ERIC ED602955: The Tip-of-the-Tongue State and Curiosity

2017·0.69 MB·English

by  ERIC

Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Cognitive Research: Principles
DOI10.1186/s41235-017-0065-4
and Implications
ORIGINAL ARTICLE Open Access
The tip-of-the-tongue state and curiosity
Janet Metcalfe1* , Bennett L. Schwartz2 and Paul A. Bloom1
Abstract
Theories of study time allocation and of curiosity suggest thatpeople are most engaged withand want to devote
their time to materials that are not completely mastered but also are not so difficult that theymight be impossible.
Their curiosityis thought to be triggered by items that are almost known, or are in what is sometimes called the
region ofproximal learning. Answers thatare onthetip-of-the-tongue (TOT)—not immediately recallable but
nevertheless evoking a feeling of imminent recall—seem, intuitively, to be materials that have this characteristic of
being almost, but not quite, fully known. We therefore,hypothesized that people would be particularly curious to
seethe answers to questions for which theanswers were on thetips of their tongues. Totest the TOT curiosity
hypothesis, wegave participants 82 general information questions and quickly asked whether theanswerswere or
were not on thetips of their tongues and whether they wanted to see theanswers later. Overwhelmingly, items
that were accompanied by a TOT feeling were those which evoked participants’curiosity, regardless of whether the
feeling occurred in conjunction with anerror of commission, an error ofomission, or even with thecorrect answer.
Keywords: Tip-of-the-tongue,Curiosity, Google, Metacognition, Region ofproximal learning
Significance feeling states may be epiphenomenal—having no func-
This article investigates metacognitive conditions that tion in human cognition. By contrast, our results indi-
induce people’s need to know. An mTurk survey re- cate that the special feeling of mild torment that is the
vealed that people think they are most curious—opera- hallmark of theTOTstate is a goad to epistemic action.
tionalized, in the survey, as having the strongest urge to Ittriggerspeople’sneedtoknow.
Google the answer—when they do not know the answer
but not when they know the answer. However, they
think they are indiscriminate among ‘not known’ states: Background
when they are sure they do not know, when they have a Most studies of the TOT state focus on either the cha-
feeling that they might be able to recognize the answer, racteristics of the information pertaining to the not-yet-
and when they are in a genuine ‘tip-of-the-tongue’ retrieved items—such as the first letter, the number of
(TOT) state. To ascertain whether they were correct syllables, the gender of the word in certain languages, or
that all ‘don’t know’ states were equal, or whether there incorrect words called ‘blockers’ (Brown, 1991, 2012;
is a special metacognitive state associated with curiosity, Gollan & Brown, 2006; Kornell & Metcalfe, 2006a;
we gave participants 82 general information questions, Miozzo & Caramazza, 1997)—or they focus on the com-
asked them to give the answers, asked them whether pelling phenomenology of the state (Cleary & Claxton,
they were in aTOTstate, and then asked them whether 2015; Schwartz & Cleary, 2016). With respect to the
they wanted to see the answer. They overwhelmingly latter, researchers often include William James’ (1890,
wanted to see the answer when they were in a TOT p. 251) description “The state of our consciousness is
state. This occurred regardless of what their response peculiar. There is a gap therein; but no mere gap. It
was, including when they had been correct, had made a is a gap that is intensely active. A sort of wraith of
commission error, or had made an omission error. Phi- the name is in it, beckoning us in a given direction,
losophers have sometimes suggested that metacognitive making us at moments tingle with the sense of our
closeness and then letting us sink back without the
longed-for term,” R. Brown and McNeill’s (1966, p. 326)
*Correspondence:[email protected] comment that people who are “seized by a TOTstate …
1ColumbiaUniversity,NewYork,NY,USA
Fulllistofauthorinformationisavailableattheendofthearticle would appear to be in mild torment, something like the
©TheAuthor(s).2017OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.0
InternationalLicense(http://creativecommons.org/licenses/by/4.0/),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinkto
theCreativeCommonslicense,andindicateifchangesweremade.
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page2of8
brink of a sneeze,” or A. Brown’s (1991) acknowledgment most effective learning algorithm involved having people
of “personal introspections of inner turmoil when grap- selectively study the ‘T’ items. Metcalfe and Kornell
plingforanelusiveword.”Normally,whenindividualsare (2003, 2005) called T items those in the Region of
engaged in nonpathological cool cognition, they are not Proximal Learning (RPL). In addition, they found that
seized by ‘inner turmoil,’ or by any kind of ‘torment.’ But college students selectively chose these particular items
TOTstates are different. They arise into consciousness as for further study. Loewenstein (1994), like Berlyne
a distinct emotional configuration that appears to be (1954), also zeroed in on such items in his theory of
universally recognizable (Schwartz, 1999). The question curiosity, pointing to them as being the items people
that underlies the research in the present article is why found particularlyinterestinganddeservingofattention.
this particular metacognitive feeling state surfaces into The items that one finds on the tip of one’s tongue fit
consciousnessasadistinctiveandemotionalstate:whatis the characterization of ‘T’ items or ‘RPL’ items very well.
itsfunction? They are not so easy that they can be readily recalled,
Schwartz and Cleary (2016; and see Metcalfe & but they are not so difficult that they are completely un-
Schwartz, 2015; Schwartz & Metcalfe, 2011) have known and have no resonance. Indeed, some partial and
suggested that the TOT state may have an adaptive, semantic information appears to be retrievable for these
evolutionary purpose. They have focused on the finding items, making them ideal candidates for optimal lear-
that people persevere in their efforts at recall when they ning. It is possible that the feeling associated with being
are in a TOTstate: the metacognitive feeling appears to in aTOTstate is a marker of these special items and has
prompt people to continue trying. Furthermore, most a function of provoking curiosity and inducing the per-
TOTstates are eventually successfully resolved, although sontoepistemicaction.
with obvious difficulty. Even so, the difficulty of the Theaction involved ininformationseeking variesfrom
retrievalitself isthought toplayarole inenhancinglater situation to situation. A person may just keep trying to
memory (see, e.g., Bjork & Linn, 2006)—a consequence retrieve rather than giving up, may look up an answer in
that has beneficial effects for the individual’s later a dictionary, or may ask a friend or teacher. Often
performance. The present study takes this idea a step Google is the option of choice. To investigate whether
further in suggesting that the TOTstate prompts curio- theTOTstate isperceivedashavingsome special role in
sity, which could entail more than just a retrieval inducingsuchaction,we asked 127 participants(61 male,
attempt—the feeling state may entail general informa- 66 female; mean age=40.6, SD=11.6) on MTurk, the
tion seeking behavior. followingquestion:
Weiner (2006), indescribing the tenants of his attribu- “You have the strongest urge to Google the answer to
tion theory, argued that consciously experienced feelings aquestionthathasbeen posedtoyouwhen:
(i.e., emotions) are ‘goads to action.’ Consistent with this
view, we propose that the conscious emotional instan- (a)youaresureyouknowtheanswer,andcanproduceit,
tiation of the TOT state has a function of inducing the (b)youaresure youdon’tknowthe answer andcan’t
experiencer to the action of trying and persisting in produceit,
efforts to gain the sought-for information. It induces an (c)youhavea‘feelingofknowing,’thatis,youthink you
eagerdesiretoknow:curiosity. couldprobablyrecognizethe answer ifitwere
Several precursors in the literature provide convergent shown toyou.,
support for the idea that the TOT state may be related (d)youareina‘tip-of-the-tongue’state, thatis,youare
to curiosity and information seeking. First, going back as unable tothink oftheanswer,now,butfeelsure that
far as Berlyne’s (1954) theory of epistemic curiosity, the- youknow itandthat itisontheverge ofcoming
orists have proposed that people’s need to know is ig- back toyou.”
nited by materials that are neither completely unknown
nor completely known, but are, instead, in an intermedi- Only 25% of our respondents chose the TOT option
ate zone (Atkinson, 1972; Kornell & Metcalfe, 2006b; ‘d’; 39% chose ‘b’ and 31% chose ‘c’. Very few people
Litman, 2009; Loewenstein, 1994; Metcalfe, 2002). These (5%)wantedtoknowtheanswerwhentheythoughtthey
theories indicate that people are best off, as far as learn- already knew it. Not knowing was associated with infor-
ing goes, if they study items that are at an intermediate mation seeking. But people did not appear to prioritize
state of learning. For example, Atkinson (1972) among ‘not known’ states. They said they had the urge
employed an early computerized learning program to Google when they did not know at all and when they
which sorted items into those that were reliably remem- had a feeling of knowing, as frequently as when they
bered, intermediate items that were somewhat learned were in aTOTstate. The mTurk survey, of course, may
but not permanently (which he called ‘T’ or transitional or may not reflect what people choose to do when
items),andthose thatwereunlearned. Hefound thatthe actuallyinsuchstates.
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page3of8
The only empirical investigation that we were able to low or high. The materials were also divided into two
find of the relation between people’s curiosity and their sets, because some questions are more likely to evince
TOT states was that by Litman, Hutchins, and Russon TOTstates than others.
(2005). They gave participants 12 general-information Finally, we asked participants to report TOTstates for
questions to answer in a self-paced manner, and had all responses, even those that were correct. Although
them indicate for each whether they knew the answer, most studies look only at TOT experiences when the
they did not know the answer, or they were in a TOT person has made an omission error, TOT states some-
state. They then asked for intensity or confidence ratings. times occur even when people have made a commission
Personality questionnaires were administered, allowing error (they may be in what is sometimes called a
examination of trait curiosity. For some participants a ‘blocked’ state, and retrieve the blocker as an error) or
forced-choicerecognitiontestwasgiven.Attheendofthe even when they were correct (presumably if they had
experiment, the participants were provided with sealed low confidence about their answer). We were interested
envelopes with the 12 questions written on the outside in all of these TOT states. Our primary question,
andtheanswersenclosed.Theresearchersinvitedpartici- throughout, was the relation of TOTstates to curiosity:
pants to open any envelopes they wished. They reported did people want to see the answers more when they
that people were more likely to open the envelopes of were inaTOTstatethan whentheywere not?
questionstowhichtheyhadbeeninaTOTstate.
Although the results of this experiment are supportive
Experiment
of a link between TOTstates and curiosity, by the time
Method
the choices were made many, if not most, TOT expe-
Participants
riences would probably already have subsided (see
The 46 participants (20 male, 25 female, one who did
Brown & Croft, 2014). To overcome this problem, in the
not answer) were Columbia University and Barnard
experiment that follows we asked participants whether
College undergraduates, ranging in age from 18 to 29
they wanted to see the answer to the questions the mo-
years (M=19.96; SD=2.32). They received partial
ment after they gave their TOT judgments for each
course credit for participating in the experiment. All
question, while they were still in the state. Furthermore,
procedures were in accordance with the ethical princi-
TOTstates are rare—occurringin thenaturalflow of life
ples of the APA, and were approved by the Columbia
(see Brown, 2012) only about once per week, and in a
UniversityInternalReviewBoard(IRB-AAAD4902).
laboratory setting about 10–20% of the time for omis-
sion errors. With only 12 questions, there probably were
not many TOT experiences per person in Litman et al.’s Materials
The stimuli were 82 general information questions taken
(2005) study. The frequency was not reported. To accu-
from the Nelson and Narens (1980) norms, and updated
mulate enoughTOT experiences to allow more detailed
and corrected (see, Metcalfe, Casal-Roscum, Radin, &
analyses, we used 82 questions rather than just 12. We
Friedman, 2015). For example, one question was “What
also speeded responding to each question. TOT expe-
is the name of the ancient warrior who was dipped in
riences often exist in the first few seconds of attempted
the River Styx?” The questions were displayed on an
retrieval, but then, upon successful retrieval, they dissi-
iMac desktopcomputer withSuperLab4.5software.
pate. Instead of waiting for this to happen, we pushed
people to give aTOTjudgment quickly, while they were
still in the unresolved state. This meant that the experi- Design
menter had to run each participant in person, and The experiment was 2 x 2 mixed design, with factors
enforce timing of the responses. He or she read each Memory Load (Load or No Load, within participant)
question aloud and then allowed only 5 s before the and Group (1 or 2, between participants). The questions
participant had to say whether or not he or she was in a (Set A) that were in the No Load condition in Group 1
TOT state. Our procedure resulted in a considerably were in the Load condition in Group 2; the questions
higher proportion of TOT states than is typically ob- that were in the No Load condition in Group 2 (Set B)
served in self-paced experiments. In addition, we used a were in the Load condition in Group 1. Twelve ques-
load manipulation on half of the questions because tions were reserved to be in the same load conditions
working memory load, in some but not all experiments in both groups, allowing us to look at overall group
(e.g., Schwartz’s, 2008, Experiments 1, 2 and 3 showed differences. On these 12 questions, no significant
the effect whereas Experiment 4 did not), has been between-group differences in the probability correct,
shown to influence the probability of TOT experiences. the probability of commission errors, or the probability
We wanted to see whether curiosity was evoked by TOT of TOT experiences were found. The order of presenta-
states regardless of whether their overall probability was tion of questions was randomized for each participant.
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page4of8
Procedure and the probability of an omission error was 0.54
Participants were instructed that they would be given a (SD=0.16). The overall probability of being in a TOT
series of general information questions to try to answer state was 0.24 (SD=0.15) and all participants exhi-
and, while trying, to say (a) whether they were in aTOT bited TOT states, but not in all treatment combina-
state or not and (b) whether they wanted to see the tions (which will account for differences in degrees of
answer later or not. They also were informed that they freedom shown in some of the following analyses).
would later be able to see the correct answer for up to
10% of the questions. We included (but did not enforce) Omissionerrors
this restriction because in pilot testing with no restriction When we examined the probability of TOT experiences
many participants had stated that they wanted to see the givenomissionerrors,whichisthestandard analysis, there
answers to all of the questions. The 10% ‘rule’allowed us wasnoeffectofeitherGrouporofMemoryLoad,butthere
to investigate differences depending upon whether people wasaninteractionbetweenthem,F(1,44)=9.19,p=0.004,
wereinaTOTstateornot.Participantsweretoldthaton η2 ¼0:173, power=0.84. This interaction appeared to be
p
halfofthequestionstheywouldseea stringoffourdigits
duetoadifferenceinTOTexperiencesforSetAandSetB.
before the question itself was presented, whereas half of
Set A was in the No Load condition for Group 1 (with a
the time they would see asterisks. If a string of numbers
TOT probability of 0.36) and in the Load condition for
appeared they would be required to recall those digits
Group 2 (p=0.35). Set B was in the Load condition for
after answering the general information question and
Group1(p=0.28)andintheNoLoadconditionforGroup
giving their judgments. When asterisks appeared, they
2 (p=0.30). This same Group×Memory Load interaction
wouldlaterhavetosayithadbeenasterisks.
wasshowninthedataonthesimpleprobabilityofwanting
Participants nearly always knew what aTOTstate was toseelater,F(1,44)=4.87,p=0.03, η2 ¼0:1,power=0.58.
(also see Schwartz & Metcalfe, 2011). If, however, a par- p
Peopletendedtowanttoseetheanswersmoreinthecon-
ticipant did not, the experimenter explained it as a state
ditionsinwhichthereweremoreTOTexperiences.
in which “you feel sure you know the answer and think
you can get it—it is imminent—but you cannot think of
itatthemoment.”Participantscompletedapractice trial ChoicetoseelaterwheninaTOTstateornot(omission
errors)
before theexperimentbegan.
We treated being or notbeing inaTOTstate as if it was
First, the asterisks or numbers appeared, followed by a
an independent variable and computed the conditional
5-sdelay.Then,thegeneralinformationquestionappeared
probability of choosing to see later as the dependent
onscreen and was read aloud by the experimenter, who
variable in a 2 (TOT or not)×2 (Group)×2 (Memory
immediately upon finishing hit a key to start the 5-s re-
Load) design. Neither Group nor Memory Load, nor the
sponse period. During this period,the experimenter typed
in the participant’s answer, if any. At the end of the 5 s, interaction, was significant. Indeed, the only thing that
wassignificantwaswhetherthepersonwasinaTOTstate
participantsindicatedwhethertheywereinaTOTstateor
or not, F(1, 43)=42.68, p=0.001, η2 ¼0:50, power=1.0.
not. The computer then immediately asked whether they p
wanted to see the answer later. The participant then People were about twice as likely to want to see the
recalled the four digits or that it had been asterisks, and answer later when they were in a TOT state (0.45) than
theprogramthenwentontothenexttrial. when they were not (0.24). This difference, as shown in
At the end of the experiment, after filling out a demo- Fig. 1, was evident for every group and for every load
graphics sheet, participants were thanked, debriefed, condition/questionset.
given course credit for participation, and allowed to see
theanswers toallofthegeneralinformation questions. Commissionerrors
Commission errors are interesting because TOT expe-
Results riences sometimes occur when people are in a ‘blocked’
In this experiment, people wanted to see the answers state, or when they have only partial information, such
more when they were in a TOTstate (0.44) than when as the first letter. Such responses would have been clas-
they were not in a TOT state (0.18), t(45)=8.06, p< sified as commission errors here. There were no effects
0.001. This is the main result of interest, and was ob- of either Group or of Memory Load, nor was there an
tainedinallconditionsthatwewereabletoexamine,as interactionbetweenGroup andMemory Load.
will be illustrated shortly. There were other differences, The simple probability of wanting to see answers later,
however, as indicated in the following data are provided however, indicated an interaction between Group and
intheAdditionalfile1. Working Memory Load (which may have been a bias to-
Theprobability ofa correct answer was0.26 (SD=0.15), ward Set A—as noted earlier), F(1, 44)=3.92, p=0.054,
theprobabilityofacommissionerrorwas0.20(SD=0.09), η2 ¼0:08,power=0.49.
p
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page5of8
Fig.1Probabilityofanswerseekinggiventhattheresponsewasanomissionerror.Errorbars:standarderrorofthemean.TOT:tip-of-the-tongue
ChoicetoseelaterwheninaTOTstateornot ChoicetoseelaterwheninaTOTstateornot(onall
(commissionerrors) errors)
For this analysis, we looked at the subset of data in As would be expected, because the separate error-type
whichtheinitialresponsetothequestion wasacommis- analyses were significant, when we collapsed over all
sion error—a response that was either partial or wrong errors there was a significant main effect for being in a
(note that ‘don’t know’ responses were counted as omis- TOTstateonchoicetoseelater,F(1,44)=54.65,p<0.001,
sion errors)—and computed the conditional probability η2 ¼0:55, power=1.0. No other main effects or interac-
p
of wanting to see later as a function of whether the per- tionsweresignificant.
son was in a TOT state or not, conducting a 2×2×2
ANOVA with factors TOT, Group, and Memory Load. Correctresponses
There was a significant effect of TOTstate: people more In this experiment, participants were not given feedback
often chose to see the answer later when they were in a about whether their responses were correct or incorrect,
TOT state than when they were not, F(1, 19)=8.44, and we observed a number of cases in which people
p=0.009, η2 ¼0:308, power=0.78. No other effects expressed that they were in a TOT state even though
p
or interactions were significant. The conditional pro- theyhad,infact,giventhe correctanswer.Toourknow-
bability of choice to see later for each of the eight ledge, this phenomenon has not been observed pre-
cells for commission errors is shown in Fig. 2. viously. The dynamics of uncertainty resolution and the
Fig.2Probabilityofanswerseekinggiventhattheresponsewasacommissionerror.Errorbars:standarderrorofthemean.TOT:tip-of-the-tongue
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page6of8
feeling of TOT are not well understood. However, it is TOTstate. Thus, although theTOTstate affects people’s
plausible that a person might feel they were in a TOT pursuit of the answers, people are not aware that it does
state if they were uncertain about their correct response. so. Interestingly, the effect of theTOTstate on informa-
In that case, these responses, like TOT responses made tion seeking occurred at a variety of different levels of
after either omission or commission errors, should be TOT experiences, for both groups, and regardless of
indicative thatpeople were in their RPL—close to having whether people were in a memory load condition or not.
the right answer—and might evoke curiosity. Accor- It occurred regardless of the type of error people had
dingly, weexaminedthe correctresponses. made—omission or commission. Indeed, it even oc-
There was an effect of Working Memory Load, F(1,44) curred when people had been correct. We conclude that
=8.89, p=0.005, η2 ¼0:17, power=0.83, showing that thenagging feeling of beinginthisparticularmetacogni-
p
people were more likely to be in aTOTstate when they tive stateisagoadtoepistemic action.
had been in the Load condition than in the Asterisks There remain many unaddressed complexities and un-
condition. There was also an interaction between Group answered questions. Curiosity is often taken to be a
and Memory Load, F(1, 44)=35.18, p<0.001, η2 ¼0:44, pleasant experience—the happy state of an enquiring
p
mind. This perspective, concerning the positive emo-
power=1.0,whichwas,again,probablyduetosetdifficulty.
tional quality of curiosity, contrasts starkly with the nag-
ging, tormented quality so frequently ascribed to TOT
ChoicetoseelaterwheninaTOTstateornot(correct
states. Interestingly, Litman, Crowson and Kolinski
responses)
(2010; and see Litman, 2009) have proposed that there
As found earlier, the only effect that was significant
may be two kinds of curiosity, which they call interest-
was whether the person was in a TOT state or not,
based curiosity, which is largely positive, and
F(1, 12)=8.24, p=0.014, η2 ¼0:407, power=0.749.
p deprivation-type epistemic curiosity, which is not. The
People were more likely to want to see the answer need to know induced by the TOTstate examined here
later when they had been correct but were in a TOT appears to be of the latter type, although more studies
state (0.24) than when they were correct but not in a are needed to investigate the nuances of this emotional
TOT state (0.03). This difference, as shown in Fig. 3, distinction. To complicate matters even further, the tor-
was evident for every group and for every load condi- mented feeling that accompanies the TOTstate prior to
tion/question set for which we had data. resolution may be modulated by anticipation of the
highly positive emotionally charged feelings that arise
Conclusion upon solution—feelings of pleasure that may be similar
The results of this experiment indicated that when to the delight experienced when a person solves an
people were in aTOTstate they wanted to see the cor- insight or a magic problem (see, e.g., Danek, Fraps, von
rect answer more than when they were not in a TOT Muller, Grothe, & Ollinger, 2014; Hedne, Norman, &
state. The results of our mTurk survey indicated that Metcalfe, 2016), or experiences a scientific epiphany.
people do not realize there is any difference in their own The motivation to overcome the torment may go
curiosity between merely not knowing and being in a hand in hand with the anticipation of the intense
Fig.3Probabilityofanswerseekinggiventhattheresponsewascorrect.Errorbars:standarderrorofthemean.TOT:tip-of-the-tongue
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page7of8
pleasure felt when the state is resolved. Both may Received:25November2016Accepted:25April2017
contribute to people’s curiosity.
Sometimes metacognitive states have been thought to
be epiphenomenal—having no ramifications for cogni- References
Atkinson,R.C.(1972).Ingredientsforatheoryofinstruction.American
tion. The TOT data presented in this article contravene Psychologist,27,921–931.
this conjecture: the presence of the TOT feeling state Berlyne,D.E.(1954).Atheoryofhumancuriosity.BritishJournalofPsychology,
has implications for people’s choice to seek more infor- 45,180–191.
mation—a propensity with deep consequences for our Bjork,R.A.,&Linn,M.C.(2006).Thescienceoflearningandthelearningof
science:Introducingdesirabledifficulties.AssociationforPsychologicalScience
curiosity, for our engagement in learning, and, ulti- Observer,19,1–2.
mately, forourunderstanding. Brown,A.S.,&Croft,K.(2014).Thereitisagainonmytongue:Trackingrepeat
TOTs.InB.L.Schwartz&A.S.Brown(Eds.),Tipofthetonguestatesand
relatedphenomena(pp.32–49).Cambridge:CambridgeUniversityPress.
Additional file Brown,A.S.(1991).Areviewofthetip-of-the-tongueexperience.Psychological
Bulletin,109,204–223.
Brown,A.S.(2012).Tipofthetonguestate.Hove,UK:PsychologyPress.
Additionalfile1:Data.(XLSX264kb) Brown,R.,&McNeill,D.(1966).The‘tipofthetongue’phenomenon.Journalof
VerbalLearningandBehavior,5,325–337.
Cleary,A.M.,&Claxton,A.B.(2015).Thetip-of-the-tongueheuristic:Howtip-of-
Abbreviation
the-tonguestatesconferperceptibilityoninaccessiblewords.Journalof
TOT:Tip-of-the-tongue ExperimentalPsychologyLearningMemoryandCognition,41,1533–1539.
Danek,A.H.,Fraps,T.,vonMuller,A.,Grothe,B.,&Ollinger,M.(2014).It’sakindof
Acknowledgements magic—Whatself-reportscanrevealaboutthephenomenologyofinsight
TheauthorsthankTealEich,RachelBurris,KarenKellyandJudyXu.They problemsolving.FrontiersinPsychology,5,1408.
haveallowedustothankthem. Gollan,T.H.,&Brown,A.S.(2006).Fromtip-of-the-tongue(TOT)datato
theoreticalimplicationsintwosteps:WhenmoreTOTsmeansbetter
Funding retrieval.JournalofExperimentalPsychology:General,135,462–483.doi:10.
TheauthorsaregratefultotheInstituteofEducationScience(R305A150467) 1037/0096-3445.135.3.462.
andJamesS.McDonnellFoundationCollaborativegrant(ledbyRoddy Hedne,M.R.,Norman,E.,&Metcalfe,J.(2016).Intuitivefeelingsofwarmthand
Roediger,#484800),withJM,asColumbiaPI,fortheirsupport.Theauthors confidenceininsightandnoninsightproblemsolvingofmagictricks.
aresolelyresponsibleforthecontentofthisarticle. FrontierinPsychologySectionCognitiveScience,7(1314),1–13.doi:10.3389/
fpsyg.2016.01314.
Availabilityofdataandmaterials James,W.(1890).Principlesofpsychology.NewYork:Holt.
DataforindividualsubjectsandinsummaryformareavailableinanExcel Kornell,N.,&Metcalfe,J.(2006a).‘Blockers’donotblockrecallduringtip-of-the-
filesupplementarytothisarticle.Rawdata,intextfiles,fromthecomputer- tonguestates.MetacognitionandLearning,1,248–261.
generatedsubjectfilesaswellassourceprogramsareavailableonrequest Kornell,N.,&Metcalfe,J.(2006b).Studyefficacyandtheregionofproximal
fromtheauthors. learningframework.JournalofExperimentalPsychologyLearningMemoryand
Cognition,32,609–622.
Authors’contributions Litman,J.A.,Hutchins,T.L.,&Russon,R.K.(2005).Epistemiccuriosity,feeling-of-
knowing,andexploratorybehaviour.CognitionandEmotion,19,559–582.
JMandBLSconceived,planned,anddesignedtheexperiment.JMoversaw
Litman,J.A.(2009).Curiosityandmetacognition.InC.B.Larson(Ed.),
theconductoftheexperimentwhichwasruninherlaboratory,trainedthe
Metacognition:NewResearchDevelopments(pp.105–116).Hauppauge,NY:
experimenters,supervisedthecodingofthedataandtheanalysis,and
NovaSciencePublishers.
wrotetheinitialdraft.BLSrevisedtheinitialdraftandcontributedtothe
Litman,J.A.,Crowson,H.M.,&Kolinski,K.(2010).Validityoftheinterest-and
programmingandanalysisoftheexperiment.PABranthesubjectsforthe
deprivation-typeepistemiccuriositydistinctioninnon-students.Personality
secondhalfoftheexperiment(Group2),codedandcheckedthedatafrom
andIndividualDifferences,49,531–536.
therawcomputeroutputs,contributedtotheanalyses,createdthefigures,
Loewenstein,G.(1994).Thepsychologyofcuriosity:Areviewand
andwroteadraftofthemethodssection.Dataanalysiswasconductedby
reinterpretation.PsychologicalBulletin,116,75–98.
TealEich(seeAcknowledgements).Allauthorsreadandapprovedthefinal
Metcalfe,J.(2002).Isstudytimeallocatedselectivelytoaregionofproximal
manuscript.
learning?JournalofExperimentalPsychology:General,131,349–363.doi:10.
1037/0096-3445.131.3.349.
Competinginterests
Metcalfe,J.,Casal-Roscum,L.,Radin,A.,&Friedman,D.(2015).Onteachingold
Theauthorsdeclarethattheyhavenocompetinginterests. dogsnewtricks.PsychologicalScience,26,1833–1840.
Metcalfe,J.,&Kornell,N.(2003).Thedynamicsoflearningandallocationofstudy
Consentforpublication timetoaregionofproximallearning.JournalofExperimentalPsychology:
Notapplicable. General,132,530–542.doi:10.1037/0096-3445.132.4.530.
Metcalfe,J.,&Kornell,N.(2005).Aregionofproximallearningmodelofstudy
Ethicsapproval timeallocation.JournalofMemoryandLanguage,52,463–477.
AllprocedureswereinaccordancewiththeethicalprinciplesoftheAPA, Metcalfe,J.,&Schwartz,B.L.(2015).Theghostinthemachine:Self-reflectiveconsciousness
and were approved by the Columbia University Internal Review Board andtheneuroscienceofmetacognition.InJ.Dunlosky&S.K.Tauber(Eds.),TheOxford
(IRB-AAAD4902). Participants signed informed consent. HandbookofMetamemory(pp.407–424).Oxford:OxfordUniversityPress.
Miozzo,M.,&Caramazza,A.(1997).Retrievaloflexical–syntacticfeaturesintip-of-
the-tonguestates.JournalofExperimentalPsychologyLearningMemoryand
Publisher’s Note Cognition,23,1410–1423.
SpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublished Nelson,T.O.,&Narens,L.(1980).Normsof300general-informationquestions:
mapsandinstitutionalaffiliations. accuracyofrecall,latencyofrecall,andfeeling-of-knowingratings.Journalof
VerbalLearning&VerbalBehavior,19,338–368.
Authordetails Schwartz,B.L.,&Cleary,A.M.(2016).Tip-of-the-tonguestates,déjàvuandother
1ColumbiaUniversity,NewYork,NY,USA.2FloridaInternationalUniversity, metacognitiveoddities.InJ.Dunlosky&S.Tauber(Eds.),OxfordHandbookof
Miami,FL,USA. Metamemory(pp.95–108).NewYork:OxfordUniversityPress.
Metcalfeetal.CognitiveResearch:PrinciplesandImplications_#####################_ Page8of8
Schwartz,B.L.(1999).Sparklingattheendofthetongue:Theetiologyand
phenomenologyoftip-of-the-tonguestates.PsychonomicBulletin&
Review,6,379–393.
Schwartz,B.L.(2008).Workingmemoryloaddifferentially affectstip-of-the-
tonguestates and feeling-of-knowing judgment. Memory & Cognition,
36, 9–19.
Schwartz,B.L.,&Metcalfe,J.(2011).Tip-of-the-tongue(TOT)states:Retrieval,
behavior,andexperience.Memory&Cognition,39,737–749.
Weiner,B.(2006).Socialmotivation,justice,andthemoralemotions:Anattributional
approach.Mahwah,NJ:LawrenceErlbaum.
Submit your manuscript to a
journal and benefi t from:
7 Convenient online submission
7 Rigorous peer review
7 Immediate publication on acceptance
7 Open access: articles freely available online
7 High visibility within the fi eld
7 Retaining the copyright to your article
Submit your next manuscript at 7 springeropen.com