In Defense of Eye Candy

[ñͤͣ̄ͦ̌̑͗͊͛͂͗ ̸̨̨̣̺̼̣̜͙͈͕̮̊̈́̈͂͛̽͊ͭ̓͆ͅé ̰̓̓́ͯ́́͞]

Stephen P. Anderson makes some very counter-intuitive points about the relationship between aesthetics and functionality. A ton of high quality references to the specific studies that back up his arguments separate this article from the reams of drivel that can be found floating around the more visual areas of design.

Researchers in Japan set up two ATMs, "identical in function, the number of buttons, and how they worked." The only difference was that one machine's buttons and screens were arranged more attractively than the other. In both Japan and Israel (where this study was repeated) researchers observed that subjects encountered fewer difficulties with the more attractive machine. The attractive machine actually worked better.

So now we're left with this question: why did the more attractive but otherwise identical ATM perform better?

Norman offers an explanation, citing evolutionary biology and what we know about how our brains work. Basically, when we are relaxed, our brains are more flexible and more likely to find workarounds to difficult problems. In contrast, when we are frustrated and tense, our brains get a sort of tunnel vision where we only see the problem in front of us. How many times, in a fit of frustration, have you tried the same thing over and over again, hoping it would somehow work the seventeenth time around?

Another explanation: We want those things we find pleasing to succeed. We're more tolerant of problems with things that we find attractive.

Source

[Cainad (dec.)]

Well that pretty much just explained the reason for the success of Apple computers.

[Bruno]

Well that pretty much just explained the reason for the success of Apple computers.

And all Steve Jobs had to do to "get it" was eat a bunch of acid.

[Cainad (dec.)]

Well that pretty much just explained the reason for the success of Apple computers.

And all Steve Jobs had to do to "get it" was eat a bunch of acid.

Worse things have been done in pursuit of becoming filthy rich. Good on 'im.

[Triple Zero]

"more attractively" or "more in line with gestalt theory"?

I'd like to compare the designs of those machines myself.

[ñͤͣ̄ͦ̌̑͗͊͛͂͗ ̸̨̨̣̺̼̣̜͙͈͕̮̊̈́̈͂͛̽͊ͭ̓͆ͅé ̰̓̓́ͯ́́͞]

"more attractively" or "more in line with gestalt theory"?

I'd like to compare the designs of those machines myself.

I thought you said, "I'd like to compare the designs of those machines to myself." Well, I hate to break it to ya, but...

Yeah, how they defined "more attractively" and what that was comprised of is very interesting. I think principles of gestalt theory can be present in a design without it being necessarily attractive though. That certainly seems like an important conflating factor. If I weren't mired in Schoolesian Duality, enjoying my girlfriend and a full time job playing with fonts I'd take a much harder look at their experimental design.

I think he makes a decent case against the popular, knee-jerk sour grapes reaction of "OMG it's beautiful, must be useless/vapid/crass marketing." I'm also curious as to how much of this generalizes to other areas such as book design and so on.

I just edited the OP to make the link to the article more obvious, in case you missed where I put it.

[Triple Zero]

ah! hahaha I was just about to link you cause I remembered "hm there was an article on ALA about just this topic"

anyway, the part where they said they rearranged the buttons to appear more attractively rang some alarm bells with me. if they rearrange the buttons, this has a profound impact on usability, regardless (more or less) of aesthetics.

and indeed this appears to be the case if you inspect the layouts of the two machines as shown in the paper: http://sigchi.org/chi97/proceedings/paper/nt.htm#U8
from what I can see, they skewed the grid of buttons, took one button out of the grid entirely and reversed the numbers.

dont get me wrong, I still think it's a useful experiment, but this merely proves that layout has an effect on usability. in order to purely measure the effect of aestethics on usability, I'd rather see an experiment where the layouts are equal, but one machine has a more pleasant colour scheme, the buttons appear more "rich", smooth, rounded, it's cleaner, etc. whereas the other will have sharp edges, bolts, default factory metal colours, squeaky cheap dated plastic buttons and hints of rust. maybe even give one a high-res colour display and the other a lower-res monochrome one, as long as the icons and imagery on the monochrome one are clear and easy to recognize.

[Reginald Ret]

Wouldn't the low res one work faster?

easily compensated ofcourse.

interesting.

[Template]

ah! hahaha I was just about to link you cause I remembered "hm there was an article on ALA about just this topic"

anyway, the part where they said they rearranged the buttons to appear more attractively rang some alarm bells with me. if they rearrange the buttons, this has a profound impact on usability, regardless (more or less) of aesthetics.

and indeed this appears to be the case if you inspect the layouts of the two machines as shown in the paper: http://sigchi.org/chi97/proceedings/paper/nt.htm#U8
from what I can see, they skewed the grid of buttons, took one button out of the grid entirely and reversed the numbers.

Yes, the second layout seems to represent "strikingly unfamiliar" as much as it represents "aesthetically poor".  The layouts in both cases should probably be the same, be they traditional or novel.  Ugly and incomprehensible are different!

[Golden Applesauce]

ah! hahaha I was just about to link you cause I remembered "hm there was an article on ALA about just this topic"

anyway, the part where they said they rearranged the buttons to appear more attractively rang some alarm bells with me. if they rearrange the buttons, this has a profound impact on usability, regardless (more or less) of aesthetics.

and indeed this appears to be the case if you inspect the layouts of the two machines as shown in the paper: http://sigchi.org/chi97/proceedings/paper/nt.htm#U8
from what I can see, they skewed the grid of buttons, took one button out of the grid entirely and reversed the numbers.

Yes, the second layout seems to represent "strikingly unfamiliar" as much as it represents "aesthetically poor".  The layouts in both cases should probably be the same, be they traditional or novel.  Ugly and incomprehensible are different!

Also note that on the poorer performing interfaces, the word/symbol keys are split up - they're all together in the "more aesthetic" one, which would seem to make it more usable right there.  "Hey, what fuckhead decided to hide the **** key above the 8?"

[ñͤͣ̄ͦ̌̑͗͊͛͂͗ ̸̨̨̣̺̼̣̜͙͈͕̮̊̈́̈͂͛̽͊ͭ̓͆ͅé ̰̓̓́ͯ́́͞]

I'd question the division between aesthetics and layout though.

Imagine someone you've fapped to on multiple occasions with their eyeball and armpit swapped positions.

"Layout" is the design way of saying "composition" in art terms, it just sounds more technical and devoid of importance that way. It's an obvious, duh, kind of study, but a lot of research doesn't support what we think is so obvious. I think there was some psychologists who study this kind of reaction to studies, where half of the people were told the study went one way and the other half were told it went another way and most people in both groups were all, "Duh, that's obvious."

The mismatch between how people say they attribute credibility is very interesting as well:

According to a 2002 study, the "appeal of the overall visual design of a site, including layout, typography, font size, and color schemes," is the number one factor we use to evaluate a website's credibility.

[Triple Zero]

okay now i can't think of anything else except fucking Angelina Jolie's unshaven eyesocket ...

[LMNO]

Oh, that's HAWT.

[Triple Zero]

heh

"eye candy".

[ñͤͣ̄ͦ̌̑͗͊͛͂͗ ̸̨̨̣̺̼̣̜͙͈͕̮̊̈́̈͂͛̽͊ͭ̓͆ͅé ̰̓̓́ͯ́́͞]

It had to be done.