Halle R. Zucker1, Patricia A. Reuter-Lorenz2, & Elizabeth ...
Aging and false memories: Comparing effects of itemrelatedness and list position Halle R.
Patricia A.
2 Reuter-Lorenz ,
& Elizabeth A.
1 Kensinger
College, Department of Psychology; 2University of Michigan, Departments of Psychology & Neuroscience
Introduction
Method
In list-learning tasks, memory distortions are usually meaning-based (semantic) or sound-based (phonological).
Younger adults 42 (female = 25)
Older adults 42 (female = 27)
Mean age (SD)
19 (.89) years
76 (6.7) years
MMSE
29.43 (.83)
28.76 (1.36)
Such false memory (FM) errors reveal the interplay between verbatim and gist traces and may elucidate how memoranda are processed and stored in memory. Prior work has shown that FM can be delay-invariant (Flegal, Atkins, & Reuter-Lorenz, 2010; Zucker, Flegal, Atkins, & Reuter-Lorenz, 2010).
Primacy positions were more likely to show age-differences in terms of correct responses, however YA and OA had similar FM performance across positions. These relationships were not affected by item-relatedness for either age group.
+
Semantic list
blouse “blouse” mop “mop” SHIRT PRIMACY
half “half”
XXXXX Phonological list
dancer “dancer”
BROOM MIDDLE
better “better” ANSWER PRIMACY
Hypotheses Semantically organized lists should promote better memory than phonologically organized lists, and this advantage should be more evident in YA than OA because YA are more likely to use and benefit from semantic codes.
The continuity of memory performance across positions and item-relatedness is inconsistent with the multiple memory systems view.
Free recall (1000 ms ITI)
WHOLE RECENCY
Results
planet “planet”
XXXXX
Free recall 30 sec minimum
LETTER MIDDLE
CARPET RECENCY Younger adults/semantic trials Younger adults/phono trials Older adults/semantic trials Older adults/phono trials
Correct responses: Position x Item-relatedness
46 trials
Error responses : Position x Item-relatedness
What about memory performance at other list positions? Despite similarities at primacy and recency positions, error performance from middle positions shows a divergent pattern between YA and OA. Error responses: Position x Item-relatedness
1
0.5
0.35
0.95
0.45
0.3
0.9
0.4
FM / (Correct + FM)
Correct / (Correct + FM)
Age differences should be greater in primacy than recency lists to the extent that age affects LTM more than STM and this difference may be most pronounced for semantic lists where age differences may be most evident.
We confirmed our hypothesis that semantically organized lists would support better memory performance, however, contrary to our prediction, the effect of item-relatedness was not different for YA & OA.
1000 ms (32 ms ISI)
Here, we investigated the prevalence of semantic and phonological codes across putatively different memory systems (STM, LTM) and examined potential differences due to age.
Further, the advantage of semantic lists should be greater in the primacy than recency position, if semantic codes prevail in LTM, and phonological codes in STM, as posited by the separate memory system view.
Conclusions
0.85 0.8 0.75 0.7 0.65
0.35 0.3 0.25 0.2
0.1
0.55
0.05
0.5
0 Recency
More correct responses came from recency than primacy positions, p < .001. More responses were made on semantic than phono trials, p < .001. YA made more responses than OA, p = .029. There was a trending age x position interaction, p = .089, with a larger age effect in the primacy position.
0.25 0.2 0.15 0.1 0.05 0
0.15
0.6
Primacy
FM / (Correct + FM)
1Boston
1 Zucker ,
Primacy
Primacy
Recency
More FM responses came from primacy than recency positions, p = .026. OA made more FM responses than YA, p = .026. There were no significant interactions with position, age, or item-relatedness.
Recency
There is a trending interaction between position and item-relatedness, and age, p = .101.
These list positions are an open area for future studies of YA & OA memory processes.
Acknowledgements & References Thanks to Maite Balda, Kelley Durham, Stephen Lee, Rachel Newmiller, and Meymune Topcu for their help with stimulus generation, data collection, and data analysis. This work was supported by Boston College internal funding awarded to EAK. Flegal, K.E., Atkins, A.S., & Reuter-Lorenz, P.A. (2010). False memories seconds later: The rapid and compelling onset of illusory recognition. Journal of Experimental Psychology: Learning, memory, and cognition, 36(5), 1131-1338. Zucker, H.R., Flegal, K.E., Atkins, A.S., & Reuter-Lorenz, P.A. (2010, April). Distinguishing between memory systems: False memories depend on serial position. Poster presented at Cognitive Neuroscience Society, Montreal, Canada.
Middle
For more information, contact Halle ([email protected])
Patricia A.
2 Reuter-Lorenz ,
& Elizabeth A.
1 Kensinger
College, Department of Psychology; 2University of Michigan, Departments of Psychology & Neuroscience
Introduction
Method
In list-learning tasks, memory distortions are usually meaning-based (semantic) or sound-based (phonological).
Younger adults 42 (female = 25)
Older adults 42 (female = 27)
Mean age (SD)
19 (.89) years
76 (6.7) years
MMSE
29.43 (.83)
28.76 (1.36)
Such false memory (FM) errors reveal the interplay between verbatim and gist traces and may elucidate how memoranda are processed and stored in memory. Prior work has shown that FM can be delay-invariant (Flegal, Atkins, & Reuter-Lorenz, 2010; Zucker, Flegal, Atkins, & Reuter-Lorenz, 2010).
Primacy positions were more likely to show age-differences in terms of correct responses, however YA and OA had similar FM performance across positions. These relationships were not affected by item-relatedness for either age group.
+
Semantic list
blouse “blouse” mop “mop” SHIRT PRIMACY
half “half”
XXXXX Phonological list
dancer “dancer”
BROOM MIDDLE
better “better” ANSWER PRIMACY
Hypotheses Semantically organized lists should promote better memory than phonologically organized lists, and this advantage should be more evident in YA than OA because YA are more likely to use and benefit from semantic codes.
The continuity of memory performance across positions and item-relatedness is inconsistent with the multiple memory systems view.
Free recall (1000 ms ITI)
WHOLE RECENCY
Results
planet “planet”
XXXXX
Free recall 30 sec minimum
LETTER MIDDLE
CARPET RECENCY Younger adults/semantic trials Younger adults/phono trials Older adults/semantic trials Older adults/phono trials
Correct responses: Position x Item-relatedness
46 trials
Error responses : Position x Item-relatedness
What about memory performance at other list positions? Despite similarities at primacy and recency positions, error performance from middle positions shows a divergent pattern between YA and OA. Error responses: Position x Item-relatedness
1
0.5
0.35
0.95
0.45
0.3
0.9
0.4
FM / (Correct + FM)
Correct / (Correct + FM)
Age differences should be greater in primacy than recency lists to the extent that age affects LTM more than STM and this difference may be most pronounced for semantic lists where age differences may be most evident.
We confirmed our hypothesis that semantically organized lists would support better memory performance, however, contrary to our prediction, the effect of item-relatedness was not different for YA & OA.
1000 ms (32 ms ISI)
Here, we investigated the prevalence of semantic and phonological codes across putatively different memory systems (STM, LTM) and examined potential differences due to age.
Further, the advantage of semantic lists should be greater in the primacy than recency position, if semantic codes prevail in LTM, and phonological codes in STM, as posited by the separate memory system view.
Conclusions
0.85 0.8 0.75 0.7 0.65
0.35 0.3 0.25 0.2
0.1
0.55
0.05
0.5
0 Recency
More correct responses came from recency than primacy positions, p < .001. More responses were made on semantic than phono trials, p < .001. YA made more responses than OA, p = .029. There was a trending age x position interaction, p = .089, with a larger age effect in the primacy position.
0.25 0.2 0.15 0.1 0.05 0
0.15
0.6
Primacy
FM / (Correct + FM)
1Boston
1 Zucker ,
Primacy
Primacy
Recency
More FM responses came from primacy than recency positions, p = .026. OA made more FM responses than YA, p = .026. There were no significant interactions with position, age, or item-relatedness.
Recency
There is a trending interaction between position and item-relatedness, and age, p = .101.
These list positions are an open area for future studies of YA & OA memory processes.
Acknowledgements & References Thanks to Maite Balda, Kelley Durham, Stephen Lee, Rachel Newmiller, and Meymune Topcu for their help with stimulus generation, data collection, and data analysis. This work was supported by Boston College internal funding awarded to EAK. Flegal, K.E., Atkins, A.S., & Reuter-Lorenz, P.A. (2010). False memories seconds later: The rapid and compelling onset of illusory recognition. Journal of Experimental Psychology: Learning, memory, and cognition, 36(5), 1131-1338. Zucker, H.R., Flegal, K.E., Atkins, A.S., & Reuter-Lorenz, P.A. (2010, April). Distinguishing between memory systems: False memories depend on serial position. Poster presented at Cognitive Neuroscience Society, Montreal, Canada.
Middle
For more information, contact Halle ([email protected])
