View in EDS HTML Full Text PDF Full Text

Age Affects Strategic but Not Spontaneous Recall in 35- and 46-Month-Old Children

Saved in:
Bibliographic Details
Title: Age Affects Strategic but Not Spontaneous Recall in 35- and 46-Month-Old Children
Language: English
Authors: Sonne, Trine, Kingo, Osman S., Berntsen, Dorthe, Krøjgaard, Peter
Source: Journal of Cognition and Development. 2020 21(4):603-621.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 19
Publication Date: 2020
Document Type: Journal Articles
Reports - Research
Descriptors: Age Differences, Young Children, Memory, Recall (Psychology), Developmental Stages, Performance
DOI: 10.1080/15248372.2020.1797748
ISSN: 1524-8372
Abstract: It is well documented that young children have difficulties with strategically remembering past events. Recent evidence on event memory in 35- and 46-month-old children suggests that strategic retrieval (yes/no questions) improves with age, whereas spontaneous retrieval is relatively unaffected by age. We here replicate and extend those findings (N = 124): First, a novel free (strategic) recall test was added to improve ecological validity. Second, the free recall procedure allowed us to make direct comparisons between spontaneous and free strategic recall relative to age. The free recall test revealed similar results in the standard yes/no questions (older children outperformed younger). The direct comparison between spontaneous and free recall revealed a reliable interaction between age and retrieval mode: While the children's age did not affect spontaneous recall, the 46-month-olds outperformed the 35-month-olds on the free recall test. The results add to the accumulating evidence that spontaneous recall of events is an early developmental achievement.
Abstractor: As Provided
Entry Date: 2020
Accession Number: EJ1270504
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGQSQApgil_WSPEDvfK_cSLAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDJ9Yu1Hccn5FTwTbpAIBEICBmwmcZXX7Qjk-6ej6dmMAracaTt6wNkhzWb-QOdDvIMOUvOEyJwa2Zs1UG5SCVXXixTWKlf_Mk8zJjLldt1MYCi8TLs4MrJeun3Gq8pSGK4x1sBA0iLMcSVe_9pKcLVCjPoUXR_HjsGGrpjGD2q5XqlQaTOEp46EJWlcjXXfW-9XxpDKVJlrom3L-Damc8UJFPRObw8LdDshJ7CxQ
Text:
  Availability: 1
  Value: <anid>AN0146083225;7m701aug.20;2020Sep28.05:47;v2.2.500</anid> <title id="AN0146083225-1">Age Affects Strategic But Not Spontaneous Recall in 35- and 46-Month-old Children </title> <sbt id="AN0146083225-2">Introduction</sbt> <p>It is well documented that young children have difficulties with strategically remembering past events. Recent evidence on event memory in 35- and 46-month-old children suggests that strategic retrieval (yes/no questions) improves with age, whereas spontaneous retrieval is relatively unaffected by age. We here replicate and extend those findings (N = 124): First, a novel free (strategic) recall test was added to improve ecological validity. Second, the free recall procedure allowed us to make direct comparisons between spontaneous and free strategic recall relative to age. The free recall test revealed similar results in the standard yes/no questions (older children outperformed younger). The direct comparison between spontaneous and free recall revealed a reliable interaction between age and retrieval mode: While the children's age did not affect spontaneous recall, the 46-month-olds outperformed the 35-month-olds on the free recall test. The results add to the accumulating evidence that spontaneous recall of events is an early developmental achievement.</p> <p>We spend a considerable amount of time reminiscing about events we have experienced. Episodic memories of such events can be retrieved by two different routes: one can search actively for a specific memory, for instance when asked to clarify what happened on a specific occasion (e.g., when someone asks "do you remember any family trips from your childhood?"). In that case, the person is engaging in strategic recall, deliberately trying to search for a memory. Another route to our memories is when we involuntarily, with no direct attempt at remembering, recall an event. Involuntary recall refers to the type of memories that spontaneously pop-up in our minds, often triggered by cues in the environment (e.g., a specific kind of trees), for instance going for a walk in the park while suddenly remembering a camping trip in the forest with your family as a child (for a review, see Berntsen, [<reflink idref="bib2" id="ref1">2</reflink>]). Compared with strategic recall, this way of remembering is less cognitively demanding, probably due to the highly associative mechanisms involved when involuntarily remembering an event (Hall et al., [<reflink idref="bib6" id="ref2">6</reflink>]). Involuntary recall has been described as a basic mode of remembering that has been suggested to developmentally precede strategic recall (e.g., Berntsen, [<reflink idref="bib2" id="ref3">2</reflink>]; Berntsen, Staugaard, & Sørensen, [<reflink idref="bib4" id="ref4">4</reflink>]).</p> <p>Involuntary recall has not received much attention in experimental research on memory in early childhood. Observational studies are also scarce (for observational studies using a structured diary, see for example Nelson & Ross, [<reflink idref="bib13" id="ref5">13</reflink>]; Reese, [<reflink idref="bib16" id="ref6">16</reflink>]). Recently, Krøjgaard and colleagues developed a novel experimental paradigm making it possible to assess, from an observer perspective, whether or not a memory expressed by a child was involuntarily recalled (Krøjgaard, Kingo, Dahl, & Berntsen, [<reflink idref="bib9" id="ref7">9</reflink>]; Krøjgaard, Kingo, Jensen, & Berntsen, [<reflink idref="bib10" id="ref8">10</reflink>]). Using an observer perspective, such memories were operationalized as <emph>spontaneous memories</emph>, that is, memories that are (i) verbally produced, (ii) have not been the result of any kind of prompting, and (iii) are typically triggered by cues in the environment (Krøjgaard et al., [<reflink idref="bib9" id="ref9">9</reflink>], [<reflink idref="bib10" id="ref10">10</reflink>]). These criteria were used to ensure that the memories in question were an expression of conscious recall (hence the verbal criterion), and at the same time <emph>not</emph> the result of strategic retrieval (hence the <emph>unprompted and environmentally cued</emph> criteria).[<reflink idref="bib1" id="ref11">1</reflink>] To investigate spontaneous recall, the children visited the same distinctive lab setting twice with a one-week retention interval. In the first visit, the children experienced a highly memorable event – either a Teddy event hidden in a red metal box <emph>or</emph> a Game event hidden in a gray plastic box. In the second visit, the children's verbal utterances were recorded while waiting with their parent in front of the two distinct boxes containing the props from the events. A high degree of overlap in cues between the encoding session and the test session was ensured in order to increase the likelihood of remembering (cf., the encoding specificity principle, Tulving & Thomson, [<reflink idref="bib22" id="ref12">22</reflink>]). This overlap was highly distinct and thus likely to help discriminate the event from alternatives (Berntsen, [<reflink idref="bib2" id="ref13">2</reflink>]; Berntsen et al., [<reflink idref="bib4" id="ref14">4</reflink>]; Hunt & Smith, [<reflink idref="bib8" id="ref15">8</reflink>]; Nairne, [<reflink idref="bib12" id="ref16">12</reflink>]). This novel paradigm has been used in a series of studies investigating different aspects of spontaneous recall. These studies have all documented that it is indeed possible to induce spontaneous recollections (e.g. uttered like: "last time I saw a Teddy" or "She did that last time too") in a lab setting even in young children, when using cues with high levels of discriminability (Hjuler, Krøjgaard et al., [<reflink idref="bib10" id="ref17">10</reflink>]; Sonne, Kingo, Berntsen, & Krøjgaard, [<reflink idref="bib20" id="ref18">20</reflink>], [<reflink idref="bib19" id="ref19">19</reflink>]).</p> <p>Only two studies have examined age differences in children's performance. Both showed that spontaneous memories could be induced in 35- and 46-month-olds and that the two age groups performed equally well when spontaneously recalling the event (Krøjgaard et al.,Sonne et al., [<reflink idref="bib20" id="ref20">20</reflink>]). In contrast, when Krøjgaard et al. ([<reflink idref="bib10" id="ref21">10</reflink>]) assessed how children performed when asked questions requiring <emph>strategic</emph> recall of the same event, the youngest age group was outperformed by their older peers (Krøjgaard et al., [<reflink idref="bib10" id="ref22">10</reflink>]). This apparent interaction between retrieval type and age is important, in part because the majority of the research done on memory in childhood has focused on strategic recall (e.g., Hayne, Scarf, & Imuta, [<reflink idref="bib7" id="ref23">7</reflink>]). Only very few studies have examined spontaneous recall. Thus, if young children are better at spontaneous recall, but this type of recall is neglected in research, there is a risk of underestimating children's mnemonic abilities, when assessing the latter using strategic recall tasks only (Krøjgaard et al., [<reflink idref="bib10" id="ref24">10</reflink>]). In other words, young children's difficulties recalling past events as evidenced in numerous studies (e.g., Dahl, Kingo, & Krøjgaard,Simcock & Hayne, [<reflink idref="bib18" id="ref25">18</reflink>]) may, at least in part, be due to the strategic recall tasks employed and hence may not exclusively reflect a memory trace problem.</p> <p>Nonetheless, previous studies comparing spontaneous and strategic recall in young children have limitations. So far, the only two studies comparing the children's performance on a spontaneous recall versus a strategic recall test have used direct yes/no questions for prompting strategic recall of the content of the boxes (Krøjgaard et al., [<reflink idref="bib10" id="ref26">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref27">19</reflink>]). One potential limitation related to the use of "closed" questions is that children tend to provide an answer to such questions even if they do not know the correct information (or even when the question does not make sense), and that they tend to favor "yes" as an answer over the other option "no" (Peterson, Dowden, & Tobin, [<reflink idref="bib14" id="ref28">14</reflink>]). Consistent with this concern, in the paradigm developed by Krøjgaard et al. ([<reflink idref="bib10" id="ref29">10</reflink>]), the correct "yes"/"no"-responses were unevenly distributed across the two boxes: When asking the child whether she (or he) had seen the contents of the box, she was presented with at the first visit (the Known Box), a "yes" would be the correct answer, whereas for the box she had never seen the contents of (the Unknown Box), a "yes" would be the wrong answer. Evidence that children often are less correct on questions requiring a "no" for an answer suggests that the children may have been especially challenged when asked about the Unknown Box (see e.g., Peterson et al., [<reflink idref="bib14" id="ref30">14</reflink>]). The results from several studies suggest that this way of asking children can be problematic (for an overview see e.g., Pipe & Salmon, [<reflink idref="bib15" id="ref31">15</reflink>]), especially for younger children (e.g., Fritzley, Lindsay, & Lee, [<reflink idref="bib5" id="ref32">5</reflink>]). This is also reflected in forensic settings where the use of free recall questions is the recommended strategy to be used instead of, or at least prior to, the use of yes/no questions (Pipe & Salmon, [<reflink idref="bib15" id="ref33">15</reflink>]). This is also consistent with the results from the previous studies from our lab using this design, which showed that the youngest children in particular struggled with providing the correct answers regarding the Unknown Box (Krøjgaard et al., [<reflink idref="bib10" id="ref34">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref35">19</reflink>]). As a consequence, the documented difficulties exhibited by the 35-month-olds regarding strategic recall (Krøjgaard et al., [<reflink idref="bib10" id="ref36">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref37">19</reflink>]) may have been amplified by the specific way in which the questions were asked during the strategic recall test.</p> <hd id="AN0146083225-3">The present study</hd> <p>The present study was undertaken to examine whether the previously documented age differences evident in the children's performance on the Strategic Recall test (but not in the Spontaneous Recall test) replicated when using a different and possibly less biased strategic recall task than the yes/no task in previous studies. Accordingly, we examined how a group of 35- and 46-month-olds would perform on two different tests of Strategic Recall: The standard yes/no questions (henceforth Strategic Recall yes/no questions) used in the previous studies, <emph>and</emph> an additional and potentially more ecologically valid Free Recall session (henceforth Strategic Free Recall). Adding the Strategic Free Recall part also allowed us to analyze the strategic recall by use of the same coding procedures as used for the Spontaneous Recall test. This would allow for a <emph>direct</emph> comparison of the two types of retrieval relative to age, which has never been done before. This research is important for our understanding of the differences between spontaneous and strategic recall in young children and would have direct influences on our understanding of the developmental trajectories of the two types of recall. The specific age groups matched the ones from similar studies from our lab (e.g. Krøjgaard et al., [<reflink idref="bib10" id="ref38">10</reflink>]), and were originally chosen to challenge Tulving's ([<reflink idref="bib21" id="ref39">21</reflink>]) claim that episodic memory is a developmentally late achievement not available before the age of 4 years or maybe even later.</p> <p>We used the paradigm developed by Krøjgaard et al. ([<reflink idref="bib10" id="ref40">10</reflink>]) involving two visits to the lab[<reflink idref="bib2" id="ref41">2</reflink>] including the same two age groups as before in order to see whether the findings from this earlier study would be replicated. In the first visit, the children waited alone (with their parents) in the lab for two minutes while everything mentioned by the children was recorded (as a baseline measure), then they experienced a memorable event involving either teddies or games. In the second visit, they returned after one week for a test of both their spontaneous (reported in Sonne et al., [<reflink idref="bib20" id="ref42">20</reflink>]) and strategic recall of the event. First, they were again left alone for two minutes so we could record everything that they might spontaneously say about the previous visit (reported in Sonne et al., [<reflink idref="bib20" id="ref43">20</reflink>]), and then we added an extra task involving a Strategic Free Recall part to be able to test the children's ability to strategically remember the event but without the problems inherent in the standard Strategic Recall yes/no questions typically used in the paradigm. We thus simply asked the children to tell us everything that happened during the previous visit. Our hope was that this Strategic Free Recall task would circumvent some of the inherent problems when using standard Strategic Recall yes/no questions. However, we still expected the results to be similar to the previous findings from the standard Strategic Recall yes/no questions, that is, that the oldest group would outperform the youngest group of children (i.e., on the amount of hits on a fixed word-list and hits on a coding scheme, see section 2.4). This expectation was based on the previous interpretation that the differences in results are in fact due to differences in retrieval type and not caused by the way the questions were asked (e.g., Krøjgaard et al., [<reflink idref="bib10" id="ref44">10</reflink>]). In order to be able to compare the children's performance on the Strategic Free Recall test to the results from the study by Krøjgaard et al. ([<reflink idref="bib10" id="ref45">10</reflink>]), we ended the test session by also asking the same standard Strategic Recall yes/no questions as in Krøjgaard et al. ([<reflink idref="bib10" id="ref46">10</reflink>]).</p> <p>The present study is based on the same participants, the same materials, and the same data concerning children's spontaneous recall as Sonne et al. ([<reflink idref="bib20" id="ref47">20</reflink>]), but are here collapsed across two conditions reported in Sonne et al. ([<reflink idref="bib20" id="ref48">20</reflink>]) which are irrelevant for the present focus. The main focus here is on the results from the two types of questions addressing strategic recall, which have not been reported previously. For the analysis in which we attempt to directly compare the children's performance on Spontaneous Recall and Strategic Free Recall, we reanalyze the Spontaneous Recall data from the Sonne et al. ([<reflink idref="bib20" id="ref49">20</reflink>]) study. Based on previous indirect evidence (Krøjgaard et al., [<reflink idref="bib10" id="ref50">10</reflink>]), we expected the direct comparison between Spontaneous Recall and Strategic Free Recall to show an interaction between age and retrieval type; that is, while age should not affect Spontaneous Recall, we expected the older participants to reliably outperform the younger ones on Strategic Free Recall (i.e., have more hits on our word-list and coding scheme). For a more detailed description of the basic design, we refer to the study by Krøjgaard et al. ([<reflink idref="bib10" id="ref51">10</reflink>]).</p> <hd id="AN0146083225-4">Method</hd> <p></p> <hd id="AN0146083225-5">Participants</hd> <p>A total of 124 children participated in the present study. Sixty-two of these were 35-month-olds (31 girls, <emph>M</emph><subs>age</subs> = 35.19 months, SD =.50, range 34.4–36.7), and sixty-two were 46-month-olds (35 girls, <emph>M</emph><subs>age</subs> = 46.47 months, SD =.85, range 44.1–48.1). The children were full-term and healthy and they were all recruited from national birth registries. Predominantly Scandinavian Caucasian children participated from families with predominately higher educations (Higher education 66.1%, High School level 17.7%, Occupational or other 16.1%). The only compensation offered was a small present to the participating child.</p> <p>The children were randomly assigned to either the Teddy (n = 62) or the Game event (n = 62) stratified by age and gender. Thirteen children were tested but later excluded due to: fussiness (<reflink idref="bib5" id="ref52">5</reflink>), experimental error (<reflink idref="bib1" id="ref53">1</reflink>), parental disturbances or not adhering to the protocol (<reflink idref="bib4" id="ref54">4</reflink>), being impossible to code because of speaking a foreign language at test (<reflink idref="bib1" id="ref55">1</reflink>) and technical error (<reflink idref="bib2" id="ref56">2</reflink>). The present study was approved by the local ethics committee. Written and informed consent was obtained by the parents of all of the participating children.</p> <hd id="AN0146083225-6">Materials</hd> <p>The children were presented with one of two memorable events: either involving two mechanical teddies or two games. The Teddy event involved two mechanical teddies (that were not commercially available in the country of data collection): Elly, the elephant that could sing and wiggle its ears and Alfred, a dog that could sing, wiggle its ears and clap. The Game event included a bowling game and a homemade throwing game in which the children had to throw three balls into three buckets. All of the props related to each event were hidden in two locked boxes (one red and one gray) when not being demonstrated to the children.</p> <hd id="AN0146083225-7">Design and procedure</hd> <p>The children took part in the typical set-up involving two visits separated by one week (<emph>M</emph> = 6.98 days, <emph>SD </emph>=.49, range = 6–9). Parents had been carefully instructed not to talk about the event at home or during the test, and not to prompt the children in any way. In the first visit, the experimenter brought the child and the parent to the lab, obtained written consent before excusing herself and left the child and parent alone for two minutes in front of the two boxes containing the props for the events. This was done to make a baseline recording for later coding of anything the child might spontaneously say while waiting. Subsequently, the child was presented with one of the two events (Teddy or Game), each lasting about six minutes. For the Teddy event, the experimenter would demonstrate to the child how the two teddies could sing, clap, and wiggle their ears. For the Game event, the experimenter presented the child first with a throwing game and then a bowling game. During both events, the children were encouraged to participate actively and to name the props involved.</p> <p>At the second visit, the child's memory was tested in three different ways: Spontaneous Recall, Strategic Free Recall, and Strategic Recall yes/no questions.</p> <p> <bold>Spontaneous Recall</bold>. At the second visit, the child (and parent) again waited for two minutes during which any utterances that the child might produce were recorded. This part of the session was identical to the first visit (except for the fact that the child now had seen <emph>one</emph> of the events hidden in one of the locked boxes) to be able to compare the results in the baseline measure. After this, the experimenter returned to ask questions aimed at assessing the children's strategic recall.</p> <p> <bold>Strategic Free Recall</bold>. The children were asked different questions probing strategic recall at the very end of the second visit, that is, <emph>after</emph> collecting possible spontaneous recollections. This was done to avoid interfering with the Spontaneous Recall test in which it is important that the child is not prompted to recall anything, and that the child does not sense that his or her memory is being tested. As an extra element compared to the previous studies using the same design, the children in the present study were initially asked an open-ended question regarding what had happened during their previous visit. The experimenter put on a hat called <emph>the Forgetting Hat</emph> and told the child that while wearing the hat, she had forgotten what happened during the previous visit. The exact wording (translated) was:</p> <p>Now look here! This is my Forgetting Hat. Isn't it nice? When I wear it, I don't know anything about what happened last time you were here (puts the hat on). But the forgetting hat actually only works until this clock rings (shows the clock and sets it for two minutes). Now look! Now I am wearing the hat, and because I don't know anything about what happened last time you were here, I would like to hear: Do you remember last time you were here – What happened last time you were in here?</p> <p>The story about the hat was introduced to make the question "What happened last time you were in here" meaningful. The child had a maximum of two minutes to respond, and if the situation was assessed to be too unpleasant for the child, the session would be stopped (but this never happened), otherwise the experimenter kept prompting the child to talk about what happened at the previous visit. Prompts were additional questions following the initial request of a free recall of the event, such as "what can you tell me about your previous visit" or "the Forgetting Hat is still on, therefore I want to know everything you can tell me about your previous visit" with a minimum of 10 seconds between each prompt. Prompting was considerably more prominent here than in the standard Strategic Recall yes/no questions (to be described shortly), as the questions in that procedure were only repeated once before the test was ended. To be able to take into account the potential noise the prompting might add, we registered the number of additional prompts used in the Forgetting Hat procedure.</p> <p>The two-minute duration for the Strategic Free Recall test allowed us to compare the children's performance on this task to their performance in the two-minute Spontaneous Recall test. We added this part to (a) avoid the potential problems inherent when asking children yes/no questions (for an overview see e.g., Pipe & Salmon, [<reflink idref="bib15" id="ref57">15</reflink>]), and (b) to allow us to use the same coding manuals as for the Spontaneous Recall test thereby also making it possible to analyze the children's performance in one combined analysis.</p> <p> <bold>Strategic Recall Yes/No Questions</bold>. Following the Forgetting Hat procedure, for each box (of which the child had previously seen the contents of only one) the child was asked the standard yes/no questions adhering to the procedure from previous studies using this design (Krøjgaard et al., [<reflink idref="bib10" id="ref58">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref59">19</reflink>]). This procedure was maintained to be able to compare spontaneous recall to both of the two different strategic recall tasks (Strategic Free Recall and Strategic Recall yes/no questions), and to make connections to previous studies using the yes/no questions (i.e. Krøjgaard et al., [<reflink idref="bib10" id="ref60">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref61">19</reflink>]). We balanced across subjects which box they were asked about first (i.e., half of the children were asked about the red box first, whereas the other half were asked about the gray box). The experimenter asked (after briefly clarifying that some of the questions might be asking about the same information as provided by the child earlier):</p> <p>"Remember last time you were here ... :</p> <p>Q1: Did you see what was inside the red cabinet/the gray plastic box? (In case the response was "no" or if a "yes" was followed by a description of the contents, Q2 was left out. Only if "yes" or no response was obtained was the child asked about Q2.</p> <p>Q2: What was inside the red cabinet/gray plastic box?</p> <p>Importantly, the Strategic Free Recall part and the Strategic yes/no questions were always presented in this particular order, in order to avoid the Strategic Free Recall question to be influenced by the standard Strategic Recall yes/no questions.</p> <hd id="AN0146083225-8">Coding</hd> <p>Data were coded following the manual employed in the previous studies using this design. For a more detailed description see Krøjgaard et al. ([<reflink idref="bib10" id="ref62">10</reflink>]). The coders were blinded with regard to which event the child had previously seen, and whether it was the child's first or second visit. The data were coded by a primary coder and a secondary coder (re-coding 20% of the data). Interrater agreement was high: 98.7% (90.5% – 100%). Cohens's κ =.74, <emph>p</emph> <.001 also revealed substantial agreement.</p> <p>The data from both the Spontaneous Recall test and the Strategic Free Recall test were coded following a Word List and a Coding Scheme. The coders thus assessed by use of a predefined Word List (see Table 1) whether the children mentioned any event-specific words (about the Teddy or the Game event) or any unspecific words indicating that the child had been here before but without directly mentioning the specific event. Sum scores were created for each child in the end.</p> <p>Table 1. Word list concerning words from the two events separately and the list of "Unspecific" words including words indicating that the child had been to the lab before but without referring to specific event details.*.</p> <p> <ephtml> <table><thead><tr><td>Teddy condition</td><td>Game condition</td><td>Unspecific</td></tr></thead><tbody><tr><td>Alfred</td><td>Ball</td><td>Always</td></tr><tr><td>Animal</td><td>Blue</td><td>Have been</td></tr><tr><td>Button</td><td>Bowling</td><td>Again</td></tr><tr><td>Clap</td><td>Bucket</td><td>Last time</td></tr><tr><td>Dog</td><td>Game</td><td>Key</td></tr><tr><td>Elephant</td><td>Green</td><td>Unlock</td></tr><tr><td>Elly</td><td>Hit</td><td>New/other girl/lady</td></tr><tr><td>Ear</td><td>Medal</td><td>Toys</td></tr><tr><td>Hug</td><td>Pin</td><td /></tr><tr><td>Lives</td><td>Play</td><td /></tr><tr><td>Music</td><td>Prize</td><td /></tr><tr><td>Push</td><td>Red</td><td /></tr><tr><td>Sing</td><td>Roll</td><td /></tr><tr><td>Take/bring out or get</td><td>Throw</td><td /></tr><tr><td>Teddy</td><td>Turn over</td><td /></tr><tr><td>Turn on</td><td>Win</td><td /></tr><tr><td>Wiggle</td><td>Yellow</td><td /></tr></tbody></table> </ephtml> </p> <p>1 * Synonyms and conjugations of the words presented in the table were also coded.</p> <p>Furthermore, the coders analyzed the same utterances according to a Coding Scheme. Here every two-minute clip was divided into twelve time slots with a duration of ten seconds each. Inspired by Levine, Svoboda, Hay, Winocur, and Moscovitch ([<reflink idref="bib11" id="ref63">11</reflink>]) how the child performed in relation to six questions (resulting in either a single score or no score) was assessed:</p> <p></p> <ulist> <item> Does the child by means of <emph>language</emph> refer to knowledge that originates from a previous visit?</item> <p></p> <item> Does the child by means of <emph>gestures</emph> refer to knowledge that originates from a previous visit?</item> <p></p> <item> Does the child's verbal and/or non-verbal behavior indicate that the child mentally <emph>relives</emph> parts of a previous visit?</item> <p></p> <item> Does the child refer to specific <emph>action details</emph> from a previous visit?</item> <p></p> <item> Does the child by means of specific <emph>spatial details</emph> refer to knowledge that originates from a previous visit?</item> <p></p> <item> Does the child by means of specific <emph>social details</emph> refer to knowledge that originates from a previous visit?</item> </ulist> <p>For instance, one child said: "We will try that game [pointing to the gray box] when the lady comes back!" This statement resulted in a score on the dimensions 1, 2, 3, 4, and 6 (but not 5). Another child just stated: "Dog, elephant!" This statement resulted in a score on the dimensions 1 and 3 (but not 2, 4, 5, and 6). For every child, we ended up having six sum scores combining the results from all of the ten second time slots (range 0–12). For the standard Strategic Recall yes/no questions it was simply noted by the experimenter whether the answer was correct or not. Reliability assessment of this part was not deemed necessary (see Krøjgaard et al., [<reflink idref="bib10" id="ref64">10</reflink>]).</p> <hd id="AN0146083225-9">Results</hd> <p>Although the standard Strategic Recall yes/no questions were administered at the very end of the procedure, we will start by looking into the results from these questions to be able to assess whether our dataset is comparable to previous findings (Krøjgaard et al., [<reflink idref="bib10" id="ref65">10</reflink>]). This replication is a prerequisite to meaningfully interpret the children's performance on the Strategic Free Recall test.[<reflink idref="bib3" id="ref66">3</reflink>] We then present findings related to the Strategic Free Recall test, and compare these findings with the Spontaneous Recall test.</p> <hd id="AN0146083225-10">Strategic recall yes/no questions</hd> <p>The Strategic Recall yes/no questions were designed to assess whether the children were able to remember the event. Therefore, at the second visit they were asked directly about the contents of both of the boxes (including the box of which the children had already seen the content [the Known Box], as well as the box that the child had not seen the contents of [the Unknown Box]. Non-parametric binomial tests (tested against the value of <emph>p</emph> =.50, two-tailed) allowed us to investigate how well the children performed on these questions (See Table 2).</p> <p>Table 2. Descriptives and binomial statistics from the standard strategic recall yes/no questions.</p> <p> <ephtml> <table><thead><tr><td>Age Group</td><td>Box asked about first?</td><td /><td>Correct replies regarding Known Box</td><td>Correct replies regarding Unknown Box</td></tr></thead><tbody><tr><td>35 month-olds</td><td /><td><italic>N</italic></td><td>Count</td><td>%</td><td><italic>p*</italic></td><td>Count</td><td>%</td><td><italic>p*</italic></td></tr><tr><td /><td>Known</td><td>34</td><td>27</td><td>79</td><td>=.001</td><td>22</td><td>65</td><td>=.121</td></tr><tr><td /><td>Unknown</td><td>27</td><td>22</td><td>81</td><td>=.002</td><td>8</td><td>30</td><td>=.052<xref ref-type="fn" rid="tfn2" /></td></tr><tr><td /><td>(Both boxes)</td><td>61<xref ref-type="fn" rid="tfn3" /></td><td>49</td><td>80</td><td><.001</td><td>30</td><td>49</td><td>=1</td></tr><tr><td>46-month-olds</td><td /><td /><td /><td /><td /><td /><td /><td /></tr><tr><td /><td>Known</td><td>34</td><td>32</td><td>94</td><td><.001</td><td>33</td><td>97</td><td><.001</td></tr><tr><td /><td>Unknown</td><td>28</td><td>25</td><td>89</td><td><.001</td><td>16</td><td>57</td><td>=.572</td></tr><tr><td /><td>(Both boxes)</td><td>62</td><td>57</td><td>92</td><td><.001</td><td>49</td><td>79</td><td><.001</td></tr><tr><td>All</td><td /><td /><td /><td /><td /><td /><td /><td /></tr><tr><td /><td>Known</td><td>68</td><td>59</td><td>87</td><td><.001</td><td>55</td><td>81</td><td><.001</td></tr><tr><td /><td>Unknown</td><td>55</td><td>47</td><td>85</td><td><.001</td><td>24</td><td>44</td><td>=.419</td></tr><tr><td /><td>(Both boxes)</td><td>123</td><td>106</td><td>86</td><td><.001</td><td>79</td><td>64</td><td>=.002</td></tr></tbody></table> </ephtml> </p> <ulist> <item>2 Although this number is approaching significance, it actually reflects the fact that the 35-month-olds were performing close to significantly worse than chance level.</item> <item>3 Please note that we failed to get responses from one child in this part of the test.</item> </ulist> <p>Taken together, the children performed above chance level on the questions regarding the contents of both boxes (see Table 2). However, when inspecting findings related to the age groups separately, it appears that especially the youngest children had problems answering the question regarding the Unknown Box.</p> <p>To investigate directly whether the results differed across age groups, a sum score was created for the number of correct responses to the two standard questions (i.e., both the question referring to the Known Box as well as the Unknown Box, resulting in a range between 0 and 2). An ANOVA with Age Group (35 vs. 46) as between-subjects variable and the Box sum score as the dependent variable revealed a main effect of Age Group: <emph>F</emph>(<reflink idref="bib1" id="ref67">1</reflink>, 121) = 21.706, <emph>p</emph> <.001, <emph>η<subs>p</subs><sups>2 </sups></emph>=.152. Thus, the 46-month-olds obtained reliably larger sum scores (<emph>M</emph> = 1.71, <emph>SD</emph> =.492) compared to the 35-month-olds (<emph>M</emph> = 1.30, <emph>SD</emph> =.495), thereby replicating the pattern of results from previous findings (Krøjgaard et al., [<reflink idref="bib10" id="ref68">10</reflink>]). This replication allowed us to meaningfully compare the children's performance on this Strategic Recall yes/no test to the Strategic Free Recall test, added in the present study.</p> <hd id="AN0146083225-11">Strategic free recall</hd> <p></p> <hd id="AN0146083225-12">Preliminary analyzes</hd> <p>Table 3 displays the descriptive statistics from the Word List and the Coding Scheme measures from the Strategic Free Recall test (and Spontaneous Recall test). Because the children in the Teddy event served as control for the children in the Game event and <emph>vice versa</emph>, the obtained hits were collapsed across the two events (cf., Krøjgaard et al.,Sonne et al., [<reflink idref="bib19" id="ref69">19</reflink>], [<reflink idref="bib20" id="ref70">20</reflink>]).</p> <p>Table 3. Displays the descriptive statistics and main effects based on the Word List and Coding Scheme. Congruent hits include both event-specific hits (e.g., "teddy," "elephant") and unspecific hits indicative of memory (e.g., "last time," "again").*.</p> <p> <ephtml> <table><thead><tr><td /><td>Strategic Free Recall (The Forgetting Hat)</td><td>Spontaneous Recall (Data collapsed from Sonne et al., <xref ref-type="bibr" rid="bibr20">2020</xref>)</td></tr><tr><td>35-month-olds (n = 62)</td><td>Descriptives</td><td>Descriptives</td><td>One-sample t-tests</td></tr><tr><td>Measure</td><td><italic>M</italic></td><td><italic>SD</italic></td><td><italic>M</italic></td><td><italic>SD</italic></td><td><italic>t</italic></td><td><italic>df</italic></td><td><italic>p</italic></td><td><italic>r</italic></td></tr></thead><tbody><tr><td>Word List</td><td>0.40</td><td>0.80</td><td>0.82</td><td>1.42</td><td>4.56</td><td>61</td><td><.001</td><td>.50</td></tr><tr><td>Word List, Adjusted score</td><td>0.29</td><td>0.60</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr><tr><td>Coding Scheme</td><td /></tr><tr><td> Language</td><td>0.34</td><td>0.68</td><td>0.58</td><td>0.88</td><td>5.20</td><td>61</td><td><.001</td><td>.55</td></tr><tr><td> Gesture</td><td>0.16</td><td>0.45</td><td>0.23</td><td>0.42</td><td>4.22</td><td>61</td><td><.001</td><td>.48</td></tr><tr><td> Reliving</td><td>0.34</td><td>0.68</td><td>0.60</td><td>0.88</td><td>5.36</td><td>61</td><td><.001</td><td>.57</td></tr><tr><td> Action</td><td>0.08</td><td>0.28</td><td>0.24</td><td>0.50</td><td>3.80</td><td>61</td><td><.001</td><td>.44</td></tr><tr><td> Spatial</td><td>0.06</td><td>0.25</td><td>0.15</td><td>0.40</td><td>2.87</td><td>61</td><td>=.006</td><td>.34</td></tr><tr><td> Social</td><td>0.00</td><td>0.00</td><td>0.02</td><td>0.13</td><td>1.00</td><td>61</td><td>=.321</td><td>.13</td></tr><tr><td>SUM Score (Coding Scheme)</td><td>0.98</td><td>1.98</td><td>1.79</td><td>2.65</td><td>5.32</td><td>61</td><td><.001</td><td>.56</td></tr><tr><td>SUM Score, Adjusted</td><td>0.70</td><td>1.47</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr><tr><td>46-month-olds (n = 62)</td><td>Descriptives</td><td>Descriptives</td><td>One-sample t-tests</td></tr><tr><td>Measure</td><td><italic>M</italic></td><td><italic>SD</italic></td><td><italic>M</italic></td><td><italic>SD</italic></td><td><italic>t</italic></td><td><italic>df</italic></td><td><italic>p</italic></td><td><italic>r</italic></td></tr><tr><td>Word List</td><td>1.24</td><td>1.70</td><td>1.08</td><td>1.42</td><td>6.00</td><td>61</td><td><.001</td><td>.61</td></tr><tr><td>Word List, Adjusted Score</td><td>1.46</td><td>2.15</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr><tr><td>Coding Scheme</td></tr><tr><td> Language</td><td>0.74</td><td>0.96</td><td>0.77</td><td>0.82</td><td>7.45</td><td>61</td><td><.001</td><td>.69</td></tr><tr><td> Gesture</td><td>0.18</td><td>0.39</td><td>0.37</td><td>0.55</td><td>5.31</td><td>61</td><td><.001</td><td>.56</td></tr><tr><td> Reliving</td><td>0.74</td><td>0.96</td><td>0.82</td><td>0.88</td><td>7.38</td><td>61</td><td><.001</td><td>.69</td></tr><tr><td> Action</td><td>0.55</td><td>0.78</td><td>0.24</td><td>0.43</td><td>4.41</td><td>61</td><td><.001</td><td>.49</td></tr><tr><td> Spatial</td><td>0.16</td><td>0.41</td><td>0.19</td><td>0.40</td><td>3.83</td><td>61</td><td><.001</td><td>.44</td></tr><tr><td> Social</td><td>0.15</td><td>0.36</td><td>0.02</td><td>0.13</td><td>1.00</td><td>61</td><td>=.321</td><td>.13</td></tr><tr><td>SUM Score (Coding Scheme)</td><td>2.37</td><td>2.93</td><td>2.40</td><td>2.40</td><td>7.89</td><td>61</td><td><.001</td><td>.71</td></tr><tr><td>SUM Score, Adjusted</td><td>2.88</td><td>4.14</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr></tbody></table> </ephtml> </p> <p>4 <bold>*</bold> Please note that we have included both the raw data as well as adjusted scores for both the Word List and the Coding Scheme. The SUM Score from the Coding Scheme only reflects the five dimensions with the social dimension left out. Only in relation to the Spontaneous Recall data was it possible to run t-tests tested against "0," as we only have baseline data from this test.</p> <p>To simplify the analyzes involving the results of the Strategic Free Recall test from the comprehensive Coding Scheme measures, we first examined whether the dimensions from the Coding Scheme would constitute a scale (for a similar procedure, see Krøjgaard et al., [<reflink idref="bib10" id="ref71">10</reflink>]; Sonne et al., [<reflink idref="bib19" id="ref72">19</reflink>]). An item analysis of the different dimensions from the Coding Scheme resulted in a Cronbach's <emph>α </emph>=.841. Although the Cronbach's <emph>α</emph> obtained here is lower than the value obtained in the previous studies (see e.g., Krøjgaard et al., [<reflink idref="bib10" id="ref73">10</reflink>]), the Cronbach's <emph>α</emph> still indicates acceptable internal consistency meaning that we can base the coming analyzes on the sum scores from the Coding Scheme. As in previous studies we decided to include five out of the six dimensions since removing the Social dimension increased the Cronbach's <emph>α</emph> (=.851). Based on these five dimensions, a sum score was calculated. This sum score was used as raw score in all following analyzes involving the Coding Scheme measures.</p> <p>To be able to extend the Strategic Free Recall procedure to make it last two minutes (which was necessary to use the same coding material as for Spontaneous Recall), the use of prompts was a necessary addition to make the procedure tolerable for the children. This, however, included the risk that the number of prompts would be applied disproportionally across participants and more specifically unevenly used across the two age groups. Potentially, the children having most difficulties recalling what had happened could end up receiving relatively more help (i.e., the younger children receiving more prompts), which would confound the results as all children were then not tested on equal terms. We therefore first examined whether the use of prompts affected the results. An independent samples <emph>t</emph>-test indeed showed a statistical significance in the number of prompts given to the two age groups, <emph>t</emph>(<reflink idref="bib122" id="ref74">122</reflink>) = 4.473, <emph>p < </emph>.001, <emph>r</emph> =.38. As expected, the youngest age group received reliably more prompts (<emph>M</emph> = 4.95, <emph>SD</emph> = 1.35) compared to their older peers (<emph>M</emph> = 3.92, <emph>SD</emph> = 1.22). To investigate whether prompts additionally had an effect on the children's performance, we correlated the number of prompts with the two age groups' performance on both the Word List results and the Coding Scheme sum score results. For the youngest age group we found a significant positive correlation between the number of prompts and the children's performance on both the Word List (<emph>r</emph> =.506, <emph>p</emph> <.001) as well as in relation to the Coding Scheme sum score measures (<emph>r</emph> =.522, <emph>p</emph> <.001). For the oldest age group however, no significant correlations were found (the Word List [<emph>r</emph> =.144, <emph>p</emph> =.263] and the Coding Scheme sum score measures [<emph>r</emph> =.133, <emph>p</emph> =.304]). Thus, the youngest children did receive more prompts than their older peers, and the numbers of prompts provided were systematically positively associated with the young children's performance.</p> <p>Based on these preliminary analyzes, we deemed it necessary to control for the number of prompts provided in relation to both coding measures (the Strategic Free Recall Word List measures and the Strategic Free Recall Coding Scheme Sum Score measures). Two adjusted measures (one for the Free Recall Word List, and one for the Free Recall Coding Scheme Sum Score) were computed. For each child, we first computed a factor assessing the <emph>relative prompting</emph> he or she had received by calculating the mean amount of prompting received by <emph>all</emph> children divided by the number of prompts <emph>this</emph> specific child had received, thus resulting in a We then computed two adjusted measures for each child by multiplying the individual relative prompting factor with the raw scores each child had received from the Free Recall Word List as well as from the Free Recall Coding Scheme Sum Score. The adjusted scores thus controlled for the number of prompts received while maintaining scores within the same numerical magnitude as the raw scores. Note that multiplying with the mean number of prompts (or with any other constant) does not affect the relation between the individual scores, while it does have the benefit of bringing the measures from the Strategic Free Recall procedure within the same numerical magnitude as the data from spontaneous recall which was crucial in order to be able to look at spontaneous and strategic recall in a combined analysis in relation to Age Group (see below). This allowed us to test for the possible interaction between Retrieval Type and Age Group while still controlling for the noise induced by the use of prompts. Thus, all subsequent analyzes involving the Strategic Free Recall measures were based on these adjusted measures (see Table 3).</p> <p>In the following, we first look at the adjusted results from the Word List followed by the adjusted results using the Coding Scheme sum scores and finally the combined direct comparison.</p> <hd id="AN0146083225-13">Strategic free recall: word list</hd> <p>Running an ANOVA with Age Group (35 vs. 46) as between-subjects variable and the adjusted Word List Free Recall as the dependent variable revealed a main effect of Age Group, <emph>F</emph>(<reflink idref="bib1" id="ref75">1</reflink>, 122) = 17.09, <emph>p</emph> <.001, <emph>η<subs>p</subs><sups>2 </sups></emph>=.123, as the 46-month-olds (<emph>M</emph> = 1.46, <emph>SD</emph> = 2.15, range 0.0– 10.36) reliably outperformed their younger peers (<emph>M</emph> =.29, <emph>SD</emph> =.60, range 0.0– 2.66). This was confirmed with a nonparametric test (Mann-Whitney: Z = − 3.437, <emph>p</emph> <.001, <emph>r</emph> = − 0.308). These results thus replicated the results from the standard Strategic Recall yes/no questions from the previous study in which the same two age groups were compared (see Krøjgaard et al., [<reflink idref="bib10" id="ref76">10</reflink>]), as well as the results from the standard Strategic Recall yes/no questions from the present study.</p> <hd id="AN0146083225-14">Strategic free recall: coding scheme</hd> <p>Running an ANOVA with Age Group (35 vs. 46) as between-subjects variable and the adjusted sum scores computed from the Strategic Free Recall coding scheme as the dependent variable again revealed a main effect of Age Group: <emph>F</emph>(<reflink idref="bib1" id="ref77">1</reflink>, 122) = 15.33, <emph>p</emph> <.001, <emph>η<subs>p</subs><sups>2 </sups></emph>=.112. Thus, resembling the results obtained from the adjusted Word List, we again found that the 46-month-olds (<emph>M</emph> = 2.88, <emph>SD</emph> = 4.14, range 0.0– 17.76) outperformed the 35-month-olds (<emph>M</emph> =.70, <emph>SD</emph> = 1.47, range 0.0– 7.10) in relation to Strategic Recall. This was confirmed with a nonparametric test (Mann-Whitney: Z = − 3.364, <emph>p</emph> <.001, <emph>r</emph> = − 0.302).</p> <p>Importantly, these findings combined with the results from the standard Strategic Recall yes/no questions suggest that regardless of the type of questions asked requiring strategic recall (Free Recall <emph>or</emph> standard yes/no questions), the 46-month-olds reliably outperformed their younger peers on strategic recall.</p> <hd id="AN0146083225-15">Spontaneous and strategic free recall combined</hd> <p>Since the Strategic Free Recall part was coded in the exact same way as the spontaneous utterances (reported in Sonne et al., [<reflink idref="bib20" id="ref78">20</reflink>], see Table 3 for descriptives and related <emph>t</emph>-tests) it allowed us to examine whether Retrieval Type would interact with age of the children using two mixed-model ANOVAs with Age Group (35 vs. 46) as between-subjects factor, and Retrieval Type (Spontaneous vs. Strategic) as within-subjects factor. We ran one for each coding measure as the dependent variable – one with Word List codings, and one with the Coding Scheme sum score.</p> <p>The ANOVA with Word List measures as the dependent variable revealed a main effect of Age Group, <emph>F</emph>(<reflink idref="bib1" id="ref79">1</reflink>, 122) = 11.75, <emph>p</emph> =.001, <emph>η<subs>p</subs><sups>2 </sups></emph>=.088 as the 46-month-olds generally outperformed their younger peers regardless of Retrieval Type. Importantly, the analysis also revealed a significant interaction between Retrieval Type and Age Group, <emph>F</emph>(<reflink idref="bib1" id="ref80">1</reflink>, 122) = 7.16, <emph>p</emph> =.008, <emph>η<subs>p</subs><sups>2 </sups></emph>=.055 (see Figure 1a). Thus, whereas the children's performance across the two age groups was quite similar in relation to Spontaneous Recall (<emph>M</emph><subs>35</subs> = 0.82, <emph>SD</emph> = 1.42, range 0.0– 6.0; <emph>M</emph><subs>46</subs> = 1.08, <emph>SD</emph> = 1.42, range 0.0– 6.0; <emph>t</emph>(<reflink idref="bib122" id="ref81">122</reflink>) = − 1.01, <emph>p</emph> =.313, <emph>d</emph> =.18), the 46-month-old outperformed their younger peers in relation to the Strategic Free Recall task (<emph>M</emph><subs>35</subs> = 0.29, <emph>SD</emph> = 0.60, range 0.0– 2.66; <emph>M</emph><subs>46</subs> = 1.46, <emph>SD</emph> = 2.15, range 0.0– 10.36; <emph>t</emph>(<reflink idref="bib122" id="ref82">122</reflink>) = − 4.134, <emph>p</emph> <.001, <emph>d</emph> =.77).</p> <p>Graph: Figure 1. a, b: The figures display the comparisons of retrieval performance (Spontaneous vs. Strategic Free Recall) divided by Age Group for each of the two coding measures (Figure 1a: Word List measure; Figure 1b: Coding scheme measure). Dotted line: 46 months old, full line: 35 months old.</p> <p>The equivalent mixed-model ANOVA with the sum scores from the Coding Scheme measures as the dependent variable led to similar results. Again, we obtained a main effect of Age Group, <emph>F</emph>(<reflink idref="bib1" id="ref83">1</reflink>, 122) = 12.51, <emph>p</emph> =.001, <emph>η<subs>p</subs><sups>2 </sups></emph>=.093, indicating that the 46-month-olds fared better than their younger peers regardless of Retrieval Type. And again, there was a significant interaction between Retrieval Type and Age Group, <emph>F</emph>(<reflink idref="bib1" id="ref84">1</reflink>, 122) = 6.03, <emph>p</emph> =.015, <emph>η<subs>p</subs><sups>2 </sups></emph>=.047 (see Figure 1b). Thus, whereas the children performed quite similarly when recalling spontaneously (<emph>M</emph><subs>35</subs> = 1.79, <emph>SD</emph> = 2.65, range 0.0– 9.0; <emph>M</emph><subs>46</subs> = 2.4, <emph>SD</emph> = 2.40, range 0.0– 9.0; <emph>t</emph>(<reflink idref="bib122" id="ref85">122</reflink>) = − 1.350, <emph>p</emph> =.179, <emph>d</emph> =.82), the 46-month-olds reliably outperformed their younger peers on Strategic Recall in the Free Recall test (<emph>M</emph><subs>35</subs> = 0.70, <emph>SD</emph> = 1.47, range 0.0– 7.10; <emph>M</emph><subs>46</subs> = 2.88, <emph>SD</emph> = 4.14, range 0.0– 17.76; <emph>t</emph>(<reflink idref="bib122" id="ref86">122</reflink>) = − 3.916, <emph>p</emph> <.001, <emph>d</emph> =.91).</p> <p>To summarize, the results display two features: First, the results from the Strategic Free Recall questions conceptually replicate the results from the standard Strategic Recall yes/no questions: In both types of strategic recall, the 46-month-olds clearly and reliably outperformed their younger peers. Second, the combined analyzes of the Spontaneous Recall and the Strategic Free Recall test revealed reliable interactions between Retrieval Type and Age Group for both measures (Word List and Coding Scheme sum scores). These interactions show that while the children across the two age groups obtained similar results when recalling spontaneously, the 46-month-olds reliably and consistently outperformed their younger peers on strategic recall.</p> <hd id="AN0146083225-16">Discussion</hd> <p>In the present study, we investigated 35- and 46-month-olds' ability to strategically recall a unique event using both free recall and standard closed yes/no questions. The current research was inspired by an interest in shedding new light on previous findings documenting differences in young children's performance on different memory tasks, and suggesting that 35- and 46-month-old children's performance is effected by retrieval type in that especially strategic recall was difficult for the youngest age group whereas spontaneous recall was possible for both age groups (Krøjgaard et al., [<reflink idref="bib10" id="ref87">10</reflink>]). The purpose of the present study was therefore twofold: First, by using the novel Forgetting Hat procedure we wanted to extend previous findings on strategic retrieval derived from the Strategic Recall yes/no question task to a potentially more ecologically valid Strategic Free Recall task that would also be more similar to recommended interviewing procedures (Pipe & Salmon, [<reflink idref="bib15" id="ref88">15</reflink>]). Second, using the same two-minute coding procedure for the Strategic Free Recall test as has been used for Spontaneous Recall, we wanted to conduct a direct test of the possible interaction between age of the children and retrieval type, only inferred in previous studies using the same design (cf., Krøjgaard et al., [<reflink idref="bib10" id="ref89">10</reflink>]). We begin by discussing the obtained results in the same order as outlined above. Then we discuss the pros and cons of using the Strategic Free Recall test. Finally, we discuss how the obtained results may impact the field in a broader perspective.</p> <p>Overall using a question probing Strategic Free Recall of the event produced a pattern of results (in terms of observed age differences) comparable to when asking the children to recall the event by use of the standard Strategic Recall yes/no questions. The results concerning Strategic Recall (both methods) thus replicate and extend previous findings: First, when using standard Strategic Recall yes/no questions, we again found that the 35-month-olds were outperformed by the 46-month-olds. Furthermore, the use of more ecologically valid open-ended Strategic Free Recall questions did not change this pattern in the results. Again, the youngest age group was outperformed by their older peers. One potential concern might be that being exposed to the Strategic Free Recall procedure just <emph>before</emph> being asked the standard yes/no questions, would have affected the children's performance on the latter. However, as the pattern and magnitude of the present results from the standard yes/no questions for both age groups were very similar to those obtained in the Krøjgaard et al. ([<reflink idref="bib10" id="ref90">10</reflink>]) study, in which the Strategic Free Recall procedure was not administered, this concern does not seem warranted. Thus, even when using the Strategic Free Recall task – in our view a more ecologically valid task – we still obtained the same overall pattern of results as previously documented with the use of yes/no questions. Hence, these results suggest that strategic retrieval is largely dependent on age – regardless of the kinds of questions asked.</p> <p>The format of the Strategic Free Recall test had the advantage of allowing for a direct statistical comparison with the same children's performance on a Spontaneous Recall test. Such direct comparison assessing potential interactions between Age Group and Retrieval Type has never been conducted before among children. The results from both the Word List and the Coding Scheme measures were clear and systematic. On both measures, the combined analyzes revealed a reliable interaction between Age Group and Retrieval Type: Whereas the 35- and the 46-month-olds performed similarly when spontaneously recalling the first lab visit, the 46-month-olds systematically outperformed their younger peers on the Strategic Free Recall questions, requiring strategic recall. Thus, the results from the present study replicate previous studies using Strategic Recall yes/no questions and provide the first direct comparison of Spontaneous and Strategic Free Recall tested against each other using the same statistical analysis.</p> <p>Taken together, the results (a) lend further credibility to the basic design in which spontaneous recall can be induced experimentally in young children, (b) reveal young children's substantial difficulties with strategic recall relative to spontaneous recall when tested on the same to-be-remembered events, and therefore (c) provide empirical evidence supporting the claim that spontaneous recall may be a basic mode of retrieval that may developmentally precede strategic retrieval (Berntsen, [<reflink idref="bib2" id="ref91">2</reflink>], [<reflink idref="bib3" id="ref92">3</reflink>]).</p> <p>It should be noted that while the employment of the novel Strategic Free Recall test using the Forgetting Hat procedure obviously had advantages, it also had drawbacks. First of all, in order to obtain results from a two-minute time span comparable to the testing period for Spontaneous Recall, we had to use prompts. And since the use of these prompts was not evenly distributed across the two age groups, this procedure added noise to the design that we subsequently had to control for by computing scores adjusted for the amount of prompting. In addition, the coding of the Strategic Free Recall procedure was comprehensive and considerably more time consuming than the coding procedure related to the standard yes/no questions. Finally, some of the children were somewhat uncomfortable being repeatedly prompted in a, for them, rather difficult task. Considering, that the results from these two procedures (standard Strategic Recall yes/no questions and the Strategic Free Recall test) both testing strategic recall, basically led to the same results (the older children reliably outperforming their younger peers), one might ask whether the Strategic Free Recall using the Forgetting Hat procedure in its current form was the optimal version. For future studies, the best compromise between the Strategic Recall standard yes/no questions and the Strategic Free Recall test might be to ask open-ended questions as in the Strategic Free Recall task, but without using repeated prompts. In this manner, the questions will still be more ecologically valid, while avoiding adding noise.</p> <p>The results from the present study raise several important issues to be considered in a broader perspective. First, although we here administered supposedly more ecologically valid questions requiring strategic recall, there still was a clear effect of age, and the youngest children had severe difficulties providing relevant information (for the Strategic Free Recall) or correct answers (for the Strategic Recall yes/no questions). These results suggest that strategic recall is <emph>generally</emph> difficult for young children. Second, and closely related, considering the fact that Age Groups and Retrieval Type interacted systematically in both direct comparisons, we may ask whether spontaneous recall has received the attention it seems to deserve. The background for this question obviously is that the great majority of studies on event memory have employed questions asked directly to children (Hayne et al., [<reflink idref="bib7" id="ref93">7</reflink>]) requiring strategic recall – clearly inspired by the default methodological approach when examining event memory in adults.</p> <p>When considering this question, it should be taken into consideration that the criteria currently used for spontaneous recall (i.e., verbally produced, unprompted, and environmentally cued, Krøjgaard et al., [<reflink idref="bib9" id="ref94">9</reflink>]) are quite conservative. For instance, the experimental paradigm used here only captures the spontaneous recollections that are <emph>de facto</emph> verbally stated by the children. Meanwhile, the parents are carefully instructed not to ask questions and only to reply with single-word responses to any utterances from the children, which all things being equal probably diminishes the children's inclination to say anything at all during the test situation. Thus, if the children actually experience remindings of the earlier visit during the test, but refrain from verbalizing them (which, in our view, certainly is possible given the test conditions), these will (obviously) not be registered. In other words, we have reason to believe that the way we induce and assess spontaneous recall experimentally in the present paradigm, may underestimate the actual prevalence of "spontaneous" recall. Consequently, the prevalence of spontaneous recall may be even more pronounced than the results from the present study seem to imply.</p> <p>At least in part due to the nature of spontaneous recollections, more research is needed to fully understand the conditions under which children are capable of spontaneous recall. For instance, we have no knowledge of how different retention intervals might affect spontaneous recall. Nor do we know in detail under what conditions this way of recalling events is promoted or hindered. In addition, as we move down in age, children's language abilities will eventually become a delimiting factor. To pursue and test the claim that spontaneous recall may be a basic mode of remembering (Berntsen, [<reflink idref="bib2" id="ref95">2</reflink>], [<reflink idref="bib3" id="ref96">3</reflink>]), we probably need to investigate whether non-verbal "spontaneous" recall could be induced experimentally in even younger children than the ones tested here. However, this would require a modification to the operationalization of spontaneous recall, as motor indices are not verbal.</p> <p>In summary, the present results reveal the difficulties especially young children have in relation to strategically recalling an event even when using open-ended questions, and document how spontaneous recall could be a way of bypassing these difficulties. From a forensic point of view, these findings could be important and might be used to inform interview practices when for instance relevant environmental cues are available.</p> <hd id="AN0146083225-17">Acknowledgments</hd> <p>We would like to thank Amalie Juul Harregaard, Anne Rytter Christensen, Louise Jensen, and Sarah Jakobsen for conducting all of the test sessions. Furthermore, we would like to thank Yibin Zhang for editing the video material for later coding, Caroline Beyer and Mette Sørensen for coding the data and finally Emilie Kristensen for entering data into SPSS. Finally, we want to thank all of the participating children as well as their parents for bringing them to our lab.</p> <p>This study has been supported by the Danish National Research Foundation (DNRF89) and the VELUX FOUNDATION (VELUX10386).</p> <hd id="AN0146083225-18">Disclosure statement</hd> <p>No potential conflict of interest was reported by the authors.</p> <ref id="AN0146083225-19"> <title> References </title> <blist> <bibl id="bib1" idref="ref11" type="bt">1</bibl> <bibtext> Bauer, P. J., Kroupina, M. G., Schade, J. A., Dropnik, P. L., & Wewerka, S. S. (1998). If memory serves, will language? Later verbal accessibility of early memories. Development and Psychopathology, 10 (4), 655 – 679. doi: 10.1017/S0954579498001801</bibtext> </blist> <blist> <bibl id="bib2" idref="ref1" type="bt">2</bibl> <bibtext> Berntsen, D. (2009). Involuntary autobiographical memories: An introduction to the unbidden past. Cambridge : Cambridge University Press.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref66" type="bt">3</bibl> <bibtext> Berntsen, D. (2012). Spontaneous recollections: Involuntary autobiographical memories are a basic mode of remembering. In D. Berntsen & D. C. Rubin (Eds.), Understanding autobiographical memory: Theories and approaches (pp. 290 – 310). Cambridge : Cambridge University Press.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref4" type="bt">4</bibl> <bibtext> Berntsen, D., Staugaard, S. R., & Sørensen, L. M. T. (2013). Why am I remembering this now? Predicting the occurrence of involuntary (spontaneous) episodic memories. Journal of Experimental Psychology. General, 142 (2), 426 – 444. doi: 10.1037/a0029128</bibtext> </blist> <blist> <bibl id="bib5" idref="ref32" type="bt">5</bibl> <bibtext> Fritzley, V. H., Lindsay, R. C. L., & Lee, K. (2013). Young children's response tendencies toward yes-no questions concerning actions. Child Development, 84 (2), 711 – 725. doi: 10.1111/cdev.12006</bibtext> </blist> <blist> <bibl id="bib6" idref="ref2" type="bt">6</bibl> <bibtext> Hall, S. A., Rubin, D. C., Miles, A., Davis, S. W., Wing, E. A., Cabeza, R., & Berntsen, D. (2014). The neural basis of involuntary episodic memories. Journal of Cognitive Neuroscience, 26 (10), 2385 – 2399. doi: 10.1162/jocn_a_00633</bibtext> </blist> <blist> <bibl id="bib7" idref="ref23" type="bt">7</bibl> <bibtext> Hayne, H., Scarf, D., & Imuta, K. (2015). Childhood memories. In J. D. Wright (Ed.), International encyclopedia of the social & behavioral sciences (Vol. 3, 2nd ed., pp. 465 – 470). Oxford, UK : Elsevier.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref15" type="bt">8</bibl> <bibtext> Hunt, R. R., & Smith, R. E. (1996). Accessing the particular from the general: The power of distinctiveness in the context of organization. Memory & Cognition, 24 (2), 217 – 225. doi: 10.3758/BF03200882</bibtext> </blist> <blist> <bibl id="bib9" idref="ref7" type="bt">9</bibl> <bibtext> Krøjgaard, P., Kingo, O. S., Dahl, J. J., & Berntsen, D. (2014). "That one makes things small": Experimentally induced spontaneous memories in 3.5-year-olds. Consciousness and Cognition, 30, 24 – 35. doi: 10.1016/j.concog.2014.07.017</bibtext> </blist> <blist> <bibtext> Krøjgaard, P., Kingo, O. S., Jensen, T. S., & Berntsen, D. (2017). By-passing strategic retrieval: Experimentally induced spontaneous episodic memories in 35- and 46-month-old children. Consciousness and Cognition, 55, 91 – 105. doi: 10.1016/j.concog.2017.08.001</bibtext> </blist> <blist> <bibtext> Levine, B., Svoboda, E., Hay, J. F., Winocur, G., & Moscovitch, M. (2002). Ageing and autobiographical memory: Dissociating episodic from semantic retrieval. Psychology and Aging, 17 (4), 677 – 689. doi: 10.1037//0882-7974.17.4.677</bibtext> </blist> <blist> <bibtext> Nairne, J. S. (2002). The myth of the encoding-retrieval match. Memory, 10 (5–6), 389 – 395. doi: 10.1080/09658210244000216</bibtext> </blist> <blist> <bibtext> Nelson, K., & Ross, G. (1980). The generalities and specifics of long-term memory in infants and young children. New Directions for Child and Adolescent Development, 10 (10), 87 – 101. doi: 10.1002/cd.23219801008</bibtext> </blist> <blist> <bibtext> Peterson, C., Dowden, C., & Tobin, J. (1999). Interviewing preschoolers: Comparisons of yes/no and Wh-questions. Law and Human Behavior, 23 (5), 539 – 555. doi: 10.1023/A:1022396112719</bibtext> </blist> <blist> <bibtext> Pipe, M., & Salmon, K. (2009). Memory development and the forensic context. In M. L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (2nd ed., pp. 241 – 282). Hove, United Kingdom & New York: Psychology Press.</bibtext> </blist> <blist> <bibtext> Reese, E. (1999). What children say when they talk about the past. Narrative Inquiry, 9 (2), 215 – 241. doi: 10.1075/ni.9.2.02ree</bibtext> </blist> <blist> <bibtext> Rovee-Collier, C., & Cuevas, N. (2009). The development of infant memory. In M. L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 11 – 41). Hove, United Kingdom & New York : Psychology Press.</bibtext> </blist> <blist> <bibtext> Simcock, G., & Hayne, H. (2002). Breaking the barrier? Children fail to translate their preverbal memories into language. Psychological Science, 13 (3), 225 – 231. doi: 10.1111/1467-9280.00442</bibtext> </blist> <blist> <bibtext> Sonne, T., Kingo, O. S., Berntsen, D., & Krøjgaard, P. (2019). Thirty-five-month-old children have spontaneous memories despite change of context for retrieval. Memory, 27 (1), 38 – 48. doi: 10.1080/09658211.2017.1363243</bibtext> </blist> <blist> <bibtext> Sonne, T., Kingo, O. S., Berntsen, D., & Krøjgaard, P. (2020). Noting a difference: Change in social context prompts spontaneous recall in 46-month-olds but not in 35-month-olds. Psychological Research. doi: 10.1007/s00426-020-01310-7</bibtext> </blist> <blist> <bibtext> Tulving, E. (2005). Episodic memory and autonoesis: Uniquely human? In H. S. Terrace & J. Metcalfe (Eds.), The missing link in cognition: Origins of self-reflective consciousness (pp. 3 – 56). Oxford : Oxford University Press.</bibtext> </blist> <blist> <bibtext> Tulving, E., & Thomson, D. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80 (5), 352 – 373. doi: 10.1037/h0020071</bibtext> </blist> </ref> <ref id="AN0146083225-20"> <title> Footnotes </title> <blist> <bibtext> Note that extensive lines of existing infant memory research do not qualify as spontaneous recall as operationalized here. First, imitation studies are motor based and not verbal, and if they register verbal indices of memory, the procedures involve explicit prompts (i.e., "show me how to make a windmill," see Bauer, Kroupina, Schade, Dropnik, & Wewerka, [1]), and hence do not qualify as spontaneous recall. Second, while reinforcement paradigms like the conjugate reinforcement paradigm (for a review, see Rovee-Collier & Cuevas, [17]) do not involve prompts, recall is not spontaneous as defined here, as the responses are purely motor-based and not verbal as required here, and do not involve a communicative aspect either.</bibtext> </blist> <blist> <bibtext> Please note that the data on the children's spontaneous recollections in this set-up has been reported in a different study, broken down by social cue (see Sonne et al., [20]). This manipulation was theoretically unrelated to the primary questions of the present paper, and data from this turned out to also be statistically unrelated. Consequentially, they were presented in a separate paper. The spontaneous recollections are not the primary focus in the present paper, but will however be employed in a combined analysis (see the Results section).</bibtext> </blist> <blist> <bibtext> In the study by Sonne et al. ([20]) we manipulated whether the children were met by the same or a novel experimenter at test: However, this manipulation showed no main effect and no interaction with age group for any of the strategic retrieval tasks. Therefore, the results from the two experimental groups were collapsed in all of the following analyzes.</bibtext> </blist> </ref> <aug> <p>By Trine Sonne; Osman S. Kingo; Dorthe Berntsen and Peter Krøjgaard</p> <p>Reported by Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib13" firstref="ref5"></nolink> <nolink nlid="nl2" bibid="bib16" firstref="ref6"></nolink> <nolink nlid="nl3" bibid="bib10" firstref="ref8"></nolink> <nolink nlid="nl4" bibid="bib22" firstref="ref12"></nolink> <nolink nlid="nl5" bibid="bib12" firstref="ref16"></nolink> <nolink nlid="nl6" bibid="bib20" firstref="ref18"></nolink> <nolink nlid="nl7" bibid="bib19" firstref="ref19"></nolink> <nolink nlid="nl8" bibid="bib18" firstref="ref25"></nolink> <nolink nlid="nl9" bibid="bib14" firstref="ref28"></nolink> <nolink nlid="nl10" bibid="bib15" firstref="ref31"></nolink> <nolink nlid="nl11" bibid="bib21" firstref="ref39"></nolink> <nolink nlid="nl12" bibid="bib11" firstref="ref63"></nolink> <nolink nlid="nl13" bibid="bib122" firstref="ref74"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1270504
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Age Affects Strategic but Not Spontaneous Recall in 35- and 46-Month-Old Children
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Sonne%2C+Trine%22">Sonne, Trine</searchLink><br /><searchLink fieldCode="AR" term="%22Kingo%2C+Osman+S%2E%22">Kingo, Osman S.</searchLink><br /><searchLink fieldCode="AR" term="%22Berntsen%2C+Dorthe%22">Berntsen, Dorthe</searchLink><br /><searchLink fieldCode="AR" term="%22Krøjgaard%2C+Peter%22">Krøjgaard, Peter</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Cognition+and+Development%22"><i>Journal of Cognition and Development</i></searchLink>. 2020 21(4):603-621.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 19
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2020
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Age+Differences%22">Age Differences</searchLink><br /><searchLink fieldCode="DE" term="%22Young+Children%22">Young Children</searchLink><br /><searchLink fieldCode="DE" term="%22Memory%22">Memory</searchLink><br /><searchLink fieldCode="DE" term="%22Recall+%28Psychology%29%22">Recall (Psychology)</searchLink><br /><searchLink fieldCode="DE" term="%22Developmental+Stages%22">Developmental Stages</searchLink><br /><searchLink fieldCode="DE" term="%22Performance%22">Performance</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/15248372.2020.1797748
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1524-8372
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: It is well documented that young children have difficulties with strategically remembering past events. Recent evidence on event memory in 35- and 46-month-old children suggests that strategic retrieval (yes/no questions) improves with age, whereas spontaneous retrieval is relatively unaffected by age. We here replicate and extend those findings (N = 124): First, a novel free (strategic) recall test was added to improve ecological validity. Second, the free recall procedure allowed us to make direct comparisons between spontaneous and free strategic recall relative to age. The free recall test revealed similar results in the standard yes/no questions (older children outperformed younger). The direct comparison between spontaneous and free recall revealed a reliable interaction between age and retrieval mode: While the children's age did not affect spontaneous recall, the 46-month-olds outperformed the 35-month-olds on the free recall test. The results add to the accumulating evidence that spontaneous recall of events is an early developmental achievement.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2020
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1270504
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1270504
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/15248372.2020.1797748
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 19
        StartPage: 603
    Subjects:
      – SubjectFull: Age Differences
        Type: general
      – SubjectFull: Young Children
        Type: general
      – SubjectFull: Memory
        Type: general
      – SubjectFull: Recall (Psychology)
        Type: general
      – SubjectFull: Developmental Stages
        Type: general
      – SubjectFull: Performance
        Type: general
    Titles:
      – TitleFull: Age Affects Strategic but Not Spontaneous Recall in 35- and 46-Month-Old Children
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Sonne, Trine
      – PersonEntity:
          Name:
            NameFull: Kingo, Osman S.
      – PersonEntity:
          Name:
            NameFull: Berntsen, Dorthe
      – PersonEntity:
          Name:
            NameFull: Krøjgaard, Peter
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2020
          Identifiers:
            – Type: issn-print
              Value: 1524-8372
          Numbering:
            – Type: volume
              Value: 21
            – Type: issue
              Value: 4
          Titles:
            – TitleFull: Journal of Cognition and Development
              Type: main
ResultId 1