Acknowledgements 

Sponsored by the letter S, the number .05 and the word connectivity

Of course, a special thanks goes out to the staff of Hugos, who serve up the most delectable frosty beverages.

 

Now for our feature presentation

McEvoy, C.L., Nelson, D.L. & Komatsu, T. (1999).What is the connection between true and false memories? The differential roles of interitem associations in recall and  recognition. Journal of Experimental Psychology: Learning, Memory, & Cognition, 25, 1177-1194.

 

McEvoy et al. are noted for their work with cued recall and the semantic connectivity of words. These researchers have hypothesized that similar semantic mechanisms to those controlling cued recall are operating in false memory for semantically related items.

 

Theoretical Assumptions underlying the Experimental Method:

The theoretical inquiry focused on two properties of semantic networks, associative strength and interconnectivity of related items. Associative strength was manipulated in study 1 and interconnectivity was varied in studies 2 and 3.

 

Assumption 1: Associative Strength

The more strongly associated critical lures are words presented in the study lists the more likely the critical lures will be falsely remembered as having been presented. 

Assumption 2: Associative Strength

The more highly interconnected a list of words is the fewer false memories that will be produced under recall testing. However, the opposite should be found for old, new recognition testing. 

 

Study 1

Half of the participants (n = 20) were presented 12 lists of words highly related to the not presented critical lures. Remaining participants viewed 12 lists of words weakly related to the not presented critical lure. 

 

Participants were tested for free recall after the presentation of each list, and following the presentation of the 12th list participants completed the recognition test. 

 

Results:

Excerpt from Table 2  
 
 
Experiment 1: Associative Strength
Dependent Measures
High 
Low
Free recall
 
 
 Critical lures
.24
.09
 List words
.66
.60
Final Recognition
 
 
 Critical items
.59
.47
 Critical items control
.03
.10
 List words
.88
.80
 List words control
.02
.07
Final Recognition d’
 
 
 Critical items
2.59
1.59
 List words
3.87
2.78

 

As predicted the amount of associative strength between the critical lure and the list words led to greater rates of false recall and recognition of the critical item as old. Notable are the very similar levels of d’ for critical items recognized in the high group and the list words recognized from the low group. 

 

Study 2

 

The design of this study was interested in testing how false recall and recognition are influenced by highly interrelated list words.

 

Half of the participants (n = 20) were presented 12 lists of interrelated words. Remaining participants viewed 12 lists of weakly interrelated words. 

 

Participants were tested for free recall after the presentation of each list, and following the presentation of the 12th list participants completed the recognition test. 

 

Results 

 

Excerpt from Table 2  
 
 
Experiment 2 Interconnectivity Strength
Dependent Measures
High 
Low
Free recall
 
 
 Critical lures
.13
.25
 List words
.64
.58
Final Recognition
 
 
 Critical items
.70
.63
 Critical items control
.05
.06
 List words
.88
.79
 List words control
.06
.06
Final Recognition d’
 
 
 Critical items
2.68
2.37
 List words
3.46
2.98

 

As predicted the rates of falsely recalling a critical lure were lower in the high group opposed to the low group, and rates of falsely recognizing critical lures is higher in the high group versus the low group. A very interesting result is the decreased level of list word recall in the low group even though the rates of false recalling of critical lures was increased. 

 

Study 3 

 

The design of study three remain identical to that of study 2 with the exception of the addition of a delay of 1 minute between free recall and recognition for half of the participants.

 

Results 

 

Excerpt from Table 2  
 
 
Immediate initial test
Delayed initial test
Dependent Measures
High 
Low
High
Low
Initial Free recall
 
 
 
 
 Critical lures
.14
.30
.18
.26
 List words
.66
.61
.59
.53
Final Free Recall
 
 
 
 
 Critical items
.06
.19
.11
.28
 List words
.27
.28
.30
.30
 List words (as portion of  

  lists recalled) 
.43
.35
.38
.36

 

As predicted critical lures were recall more often in both the immediate and delay low groups, than the either of the high groups. These results held constant and were still present at the final recall test. As found in study two more list words were recalled in the on the initial free recall test for the high group. However, on the final recall test levels of recalling lists words were equal between high and low groups. 

 

Discussion

 

The researchers feel that false memories are a product of associated words being implicitly cued during encoding and at retrieval. 

 

The researchers feel that levels of false recall and false recognition between the high and low interconnectivity lists is difficult for models that explain false memories as the product of a single memory trace (e.g. gist). However, these predictions are based on Fuzzy version 1.0 and not Fuzzy 15.5 (now with a turbo charger Nonidentity judgment accounting for the increased levels of verbatim leading to decreased levels of false memories)

 

The researchers circa 1997 suggest that an increase in veridical judgments may cause a decrease in false judgments. 
 


 
University of Arkansas
Department of Psychology
Lampinen Lab
False Memory Reading Group
False Memory Reading Group Summer 2000