Content uploaded by Tamra J Bireta
Author content
All content in this area was uploaded by Tamra J Bireta
Content may be subject to copyright.
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
Short article
Age-related differences in the von Restorff
isolation effect
Tamra J. Bireta
The College of New Jersey, Ewing, NJ, USA
Aime
´
e M. Surprenant and Ian Neath
Memorial University of Newfoundland, St. John’s, Newfoundland, Canada
When one item is made distinct from the other items in a list, memory for the distinctive item is
improved, a finding known as the isolation or von Restorff effect (after von Restorff, 1933).
Although demonstrated numerous times with younger adults and children, this effect has not been
found with older adults (Cimbalo & Brink, 1982). In contrast to the earlier study, we obtained a
significant von Restorff effect for both younger and older adults using a physical manipulation of
font colour. The effect size for older adults was smaller than that obtained for younger adults, con-
firming a prediction of Naveh-Benjamin’s (2000) associative deficit hypothesis, which attributes
age-related differences in memory performance to older adults’ reduced ability to form associations.
The findings are consistent with related research in which older adults demonstrate similar—but
smaller—benefits for distinctive information to those for younger adults.
Keywords: Ageing; Distinctiveness; Free recall; Isolation effect; von Restorff.
Normal ageing is typically associated with declines
in memory performance across a wide variety of
tasks and for many different types of informat ion
(Zacks, Hasher, & Li, 2000). Explanations for
these changes in memory performance usually
cite a deficit in some process involved in memory
formation/retrieval (e.g., forming associations;
Naveh-Benjamin, 2000) or a deficit in a general
cognitive factor (e.g., processing speed; Salthouse,
1996). None of these theories, however, propos e
that memory mechanisms operate in a fundamen-
tally different manner in older adults. Consistent
with this assumption, younger and older adults
usually show similar benefits from experimental
manipulations that improve memory, including
generation (Taconnat & Isingrini, 2004), and
levels of processing (Troyer, Ha
¨
fliger, Cadieux, &
Craik, 2006). One factor that may be an exception
Correspondence should be addressed to Tamra J. Bireta, Psychology Department, Social Sciences Building, The College of New
Jersey, 2000 Pennington Rd., Ewing, NJ, 08628, USA. E-mail: tbireta@tcnj.edu
This research was supported, in part, by National Institute on Aging Grant AG021071 awarded to AMS and IN. Portions of this
paper are based on a dissertation presented by TJB to the faculty of the Graduate School of Purdue University. We thank
E. J. Capaldi, J. S. Nairne, and R. W. Proctor for comments and discussions. Part of this work was presented at the 47th Annual
Meeting of the Psychonomic Society, Houston, Texas, November 2006.
# 2007 The Experimental Psychology Society 1
http://www.psypress.com/qjep DOI:10.1080/17470210701626608
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY
0000, 00 (0), 1–8
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
to this rule is distinctiveness: The available data
suggest that the benefit of distinctiveness may be
reduced—or even absent—for older adults.
There are two purposes of the research reported
here: (a) to examine age-related differences in
the effects of distinctiveness and (b) to ev aluate
the hypothesis that an associative deficit in older
adults (Naveh-Benjamin, 2000) underlies age-
related deficits in memory.
When a particular item is made distinct from
other items in a list, memory for the distinctive
item is improved. This is known as the von
Restorff effect or the isolation effect (von
Restorff, 1933). Recall of the distinctive item is
typically compared to that of a control item that
is consistent with the background items .
Isolation effects can be achieved with both physical
manipulations (e.g., size, colour, or font type) and
semantic changes (e.g., meaningfulness or cat-
egory). Numerous studies have demonstrated the
robustness of this effect using various stimul i and
methodologies (for a review, see Hunt, 1995),
and the isolation effect is reliably obtained in
college-age adults as well as in chil dren (Howe,
Courage, Vernescu, & Hunt, 2000). The size of
the isolation effect depends on several fa ctors,
including the manner in which the isolate differs
from the other items in a list. Large effects are
obtained using size, colour, and spacing manipula-
tions, and isolati on effects increase in magnitude as
the difference between the isolate and the other list
items increases (Gumenik & Levitt, 1968; see also
Hunt, 1995).
Despite the la rge liter ature on isolation effects
in children and young adults, only one published
study has examined the isolation effect in older
adults. Cimbalo and Brink (1982) displayed lists
of nine consonants to younger and older adults
for immediate written serial recall, with the
isolate lists containing a consonant in a larger
font in the fifth position. Younger, but not older,
adults recalled the isolates better than the control
items. The authors argued that older adults
noticed the size difference, but suggested that
the lack of an isolation effect reflected a deficit in
using structural information to organize the list.
The isolate, then, “may have been considered as
analogous to noise, something to be ignored” by
the older adults (p. 76).
Most current theories of ageing and memory,
including theories based on inhibition, reduced
resources, and slower speed of processing, predict
that older adults will show lower overall perform-
ance than younger adults on most episodic
memory tasks (Zacks et al., 2000). They lack,
however, a specific reason why older adults
might show a reduced or nonexistent benefit for
the isolate. In contrast, the associative deficit
hypothesis (Naveh-Benjamin, 2000) offers a
specific explanation for age-related differences in
the isolation effect. Naveh-Benjamin argued that
it is important to distinguish between memory
for a single unit and memory for associations
among units. Associations between two units can
include two items, an item and its context, two
contextual features, or “the representation of two
mental codes” (p. 1170). He proposed that the
reason older adults show poorer memory in most
episodic memory tasks is that older adults have
difficulty “merging different aspects of an episode
into a cohesive unit” (p. 1185).
Naveh-Benjamin and colleagues (Naveh-
Benjamin, 2000; Naveh-Benjamin, Hussain,
Guez, & Bar-On, 2003; Naveh-Benjamin, Guez,
Kilb, & Reedy, 2004) have demonstrated that an
important predictor of when older adults will
show worse performance than younger adults is
whether the task requires memory for units
versus memory for the association between units.
For example, compared to younger adults, older
adults recognized fewer word–nonword and
word–word pairs despite similar levels of perform-
ance for recognition of the individual items
(Naveh-Benjamin, 2000). This type of associative
deficit was replicated using pictorial stimuli
(Naveh-Benjamin et al., 2003) and face/name
pairs (Naveh-Benjamin et al., 2004). This view
has also been successful in accounting for older
adults’ reduced memory for contextual details
(for a meta-analysis, see Spencer & Raz, 1995)
using the idea that older adults can remember
individual units (i.e., the items or the contextual
features), but suffer when they must also remem-
ber associations amongst them (i.e., which
2 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0)
BIRETA, SURPRENANT, NEATH
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
contextual feature went with which items). Naveh-
Benjamin (2000, Exp. 3) ruled out the possibility
that older adults simply lack information about
contextual details in general by presenting words
in different fonts, followed by tests for the
words, fonts, and words–font relations. Older
adults showed equivalent memory for the words
and for the fonts, but a large deficit in recalling
the word –font relations.
According to the associative deficit hypothesis,
age-related differences in the isolation effect are
due to the difficulty that older adults have in
associating contextual features with items such as
the colour in which a word was presen ted. The
association of the contextual element (colour) to
an item is necessary for an isolation effect to
occur because it is this contextual element that
causes the isolate to have increased distinctiveness
at retrieval (due either to items being placed in
different categories, e.g., Hunt & Lamb, 2001,
or cues uniquely specifying the isolate, e.g.,
Nairne, 2006). If older adults do not associate
contextual elements to items as well as younger
adults, then a smaller or nonexistent isolation
effect should be found for older adults.
The current experiment was designed to compare
isolation effects in younger and older adults and to
test a prediction of the associative deficit hypothesis
that older adults will show reduced isolation effects.
Participants viewed lists of 12 unrelated nouns, with
the isolate lists consisting of one item in a red font
and all other items in a black font. Immediately fol-
lowing presentation, participants recalled the items
using free written recall. In addition, two different
presentation rates were used to determine whether
older adults would show greater isolation effects
when given an increased opportunity to encode
the information.
Method
The design of the current experiment differed
from the Cimbalo and Brink (1982) study in the
following ways: First, the list items were presented
one at a time rather than simultaneously to ensure
that all items were given equal opportunity to be
encoded. Second, presentation time was held
constant across lists for a given participant rather
than varied unpredictably. Third, the isolate
manipulation involved a perceptually obvious
manipulation of colour rather than a slight size
difference (1 mm) between controls and isolates.
Finally, the test was free rather than serial recall,
a task that typically yields larger isolation effects
(e.g., Kelley & Nairne, 2001).
Participants
A total of 80 younger adults (M ¼ 19.3 years,
range 18–26) and 80 older community-dwelling
adults (M ¼ 70.1 years, rang e 60–89) participated
in this experiment. The younger adults were
Purdue University undergraduates who partici-
pated in exchange for course credit. The older
adults were paid $10 per hour and were recruited
from the community. The education levels of
older adults ranged from high sch ool to gradu-
ate/professional degrees, with 79% of the partici-
pants completing at least some college
coursework (highest level of education was a
high-school degree for 21% of the older adults;
some college for 15%; a college degree for 21%;
some gradu ate or professional education for 3%;
and a graduate or professional degree for 40%).
All participants reported themselves to be in
good health, and none reported using any medi-
cations that migh t interfere with cognitive
functioning.
All participants were admin istered a 20-item
vocabulary test adapted from Salthouse (1993)
and a computerized memory span task based
upon the reading span test in Kane et al. (2004)
in order to obtain an estimate of each participant’s
overall level of cognitive function ing. Typical
results obtained: There was a slight difference in
favour of the younger adults for memory span
(younger M ¼ 7.6, SD ¼ 1.1; older M ¼ 6.7,
SD ¼ 1.0), t (158) ¼ 5.35, p , .01, but older
adults did significantly better on the vocabulary
test (younger M ¼ 7.9, SD ¼ 3.4; older M ¼
13.6, SD ¼ 5.2), t(158) ¼ 8.24, p , .01.
Materials
The stimuli were 241 nouns selected from Clark
and Paivio (2004). The nouns selected were of
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0) 3
AGEING AND THE ISOLATION EFFECT
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
medium to high imageability (M ¼ 6.03, range
4.00–6.90 out of 6.90), familiarity (M ¼ 5.97,
range 4.21–6.92 out of 7.00), and frequency
(M ¼ 1.75, range of 1.00 to 2.00 using log
Thorndike–Lorge frequency; M ¼ 1.70, range of
1.00 to 2.77 using log Kucera–Francis frequency).
Design
Age (younger, older) and presentation rate (1.5 s,
3 s) were between-subjects variables, and the
isolation manipulation of list type (isolate,
control) and serial position (1–12) were within-
subjects variables. Participants viewed the to-be-
remembered items one at a time in the middle of
a computer screen and were asked to read each
word silently. Each list contained 12 different
items, with each item presented for either 1.5 or
3 s and no delay between items. The items on
each list were randomly selected without replace-
ment from the 241 stimuli; thus no word was
repeated within or across lists for any individual.
All of the items in the control lists were presented
in black against a white background. In the isolate
lists, the 7th item was presented in red, and the
other 11 items were black. The isolate always
appeared in the 7th serial position to maximize
the possibility of obtaining an isolation effect for
older adults. First, fewer trials are required to
obtain reliable data than if all positions are
tested, and, second, effect sizes are likely to be
larger in middle serial positions than in earlier
and later positions (because of the absence of
primacy and recency effects). There were 20
trials in total: 10 control lists and 10 isolate lists.
List type on any given trial was random.
Procedure
Participants were told that they would see 12
words one at a time followed by the cue “please
recall the words”. They were asked to recall as
many words as possible by writing the words
down on a response sheet that contained trial
numbers and 12 numbered lines per trial. They
were informed that they could recall the items in
any order and could take as much time as necess-
ary. To proceed to the next list, the participant
clicked on a button labelled “next trial”.
Participants were tested one at a time, and the
experimenter remained in the room to ensure
compliance with the instructions.
Results
As Figure 1 illustrates, older adults clearly showed
a von Restorff isolation effect regardless of the
presentation rate. When questioned after the
experiment, all of the younger adults and all but
2 of the 80 older participants reported awareness
of the isolate.
Free recall
The data were first analysed with a 2 (age: younger,
older) ! 2 (presentation rate: fast, slow) ! 2 (list
type: control, isolate) ! 12 (serial position: 1–12)
Figure 1. Recall of control and isolate lists as a function of group
and serial position. The top panel shows fast presentation, and
the bottom panel shows slow presentation.
4 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0)
BIRETA, SURPRENANT, NEATH
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
analysis of variance (ANOVA). For this and all sub-
sequent analyses, alpha was set to .05. Younger
adults recalled more list items than older adults
(5.28 out of 12 vs. 3.99), yielding a main effect of
age, F(1, 156) ¼ 60.12, MSE ¼ 26.24. More
words were recalled with the slower presentation
rate than with the faster rate (5.05 vs. 4.22), result-
ing in a main effect for presentation rate, F(1, 156)
¼ 25.13, MSE ¼ 26.24. Overall, lists containing
the isolate were not recalled differently from
control lists (4.63 vs. 4.64), F(1, 78) , 1. Typical
primacy and recency effects were obtained, resulting
in a main effect of serial position, F(11, 1716) ¼
121.96, MSE ¼ 4.45.
The difference in recall between younger and
older adults increased from earlier to later serial
positions (collapsed across presentation rates),
with a significant interaction between age and pos-
ition, F (11, 1716) ¼ 4.66, MSE ¼ 4.45. This
occurred primarily because of the increased
recency effects that younger adults displayed with
fast than with slow presentation. List type inter-
acted with position, reflecting the isolation effect
in which the item in the 7th serial position was
better recalled in the isolate lists than the control
lists (4.94 vs. 3.33), F(11, 1716) ¼ 10.35,
MSE ¼ 2.24. Slow presentation resulted in a
memory advantage over fast presentation for
earlier, but not for later, serial positions, yielding
a significant interaction between position and
presentation rate, F(11, 1716) ¼ 4.91, MSE ¼
4.45. There were no other significant interactions.
Isolation effect
To examine the effects of age on the isolation
effect, additional analyses were performed on
recall of just the items occurring in the 7th serial
position. A 2 (age: younger, older) ! 2 (presen-
tation rate: fast, slow) ! 2 (list type: control,
isolate) ANOVA yielded main effects for all vari-
ables. More items were recalled by younger than
older adults (5.18 vs. 3.09), F(1, 156) ¼ 82.76,
MSE ¼ 4.19, and with slow than fast presentation
(4.54 vs. 3.73), F(1, 156) ¼ 12.81, MSE ¼ 4.19,
and the isolates were recalled more often than
control items (4.94 vs. 3.23), F(1, 156) ¼ 78.03,
MSE ¼ 2.69. Younger adults demon strated a
larger difference between isolates and controls
(6.26 vs. 4.09) than did older adults (3.63 vs.
2.56), resulting in a significant interaction
between age and list type, F(1, 156) ¼ 9.21,
MSE ¼ 2.69. A t test confirmed an isolation
effect for older adults, with the means for isolates
(3.63) and controls (2.56) differing significantly,
t(80) ¼ 4.29. No other interactions were signifi-
cant. The lack of a significant three-way inter-
action between age, presentation rate, and list
type, F(1, 156) ¼ 0.62, MSE ¼ 2.69, suggests
that the difference in the size of the isolation
effect for younger compared to older adults does
not depend on presentation rate.
Participant variables
The sample of older adults included many who
had graduate or professional degrees. One possible
concern is that the von Restorff effect that was
observed was driven by a subgroup of highly
educated and potentially higher performing older
participants. The older participants were divided
into two groups based on level of education. The
closest we could come to a median split that
resulted in groups with equivalent ages was to
define the “high education” group as those with
at least some graduate school (N ¼ 34, M ¼
70.41, range 60 to 82) and the “low education”
group as everyone else (N ¼ 46, M ¼ 69.85,
range 60 to 89). Both groups showed a significant
von Restorff effect, t(32) ¼ 4.02 and t(44) ¼ 2.41
for the “high” and “low”, respectively, although the
high education group had a larger isolation effect
than the low education group (2.56 vs. 3.91 com-
pared to 2.56 vs. 3.41). However, the age-related
difference remains even when the high education
group was compared to the younger adults
(4.09 vs. 6.26). Thus, high levels of education in
older adults reduces, but does not eliminate, the
age-related differences seen in the magnitude of
the isolation effect.
Because our sample included a wide range of ages
forolder adults, it was possible to examine differences
in the isolation effect between the young–old (ages
60–74, N ¼ 59) and the old–old (ages 75–89, N
¼ 21). Each age group showed a significant isolation
effect: 3.78 isolates versus 2.69 controls for the
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0) 5
AGEING AND THE ISOLATION EFFECT
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
young–old, t(19) ¼ 2.53, compared to 2.19 isolates
versus 3.19 controls for the old–old, t(57) ¼ 3.97.
The magnitude of the isolation effect did not
differ, t(78) ¼ 0.15. Thus, the benefit of the isolate
appears relatively stable across older age.
Output order
One possible explanation for the higher recall of
the isolate may be that it is output earlier than
the control item, which reduces output interfer-
ence. To check for this, mean output position for
recalling the item from the 7th position (when it
was recalled) in both the control and isolate lists
was calculated (see Figure 2). A 2 (age: younger,
older) ! 2 (presentation rate: fast, slow) ! 2
(item type: control, isolate) ANOVA was con-
ducted on average output position for isolates
and controls. Of most importance, isolate items
were not output at earlier positions than control
items, F(1, 56) , 1. However, older adults
output the 7th items (collapsed across control
and isolate lists) significantly earlier than did
younger adults (position 3.7 vs. 4.4), resulting in
a main effect of age, F(1, 156) ¼ 14.22, MSE ¼
2.48. This is due to the fewer total number of
items recalled by older adults than by younger
adults. The isolated items were also output
earlier with fast presentation compared to slow
presentation (3.7 vs. 4.3), F(1, 156) ¼ 12.27,
MSE ¼ 2.48. Again, this effect occurred as a
result of the fewer items recalled with fast than
with slow presentation. There were no significant
interactions. This analysis was repeated on nor-
malized data, which takes into account the
differential levels of recall; again, no evidence of
early output of the is olated item was observed.
GENERAL DISCUSSION
Younger adults demonstrated better memory for
isolated items than did controls in lists of homo-
geneous background items, replicating the well-
known isolation effect (von Restorff, 1933). In con-
trast to the results obtained by Cimbalo and Brink
(1982), the results of the current study clearly
demonstrate a significant isolation effect for older
adults. Because of the numerous methodological
differences between the Cimbalo and Brink study
and the current experiment, it is not clear why the
effect was not obtained in their study. However,
their combination of serial recall, simultaneous
presentation, and a relatively small difference
between isolate and control items may all have con-
tributed to the difference. The current study used a
more typical methodology and found that older
adults are affected by distinctiveness in a qualitat-
ively similar way to younger adults.
Although the older adults showed a signi ficant
isolation effect, it was smaller than that of the
younger adults. These results support the predic-
tions of Naveh-Benjamin’s (2000) associati ve
deficit hypothesis. According to this view, older
adults show poorer memory in most episodic
memory tasks due to difficulty “merging different
aspects of an episode into a cohesive unit” (p.
1185). In the current experiment, memory for
each list requires associations amongst the items
and/or associations between each item and the
surrounding context, both of which will be
deficient for older adult s. This v iew, therefore,
correctly predicts overall lower performance for
older adults. Further, because older adults are
not able to assoc iate contextual elements with
specific items as well as younger adults, th ey are
not able to use these contextual features as retrieval
cues as effectively. The weaker association between
the unique colour information and the item will
reduce the isolation effect because it is this contex-
tual element that causes the isolate to have
increased distinctiveness at retrieval.
Figure 2. Proportion of responses for isolates and controls (7th
position) by output position.
6 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0)
BIRETA, SURPRENANT, NEATH
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
The finding of a reduced isolation effect for older
adults could also be explained by accounts that pos-
tulate processing differences between younger and
older adults, as processing differences have been
shown to affect the magnitude of the isolation
effect (e.g., Fabiani, Karis, & Donchin, 1990). For
example, one view states that distinctive processing,
defined as “processing the difference in the context
of similarity”, is more difficult for older adults (see
Smith, 2006, p. 279). This difficulty arises from the
need to process information about the item as well
as information about the relationships between the
items, which draws upon cognitive resources that
are limited for older adults. According to this view,
older adults are less likely to engage in distinctive
processing when the task requires a large amount
of cognitive resources. Therefore, in a task such as
the one in the current study that requires a large
amount of cognitive resources, older adults would
be less able to engage in the distinctive processing
that results in the isolation effect.
Most current accounts of the isolation effect
attribute it to the increased distinctiveness at
retrieval for the isolate relative to the background
items (e.g., Hunt & Lamb, 2001; Kelley &
Nairne, 2001; Nairne, 2006). Research in other
areas has found that older adults do benefit from
enhanced distinctiveness at retrieval, albeit to a
lesser extent than younger adults (e.g., Ma
¨
ntyla
¨
& Ba
¨
ckman, 1992; Smith, 2006), and it would
therefore be surprising if older adults did not
show an isolation effect. Contrary to the only
other published study, the result s reported here
show that older adults do show a typical isolation
effect, and, in parallel with findings in other
areas, the effect is reduced for older adults.
Original manuscript received 4 April 2007
Accepted revision received 27 July 2007
First published online day month year
REFERENCES
Cimbalo, R. S., & Brink, L. (1982). Aging and the von
Restorff isolation effect in short/term memory. The
Journal of General Psychology, 106, 69–76.
Clark, J. M., & Paivio, A. (2004). Extensions of the
Paivio, Yuille, and Madigan (1968) norms.
Behavior Research Methods, Instruments &
Computers, 36, 371–383.
Fabiani, M., Karis, D., & Donchin, E. (1990). Effects of
mnemonic strategy manipulation in a von Restorff
paradigm. Electroencephalography and Clinical
Neurophysiology, 75, 22–35.
Gumenik, W. E., & Levitt, J. (1968). The von Restorff
effect as a function of the difference of the isolated
item. American Journal of Psychology, 81, 247–252.
Howe, M. L., Courage, M. L., Vernescu, R., &
Hunt, M. (2000). Distinctiveness effects in children’s
long-term retention. Developmental Psychology, 36,
778–792.
Hunt, R. R. (1995). The subtlety of distinctiveness:
What von Restorff really did. Psychonomic Bulletin
& Review, 2, 105–112.
Hunt, R. R., & Lamb, C. A. (2001). What causes the
isolation effect? Journal of Experimental Psychology:
Learning, Memory, and Cognition, 27, 1359– 1366.
Kane, M. J., Hambrick, D. Z., Tuholski, S. W.,
Wilhelm, O., Payne, T. W., & Engle, R. W.
(2004). The generality of working memory capacity:
A latent-variable approach to verbal and visuospatial
memory span and reasoning. Journal of Experimental
Psychology: General, 133, 189–217.
Kelley, M. R., & Nairne, J. S. (2001). Von Restorff
revisited: Isolation, generation, and memory for
order. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 27, 54–66.
Ma
¨
ntyla
¨
, T., & Ba
¨
ckman, L. (1992). Aging and memory
for expected and unexpected objects in real-world
settings. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 18, 1298– 1309.
Nairne, J. S. (2006). Modeling distinctiveness:
Implications for general memory theory. In
R. R. Hunt & J. B. Worthen (Eds.), Distinctiveness
and memory (pp. 27–46). New York: Oxford
University Press.
Naveh-Benjamin, M. (2000). Adult age differences in
memory performance: Tests of an associative deficit
hypothesis. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 26, 1170– 1187.
Naveh-Benjamin, M., Guez, J., Kilb, A., & Reedy, S.
(2004). The associative memory deficit of older
adults: Further support using face–name associ-
ations. Psychology & Aging, 19, 541–546.
Naveh-Benjamin, M., Hussain, Z., Guez, J., &
Bar-On, M. (2003). Adult age-differences in episodic
memory: Further support for an associative-deficit
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0) 7
AGEING AND THE ISOLATION EFFECT
Downloaded By: [Memorial University of Newfoundland] At: 17:29 3 November 2007
hypothesis. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 29, 826–837.
Salthouse, T. A. (1993). Speed and knowledge as deter-
minants of adult age differences in verbal tasks.
Journal of Gerontology: Psychological Sciences, 48B,
29–36.
Salthouse, T. A. (1996). The processing speed theory of
adult age differences in cognition. Psychological
Review, 103, 403–428.
Smith, R. E. (2006). Adult age differences in episodic
memory: Item-specific, relational, and distinctive
processing. In R. R. Hunt & J. B. Worthen (Eds.),
Distinctiveness and memory (pp. 259–287).
New York: Oxford University Press.
Spencer, W. D., & Raz, N. (1995). Differential effects
of aging on memory for content and context: A
meta-analysis. Psychology & Aging, 10, 527–539.
Taconnat, L., & Isingrini, M. (2004). Cognitive oper-
ations in the generation effect on a recall test: Role
of aging and divided attention. Journal of
Experimental Psychology: Learning, Memory, and
Cognition, 30, 827–837.
Troyer, A. K., Ha
¨
fliger, A., Cadieux, M. J., &
Craik, F. I. M. (2006). Name and face learning in
older adults: Effects of levels of processing, self-
generation, and intention to learn. Journal of
Gerontology: Psychological Sciences, 61B, P67–P74.
von Restorff, H. (1933). Analyse von Vorgangen in
Spurenfeld. I. U
¨
ber die Wirkung von
Bereichsbildungen im Spurenfeld [Analysis of pro-
cesses in the memory trace. I. On the effect of
group formations on the memory trace].
Psychologische Forschung, 18, 299–342.
Zacks, R. T., Hasher, L., & Li, K. A. H. (2000). Human
memory. In F. I. M. Craik & T. A. Salthouse (Eds.),
The handbook of aging and cognition (2nd ed; pp.
293–357). Mahwah, NJ: Lawrence Erlbaum
Associates.
8 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 0000, 00 (0)
BIRETA, SURPRENANT, NEATH