Brainerd, C.J, & Reyna, V.F. (1998). When things that were never experienced are easier to "remember" than things that were. Psychological Science, 9, 484-489.

Background Information:

Theory based:

Based on Fuzzy trace theory Brainerd and Reyna suggest that targets should be remembered by verbatim memories and that distractors should be falsely remembered due to gist memories. This also led them to the assumption that over time verbatim memories should decrease and gist memories should constant leading to increased distractor acceptances and decreased target acceptances. They propose to use a meaning recognition test to measure the idea that distractors are more laden with gist memories. A meaning recognition test asks participants to not worry about an item being actually studied but rather to accept the item if it keeps the gist of the studied list.

Applied based:

Interested in explaining the role of psychotherapy and interview techniques used by clinicians and police. In certain types of psychotherapy and interrogation settings clients and suspects are encouraged to rely on their gist memory for events. These techniques could lead to false memories of events that never happened because the false events evoke so many gist memories of related real events that the false event seems very real.

Experiment One:

Participants: Sixty Undergraduate Students

Materials: 24 Roediger McDermott lists

Procedure:

Participants listened to audio recording of 12 lists, randomly selected out of the 24 lists. Presentation rate of each word was 3 seconds. Stronger associates were presented before weaker associates. Next they read instructions on how to respond to the recognition test. 30 participants received standard instructions and 30 participants received meaning recognition instructions. Meaning recognition instructions instructed the participants to ignore whether or not the item had been on the first list and to select any item, which was consistent with the theme of the list. Finally subjects listened to an audio recording of 72 items: 36 targets, 12 critical distractors for the old lists, 24 critical distractors for new lists. Results:
 
 
Type of recognition
Type of item
Standard
Meaning
Target 
.61 (.08)
.68 (.09)
Distractor    
Critical 
.63 (.13)
.88 (.08)
Unrelated critical 
.19 (.08)
.28 (.14)
Unrelated exemplar 
.16 (.11)
.25 (.15)
  Planned comparisons ran on the mean differences:

Standard recognition: critical distractors = target > unrelated distractors

Meaning recognition: critical distractors > targets > unrelated distractors

 The researchers claim that these results were not due to a difference in response criterion set by the meaning instructions in relationship to the standard instructions. Instead, based on the findings of the first study the researchers concluded that on meaning-recognition tests distractors provided a superior access to strong gist memories.

Experiment 2

Participants: 60 undergraduate students

Methods:

Creation of Materials

74 undergraduate students listened to the 24 lists of words used by Roediger and McDermott. These students wrote a single word that captured the meaning of most of the list words (list-associate distractors). Also they wrote 3 words that could have been on the list but were not (missing-exemplar distactors). Over half of the subjects gave the same list-associates for 12 of the lists, and these lists were used in the study.

Procedure

Procedure was the same as experiment one except that test lists included 60 targets, 12 critical distractors of the lists, 12 list-associate distractors, 12 missing-exemplar distractors, 12 critical distractors for the 12 unpresented lists, and 12 additional distractors from the non-presented lists.

 Hypotheses Critical distractors should have the most repeatedly cued gist memories and therefore be the easiest to remember (because they are associates of all the targets). List associates should be the next most falsely remembered items because they are related to more items than the missing exemplar items. Brainerd and Reyna assume that the an item that has many related items on the old list will produce more gist memories increasing the likelihood of an item being accepted under meaning instructions. Results
 
 
Type of recognition
Type of item
Standard
Meaning
Target 
.58 (.07)
.67 (.06)
Distractor    
Critical 
.54 (.14)
.86 (.11)
List associate
.36 (.13)
.76 (.14)
Missing exemplar 
.24 (.10)
.48 (.16)
Unrelated critical 
.16 (.10)
.24 (.14)
Unrelated exemplar 
.13 (.11)
.20 (.09)
   

Planned comparisons ran on the mean differences

Standard recognition: targets = critical distractors > list-associate distractors > missing-exemplar

distractors > unrelated distractors

meaning recognition: critical distractors > list-associate distractors > targets > missing-exemplar distractors > unrelated distractors

 All of the hypotheses were supported by the data.

Experiment 3

Purpose: Wanted to look at the effects of manipulation of the strength of the to-be-accessed gist.

Subjects: 80 undergraduate students

Subjects studied a single list of familiar words. Some of the words in the list were exemplars of that list. On the recognition test there were strong gist items (category names that had three exemplars on the original list) and weak gist items (category names for which one exemplar had been studied). Half of the participants received standard instruction and the other half received meaning instructions.

Presentation List

48 category lists were selected. A list of words was constructed based on the 48 category lists (36 category exemplars, 3 each from 12 of the 48 category lists) (12 category exemplars, 1 from each of the reaming 12 lists) (48 filler targets)

Recognition List

84 items (12) targets for the strong gist categories, (12) targets for the weak gist categories, (12) filler targets (12), distractors that were names of the strong-gist categories (12), distractors that were names of categories of the weak-gist categories (12), distractors that were names of categories that had been exemplified on the study list (12), distractors each of which was an exemplar of those categories in the previous group

Results
 
 
Type of recognition
Type of item
Standard
Meaning
 
Target     
Strong gist
8.38 (1.63)
9.83 (1.39)
Weak gist
7.30 (1.77)
8.48 (1.60)
Filler
6.70 (2.03)
7.80 (1.77)
Distractor    
Strong-gist category name
3.20 (1.57)
11.08 (.94)
Weak-gist category name 
2.23 (1.39)
7.45 (1.74)
Unrelated category name 
1.33 (1.16)
2.02 (1.31)
Unrelated category exemplar
1.13 (.99)
1.95 (1.11)
 

Planned mean difference comparisons:

standard recognition: strong-gist targets > weak-gist targets = filler targets > strong-gist category names > weak-gist category names > unrelated distractors

meaning recognition: strong-gist category names > strong-gist targets > weak-gist targets > filler targets = weak-gist category names > unrelated distractors

General Discussion
  1. Category-name distractors and the critical distractors were remembered more than frequently than the targets on the meaning instructions.
  2. The rate at which distractors were accepted varied by the ability of the distractors to access gist memories
  3. Individuals willingness to accept distractors at higher rates than targets varied based on the strengths of the relevant gist memories for the distractors versus the targets.
The Brainerd and Reyna claim that the fuzzy trace theory predicts the findings but implicit associative response (IAR) theory does not. They reject the ability of IAR to explain the data based on the fact that targets and distractors were equally remembered in the standard set of instructions. Therefore, they hold that IAR predicts that the levels should stay equal in the meaning instruction design.

They also claim they answer an important applied question. How psychotherapy and interrogation interviews can create a belief that never-experienced things happened.


  
University of Arkansas
Department of Psychology
Lampinen Lab
False Memory Reading Group
False Memory Reading Group Fall 1999