On Mon, 13 Dec 1999, George Bailey wrote:
> Would you exptect to get the same results if the question were: which one
> looks better? - George
>
> >When the
> >two images are presented, the subjects will tell you that they don't
> >recognize either of them. Yet when choosing a prefered image, they almost
> >always choose the one that had been presented in a masked format.
Well, that's a subtle point, and I'm not sure it's germane. Your average
person would wonder why you were splitting between which they preferred
and which they thought looked better. "Why would I prefer it if I didn't
think it 'looked better'?"
Wait. Let me reverse myself. It could be quite germane to the
experiment.
Working with human subjects is notoriously difficult because people lie.
If asked which figure looked better, the subject might be trying to
conform to whatever they think is the "right" answer. They may try to
shape their answers toward what they think is supposed to look better.
I prefer my hair to be an electric shade of royal blue, but I've conformed
to the notion that my natural coloring "looks better" based on social
cues. Similarly, subjects in psychology experiments often cue off the
experimenter's unconscious signals for the 'correct' answer.
However, the way the experiment I described was done, the image pairs
were very similar -- two kanji, two finiels, two different star patterns.
They were designed to be distinguishable, but not contrasting, and were
presented without the experimenter in view. Certainly some concerns will
over-ride simple familiarity. Were the choices a sad-eyed puppy vs a
Vermeer portrait the interpretations might be a bit different. Certainly
personal taste would influence the answer in that case more than simple
pre-presentation of one of the images.
Peg.
M. S. AtKisson
Department of Neuroscience
Tufts Medical School
Boston, MA
==============================================================
Post Message: Send a message to artwithbraininmind-l@pks.bu.edu
Web Archive: http://pks.bu.edu/artwithbraininmind-l
Unsubscribe: Send message to majordomo@pks.bu.edu with
'unsubscribe artwithbraininmind-l' in body.
This archive was generated by hypermail 2b30 : Fri Jul 04 2008 - 04:03:02 EDT