vs twice a year...)
posted by bleary at 7:12 AM on August 27, 2010


back in school I recall people working on experiments to affect perfomance on recognition tasks. compare to experiments where the task is to test recall.

so I would expect the choice blindness results to alter if the blindness is in part due to poor recogniton task performance.
posted by bleary at 7:15 AM on August 27, 2010


Full text of paper at co-author page.

ps. why do I always get to these FPP after everyone is already done discussing them. feh.
posted by bleary at 5:35 PM on August 27, 2010


Reminds me of a story I've told before...

I used to work at a store in a mall. My store opened up to a small court where many years ago, the Pepsi Challenge set up shop one busy weekend.

The Pepsi Challenge was marketed by Pepsi to prove that people just really preferred the taste of Pepsi over Coke. People are invited to taste two unmarked samples of cola and asked to choose which taste they preferred.

At one point, I had some free time and was watching one woman take "the Challenge". She made her cola choice and it was revealed that she had picked Coke. Her response was: "Oh, I was wrong!"

It then occurred to me that what she might have been doing was trying to pick which cola was Pepsi, not which taste she preferred. Then I wondered if that was what most people were (unconsciously?) challenged to do Hmmm...wouldn't that skew Pepsi's results because, most people can tell the difference, especially experienced cola drinkers, as most Americans are.

So, the next day I had a little time to take this challenge myself. I sat down and was invited to taste the two samples. I was asked which cola taste I preferred. I chose "No preference". (Because I really have no preference.) The Pepsi person let me know that I was the first person to choose "No preference".
posted by jaronson at 7:54 AM on August 28, 2010


« Older There are 100 ways to love a cat....  |  THQ's Cory Ledesma opened the ... Newer »


This thread has been archived and is closed to new comments