Anda di halaman 1dari 8

Acta Psychologica 157 (2015) 8592

Contents lists available at ScienceDirect

Acta Psychologica
journal homepage: www.elsevier.com/ locate/actpsy

Selective decits in episodic feeling of knowing in ageing: A novel use of


the general knowledge task
Suzannah M. Morson a, Chris J.A. Moulin b, Cline Souchay b,
a
Institute of Psychological Sciences, University of Leeds, Leeds LS2 9JT, United Kingdom
b
LEAD UMR CNRS 5022, Universit de Bourgogne, 21065 Dijon, France

a r t i c l e i n f o a b s t r a c t

Article history: Failure to recall an item from memory can be accompanied by the subjective experience that the item is known
Received 10 March 2014 but currently unavailable for report. The feeling of knowing (FOK) task allows measurement of the predictive ac-
Received in revised form 19 February 2015 curacy of this reective judgement. Young and older adults were asked to provide answers to general knowledge
Accepted 22 February 2015
questions both prior to and after learning, thus measuring both semantic and episodic memory for the items. FOK
Available online 5 March 2015
judgements were made at each stage for all unrecalled responses, providing a measure of predictive accuracy for
PsycINFO classication:
semantic and episodic knowledge. Results demonstrated a selective effect of age on episodic FOK resolution, with
2343 older adults found to have impaired episodic FOK accuracy while semantic FOK accuracy remained intact. Al-
2860 though recall and recognition measures of episodic memory are equivalent between the two age groups, older
adults may have been unable to access contextual details on which to base their FOK judgements. The results sug-
Keywords: gest that older adults are not able to accurately predict future recognition of unrecalled episodic information, and
Metamemory consequently may have difculties in monitoring recently encoded memories.
Ageing
2015 Elsevier B.V. All rights reserved.
Feeling of knowing
Episodic memory
Semantic memory

1. Introduction 1965; Nelson & Narens, 1990), or from episodic memory (Schacter,
1983; Souchay, Isingrini, & Espagnet, 2000). Thus, FOK judgements are
Many studies have now found that age has differential effects on predictions about material that participants failed to retrieve and, al-
episodic and semantic memory. For tasks requiring a large degree of though not perfect, these judgements have been found to be relatively
involvement from semantic memory processes (e.g., general knowledge accurate in young adults (Gruneberg & Monks, 1974; Kelemen, Frost,
tasks), older adults are typically able to perform at similar levels to young & Weaver, 2000; Nelson, Gerler, & Narens, 1984; Schacter, 1983).
adults. However for tasks measuring episodic memory (e.g., word list In the ageing metamemory literature, results are consistent with the
learning), older adults are not able to match younger adult performance; memory literature in that no age effect is observed on semantic FOK
problems at encoding and/or retrieval lead to lower levels of accuracy in judgements. Buttereld, Nelson, and Peck (1988), Lachman, Lachman,
older adult samples (see Anderson & Craik, 2000; Zacks, Hasher, & Li, and Thronesbery (1979), Bckman and Karlsson (1985) and Marqui
2000 for reviews). Related memory processes involving the knowledge and Huet (2000) all used a general knowledge task similar to that of
and control of memory, termed metamemory (Flavell, 1979), may also the original Hart (1965) task. Participants were asked to recall answers
be affected by ageing. Whether all metamemory processes (e.g., judge- to a series of general knowledge questions and, upon failure to recall, to
ments of condence, feeling of knowing; Nelson & Narens, 1990; 1994) give a prediction about how likely they were to recognise the correct an-
are equally affected is still a matter of debate (Souchay & Isingrini, 2012). swer if shown it, i.e., the FOK judgement. This could be in the form of a
The feeling of knowing (FOK) metamemory paradigm can be used to simple binary FOK (Yes I will recognise or No I will not recognise), as a
examine peoples' ability to predict future recognition for both semantic percentage rating of the likelihood of recognition, or through a series of
and episodic items, and so is a useful tool to explore age effects on both relative judgements for each unrecalled item. Whether using a binary
types of memory awareness. In this procedure, participants are asked to FOK (Buttereld et al., 1988), a rating scale FOK (Lachman et al., 1979;
estimate the likelihood that they will recognise a piece of information Marqui & Huet, 2000), or relative FOK judgements (Buttereld et al.,
they have failed to recall earlier, either from semantic memory (Hart, 1988) no differences were observed in the predictive accuracy of
young and older adult judgements for future recognition. Likewise,
Corresponding author at: LEAD UMR CNRS 5022, Universit de Bourgogne, Ple AAFE
Allen-Burge and Storandt (2000) examined semantic memory and
Esplanade Erasme, 21065 Dijon cedex, France. Tel.: +33 3 80 39 90 25. FOKs for rare word denitions, again nding similar levels of predictive
E-mail address: celine.souchay@u-bourgogne.fr (C. Souchay). accuracy of FOK judgements in young and older adults. The preservation

http://dx.doi.org/10.1016/j.actpsy.2015.02.014
0001-6918/ 2015 Elsevier B.V. All rights reserved.
86 S.M. Morson et al. / Acta Psychologica 157 (2015) 8592

of semantic memory processes in ageing would thus appear to extend to However, the accuracy of the older adult group was shown to be signif-
memory monitoring processes of semantic memory. icantly below that of the young adult group. Despite exhibiting some
When considering episodic FOK accuracy, the effect of ageing is an ability to judge their recognition of unrecalled items, older adults were
issue of some debate. In agreement with the ndings of the memory lit- unable to do this to the same level of accuracy as young adults. This
erature, Perrotin, Isingrini, Souchay, Clarys, and Taconnat (2006), would therefore support previous ndings observing a selective episodic
Souchay et al. (2000), and Thomas, Bulevich, and Dubois (2011) have FOK decit in older adults (Perrotin et al., 2006; Souchay et al., 2000;
all observed decits in episodic FOK accuracy in older adults for word Thomas et al., 2011).
pair learning. However, Eakin and Hertzog (2012), Hertzog, Dunlosky, Although Souchay et al. (2007) ensured the target items were
and Sinclair (2010) and MacLaverty and Hertzog (2009) failed to nd identical in the semantic and episodic task, thus removing target charac-
an episodic FOK decit. Although a comparable word pair learning teristics as a potential confound on FOK accuracy, task characteristics
paradigm was used, older adults in these studies were able to predict could still be a factor. The prompt used to elicit the recall attempt in
their future recognition with a similar level of accuracy to that of each task is different, with the semantic task involving a more concep-
young adult participants. Several hypotheses have been proposed to tual cue whereas the episodic task involves a more contextual cue
explain potential age differences on episodic FOK, such as a decit in ex- (Koriat, 1997; Perfect & Hollins, 1999). This may inuence the strategies
ecutive functioning (Souchay et al., 2000) or a decit in recollection used at the recall attempt, thereby helping to determine the quality and
(Souchay, Moulin, Clarys, Taconnat, & Isingrini, 2007). More recently, quantity of information accessed on which the FOK judgement will be
Hertzog et al. (2010) suggested that impoverished memory representa- based. To remedy this methodological issue, Eakin, Hertzog, and Harris
tion of the items due to decient encoding in older adults could explain (2014) recently presented a new experimental paradigm in which
the age-effect observed on episodic FOK in some studies. This memory they adopted a facename associative task to equate the method of cue-
constraint hypothesis (MCH; Hertzog et al., 2010) proposes that FOK accu- ing for both episodic and semantic tasks. For both tasks, participants
racy is primarily dependent on the quality of the underlying memory were presented with a picture of a face as a cue, a famous face for the se-
processes during learning. If encoding is impaired in some way, then in- mantic task and a non-famous face for the episodic task, and asked to
sufcient or incorrect partial information will be accessed during the provide the name. Results showed a signicant age by memory task
failed recall attempt. Subsequently, the FOK judgement will be based interaction, with older adults showing poorer performance in the ep-
on these awed details, leading to lower predictive accuracy (Koriat, isodic condition for both recall and recognition tasks. Furthermore,
1993, 1997). As episodic memory is impaired in ageing, the FOK decit no group differences were observed on the semantic condition.
observed in some studies may simply be due to a lack of sufcient However, unlike Souchay et al. (2007), this study did not reveal
encoding to allow diagnostic partial information to be available to the any age effect on FOK accuracy for the episodic FOK or the semantic
participant as opposed to a decit in metacognitive ability (see Perfect FOK. Discrepancies between this study and Souchay et al. (2007)
& Stollery, 1993 for a similar argument). Indeed, by equating young and are difcult to explain. As suggested by Eakin et al. (2014), the lack
older adults' memory performance using variable delays, Hertzog et al. of age effect observed in their study on episodic FOK accuracy
(2010) were able to demonstrate equivalent levels of episodic FOK accu- could be related to the material used. Indeed, older adults may do
racy in the two age groups. However, although some of the studies that well when visual cues are used to make FOK judgements, as they may
have found age effects on FOK do also show age effects on memory per- be able to retrieve sufcient information to inform their judgement.
formance (Perrotin et al., 2006; Souchay et al., 2000, 2007), not all of However, whether or not discrepancies between the two studies are
them do. For all three experiments reported by Thomas et al. (2011), due to differences in material used is difcult to ascertain considering
memory performance on the cued recall aspect of the task was compara- the methodological issues in Souchay et al. (2007).
ble between the young and older adult participants. Despite this, older Therefore, the present study aims to remove the possible inuence
adults consistently demonstrated a decit in episodic FOK accuracy. of task characteristics on Souchay et al. (2007) ndings by utilising a
When considering whether age does lead to a selective impairment general knowledge task for both the semantic and episodic FOK tasks.
in episodic FOK accuracy it is important to consider one key limitation of To do so, a classical FOK semantic task was conducted rst by presenting
the studies discussed so far. Examination of semantic and episodic FOK participants' general knowledge items, with participants asked to make
accuracy has typically been conducted in isolation, therefore the accura- FOK judgements on questions they do not know the answer for. The
cy of each of these processes has been established in different partici- novelty of this design is to use the unknown items from the semantic
pant samples and then compared. To the best of our knowledge only FOK task as targets in the episodic FOK task. Thus, for the episodic con-
two studies have examined both semantic and episodic FOK accuracy dition, answers which are not known will subsequently be learned. The
within the same group of participants, and these studies have revealed unknown general knowledge questions become the episodic version of
contrasting ndings. Souchay et al. (2007) asked participants to com- the task, thereby ensuring that the same cues are used for both tasks,
plete both a general knowledge task and a word pair learning task, removing this as a possible confound. Episodic FOK judgements will
thereby allowing them to directly compare accuracy in the two tasks. then be made on answers that participants cannot recall.
In addition, the same target items were used in each task, allowing a
further level of control in the study. For example, in the semantic task 2. Method
the question may have been What was the subject of Magritte's famous
surrealist painting La Trahison?, the answer being Pipe. In the episodic 2.1. Participants
task, an unrelated cue word would be paired with the same item,
e.g., Birthday Pipe. The intrinsic properties of target words, such Thirty ve undergraduate students (age range 18 to 29, M = 20.23,
as frequency, can have an impact on the FOK judgement (Koriat, SD = 2.89; nine males) from the University of Leeds participated in the
1993). By using the same target items, and by counterbalancing study in return for course credit. The older adult group consisted of 21
task order, Souchay et al. (2007) were able to match the properties people aged between 60 and 85 (M = 69.86, SD = 8.00; three males)
of the required target in the semantic and episodic tasks. For the seman- recruited from the local community. Both groups reported a similar
tic task, as in previous research, comparable performance was observed number of years of education (young adults M = 15.69, SD = 2.46;
between young and older adults: both groups were able to accurately older adults M = 14.10, SD = 3.35; t(54) = 1.89, p = .067). All older
predict future recognition accuracy for unrecalled items. When examin- adults obtained scores on the Mini Mental State Examination
ing performance on the episodic task, the accuracy of both young and (MMSE; Folstein, Folstein, & McHugh, 1975) above the cutoff of 27
older adults' FOK judgements were above chance, indicating that both (M = 28.79, SD = 1.44) and were not taking medication which
age groups were able to predict their performance to a certain extent. would affect cognitive function. The research received approval
S.M. Morson et al. / Acta Psychologica 157 (2015) 8592 87

12. Which channel island is nearest to France?


from the ethical committee at the Institute of Psychological Sciences,
University of Leeds. All participants provided written informed con-
sent to take part in the study.
OR Would you recognise the answer if it was given to you? Yes No
2.2. Materials
Fig. 1. Example of response booklet for obtaining recall and FOK measures.
An initial set of 50 general knowledge questions was created and
administered to ve young adults as pilot participants who did not the correct answer to each question, and to guess where they were un-
take part in the main study. This allowed assessment of the ease of the sure of the answer, until all 60 questions had been assigned a response.
questions and the number of items that would elicit FOK responses No time limit was imposed for any phase of the Semantic task.
i.e., errors of omission. Average recall was 11 items, with 23.4 items
leading to an FOK judgement. Recognition accuracy was at 19.4 items.
2.3.2. Episodic task
Based on this, the number of items for the nal study was increased to
The procedure included a study phase, a cued recall phase, a FOK
allow greater scope for learning for the Episodic aspect of the task. The
phase and a recognition phase. In the study phase, participants were
nal materials consisted of 60 questions covering a variety of topics
presented with the same 60 questions as those in the Semantic task to-
including sport (e.g., Which Yorkshire team rst won the FA Cup), history
gether with the correct responses, irrespective of whether participants
(e.g., Which eet tried to invade England in 1588), geography (e.g., Which
successfully recalled the item or not during the semantic task. Questions
channel island is nearest to France) and general knowledge (e.g., Where
were printed on two sheets of paper with the correct answer highlight-
did the turnip originate)1. Questions were printed in two answer book-
ed in bold. Presentation was timed at 120 s, and participants were asked
lets, one for Semantic recall and recognition and one for Episodic recall
to learn the correct answers to the questions without marking the
and recognition. Question order was randomised for each recall and
sheets in any way. A response booklet was then provided to record re-
recognition test for the Semantic and Episodic tasks. A second set of
sponses. Cued recall and FOK judgements were obtained in the same
booklets was created to allow counterbalancing of the question order.
way as in the Semantic task, with the 60 questions printed in the book-
let with recall response boxes and FOK prompts, and the instruction not
2.3. Procedure
to guess. Upon completion of either a recall or FOK response for all ques-
tions, the recognition phase was completed. This again took the form of
All participants were tested individually. A standard recalljudge-
a four-alternative forced choice, with the same distracters used as in the
mentrecognition procedure was used to obtain FOKs, similar to those
Semantic task.
previously utilised in Semantic (Bckman & Karlsson, 1985) and Episodic
(Schacter, 1983) FOK tasks. The key difference in the administration of Se-
mantic and Episodic FOK tasks is the inclusion of a learning stage prior to 3. Results
recall in the Episodic FOK task only. The novelty of the current procedure
was to capture Semantic and Episodic FOK measures from the same par- Consideration of which items to class as Episodic needs clarication.
ticipants for the same target items which have also been elicited by the All items were included in the analysis of Semantic memory and
same cues i.e., the memory search for the target item Alderney is metamemory performance. However, during the Episodic task proce-
prompted by the question Which channel island is nearest to France? for dure, all items were presented for a second time, thereby including
both the Semantic FOK and the Episodic FOK. Due to this procedural ne- items which were already stored as semantic knowledge. This would
cessity, all participants completed the Semantic task followed by the Ep- therefore contaminate the Episodic measure. The following analysis
isodic task. Participants were provided with written instructions for the employs a strict criterion to identify Episodic items based on recognition
task, and two example questions were provided to ensure participants performance, whereby items which were correctly recognised during
understood the FOK judgement. Questions were presented in a different the Semantic task were subsequently removed from the Episodic
order at each test of recall and recognition and also at learning. analysis.
Five older adult participants were removed from the analysis due to
2.3.1. Semantic task no errors of omission occurring, meaning that no metamemory data
The procedure included a cued-recall combined with FOK judge- could be analysed for these participants. The subsequent analysis
ment phase, followed by a recognition phase. All recall, FOK and recog- involved 35 young adults (age M = 20.23, SD = 2.89) and 16 older
nition responses were recorded by participants in the booklet provided. adults (age M = 68.75, SD = 8.20). All signicant interactions were
In the recall test, the 60 questions were presented with a response box followed up with the appropriate t-tests, with a Bonferroni correction
beneath (see Fig. 1 for an example). Participants were asked to write the applied to control for multiple comparisons.
correct answer if known, or to leave the response box blank if the cor-
rect answer was not known. Guessing was discouraged. Immediately 3.1. Memory
after the recall attempt, the FOK judgement was made. FOK judgements
were only made for those items where a response was not attempted 3.1.1. Recall performance
i.e., omission errors. The FOK was prompted with the question Would As can be seen in Fig. 2, older adults show a distinct advantage in
you recognise the correct answer if it was given to you?. If participants Semantic memory while Episodic memory is more similar between
thought that they would be able to recognise the correct answer, they the two age groups. A main effect of age was found on recall perfor-
circled a Yes response; if participants thought that they would not rec- mance, F(1,49) = 18.40, p b .001, 2 = .27, with older adults showing
ognise the correct answer, they circled a No response. Once recall or greater accuracy than young adults. Similarly, recall on the Episodic
FOK had been completed for the question, participants then moved on task was also signicantly greater than recall for the Semantic task,
to the next question until all 60 questions had received a response. F(1,49) = 347.53, p = .001, 2 = .88. An interaction between age and
The recognition phase followed immediately after. The 60 questions task, F(1,49) = 7.34, p = .009, 2 = .13, showed that although both
were presented again with four alternative responses, the correct re- groups were signicantly more accurate in the Episodic task than the
sponse plus three plausible distracters. Participants were asked to circle Semantic task (young adults: t(34) = 22.58, p b .001; older adults:
t(15) = 7.48, p b .001), at the Semantic trial older adults are at a greater
1
The interested reader should note that the correct answers are Shefeld Wednesday, level of accuracy than young adults, t(49) = 6.50, p b .001, whereas at
the Spanish Armada, Alderney, and Greece. the Episodic trial no age difference is present t(49) = 1.58, p = .120.
88
Proportion correct recall and recognition S.M. Morson et al. / Acta Psychologica 157 (2015) 8592

Table 1
1 FOK measures.
Young adults
0.9
Older adults Young adults (N = 35) Older adults (N = 16)
0.8 M (SD) M (SD)
0.7 Yes FOKs Semantic 0.301 (0.158) 0.442 (0.146)
0.6 Episodic 0.572 (0.277) 0.545 (0.432)
Gamma scores Semantic 0.345 (0.303) 0.455 (0.214)
0.5
Episodic 0.477 (0.408) 0.177 (0.459)
0.4
Yes FOKs: mean proportions of items assigned a Yes FOK.
0.3
0.2 interaction between age and task, F(1,49) = 2.95, p = .092, 2 = .06,
0.1 was found due to young adults increasing Yes responding between
0 tasks, t(34) = 5.84, p b .001, whereas older adults do not, t(15) =
Semantic Episodic Semantic Episodic 0.99, p = .336. In addition, older adults gave more Yes FOK responses
Recall Recognition at the Semantic task than young adults, t(49) = 3.05, p = .004, whereas
no age differences in responding were present for the Episodic task,
Fig. 2. Memory measures for young and older adults. Proportions for Semantic measures t(49) = 0.22, p = .825.
were calculated by dividing the number of items the participant correctly recalled or
recognised by the total number of items on the task. Proportions for the Episodic task 3.2.2. Feeling-of-knowing accuracy
were calculated by dividing the number of items the participant correctly recalled or An indication of FOK accuracy can be obtained by examining the
recognised by the number of items the participant did not correctly recognise at the
accuracy of Yes and No FOKs i.e., the proportion of Yes FOKs subsequent-
Semantic test.
ly recognised and the proportion of No FOKs subsequently recognised.
Accuracy of Yes and No FOKs was analysed with a 2 (age) 2
3.1.2. Recognition performance (task) 2 (FOK status: Yes or No) repeated measures ANOVA. No effect
Recognition also showed a main effect of age, F(1,49) = 8.06, of age, F b 1, or task, F(1,33) = 1.08, p = .307, 2 = .03, was present. A
p = .007, 2 = .14, with older adults recognising a greater proportion main effect of status, F(1,33) = 48.99, p b .001, 2 = .60, conrms that
of items correctly than young adults as previously (see Fig. 2). A main the FOK judgements do show some diagnostic accuracy, with Yes
effect of task was also present, F(1,49) = 118.33, p b .001, 2 = .71, FOKs resulting in greater recognition accuracy than No FOKs as
again with Episodic task recognition being higher than that of Semantic would be expected (see Fig. 3). An interaction between age and
task recognition. An interaction between age and task mirrored that task, F(1,33) = 7.91, p = .008, 2 = .19, indicates that older adults
of the interaction in recall performance, F(1,49) = 31.37, p b .001, show a slight decrease in performance from the Semantic task to
2 = .39, with greater improvement between tasks in the young the Episodic task. In contrast, young adults show an increase in per-
adult group, t(34) = 16.80, p b .001, than in the older adult group, formance, with Semantic task performance lower than that of Episodic
t(15) = 2.58, p = .021. task performance.
A non-signicant interaction was found between age and FOK status,
F(1,33) = 3.32, p = .078, 2 = .09, suggesting that both groups give
3.2. Metamemory
more Yes FOKs for recognised items. If anything, the means indicate
that older adults show a greater disparity between Yes and No FOK
It is important to note that the FOK is based on unrecalled items only.
performance than young adults, indicating greater metacognitive ac-
Items which are incorrectly recalled are not included in the FOK analy-
curacy. No other interactions were signicant (task and FOK status:
sis, as these items have a different relationship to future recognition
F(1,33) = 2.01, p = .166, 2 = .06; age, task and FOK status: F b 1).
than items where no response is given (Krinsky & Nelson, 1985). First,
The association between the Yes/No FOK judgement and recognition
the proportion of Yes FOK responses were analysed, irrespective of
performance as obtained in the present study is most appropriately
later recognition accuracy, to establish whether a bias in responding
measured with the GoodmanKruskal gamma correlation (Nelson,
occurred. An indication of FOK accuracy was then considered by com-
1984; Schraw, 1995; Wright, 1996; see Masson & Rotello, 2009, for dis-
paring the proportion of Yes FOKs which were subsequently recognised
cussion of measures when using rating scales for FOK). Four possible
to the proportion of No FOKs which were subsequently recognised
outcomes can occur with a binary FOK: (a) correct recognition for Yes
(Souchay & Smith, 2013). FOK accuracy is indicated if recognition of
Yes FOK items exceeds recognition of No FOK items. While this type of
Proportion of items corectly recognised

analysis provides some evidence of the resolution of the FOKs, it is not


possible to quantify the accuracy of the FOK due to sensitivity to recog- 1
Young adults
nition accuracy (Nelson, 1984). Although subject to debate (Schraw, 0.9
Older adults
1995; Masson & Rotello, 2009), the gamma correlation (explained 0.8
below) has been identied and adopted within the metacognitive liter- 0.7
ature as one of the most appropriate measures of FOK accuracy to allow 0.6
for comparison (Nelson, 1984; Wright, 1996). Thus, gamma correlations 0.5
to examine FOK accuracy are presented last. 0.4
0.3
3.2.1. Bias in responding 0.2
To examine possible bias in responding, the proportion of unrecalled 0.1
items that were assigned a Yes FOK response was calculated, irrespec- 0
tive of recognition accuracy. A 2 (age) 2 (task) repeated measures Yes FOK No FOK Yes FOK No FOK
ANOVA on the proportion of Yes FOKs given revealed no effect of age, Semantic Task Episodic Task
F(1,49) = 0.89, p = .350, 2 = .02 (see Table 1). A main effect of task
was present, F(1,49) = 14.65, p b .001, 2 = .23, with more Yes Fig. 3. Proportion of items correctly recognised of Semantic and Episodic items assigned
FOKs given in the Episodic task than in the Semantic task. A marginal Yes and No FOKs for young and older adults.
S.M. Morson et al. / Acta Psychologica 157 (2015) 8592 89

FOKs, (b) incorrect recognition for Yes FOKs, (c) correct recognition for equivalent memory accuracy therefore suggests that decits in
No FOKs, and (d) incorrect recognition for No FOKs. The gamma encoding cannot be the sole contributor to age associated FOK
correlation compares the proportion of correct predictions (a and impairments.
d) to incorrect predictions (b and c). This leads to a coefcient ranging This conclusion is perhaps not surprising when considering the basis
from 1 to 1, with large positive values indicating a close relationship of FOK judgements. Although aimed at explaining age effects on episod-
between FOK judgement and recognition, large negative values indicat- ic FOKs rather than describing how FOKs are made more generally, the
ing an inverse relationship between FOKs and recognition, and values MCH is similar to a trace-access view, whereby FOK processes have di-
close to zero showing chance performance or guessing. However, rect access to the target memory trace leading to their predictive ability.
gamma is undened when two of the four possible outcomes (a, b, c, These approaches have largely fallen out of favour due the very issue ob-
d) are equal to 0. Therefore, an adjusted gamma score was calculated served above: if the FOK has direct access to the target, why is predictive
following Snodgrass and Corwin's (1988) recommendations (see accuracy not at or near ceiling? Inferential hypotheses answer this by
Souchay et al., 2000, 2007), whereby 0.5 was added to each frequency assuming that the FOK has only indirect access to the target memory,
and then divided by N + 1, where N is the number of judgements. and so must rely on cues as to the future accessibility of the memory.
Using this gamma score, a 2 (age) 2 (task) repeated measures These cues may be generated by the failed recall attempt (Koriat,
ANOVA revealed no effect of age, F(1, 49) = 1.56, p = .218, 2 = .03, 1993, 1997), for example the length of the memory search prior to re-
or of task, F b 1, on FOK accuracy (see Table 1). A signicant interaction trieval failure (Singer & Tiede, 2008). A longer memory search may
was found however, F(1,49) = 7.04, p = .011, 2 = .13. Young adults give the impression that the target item is within the memory system,
show no difference in accuracy between the Semantic and Episodic leading to higher (and possibly more accurate) FOKs. In addition, the
tasks, t(34) = 1.49, p = .145, whereas older adults show signicantly FOK may be based on information relating to the probe used to cue
lower accuracy in the Episodic task compared to the Semantic task, the memory: the general knowledge question in the present study, or
t(15) = 2.28, p = .037. In addition, while no effect of age was found the cue word in a paired-associates paradigm. The familiarity of this
for the Semantic task, t(49) = 1.30, p = .200, for the Episodic task pointer thus inuences the FOK, with more familiar cues leading to
older adults were signicantly lower in accuracy than the young adults, higher FOK judgements (Reder, 1987; Schwartz & Metcalfe, 1992). An
t(49) = 2.34, p = .023. Gamma correlations were also compared to inability to use any or all of these cues appropriately will lead to impair-
chance performance to establish if predictions were indicative of recog- ments in the predictive accuracy of FOK judgements. Indeed, there is
nition accuracy. For young adults, both Semantic, t(34) = 6.74, p b .001, evidence that while older adults have access to cues on which the FOK
and Episodic, t(34) = 6.92, p b .001, gammas were greater than chance. can be based, they do not spontaneously use this information for epi-
However, for older adults, only Semantic gamma was above chance, sodic FOK tasks (Thomas et al., 2011). The observed decit in episodic
t(20) = 4.90, p b .001; Episodic gamma, t(15) = 1.54, p = .144. FOKs for older adults in the present study may therefore be related to in-
efcient use of inferential cues rather than reecting episodic memory
4. Discussion decits.
It should be emphasised that the rejection of the MCH as an explana-
The inuence of ageing on episodic FOK accuracy is the subject of tion for the present ndings is not due to its proposal that encoding
much debate, with evidence for both preserved (MacLaverty & Hertzog, decits contribute to FOK inaccuracy. Rather, that it is awed in its as-
2009) and impaired (Thomas et al., 2011) predictive accuracy in older sumption that equating recall and recognition should remove the effect
adults. Reaching a consensus has been hindered by limitations with of any encoding decits and lead to accurate FOK predictions in older
study design, with previous research comparing separate participant adults. The lack of age effects on recall and recognition does not neces-
groups for semantic and episodic tasks (e.g., Allen-Burge & Storandt, sarily imply that encoding and/or retrieval processes are not impaired;
2000; Souchay et al., 2000), eliciting semantic and episodic FOKs with the decits may be so subtle that they are not detected by standard
different prompts (Souchay et al., 2007), or using visual rather than measures of memory. For example, older adults in the Thomas et al.
verbal stimuli (Eakin et al., 2014). These factors may have inuenced (2011) study achieved recall and recognition accuracy at a similar
the basis of the FOK judgement, confounding task effects with age level to young adults without the need to articially equate perfor-
effects and limiting the conclusions which can be made on the data mance via a delay (as in Hertzog et al., 2010). However, they were im-
obtained (Koriat, 1993, 1997). The present study therefore aimed to ad- paired at retrieving partial information related to the target, in this
dress these limitations with the implementation of a combined seman- case the valence of the word. Older adults who were able to retrieve par-
tic and episodic FOK task based on general knowledge questions within tial information provided more accurate FOKs than those who were un-
a single participant group. able to retrieve partial information. This suggests that the FOK, as well as
The key analysis in establishing whether a selective impairment in being reliant on encoding of the target item, is also reliant on encoding
episodic FOK accuracy occurs in ageing is the presence of an interaction the appropriate information associated with the target. Numerous stud-
between age and task. The current data clearly demonstrate such an in- ies have demonstrated that older adults are impaired at associative
teraction, with young and older adults demonstrating equivalent FOK memory tasks (e.g., Chalfonte & Johnson, 1996; Naveh-Benjamin,
accuracy for semantic items while older adults obtain lower predictive 2000; Old & Naveh-Benjamin, 2008). Thus recall and recognition could
accuracy for episodic items. Furthermore, older adults fail to reach be considered too coarse a measure of memory to ascertain the pres-
above chance levels of accuracy for their episodic FOK predictions, indi- ence of decits which may inuence FOK ability.
cating not merely a reduced predictive accuracy compared to young The importance of information associated with the target memo-
adults but an absence of predictive accuracy within this task. Interest- ry is further demonstrated by the link observed between FOK and
ingly, this age effect on episodic FOK accuracy occurs in the absence of recollective processes. Recollection is a subjective experience whereby
any age effects on episodic memory performance. This pattern of nd- the rememberer recalls thoughts, sensations and contextual details in
ings is somewhat problematic for the MCH (Hertzog et al., 2010), addition to the target memory, leading to a rich memory experience
which posits that inaccurate episodic FOKs in older adults are primarily (Tulving, 1985). Recollection therefore cannot occur without access to
due to decits in memory encoding, and that by using tasks which details associated with the target memory. Both objective (e.g., source
equate recall and recognition performance, no age effects will memory tasks, Benjamin & Craik, 2001; Dennis et al., 2008; Henkel,
occur for the FOK. In the current study episodic memory is equivalent Johnson, & De Leonardis, 1998) and subjective (e.g., the Remember
between the two age groups, as evidenced by the lack of age effects Know procedure, Jennings & Jacoby, 2003; Perfect & Dasgupta, 1997)
in recall and recognition, thus the MCH would predict preserved ep- measures of recollection have revealed a decreased level of recollective
isodic FOK resolution. The presence of an episodic FOK decit despite experience in older adults, supporting the idea that an associative
90 S.M. Morson et al. / Acta Psychologica 157 (2015) 8592

decit occurs in ageing and affects subjective memory experiences. answers is therefore unlikely to have inuenced FOK accuracy within
The relevance of this to the FOK is conrmed by a number of studies the current study. The third issue relates to experience of the task.
reporting correlations between recollective experience and FOK ac- Metamemory judgements are somewhat based on general impressions
curacy. Hicks and Marsh (2002) observed that higher levels of re- of how well you as an individual perform on tasks.
ported recollective experience were accompanied by higher levels Consider a task where the participant is asked to predict how many
of FOK accuracy. Furthermore, Souchay et al. (2007) showed that of 10 words they will successfully recall at test, with very little other in-
the age effect on recollection explained, in part, the age differences formation. They will only be able to use their general beliefs about their
observed in FOK accuracy. Therefore, while not measured in the memory to give a judgement. If the experimenter gives a recall test after
present study, a difculty to retrieve details associated to the target a 30-minute delay, when the participant made their judgement based
when recall fails may explain the FOK decits observed. on the assumption of an immediate test, this will lead to lower recall
In addition to inappropriate use of inferential cues and decits in performance than expected, affecting the accuracy of the judgement.
associative memory, it is possible that metamemory processes are When given the task a second time, the participant may well factor in
inuenced by other age-affected cognitive functions. For example, the effect of the delay on their judgement, leading to greater predictive
conceptually, metamemory is similar to executive function, whereby accuracy (see Moulin, Perfect, & Jones, 2000 for such a pattern). A sim-
information is monitored and controlled to allow for voluntary ac- ilar effect could occur for the episodic FOK in the current study. Having
tion (Miyake et al., 2000). Due to this theoretical resemblance, a already experienced the recall and recognition tasks, the participant
number of researchers have begun to examine the relationship between may use this knowledge of the criterion test to inform their FOK, affect-
metamemory and executive function (Fernandez-Duque, Baird, & ing the predictive accuracy. If young and older adults apply this knowl-
Posner, 2000; Roebers, Cimeli, Rthlisberger, & Neuenschwander, edge to a different degree, this may result in one group showing a higher
2012). In this context, Souchay et al. (2000) proposed that associated level of accuracy than the other. To date, no study has explored the im-
decits in executive function may be the driving force behind problems pact of repeated FOK procedures on the accuracy of the judgements
with FOK predictions. Young and older adults were examined on two made and so these issues remain a priority for future research. It may
measures of executive function, the Wisconsin Card Sorting Task and well be the case that we have to metacognitively reect on information
the Verbal Fluency Test, in addition to a paired associates FOK task. which is repeatedly forgotten, or difcult-to-learn information which is
Not only were executive function scores highly correlated to FOK scores presented on repeated trials. For the current study, it is reasonable to
for all ages, regression analysis revealed that the executive function presume that the effects of repetition are equal for young and older
measures accounted for 86% of the variance in episodic FOK accuracy adults, and so do not account for the group differences observed. It
within the older adult group. Further studies have conrmed executive should however be noted that the task is not process pure since the ep-
functioning as a mediator for the predictive accuracy of episodic FOKs in isodic items have already been presented in the semantic task, and there
ageing (Perrotin et al., 2006; Souchay & Isingrini, 2004; Souchay, will be episodic reections both on how they performed in that previous
Isingrini, Clarys, Taconnat, & Eustache, 2004). Therefore, in the present task, as well as presumably an episodic registration of what the
study, age-related differences in FOK accuracy could also be explained given correct response was.
by age-related differences in executive functioning. A nding of the current study requiring further consideration is the
The novel task design within the current study raises a topic which observed advantage for older adults in the semantic memory task.
has not previously been fully considered experimentally within the This higher level of recall has two possible implications: rst, that
FOK literature: how repeated presentations of the same cues/questions level of expertise may be inuencing FOK judgements and second,
in the FOK paradigm may inuence the basis or nature of the FOK that older adults will subsequently have less information to learn lead-
judgement. There are three issues related to repeated completion ing to an easier episodic task for this group. With regard to expertise,
of the FOK paradigm. First is the issue of repeated exposure to the three studies have examined the effect on FOK accuracy. Roberts and
memory cue i.e., the general knowledge question. Repetition of the Rhodes (1989) observed no effect of general knowledge expertise on
cues will lead to the items becoming more familiar to the partici- FOK accuracy when measured by self-report or when measured by an
pants, and cue familiarity has been found to affect FOK judgements objective test. Likewise, Marqui and Huet (2000) failed to nd an ef-
(Hanczakowski, Pasek, Zawadzka, & Mazzoni, 2013; Metcalfe, fect of self-efcacy beliefs on FOKs for a general knowledge task.
Schwartz, & Joaquim, 1993). However, the primary focus of these Peynirciolu and Tekcan (2000) examined language prociency
studies has been on magnitude rather than accuracy. While Metcalfe with a standardised test prior to measuring FOKs for TurkishEnglish
et al. (1993) do provide gamma correlations as a supplementary analy- translations. Again, no effect of scores on standardised testing was
sis, these are compared to zero only to establish if accuracy is above found on FOK accuracy. The difference in semantic memory observed
chance, with separate studies demonstrating both above and at chance in the current study will not therefore have had an inuence on the
levels of predictive accuracy for familiar cues. Although it is difcult to semantic FOK accuracies observed.
ascertain from previous research the effects of cue familiarity on FOK The subsequent effects of the semantic memory difference on the
accuracy in the present study, it may be that no effect occurs due to difculty level of the episodic FOK task are not so clear. The episodic
the task design. All items are presented at all stages, thus it may be as- task used a revised score to limit the effects of semantic knowledge on
sumed that familiarity increased at a similar rate for all items. Any effect the episodic measures: all items which were correctly recognised dur-
of repetition of cues would be equivalent for all participants, and may ing the semantic task were removed from the episodic analysis. This re-
not therefore be a confounding factor. Whether young and older adults sulted in young adults needing to encode an average of 37.2 items as
respond differently to differences in cue familiarity is, however, part of the episodic task, whereas older adults had 24.25 items to en-
currently unknown, and should be considered in future studies of cue code. Young and older adults correctly recalled a similar proportion of
repetition in FOKs. episodic items, and yet older adults exhibited an impairment in episodic
Secondly, participants are also exposed to the correct target on FOKs relative to young adults. While the older adult group had fewer
multiple occasions. The potential of repeated encoding trials to inu- items to learn, they were still presented with all of the items from the
ence FOK accuracy has been considered previously. When considered task. In order to take advantage of the reduced difculty level of the
between subjects, FOK accuracy does increase with repeated learning task, older adults would need to successfully identify which questions
(Nelson, Leonesio, Shimamura, Landwher, & Narens, 1982). However, they did not know the answer to and which they can ignore. However,
when considered within subjects, as with the present study, no effect older adults have repeatedly been shown to be impaired in the self-
of repeated learning on FOK accuracy is observed (Carroll & Nelson, initiation of strategies which will aid future memory performance
1993; Carroll & Simington, 1986). Repeated exposure to the correct (Dunlosky & Connor, 1997; Murphy, Sanders, Gabriesheski, & Schmitt,
S.M. Morson et al. / Acta Psychologica 157 (2015) 8592 91

1981; Murphy, Schmitt, Caruso, & Sanders, 1987). Thus despite the ob- whether the level of memory use by undergraduates is indeed a con-
jective difference in task difculty level, this may not have translated founding factor for assessments of memory awareness and memory
into benets for older adults' episodic memory, and consequently FOK, accuracy.
performance. Indeed, the present study showed that older adults re- In conclusion, the present study demonstrates the presence of a
main impaired on the epsiodic FOK despite higher performance in the selective decit in episodic FOK accuracy in older adults. This decit oc-
semantic memory task. The effect of reduced items needing to be curs despite preserved episodic memory performance, suggesting that
encoded could be examined by using a procedure similar to the selec- impairments in memory encoding processes are not responsible for
tive reminding task (Chiaravalloti, Balzano, Moore, & DeLuca, 2009), lower levels of predictive accuracy in ageing. Previous research has
whereby only items which were not successfully recalled are presented been inconclusive as to the presence or absence of an age effect within
at subsequent encoding trials. This would allow measurement of the re- the FOK. This may have been due to task and participant effects having
lationship between the difculty level of the episodic task (as dened by an effect of the metacognitive judgements made, thus masking the
the number of items to be encoded) and the accuracy of the episodic effects of age. The present study directly addresses these limitations,
FOK. providing the clearest measure to date of both semantic and episodic
The current discussion has focused on the notion that older adults FOK accuracy in ageing, and supporting previous ndings of isolated
were found to have a selective impairment of episodic FOK accuracy decits in episodic FOK resolution in older adults.
due to an inability to access or spontaneously use the available partial
information to inform their predictive judgements. It should be noted
that an alternative explanation could be proposed; that older adults References
are able to access and use partial information, but that they are utilising
Allen-Burge, R., & Storandt, M. (2000). Age equivalence in feeling-of-knowing experi-
the wrong cues to inform their judgements. As previously noted, cues ences. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences,
can be intrinsic to the task (e.g., length of memory search), but they 55(4), 214223.
may also be extrinsic to the task (Koriat, Bjork, Sheffer, & Bar, 2004), re- Anderson, N. D., & Craik, F. I. M. (2000). Memory in the ageing brain. In E. Tulving, & F. I.
M. Craik (Eds.), The Oxford handbook of memory. New York: Oxford University Press.
lying on general beliefs about how your personal memory functions, Bckman, L., & Karlsson, T. (1985). The relation between level of general knowledge and
such as a superior ability to recall faces over recalling names. Recall feeling-of-knowing: An adult age study. Scandinavian Journal of Psychology, 26(3),
and recognition in the episodic task are equivalent between the two 249258.
Benjamin, A. S., & Craik, F. I. M. (2001). Parallel effects of aging and time pressure on
age groups. This could be interpreted as showing unusually high ep- memory for source: Evidence from the spacing effect. Memory & Cognition, 29(5),
isodic memory performance in the older adult group, potentially due 691697.
to the novel method whereby cues and targets were subject to re- Buttereld, E. C., Nelson, T. O., & Peck, V. (1988). Developmental aspects of the feeling of
knowing. Developmental Psychology, 24(5), 654663.
peated encoding opportunities. If the older adult group relied pri-
Carroll, M., & Nelson, T. O. (1993). Effect of overlearning on the feeling of knowing is more
marily on extrinsic cues, in particular a belief that they have detectable in within-subject than in between-subject designs. The American Journal of
difculty learning new information (see Light, 1991, for a review of Psychology, 106(2), 227235.
self efcacy beliefs in ageing), then their judgements would not Carroll, M., & Simington, A. (1986). The effects of degree of learning, meaning, and indi-
vidual differences on the feeling-of-knowing. Acta Psychologica, 61, 316.
tally with the high level of performance achieved, resulting in inac- Chalfonte, B. L., & Johnson, M. K. (1996). Feature memory and binding in young and older
curate FOK predictions. The difculty in this interpretation lies adults. Memory and Cognition, 24(4), 403416.
with the assumption that equivalent recall and recognition reects Chiaravalloti, N. D., Balzano, J., Moore, N. B., & DeLuca, J. (2009). The Open-Trial Selective
Reminding Test (OT-SRT) as a tool for the assessment of learning and memory. The
superior memory performance in the older adult group. As described Clinical Neuropsychologist, 23(2), 231254.
above, older adults were required to encode fewer items for the ep- Connor, L. T., Dunlosky, J., & Hertzog, C. (1997). Age-related differences in absolute but not
isodic task than young adults. Recall and recognition measures are relative metamemory accuracy. Psychology & Aging, 12(1), 5071.
Dennis, N. A., Hayes, S. M., Prince, S. E., Madden, D. J., Huettel, S. A., & Cabeza, R. (2008).
presented as proportion accuracy due to differences in the raw num- Effects of aging on the neural correlates of successful item and source memory
ber of items. Studies examining the effect of list length on memory encoding. Journal of Experimental Psychology: Learning, Memory, and Cognition,
have observed that shorter lists are associated with higher propor- 34(4), 791808.
Dunlosky, J., & Connor, L. T. (1997). Age differences in the allocation of study time ac-
tion accuracy (Ward, 2002). Whether 13 fewer items is enough to count for age differences in memory performance. Memory & Cognition, 25(5),
allow this difference in proportion accuracy to manifest, and wheth- 691700.
er older adults can be expected to show the same effects of list Eakin, D. K., & Hertzog, C. (2012). Age invariance in feeling of knowing during implicit in-
terference effects. The Journals of Gerontology Series B: Psychological Sciences and Social
length, is unclear. Nonetheless, the equivalent recall and recognition
Sciences, 67(5), 555562.
performance could be interpreted as evidence of an episodic memo- Eakin, D. K., Hertzog, C., & Harris, W. (2014). Age invariance in semantic and episodic
ry decit in the older adult group. If this is the case, the inaccuracy of metamemory: Both younger and older adults provide accurate feeling-of-knowing
the FOK cannot be due to older adults' failing to account for an unex- for names of faces. Neuropsychology, Development, and Cognition. Section B, Aging,
Neuropsychology and Cognition, 21(1), 2751.
pectedly high level of performance on the task. Fernandez-Duque, D., Baird, J. A., & Posner, M. I. (2000). Executive attention and
One nal point of note is the use of an undergraduate student sample metacognitive regulation. Consciousness and Cognition, 9(2 Pt 1), 288307.
as the young adult group cohort of comparison. While this is common Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive-
developmental inquiry. American Psychologist, 34(10), 906911.
practice in the metacognitive literature (e.g., Allen-Burge & Storandt, Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini Mental State: A practical meth-
2000; Buttereld et al., 1988; Connor, Dunlosky, & Hertzog, 1997; od for grading the cognitive state of patients for the clinician. Journal of Psychiatric
Thomas et al., 2011) it does raise a number of potential issues which Research, 12, 189198.
Gruneberg, M. M., & Monks, J. (1974). Feeling of knowing and cued recall. Acta
have not been fully considered empirically. One primary point of con- Psychologica, 38, 257265.
cern is that undergraduate students are required to use and monitor Hanczakowski, M., Pasek, T., Zawadzka, K., & Mazzoni, G. (2013). Cue familiarity and
their memory on a daily basis for successful completion of their chosen don't know responding in episodic memory. Journal of Memory and Language, 69,
368383.
eld of study. While older adults may voluntarily partake in similar ac-
Hart, J. T. (1965). Memory and the feeling-of-knowing experience. Journal of Educational
tivities, such as online courses and distance learning, not all older adults Psychology, 56(4), 208216.
involved in research studies may do this, and it is unlikely to be in as Henkel, L. A., Johnson, M. K., & De Leonardis, D. M. (1998). Aging and source monitoring:
Cognitive processes and neuropsychological correlates. Journal of Experimental
regimented a fashion as an undergraduate course imposes. This more
Psychology: General, 127(3), 251268.
automatic habit of effective memory use by undergraduates may Hertzog, C., Dunlosky, J., & Sinclair, S. M. (2010). Episodic feeling-of-knowing resolution
therefore be underlying observed discrepancies in metacognitive derives from the quality of original encoding. Memory & Cognition, 38(6), 771784.
responding in young and older adults rather than there being an age ef- Hicks, J. L., & Marsh, R. L. (2002). On predicting the future states of awareness for recog-
nition of unrecallable items. Memory & Cognition, 30(1), 6066.
fect per se. Inclusion of a non-undergraduate young adult comparison Jennings, J., & Jacoby, L. L. (2003). Improving memory in older adults: Training recollec-
group in future metacognitive studies would help to characterise tion. Neuropsychological Rehabilitation, 13(4), 417440.
92 S.M. Morson et al. / Acta Psychologica 157 (2015) 8592

Kelemen, W. L., Frost, P. J., & Weaver, C. A. (2000). Individual differences in metacognition: Perfect, T. J., & Dasgupta, Z. R. (1997). What underlies the decit in reported recollective
Evidence against a general metacognitive ability. Memory & Cognition, 28(1), 92107. experience in old age? Memory & Cognition, 25(6), 849858.
Koriat, A. (1993). How do we know that we know? The accessibility model of the feeling Perfect, T. J., & Hollins, T. S. (1999). Feeling of knowing judgments do not predict subse-
of knowing. Psychological Review, 100(4), 609639. quent recognition performance for eyewitness memory. Journal of Experimental
Koriat, A. (1997). Monitoring one's own knowledge during study: A cue-utilization ap- Psychology: Applied, 5(3), 250264.
proach to judgments of learning. Journal of Experimental Psychology: General, Perfect, T. J., & Stollery, B. (1993). Memory and metamemory performance in older
126(4), 349370. adults: One decit or two ? The Quarterly Journal of Experimental Psychology, 46(1),
Koriat, A., Bjork, R. A., Sheffer, L., & Bar, S. K. (2004). Predicting one's own forgetting: The 119135.
role of experience-based and theory-based processes. Journal of Experimental Perrotin, A., Isingrini, M., Souchay, C., Clarys, D., & Taconnat, L. (2006). Episodic
Psychology: General, 133(4), 643656. feeling-of-knowing accuracy and cued recall in the elderly: Evidence for double
Krinsky, R., & Nelson, T. O. (1985). The feeling of knowing for different types of retrieval dissociation involving executive functioning and processing speed. Acta
failure. Acta Psychologica, 58, 141158. Psychologica, 122(1), 5873.
Lachman, J. L., Lachman, R., & Thronesbery, C. (1979). Metamemory through the adult life Peynirciolu, Z. F., & Tekcan, A. I. (2000). Feeling of knowing for translations of words.
span. Developmental Psychology, 15(5), 543551. Journal of Memory and Language, 43, 135148.
Light, L. L. (1991). Memory and aging: Hypotheses in search of data. Annual Review of Reder, L. M. (1987). Strategy selection in question answering. Cognitive Psychology, 19,
Psychology, 42, 333376. 90138.
MacLaverty, S. N., & Hertzog, C. (2009). Do age-related differences in episodic feeling of Roberts, L. S., & Rhodes, G. (1989). Knowing your limits: Expertise and the feeling of
knowing accuracy depend on the timing of the judgement? Memory, 17(8), 860873. knowing. New Zealand Journal of Psychology, 18, 7175.
Marqui, J. C., & Huet, N. (2000). Age differences in feeling-of-knowing and condence Roebers, C. M., Cimeli, P., Rthlisberger, M., & Neuenschwander, R. (2012). Executive
judgements as a function of knowledge domain. Psychology and Aging, 15(3), functioning, metacognition, and self-perceived competence in elementary school
451460. children: An explorative study on their interrelations and their role for school
Masson, M. E. J., & Rotello, C. M. (2009). Sources of bias in the GoodmanKruskal gamma achievement. Metacognition and Learning, 7(3), 151173.
coefcient measure of association: Implications for studies of metacognitive process- Schacter, D. L. (1983). Feeling of knowing in episodic memory. Journal of Experimental
es. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(2), Psychology: Learning, Memory, and Cognition, 9(1), 3954.
509527. Schraw, G. (1995). Measures of feeling-of-knowing accuracy: A new look at an old prob-
Metcalfe, J., Schwartz, B. L., & Joaquim, S. G. (1993). The cue-familiarity heuristic in meta- lem. Applied Cognitive Psychology, 9(4), 321332.
cognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 19(4), Schwartz, B. L., & Metcalfe, J. (1992). Cue familiarity but not target retrievability enhances
851861. feeling-of-knowing judgments. Journal of Experimental Psychology: Learning, Memory,
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. and Cognition, 18(5), 10741083.
(2000). The unity and diversity of executive functions and their contributions to Singer, M., & Tiede, H. L. (2008). Feeling of knowing and duration of unsuccessful memory
complex Frontal Lobe tasks: A latent variable analysis. Cognitive Psychology, 41(1), search. Memory and Cognition, 36(3), 588597.
49100. Snodgrass, J. G., & Corwin, J. (1988). Pragmatics of measuring recognition memory:
Moulin, C. J. A., Perfect, T. J., & Jones, R. W. (2000). Evidence for intact memory monitoring Applications to dementia and amnesia. Journal of Experimental Psychology: General,
in Alzheimer's disease: Metamemory sensitivity at encoding. Neuropsychologia, 38, 117(1), 3450.
12421250. Souchay, C., & Isingrini, M. (2004). Age related differences in metacognitive control: Role
Murphy, M. D., Sanders, R. E., Gabriesheski, A. S., & Schmitt, F. A. (1981). Metamemory in of executive functioning. Brain and Cognition, 56(1), 8999.
the aged. Journal of Gerontology, 36(2), 185193. Souchay, C., & Isingrini, M. (2012). Are feeling-of-knowing and judgment-of-learning
Murphy, M. D., Schmitt, F. a, Caruso, M. J., & Sanders, R. E. (1987). Metamemory in older different? Evidence from older adults. Acta Psychologica, 139, 458464.
adults: The role of monitoring in serial recall. Psychology and Aging, 2(4), 331339. Souchay, C., Isingrini, M., Clarys, D., Taconnat, L., & Eustache, F. (2004). Executive function-
Naveh-Benjamin, M. (2000). Adult age differences in memory performance: Tests of an ing and judgment-of-learning versus feeling-of-knowing in older adults. Experimental
associative decit hypothesis. Journal of Experimental Psychology: Learning, Memory, Aging Research, 30(1), 4762.
and Cognition, 26(5), 11701187. Souchay, C., Isingrini, M., & Espagnet, L. (2000). Aging, episodic memory feeling-of-
Nelson, T. O. (1984). A comparison of current measures of the accuracy of feeling-of- knowing, and frontal functioning. Neuropsychology, 14(2), 299309.
knowing predictions. Psychological Bulletin, 95(1), 109133. Souchay, C., Moulin, C. J. A., Clarys, D., Taconnat, L., & Isingrini, M. (2007). Diminished
Nelson, T. O., Gerler, D., & Narens, L. (1984). Accuracy of feeling-of-knowing judgments episodic memory awareness in older adults: Evidence from feeling-of-knowing and
for predicting perceptual identication and relearning. Journal of Experimental recollection. Consciousness and Cognition, 16(4), 769784.
Psychology: General, 113(2), 282300. Souchay, C., & Smith, S. J. (2013). Subjective states associated with retrieval failures in
Nelson, T. O., Leonesio, J., Shimamura, A. P., Landwher, R. F., & Narens, L. (1982). Parkinson's disease. Consciousness and Cognition, 22(3), 795805.
Overlearning and the feeling of knowing. Journal of Experimental Psychology: Thomas, A. K., Bulevich, J. B., & Dubois, S. J. (2011). Context affects feeling-of-knowing ac-
Learning, Memory, and Cognition, 8(4), 279288. curacy in younger and older adults. Journal of Experimental Psychology: Learning,
Nelson, T. O., & Narens, L. (1990). Metamemory: A theoretical framework and new nd- Memory, and Cognition, 37(1), 96108.
ings. In G. Bower (Ed.), The Psychology of Learning and Motivation: Advances in Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26(1), 112.
Research and Theory, Vol. 26. (pp. 125173). San Diego, CA: Academic Press. Ward, G. (2002). A recency-based account of the list length effect in free recall. Memory
Nelson, T. O., & Narens, L. (1994). Why investigate metacognition ? In J. Metcalfe, & A. P. and Cognition, 30(6), 885892.
Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 125). Cambridge, Wright, D. B. (1996). Measuring feeling of knowing: Comment on Schraw (1995). Applied
MA: MIT Press. Cognitive Psychology, 10(3), 261268.
Old, S. R., & Naveh-Benjamin, M. (2008). Memory for people and their actions: Zacks, R. T., Hasher, L., & Li, K. Z. H. (2000). Human memory. In F. I. M. Craik, & T. A.
Further evidence for an age-related associative decit. Psychology and Aging, Salthouse (Eds.), The handbook of ageing and cognition (2nd ed.). London: Lawrence
23(2), 467472. Erlbaum Associates.

Anda mungkin juga menyukai