Skip to main content

Full text of "ERIC ED267336: Aging and Episodic Priming: The Propositional Structure of Sentences."

See other formats


DOCUMBNT RBSUMB 



ED 267 336 

AUTHOR 
TITLE 

SPONS AGENC7 
PUB DATE 
GRANT 
NOTE 



PUB TYPE 



EDRS PRICE 
DESCRIPTORS 



CG 018 916 

Howard, Darlene V. 

Aging and Episodic Priming: The Propositional 
Structure of Sentences. 

National Inst, on Aging (dhhS/PHS), Bethesda, MD. 

Aug 85 

R01AG02751 

7p.; Paper presented at the Annual 
American Psychological Association 
Angeles, C/i, August 23-27, 1985). 
Reports - Uesearch/Technical (143) 
Speeches/Conference Papers (150) 

NFOl/PCOl Plus Postage. 
*Age Differences; Aging 
*Meaory; *01der Adults; 
*Hord Recognition 
*Priming Effects 



Convention 
(93rd, Los 



of the 



(Individuals) r Memorization; 
*Recogn i t i on ( Psychoxogy ) ; 



IDENTIFIERS 
ABSTRACT 

When presented with linguistic material, elderly 
adults are often unable to report as much material as are younger 
people. To ascertain whether elderly adults are as sensitive as young 
adults to the underlying structure of the to>be- remembered sentences, 
a study was conducted using the item recognition priming technique. 
In thxs technique, people attempt to remember sentences containing 
unassocxated nouns. The subjects are asked to decide whether or not 
/?S*ri™^ ^^^^^ occurred in the studied sentence. Thirty-six young 
(18-25 years) and 36 elderly (64-82 years) subjects were tested with 
a series of 288 item recognition trials consisting of 144 occurring 
nouns and 144 distractor nouns. When the primed trials were compared 
with the control trials, a significant and prime effect was found for 
young and elderly people with no significant differences between the 
age groups. The differences on paired-recognition of the nouns, a 
more traditional measure of memory, were significant with young 
subjects being correct for 83 percent of the trials and elderly 
subjects being correct for 72 percent of the trials. Elderly persons 
appeared to be less sensitive to the underlying propositicnal 
structure of the sentence than were younger subjects. Younger people 
showed more wi thin-proposition than between-proposition priming, 
whereas the elderly subjects did not. Using tests of memory without 
awareness may give a more complete measure of memory across the 
lifespan than do traditional measures. (ABL) 



* Reproductions supplied by EDRS are the best that can be made * 

from the original document. * 



SOD 



AGING AND EPISODIC PRIMING: THB PHOPOSniONAL STRUCTURE OF SENTENCES 
Darlena V. Hovrardy Georgetom University 
Abstract 

Thirty-six young (18-25 years) and 36 elderly (64-82 years) people studied 
36 sentences of the form N0UN1 - VERB1 - N0UN2 - oonj - N0UN3 » VERB2 - 
N0UN4. Then they made it«n recognition Judgmtjnts regarding whether single 
nouns had occurred in the sentences. Both young and elderly people showed 
priming between the nouns within the sentences; a noun was recognized 
faster when it was tested immediatey after another noun from the same 
sentence than irtien it was tested following a noun from a different 
sentence. However, young people showed more within-proposition than 
between-proposition priming, whereas the elderly did not, indicating thit 
the elderly are less sensitive than the young to the prepositional 
structure of the sentences. 



00 



CD 
O 



UM. KPAKTMCWr OF EDUCATION 

NATIONAL INSTITUTE OF EDUCATION 

7AT10NAL RES0»"1CES INFORMATION 
CENTER lERlC) 
K document hM b—n reproduced is 
received from the person or orgenizetion 
ongmeting it. 
□ Minor cheoget heve been made to improve 
reproductHX> quality. 

• Pomttof wworopmionettetedinthis. cu- 
ment do not necetMrrty represent officisl NIE 
position or policy 



"PERMISSION TO REPRODUCE THIS 
MATERIAL HAS BEEN GRANTED BY 



TO THE EDUCATIONAL RESOURCES 
INFORMATION CENTER (ERIC)." 



This work was presented at the annual meetings of the American 
Psychological Association, Los Angeles, August 1985. This research was 
supported by Grant R01AG02751 from the National Institute on Aging. 



ERIC 



AGING AND EPISODIC PRIMING: THE PROPOSniONAL STROCTUHE OF SENTENCES 

Darlentt V. Howard, Qaorgatown Onlvaralty 
Hhan elderly adults are praaented with linguiatlo natarlala such aa 
aentenoea and atorlea, upon later teatlng they are tinabla to report aa much of 
the material aa young. It la unclear, honeyer, whether thla dlfferenoe la 
only quantitative, or whether It la alao qualitative. In the aenae that the 
elderly people are leaa aenaltlve than young to the underlying ideas 
(propoaltlona) conveyed in the paaaage. Some atudlea have found age 
similarity in sonsltlvlty to prepositional atructure (e.g., Spilich, 1983; 
Petroa, Tabor, Cooney, & Chabot, 1983), whereaa othera have found age 
differences (e.g., Cohen, 1979; Meyer & Rice, 1981). One possible reaaon for 
the discrepancy is that memory la typically aaaeaaed ualng the traditional 
recall and recognition accuracy measures. Theae may be called tests of 
"memory with awareness" (Jacoby & Witherspoon, 1982) in that people must make 
an introspective Judgment about whether and what they remember. For example, 
in recognition tests people are asked, "Do you remember having seen this 
sentence?" Such measures are likely to be influenced by adult age differences 
in cautiousness and criterion, and these may vary from study to study. 

The present experiment seeks to determine whether elderly adults are as 
sensitive as young to the underlying prepositional structure of 
to-be-remembered sentences, by using a method introduced by Ratcllff and 
McKoon (1978) in work with college students. In this so-called Item 
recQgnltion prlnlng ififihniauef people are asked to memorize sentences 
consisting of previously unassociated nouns, e.g., THE KING CLOSED THE WINDOW 
AS THE TOWN BECAME CINDERS. Then they are given a series of item recognition 
trials in which they must decide whether or not individual words occurred in 
the studied set of sentences. Ratcliff and HcEoon demonstrated that among 
college atudenta, worda from the aame studied sentence prime each other, that 



Page 2 

l8 the word TOVN is recognized faster following WINDOW than following a word 
froB a different studied sentence. The fact that this prlnlng occurs between 
previously unassoolated words Indicates that people have stored and retained 
new associations among the words within studied sentences. Further » young 
people have apparently stored the prepositional structure of the sentences, 
since words taken froa within the same proposition (e.g., CINDERS-TOWN) prime 
each other more than words taken from between the two propositions (e.g., 
WINDOW-TOWN). This lt«n recognition priming method seems particularly useful 
for lifespan studies, since It Is a test of "'memory without awareness.** That 
is, memory for an association is inferred from a pattern of response times, 
rud the person need not judge whether he/she remembers the association. In 
earljer research (Howard, Heisey, & Shaw, In press) we have used this method 
to assess memory for one prepositional sentences (e.g., THE DRAGON SNIFFED THE 
FUDGE). We found that after only moderate amounts of study of the sentences, 
elderly individuals showed as much priming as young, despite the fact that 
there were large age differences in cued recall of the same sentences. 

In the present experiment, seventy-tvo people were tested, 36 young (mean 
age = 20.5 years, SD = 1.5, range = 18-25) and 36 elderly adults (mean age - 
69.9 years, SDs4.5, range = 64-82). None of the participants was 
Institutionalized and the age groups were similar in WAIS vocabulary score and 
educational level. Each participant studied a list of 36 two-propositional 
sentences of the form N0UN1 - VERB1 - N00N2 - conj - N0DN3 - VERB2 - N00N4. 
The nouns within sentences were chosen to be unassocated with each other. 
After two study periods totalling 30 seconds of study per sentence, people 
ccMnpleted a series of 288 item recognition trials consisting of the 144 nouns 
that had appeared in the sentences as well as 144 distractor nouns matched to 
the sentence nouns in length and frequency. Prime tvoe was the major 
witbin-subjects variable. Counterbalancing insured that across subjects each 



Page 3 

of the sentence nouns was testad equally often in each of three prime type 
conditions: oontrol in which a noun was tested following a noun from a 
different studied sentence, hfttwaftn^DroBoaltion prima in which a noun was 
tested followiog a noun from the other proposition of the vmme studied 
sentence, and within^proposition ju^ja^ in which a noun was tested following a 
noun from the same studied proposition. 

Each participant then completed a paired-recognition and a cued-recall 
test of the sentences, with the order of these being counterbalanced across 
subjects. The palred^raQognltion test consisted of 36 trials (18 "7/es* and 18 
"no") in which two words from the sentences were presented together, and 
people were asked to report "yes" if the two words had come from the same 
sentence and "no" otherwise. For the cued recall test there were 36 trials, 
in each of which the person was given a noun from one of the studied sentences 
and asked to recall the rest of the sentence. Across subjects, each of the 
four nouns in each sentence was used as the sentence cue the same number of 
times. 

There were three major findings. Ziiatt when primed trials (within- and 
between-proposition primes combined) are compared with control trials, young 
and elderly people both show a significant prime effect, P(2, 140) : 13.90, 
£<.001, with no significant difference in magnitude of the prime effect 
between the age groups. The prime effect (defined as response time on control 
trials minus response time on primed trials) was 96 msec for the young 
participants and 73 msec for the elderly. This indicates that both groups of 
participants had stored the associations among the words in the sentences. 
2fififind> despite the age equivalence in overall priming described above, the 
traditional measures of mesory accuracy yield significant age differences in 
memory for the sentences. The difference between the cued recall of young and 
elderly is significant, £(1,68)=21.37t Ji<.0001. Of a possible 3 nouns that 



Page 4 

could be recalled per aentencet the young recalled a mean of 1.55 and the 
elderly only •83. On paired-recognition of the nouna, there waa alao a 
aignificaat age difference, P(1,68)«88.96, a<.01. The young were correct on 
an average of S3% of the triala and the elderly on only 72%. When the above 
findings are conaidered together, they indicate that age differencea in mcoory 
are reduced Uhen priming, instead of recognition or recall accuracy, is used 
as a meaaure. Thia ?^uggeata that at leaat some age differences on recall and 
recognition tasks are due to retrieval, rather than atorage, difficulties. 

IhlCily there ia, nonetheless, an age difference in the pattern of within- 
and between-proposition priming. In keeping uith McKoon and Ratcliff'a 
results with college students, the young people show a larger 
within-proposition than between-proposition prime effect (i.e., the response 
time difference between control triala and primed trials was 114 msec for 
within-proposition primes, but only 78 msec for between-proposition primes.) 
In contrast, the elderly participants revealed slightly (though not 
significantly) less priming for within- than between-propositon primes, 
showing prime effects of 60 msec and 85 msec for within- and between- 
proposition priming, reapectively. This observation ia also supported by 
correlations, rhe correlation between an individual's age and the magnitude 
of his/her between-proposition prime effect was not aignificant (£=-.10, 
Ji>.10), but the correlation between age and the within-proposition prime 
effect was significant (ii=-.28, Si<.02). This finding indicates that when 
memory structure is assessed via priming, the elderly persor is less sensitive 
to the underlying prepositional structure of the sentences than are young 
people. Thus, this method reveala qualitative differencea between young and 
elderly in manory atructure for sentences. 

In general, the preaent findings indicate that item recognition priming 
is a sensitive means of aaaeasing the structure of stored information across 



Page 5 

the adult lifespan • Uelng such measures of Memory without awareness* may 
give a fuller picture of age differences and similarities in mMory than using 
only traditional measures that call upon ''memory with awareness.'' 

References 

Cohent 0. (1979)* Language comprehension In old age. Cognitive 

faZfibfilfigytJlt 412-429. 
Howard, D. V., Helsey, J. Q., 4 Shaw, R. J. (In press). Aging and the 

priming of newly learned associations. DftvelQ|Mnenfcal PavehQloyy . 
Jacoby, L. L., 4 Wltherspoon, D. (1982). Remembering without awareness. 

Canadian jlfiuoial sL Pavohoiogv , 300-324. 
Meyer, B. J. P., 4 Rice, G. E. (I98I). Information recalled from prose by 

young, middle, and old adult readers. Experimental A g in g Research > 2, 

253-268. 

Petros, T., Tabor, L., Cooney, T., 4 Chabot, R. J. (I983). Adult age 

differences in sensitivity to semantic structure of prose. Developmental 

PsYQhologv> ja.> 907-914. 
Ratcliff, R., 4 McKoon, G. (1978). Priming in item recognition: E\rldence 

for the proposltlonal structure of senteaces. Journal qf Verbal Learning 

and Verbal Behavior , Ji» 403-417* 
Spllich, G. J. (1983). Life-span components of text processing: Structural 

and procedural differences. Journal ^ Verbal Learning and Verbal 

Bsh^llQLf 21, 231-244.