film, ask subjects to visualize the text, and present the memory test the next
day. Source memory also tends to be poor if the misleading suggestions con-
tained in the text are repeated several times rather than presented in the text
only once (Zaragoza & Mitchell, 1996).
One other paradigm (see figure 14.4) that suggests that memory really is
affected by subsequently presented misinformation comes from Lindsay (1990;
also Weingardt, Loftus, & Lindsay, 1995; see Lindsay, 1993). Again, say that
subjects see a yield sign in the film and later read a text about a stop sign.
The description of a stop sign is misinformation. Now subjects are correctly
informed that the text did not contain any correct answers to a subsequent
memory test. In the memory test, subjects are asked to report details about the
traffic sign, and subjects know that the correct answer comes from the film and
not from the text. When it is hard to discriminate between the film and text
experiences, subjects are likely to recall incorrectly details from the text, such as
that the traffic sign was a stop sign. Such incorrect recall occurs less often when
subjects are asked to recall details from the film about which no misinformation
was given in the text. Here the demand characteristics of the experiment un-
ambiguously push subjects to recall only the information in the film, yet they
frequently recall inaccurately the information in the text.
I went through these experimental paradigms in some detail because I want
you to see exactly how researchers refine their paradigms in response to alter-
native interpretations of findings. In this case, we can say that misinformation
Figure 14.4
Experimental paradigms to investigate the impact of misleading information on eyewitness
memory.
336 R. Kim Guenther