Dewhurst,
S.A. (2001). Category repetition and false recognition: Effects of instance
frequency and category size. Journal of Memory and Language, 44,
153-167.
This study uses a category repetition
paradigm to investigate the subjective experience of false memories.
In a category repetition paradigm category exemplars are presented at study
(e.g. orange, banana, kiwi) with the finding that non-presented category
members will sometimes be falsely recognized (e.g. apple).
The subjective experience of false
memories produced in this paradigm was measured using the remember/know
judgment task. In the remember/know task subjects are told
to respond remember if their memory for the item is accompanied
by the conscious recollection of specific details of the item's presentation
(e.g. sensory qualia, where the item occurred on the list, thoughts the
subject had upon encountering the item, etc.). Subjects are told to respond
know if they are sure the item was presented, but they can't recall
anything specific about its presentation.
Studies using a variety of false
memory paradigms have found that false memories can sometimes be experienced
in the remember sense (see our review papers on the issue Lampinen,
Neuschatz & Payne, 1998; Payne, Neushatz, Lampinen & Lynn, 1997--yes,
yes, shameless self promotion, I know). One possible explanation
for this finding is that false remember judgments occur because
of associations made at encoding. So the purpose of the current study
was to determine whether encoding factors are responsible for false remember
judgments. In particular, the issue is whether one get's a large
number of false remember judgments because the lures are spontaneously
generated during encoding.
Experiment 1
Method
Visual presentation of items
from 36 categories at a rate of one word every 2 seconds (1 s + 1 s ISI).
-
For one third of the categories, only
1 exemplar was presented
-
For one third of the categories, 4
exemplars were presented
-
For one third of the categories, 8
exemplars were presented
Test included 2 exemplars from each
category, one that had been presented (Target) and one that had not been
presented (Lure). Subjects made old/new judgments as well "remember",
"know" and "guess" responses for those items they called "Old".
Results
-
Category repetition increased the proportion
of correct know judgments.
-
Category repetition increased the proportion
of false remember judgments and false know judgments, but
the effect was somewhat greater for false know judgments.
-
One interpretation of these findings
is that false remember responses occur because subjects generate
the lure at time of study resulting in an episodic trace.
Experiment 2
Purpose of this experiment was to manipulate
the typicality of the lure. The idea is that high frequency lures
are more likely to be generated at time of study than are low frequency
lures. So high frequency lures should be experienced in the "remember"
sense more often.
Method
Similar set up to the previous experiment,
except that:
-
Before each list the category label
was presented
-
Six exemplars were presented from each
of twenty five categories
-
Each list contained a high and low
frequency target, and four medium to high frequency fillers
-
Test included four exemplars from each
category
-
High Frequency Target
-
Low Frequency Target
-
High Frequency Lure
-
Low Frequency Lure
-
As before, Remember, Know, and
Guess judgments were made in addition to Old/New recognition.
Results
-
More remember judgments for
low frequency than high frequency targets.
-
More know judgments for high
frequency than low frequency targets.
-
More guesses for high frequency
than low frequency targets.
-
More false remember judgments
for high frequency than low frequency lures.
-
More false know judgments for
high frequency than low frequency lures.
Experiment 3
In this experiment category size is
manipulated. The idea is that people generate members of small categories
faster than members of large categories. So if false remember
judgments occur because the lure is generated at encoding they should be
more likely for small categories than for large ones.
Method
Like the previous experiments except
that
-
Thirty six item lists were used, half
the lists representing small categories and half the lists representing
large categories.
-
Test included a target and a lure from
each list.
Results
-
More false remember judgments
for lures from small categories.
-
More false know judgments for
lures from small categories.
Discussion
One way of interpretting these results
is that subjects spontaneously generate the lure item at time of study.
This is consistent with the finding that more false remember responses
occur with longer lists and with smaller categories.
Another interpretation discussed
is Fuzzy Trace Theory's explanation in terms of gist and verbatim.
The author argues that the effect on targets can be explained in this way,
but not the effect on false remember judgments (see Brainerd, Wright,
Reyna & Mojardin, 2001 for FTT's account of false remember judgments).
Another interpretation is the familiarity
plus corroboration account (or what we've also termed 'content borrowing').
According to this account strong feelings of familiarity lead to the borrowing
of experiential content from presented items. Dewhurst argues that
both generation of the critical lure at time of study and content borrowing
are likely mechanisms for producing false remember judgments and
that both processes could co-occur (something we agree with in this lab).