undefined | Better HN

0 pointsvacri10y ago0 comments

I'm a bit waffle-ey by nature, but really I mean two points: first, that the icon means save because we're trained to see it as the icon for save (if we want to talk unnecessary complexity, you've transformed my 'trained' into 'stimulus generalisation' :D); and second, kids know that the icon is related to floppy disks because everyone asks at some point "wtf does this have to do with 'save'", and gets a history lesson. Two separate but related events.

The unnecessarily complexity is what's in the video: the user sees the icon, recognises it as a floppy disk, understands that the floppy disk is storage from some prior experience (ha![1]), and then reverse-engineers the question "why does this icon look like a floppy disk" into "something to do with storage - probably saving the document to it". It's a very convoluted and unnecessary cognitive path.

[1] Does anyone really believe that the youth of today learned what floppy disks were before they learned about saving documents?

0 comments

3 comments · 1 top-level

travjones10y ago· 2 in thread

Hahaha I don't think putting stimulus generalization in parentheses makes my explanation any more complex. I explained the phenomenon in 2-3 simple sentences and then I used two words that could substitute for those 2-3 sentences--stimulus generalization.

If you compare our two analyses some of the claims you make cannot be tested via experimentation, such as "convoluted unnecessary cognitive path." I'm not sure what that means, how we would determine if the user has gone down that route, or how we would measure the convolution of a "cognitive path."

Nonetheless, I like this discussion. I think UX is really cool and represents a great application for applied behavior analysis.

vacriOP10y ago

That 'convoluted cognitive path' is a qualitative assessment of what the presenter is describing in the seminar. You don't need some measure of convolutedness to see it touches a number of different cognitive aspects.

But if it's robust testing you're after, then you should be railing against the seminar altogether. A self-selected, self-reported study done on the internet, with zero way to identify demographic accuracy? An assumption that puerility like 'anus@anus.anus' are only done by teenagers? An assumption that "it's really easy to tell when someone's messing with you", which just means you can't detect the subtler trolls? This isn't experimentation, it's a survey, and it's been interpreted with a lot of cognitive bias, with the researcher projecting a lot of assumed meaning. It's certainly better than nothing, but it's not robust at all.

travjones10y ago

First paragraph: You're still making major assumptions.

Second paragraph: I concur, sir!

j / k navigate · click thread line to collapse