Miguel Roig wrote:
> At 07:11 AM 5/7/99 -0700, you [me, Jeff Ricker] wrote:
> > In 1954,
> >Dement quickly trained another medical student to identify REM tracings
> >on a polygraph and served as the subject to show the student the
> >connection between REM and dreaming. Dement was awakened five times
> >during the night but could remember no dreams. On the fifth awakening,
> >"I was so embarassed and upset, I lied. I haltingly produced a phony
> >dream fragment" (p. 293). He was "overjoyed and incredibly relieved"
> >when he discovered that the student, by mistake, had awakened him only
> >during NREM. This seemingly confirmed for Dement the REM/dreaming
> >connection:: "If anyone wants to claim that subject or experimenter bias
> >plays a role in the REM sleep-dreaming relationship, I could not have
> >been more biased toward recalling a dream, and I was utterly unable to
> >dredge up even a wisp of one." (p. 294)
>
> Is this really a case of experimenter bias?
Since Dement was the subject being observed, one might say that this example
better illustrates the influence of demand characteristics (I think this is
what Dement means by the term "subject bias"): because of the cues present
in the research situation, the subject (Dement) understood what responses
were expected of him. Being a cooperative subject, he gave the researcher
(also Dement) those expected results! On the other hand, because Dement was
the researcher who was influencing the research situation in such a way that
the expected observations would be more likely to occur, a case might be
made that experimenter bias also was occurring: the researcher (Dement)
influenced the results of the study by communicating to the subject (also
Dement) that the recall of a dream was expected, and the subject complied.
And that is Dement's point: even though he knew that he was supposed to
recall dreams, Dement still wasn't able to. Therefore, regardless of what
one wants to call it, this example definitely shows that there was a serious
biasing influence present in the study, but the results were not affected by
that influence (although the subject did report a dream, the researcher was
able to show that it was a false report--I'm confused).
Nevertheless, none of this is the most important problem. The most important
problem is the anecdotal nature of this example. That is, this was not much
more than a personal experience (it definitely was NOT an observation made
in a well-controlled research situation that was compared systematically
with a set of other similarly made observations). The fact that this
experience seemed to have (if we are to trust Dement's words) a very
important influence on Dement's thinking about the connection between REM
sleep and dreaming is what bothers me.
Feeling decidedly demented,
Jeff
--
Jeffry P. Ricker, Ph.D. Office Phone: (602) 423-6213
9000 E. Chaparral Rd. FAX Number: (602) 423-6298
Psychology Department [EMAIL PROTECTED]
Scottsdale Community College
Scottsdale, AZ 85250
"For every problem, there is a solution that is neat, simple, and
wrong." H. L. Mencken
