Date: Sat Nov 20 1993 00:51:00
From: John Powell
Subj: Memory Tricks
Diagnoses of Alien
Kidnappings That Result
From Conjunction Effect
by Robyn M. Dawes and Matthew Mulford
(The Skeptical Inquirer, Vol. 18, No. 1, Fall 1993, Copyright 1993 by
the Committee for the Scientific Investigation of Claims of the
Paranormal, 3965 Rensch Road, Buffalo, NY 14228, published quarterly
with a membership/subscription rate of $25/yr.)
Events and feelings may be better recalled when they occur in
combination than singly, to the point that a conjunction of two alleged
events or feelings may be iudged to have occurred with greater frequency
in one's life than one of them alone. One part of a coniunction can
facilitate recall of the conjunction, and hence of another part of the
experience - and combinations of events can be judged to be more
probable than their components (Tversky and Kahneman 1983). The observer
to whom it is reported, however, knows that such a coniunction is
necessarily less probable than any one of its components. Thus, the
observer may attach special significance to such a conjunction.
For example, in supporting a conclusion that post-traumatic
stress from kidnapping by aliens is a major mental-health problem in
this country (allegedly affecting at least 2 percent of the population),
Hopkins and Jacobs (1992) cite the rate of affirmative responses to a
recent Roper Poll question: "How often has this happened to you: Waking
up paralyzed with a sense of a strange person or presence or something
in the room ?" Their rationale for considering affirmative responses
particularly diagnostic of alien kidnapping involves the conjunction of
the two components in the question: "A fleeting sensation of paralysis
is not unusual in either hypnogogic or hypnopompic states, but adding
the phrase 'with a sense of a strange person or presence in the room'
forcefu11y narrows the scope of the question" (p. 56).
As part of a (much) larger study, we asked that same Roper Poll
question of 144 subjects (mainly University of Oregon students and some
townspeople interested in the $20 pay for two hours). Forty percent
answered that this had happened to them at least once. A randomly
selected control group of 144 subjects in the same study were asked
simply how often they remembered waking up paralyzed. Only 14 percent
answered that this had happened to them at least once, (chi-square =
24.26; p < .001, phi = .29). (See Table 1.) The contingency was stronger
for women (phi = .44) than for men (phi = .17), significantly so
according to a Goodman-Plackett chi-square value of 4.74. Nevertheless
it was significant for both sexes (with chi-squares of 25.38 and 4.43,
Thus, due to a conjunction effect in memory, the added phrase
"with a sense of a strange person or presence . . . in the room"
actually "broadens the scope" of the question, rather than narrowing it.
Hopkins and Jacobs are, of course, correct in maintaining that the
additional phrase _should_ "narrow the scope." It's just that the phrase
doesn't. What they have discovered, therefore, is not evidence of alien
kidnappings, but of a common irrationality in the way we recall our
Hopkins, B., and D. M. Jacobs. 1992. "How
This Survey Was Designed." In Bigelow
Holding Company, _An Analysis of the
Data from Three Maior Surveys Conducted
by the Roper Organization_, pp. 55-58.
Tversky, A., and D. Kahneman. 1983.
Extensional versus intuitive reasoning:
The conjunction fallacy in probability
judgments. _Psychological Review_, 90: 293-
Robyn M. Dawes is University Professor in the Department of Social and
Decision Sciences, Carnegie Mellon University, Pittsburgh, PA
15213-3890. Matthew Mulford is in the Department of Political Science,
University of Oregon, Eugene 97403.