Bernstein, D. M., Whittlesea,
B.W. A., Loftus, E. F. (2002). Increasing confidence
in remote autobiographical memory and
general knowledge: Extensions of the revelation
effect. Memory and Cognition, 30, 432-438.
What results in a person feeling as though an event occurred
in the past? This fundamental question underlies the research conducted by
Bernstein et al. In particular, these researchers investigated the ability of
the revelation effect to increase beliefs about the occurrence of childhood
events and the truthfulness of general knowledge.
Experiment 1
Participants are trained how to unscramble anagrams. This is
important because many college undergraduates find this to be somewhat
difficult, and the researchers would like for them to unscramble as many of
these anagrams as possible. Therefore, participants are provided with rules for
how to unscramble the anagrams. Anagrams were presented within sentences (e.g.,
went to the umoanitsn). After solving the anagrams,
participants were asked to rate their belief that the event had happened to
them in the past on an 8 point scale. Fifteen such sentences were presented
during training.
The main test phase was comprised of 52 statements that were
more constrained than the statements used during training. Thirty of these
statements were taken directly from the Life Events Inventory (hyperlink to
related articles). Half of these statements contained a word that was
scrambled, while the other half did not. Like in the training phase,
participants were asked to rate their level of belief that the event in
question had happened to them in their past. Participants held higher beliefs
for the events that contained anagrams (M
= 4.17) than for those events that did not (M
= 3.89).
These results extend the revelation effect to remote
autobiographical memories. In the second experiment they attempt to further
extend the revelation effect to general knowledge (i.e., semantic knowledge).
Experiment 2
Participants are presented with general knowledge statements.
Half of the statements are completed truthfully and the rest are completed erroneously.
Half of the critical words are intact, and half are scrambled. Will
unscrambling the word make participants more likely to state that true and
false statements are true? They found that participants were indeed more likely
to contend that the statement was a factual when the word was scrambled opposed
to intact.
These results extend the revelation effect to beliefs in the
accuracy of general knowledge.
Experiment 3
This experiment used the same set up as experiment 2, expect that the anagrams were unrelated to the context or
validity of the statement. Even under these conditions, solving an anagram
increased the rate at which participants categorized truthful statements as
being true. However, the effect no longer held for false statements.
These authors argue against an activation account of the findings presented here. Familiarity arising from activating the exact or a related representation cannot account for these findings. Instead, the authors suggest that their findings resulted from a sense of discrepancy experienced on the part of the participant. The person can easily solve the anagram because of the rules that they were taught in the training phase and in the case of experiments 1 and 2 the context in which the statements are presented. However, instead of attributing the ease of processing to these aspects of the environment, they inappropriately attribute them to memory.