Research on children's prospective memory has primarily concentrated on two related and equally important questions: First, what are the effects of age on prospective memory, both in terms of the earliest age at which prospective memory skills and the development of these skills in preschool and school-age children? Second, do children possess metamemory knowledge about the best strategies for various everyday prospective memory tasks, and, if yes, how effectively can they use this knowledge in their day-to-day life? We review research addressing each of these questions in turn below.
So far, only two studies have examined prospective memory in very young (preschool) children (Guajardo & Best, 2000; Somerville et al., 1983). All other published studies have concentrated on the development of prospective memory primarily in school-age children covering relatively short developmental age spans of 2 to 3 years and using a cross-sectional methodology. Some studies have compared prospective memory performance in either preschool to school-age children from 4 to 7 years (Kvavilashvili et al., 2001) or early school-age children from 5 to 7 years (Meacham & Colombo, 1980; Meacham & Dumitru, 1976). Other studies have used older school children comparing prospective memory performance in 7- and 10-year-olds (Passolunghi et al., 1995) or in 10- and 14-year-olds (Ceci et al., 1988; Ceci & Bronfenbrenner, 1985). Finally, only a few published papers have studied the development of prospective memory across wider age ranges of 4 to 5 years. For example, Nigro et al. (2002) studied children between 7 and 11 years, and Kerns (2000) studied children between 7 and 12 years old (see also Maylor et al., 2002). All these studies have reported variable results. We briefly review these findings before we discuss possible reasons for obtaining discrepant results across the studies.
Prospective Memory in 2- to 5-Year-Old Children The question about the earliest age at which prospective memory skills can be observed has so far been addressed only in a naturalistic study of Somerville et al. (1983). In this study, 2-, 3, and 4-year-old children were assigned to eight different reminding tasks by their usual caregivers (mothers) over a period of 2 weeks. These tasks varied in the level of motivation, like "Remind me to buy candy at the store tomorrow morning" (high interest) or "Remind me to bring in the wash after the nap" (low interest), and in the length of delay between receiving these instructions and the opportunity to carry them out (5-10 minutes vs. several hours).
The results that were obtained about the motivation and delay manipulation were highly interesting and novel at the time, and were later replicated in several adult studies on prospective memory. There was a highly significant effect of motivation explaining up to 25% of variance in children's prospective memory performance (cf. Kliegel, Martin, McDaniel, & Einstein, 2001, 2004; Kvavilashvili, 1987). There was also a significant effect of delay in that performance was better with short delays of several minutes than long delays of several hours. However, this effect was much smaller and explained only 5% of the variance in performance. This finding is in line with adult studies showing only a small or no reliable effects of delays on prospective memory performance (e.g., Einstein, Holland, McDaniel, & Guynn, 1992; Harris & Wilkins, 1982; Nigro & Cicogna, 2000). However, the most important results that emerged from this study concern the absence of any age effects on children's prospective memory. Thus, 2-year-olds were as good as 4-year-olds, with 80% success in remembering tasks with high interest and short delays (the success rate was still 50% with high interest and long delay intervals of several hours).
It was this remarkable finding that prompted researchers like Meacham (1982) and Winograd (1988) to suggest that prospective memory skills may develop particularly early for a child to cope successfully in everyday social contexts. Unfortunately, no attempt has been made to replicate this finding using similar age groups. The only other published study that has compared preschool children belongs to Guajardo and Best (2000), who studied 3- and 5-year-old children's prospective memory with a laboratory task using the Einstein and McDaniel (1990) paradigm depicted in Figure 6.1(a). The ongoing task was introduced to the children as a computer game in which they received six blocks of 10 pictures of familiar objects (5 seconds per picture) and at the end of the block they had to recall as many pictures as possible. The prospective memory task consisted of pressing a key on the keyboard every time they saw a picture of a house (or a duck) as part of this "computer game."
Unlike Somerville et al. (1983), Guajardo and Best (2000) obtained a significant effect of age: 5-year-olds were reliably better at remembering to press the key than the 3-year-olds. However, postexperimental probing ofthe children showed that 52% of the 3-year-olds had difficulty remembering prospective memory instructions, as they were unable to answer the question about what it was that they were asked to do when they saw the picture of a house or duck. In addition, the ongoing free recall task was undoubtedly more difficult to the 3-year-olds, who recalled significantly fewer pictures than the 5-year-olds. Given these problems with methodology, it is difficult to draw firm conclusions about the nature of prospective memory development between the ages of 3 and 5 years. However, what is remarkable in this study is that 5-year-old children performed near ceiling in this computerized prospective memory task, with 50% of the children remembering on all six occasions and mean prospective scores ranging between 5.05 and 5.58 across different conditions.
Prospective Memory in 5- to 7-Year-Old Children Discrepant findings have been obtained in studies covering late preschool and early school years. For example, in Meacham and Dumitru (1976), 7-year-old children were reliably better at remembering to post their drawing at the end of the session than 5-year-olds. However, no age effects were obtained in a study by Meacham and Colombo (1980) in which children had to remind the experimenter, at the end of the session, to open the surprise box. One possible explanation for the discrepant findings across these two studies could be that the task of opening the surprise box was more interesting or motivating than posting the drawing and that this high level of motivation eliminated age effects.
Somewhat discrepant results for 5- and 7-year-olds were also obtained by Kvavilashvili et al. (2001), who used a modified paradigm presented in Figure 6.1(b), and described earlier. Whereas 7-year-olds performed reliably better than 5-year-olds in Experiments 1 and 2, no age effect was obtained in Experiment 3 between these age groups even though broadly similar tasks and materials were used in all three studies. Interestingly, in this study no age effects were obtained between 4-and 5-year-old children but, in both experiments, 7-year-olds were reliably better than 4-year-olds.
Prospective Memory in 7- to 14-Year-Old Children Two other studies that investigated prospective memory in older school children both found reliable age effects that, however, were qualified by interactions with some other independent variables manipulated by the researchers. For example, in Passolunghi et al. (1995), 7- and 10-year-old children were tested with a standard Einstein and McDaniel (1990) paradigm where the ongoing task consisted of 40 trials of five two-syllable words presented simultaneously on the screen for 6 seconds, which children had to read as quickly and as accurately as they could. The prospective memory tasks consisted of pressing a key on the computer keyboard whenever the word boat appeared on the screen as part of the ongoing word reading task. The encoding modality of prospective memory instructions was manipulated by showing children either a picture of the boat (pictorial encoding), the written word boat (verbal encoding), or asking them to enact the prospective memory task by actually pressing the designated key on the keyboard (motoric encoding). The results showed that age effects were present only in motoric encoding conditions, but not in pictorial and verbal encoding conditions. If anything, 7-year-olds had reliably higher scores than 10-year-olds in pictorial encoding conditions in Experiment 1, and the difference between the means in Experiment 2 was in the same direction.
Furthermore, in a study by Ceci and Bronfenbrenner (1985) on 10- and 14-year-old children, the prospective memory task was remembering to take cupcakes out of the oven (or recharge the batteries) in exactly 30 minutes while being busily engaged in an ongoing task of playing a computer game in two different contexts (laboratory vs. home). Although the primary emphasis of this study was on children's time-monitoring strategies (discussed in the next section), the results concerning prospective memory performance are equally important even though they are less well known and almost never discussed in the literature. In the laboratory, prospective memory performance was at ceiling as all but one child remembered to remove the cupcakes or recharge the batteries on time (i.e., within the first 60 seconds of the critical time). An age effect was only present when children were tested in their own homes, with 10-year-olds being more likely to be late than 14-year-olds (58% vs. 25%). One possible explanation of this age by context interaction could be differences in motivation across the two contexts in younger children. Thus, it is possible that 10-year-olds took the prospective memory task less seriously in their own homes than in the anxiety-provoking laboratory situation.
Finally, as pointed out earlier, there are very few published studies that have examined the development of prospective memory across larger age ranges of 4 to 5 years, and these studies have also produced mixed results. For example, in Nigro et al. (2002), children whose ages ranged from 7 to 11 years were busily engaged in an ongoing activity of solving problems (mathematical additions and puzzles) for 15 minutes and, in addition, had to remember to remind the experimenter to do something either at a particular time (time-based task) or when seeing another experimenter (event-based condition). Although children were more likely to remind the experimenter in the event-based than in the time-based condition, there was no effect of age (F < 1). On the other hand, Kerns (2000), who tested 7- to 12-year-old children using her novel computerized Cyber Cruiser task for studying time-based prospective memory (described earlier), did report a reliable age effect in a form of negative correlation between the chronological age and prospective memory performance assessed by the number of times children ran out of gas (r = -.29; see also Martin & Kliegel, 2003; Maylor et al., 2002).
Possible Reasons for Discrepant Findings and Conclusions What can be concluded from this brief review of findings concerning the development of prospective memory in children? At first sight this diverse set of data may seem confusing and contradictory. We would argue that there are at least two major points that need to be taken into account when trying to interpret the variable pattern of findings. The first point is methodological, and concerns the importance of equating the difficulty of ongoing tasks across the age groups in the laboratory experiments. The second point concerns the size of age effects that have been reported. These issues are now discussed in more detail.
With two exceptions (Martin & Kliegel, 2003; Nigro et al., 2002), none of the published studies have made an attempt to equate the difficulty of ongoing tasks across the age groups used. It is obvious, for example, that in Passolunghi et al. (1995), reading sets of five words in 6 seconds would have been a much more difficult task for 7-year-olds than for 10-year-olds. Similarly, in Guajardo and Best
(2000), studying lists of 10 pictures was a substantially more demanding task for 3-year-olds than for 5-year-olds. Furthermore, Kvavilashvili et al. (2001) also reported that 4- and 5-year-old children took significantly longer to name 20 pictures in each of the four stacks of cards than 7-year-olds.
In most studies, however, age effects in the performance of ongoing tasks are not even reported. On the other hand, Kerns (2000) stressed that the ongoing task of playing Cyber Cruiser was equally engaging to children of various ages who took part in her study. Even if the game was equally interesting to children aged 7 to 12 years, this still does not eliminate the possibility that the game was more difficult to 7-year-olds than to 12-year-old children. Unfortunately, Kerns did not analyze children's performance on the Cyber Cruiser to see if there were any age effects on this ongoing computer task. It is interesting that Nigro et al. (2002), who covered a similar age range (7-11 years), but at the same time adjusted the level of difficulty of problems and puzzles that children were solving as part of their ongoing activity, did not report any age effects in event-based or time-based prospective memory. Moreover, when we reanalyzed the data of Kvavilashvili et al. (2001) and entered the time spent on naming the pictures as a covariate, the effects of age reported in this paper disappeared.
This issue is obviously less important for naturalistic studies as participants would be engaged in their habitual everyday (and mostly age-appropriate) tasks. For example, naturalistic studies on aging and prospective memory have consistently failed to obtain any significant age effects between young and old (Moscovitch, 1982; Rendell & Thompson, 1999; West, 1988, Study 1). Similarly, in the only existing naturalistic study conducted by Somerville et al. (1983) on 2- to 4-year-old children, no age effect was obtained. Taken together, the evidence seems to support the idea that in many cases significant age effects may be attenuated or even disappear when children of various ages are engaged in ongoing activities that are matched in their difficulty across the age groups.
On the other hand, it would be incorrect to conclude that prospective memory is largely age invariant and that adjusting task difficulty in the developmental studies of prospective memory will always eliminate the age effects. For example, in two unpublished studies we modified the method developed by Kvavilashvili et al.
(2001) so that younger children had to process a smaller number of pictures during an ongoing activity than older children (with 3-, 5-, and 7-year-olds processing 10, 15, and 20 pictures, respectively). Nevertheless, in a study by Kvavilashvili, Kornbrot, and Messer (2002, Experiment 2) a significant age effect was found in children's prospective memory so that 7-year-olds were significantly better than 5- and 3-year-olds, who did not differ from each other. In another study using an identical ongoing activity, but a different prospective memory task (i.e., instead of putting a dog figure into a box, children had to remember to say something to the toy mole when seeing a particular picture), Kvavilashvili, Messer, and Kyle (2002) also found a significant age effect. Thus, 3-year-olds were significantly worse than 5-year-olds, who did not differ from 7-year-olds, who, in turn, did not differ from 9-year-olds.
Given that age effects can be obtained even when the length and the difficulty of ongoing activities have been controlled for, an important issue that needs to be examined is the size of these age effects. Unfortunately, very few studies have reported effect sizes and often insufficient information is provided to calculate the effect sizes in these studies. However, the examination of existing studies and available data shows that effect sizes are relatively modest, especially in comparison to often dramatic developmental changes in a variety of retrospective memory tasks covering the same age range (e.g., Gathercole, 1998; Schneider & Pressley, 1997).
For example, in two experiments reported by Kvavilashvili et al. (2001), age explained a relatively small percentage of variance in 4-, 5-, and 7-year-old children's prospective memory performance (with = .08 and .07, respectively). Moreover, in Experiment 3, the effect size was twice as large for a free recall task (^2 = .15) than for the prospective memory task (^2 = .07). Similarly, in the published study that has used the largest age range (7-12 years), Kerns (2000) reported a relatively small negative correlation between chronological age and the performance on the prospective memory task embedded in the Cyber Cruiser game (r = -.29), indicating that age explained only 8% of the variance in children's prospective memory performance.
Interestingly, when the length or the difficulty of ongoing tasks is controlled, the effect sizes can become even smaller. For example, although Kvavilashvili, Kornbrot, and Messer (2002) did find an age effect in 3-, 5-, and 7-year-old children, as described earlier, this effect explained only 3% of variance in children's prospective memory of remembering to put a dog figure in the box. In contrast, very large age effects were obtained in the same study on children's performance on standard retrospective memory tasks such as digit span (^2 = .45), picture recognition (^2 = .44), and immediate and delayed free recall (^2 = .20 and = .16, respectively; for the latter two tasks only the data of 5- and 7-year-olds were available).
Similar results were also obtained by Kvavilashvili, Messer, and Kyle (2002) in a study conducted on 3-, 5-, 7-, and 9-year-old children in which the difficulty of an ongoing task was controlled and the prospective memory task involved a verbal response instead of an overt action of putting a dog figure in the box. Initial analyses showed a large effect size (^ 2 = .16). However, this turned out to be entirely due to 3-year-olds' difficulty in remembering this verbal prospective memory task (saying something to the mole when seeing a particular picture). Indeed, when 3-year-olds were excluded from the analyses the effect of age explained only 3% of the variance in 5-, 7-, and 9-year-olds' prospective memory performance. Incidentally, the data of 3-year-olds in these two experiments seem to provide some support for the idea that "prospective memory may be superior for intentions requiring motor response than for those requiring verbal ... response" (Freeman & Ellis, 2003, p. 990). This is clearly an issue that merits further investigation in adults and especially in young children.
One final example of dramatic changes that may occur in effect sizes due to experimental manipulations was recently reported by McGann, Defeyter, Ellis, and Reid (2005). They conducted two experiments on 4-, 5-, and 7-year-old children using a modified paradigm presented in Figure 6.1(b). In Experiment 1, children had to name each of the four blocks of 20 pictures presented on the computer screen (the ongoing task), with each block being preceded by drawing a picture for Rosie the rag doll. The prospective memory task involved remembering to press a key on the computer keyboard every time they saw a food picture that Rosie "could collect for her picnic." In Experiment 2, children had to name and manually sort the four stacks of 20 cards into categories. Moreover, the length of the ongoing task was controlled by allowing each child to engage in this task for only 1 minute. Finally, the prospective memory task consisted of taking a picture with a food item and putting it into Rosie's lunch box. Significant age effects were obtained in both experiments. However, whereas age explained 20% of the variance in Experiment 1, it explained only 7% of the variance in Experiment 2, in which children were engaged in a more meaningful prospective memory task and in which the length and possibly the difficulty of the ongoing card naming and sorting task were controlled. In addition, in Experiment 2, there was an interesting age by target salience interaction so that the effect of age was present only when prospective memory targets were the same size as the nontarget pictures. When prospective memory targets were slightly larger in size than most of the nontarget pictures, there were no age effects in 4-, 5- and 7-year-old children.
Taken together, the existing evidence appears to suggest that although prospective memory does develop with age, the developmental changes are modest at best and can be reduced even further by testing children with meaningful and interesting prospective memory tasks (e.g., McGann et al., 2005) or by adjusting the difficulty of ongoing activities (Kvavilashvili, Kornbrot, & Messer, 2002). Is this relatively good ability of remembering to carry out prospective memory tasks accompanied by equally good metamemory for processes and strategies involved in successful prospective remembering?
Effects of Age on Metamemory for Prospective Memory Tasks in Children
As pointed out earlier, this question consists of two related issues. The first concerns children's knowledge of processes and strategies that can enhance performance in everyday prospective memory tasks. This issue has been examined by Kreutzer et al. (1975) and Beal (1985) by using an interview and questionnaire method. The second issue concerns children's actual ability to utilize these strategies in prospective memory tasks that they have to carry out in everyday life. This question has so far been addressed in a study by Ceci and Bronfenbrenner (1985; see also Ceci et al., 1985) and Kerns (2000), who studied children's monitoring behavior in time-based tasks.
Children's Knowledge of Strategies for Prospective Memory
Tasks Kreutzer et al. (1975) conducted the first and seminal study on children's strategic knowledge of several everyday (primarily retrospective) memory tasks such as remembering where one could have left one's jacket at school, remembering Christmas when a particular present was given, or how to memorize a categorical set of nine pictures (three pictures from three different categories). The most famous and often cited question concerned children's strategic knowledge of a typical prospective memory task. Specifically, children were asked to list every possible strategy they could think of to ensure that they would remember to take their skates to school the next morning. Four age groups were tested: kindergarten (4-5 years), first grade (6-7 years), third grade (8-9 years), and fifth grade (10-11 years)
children. Children's answers to the skates question fell into four categories, three of which referred to external strategies and one to an internal strategy (i.e., periodic rehearsal of the task in one's mind). The external strategies involved the physical manipulation of skates (e.g., putting them near the door), the use of external reminder cues other than the skates (e.g., writing a note), or soliciting help from others (e.g., asking a parent to provide a reminder).
The results showed that there were no marked age effects in the tendency to list one versus another of these four strategies. Even the kindergarteners were able to come up with at least one strategy each. There was also a clear preference for external strategies, as only 16% of the children suggested using the internal rehearsal strategy. It is interesting that in a naturalistic study, when college undergraduates had to remember to send postcards to the experimenter on prearranged dates, a similar small percentage of students (i.e., 20%) reported having actually used the internal strategy of rehearsing the task in their mind (Meacham & Singer, 1977). Overall, however, older children (the third and fifth graders) listed more strategies than younger children (kindergarteners and first graders) and the strategies that they described were more explicitly planful and means-ends-oriented than those reported by younger children. Similar results were obtained for another prospective item from the Kreutzer et al. (1975) questionnaire that asked children what they needed to do to ensure that they would not forget an upcoming event (e.g., a friend's birthday). Here again, even the youngest children could come up with a strategy or two, with more and increasingly planful strategies reported by older children.
In contrast, marked age effects were obtained in the same study with several retrospective items, such as how to remember an event from a previous Christmas or how to memorize a categorical list of nine pictures. For example, in relation to the Christmas question, 5-year-olds could hardly understand the task, whereas 7- and 9-year-olds said they would solicit help from adults. Only 11-year-olds produced more varied strategies, but even with this age group there was plenty of scope for further improvement. This contrasting pattern of findings concerning prospective and retrospective items was subsequently replicated in several other studies using similar questions (e.g., Cavanaugh & Borkowski, 1980; Kurtz & Borkowski, 1984; see also Farrant, Boucher, & Blades, 1999, for using the prospective questions in children with autism).
Another well-known study on children's metamemory of prospective memory tasks, using a similar interview method, was conducted by Beal (1985, Study 1). Unlike Kreutzer et al. (1975), Beal tested children's knowledge of the effectiveness of different types of cues in everyday prospective memory tasks. Moreover, in addition to children, she also tested a group of college undergraduates. Children (5-, 6-, and 8-year-olds) and young adults were given the descriptions of six different prospective memory task scenarios together with two alternative reminder cues that the protagonist could use to help him or her successfully remember the task (e.g., remembering to take out trash or calling a friend after school). Participants had to choose the effective reminder out of two and provide justification for their choice. The three scenarios concerned the cue informativeness (e.g., that the cue should be nonambiguous or sufficiently detailed to act as an effective reminder), and the other three, the cue placement (e.g., that the cue should be easily noticeable or that it should be encountered at the right time).
The results showed that there were no statistically reliable differences in the number of correct responses in 5-, 6-, and 8-year-old children. However, whereas 5- and 6-year-olds were significantly less accurate than adults, 8-year-olds were as good as adults in four out of six target scenarios. It is also important to note that, in comparison to adults and 8-year-olds, young children were somewhat disadvantaged by having to provide verbal justification for their choices. In addition, both 8-year-olds and especially adults performed at ceiling in several of the six target scenarios. Despite these difficulties in interpreting the results, on the whole, the results seem to be in line with the findings of Kreutzer et al. (1975) and indicate that young children, and especially 8-year-olds, may have a fairly good understanding of the basic nature and functions of reminders in prospective memory tasks.
One problem that these studies share is that they assess children's metamemory knowledge of memory situations (i.e., declarative metamemory) rather than their actual strategic behavior in everyday prospective memory tasks (i.e., procedural metamemory). There is evidence in the literature showing that although children may have an adequate knowledge of a strategy suitable for a particular retrospective memory task, they might not use it in an actual memory test situation (e.g., Fabricius & Wellman, 1983; Schneider & Pressley, 1997). None of the developmental studies have examined children's spontaneous use of external strategies in prospective memory tasks such as remembering to take skates to school. This is an interesting topic that awaits future investigation. There are, however, three published studies that have examined children's strategic monitoring behavior in time-based prospective memory tasks.
Children's Use of Strategies in Prospective Memory Tasks In their cupcake and battery recharging study, Ceci and Bronfenbrenner (1985) were primarily interested in 10- and 14-year-old children's strategic clock-monitoring behavior during the 30-minute delay interval filled with playing a computer game. An earlier study by Harris and Wilkins (1982) had shown that when young adults had to remember to carry out a time-based task at 3- or 9-minute intervals while watching a film, their clock checking prior to each critical time resembled the J-shaped pattern: Participants checked the clock initially a few times, then the clock checking dropped for some time until it dramatically increased in the period immediately preceding the critical time. Ceci and Bronfenbrenner wanted to see if 10-and 14-year-olds would also engage in this strategic clock monitoring displayed by adults and whether this behavior would vary as a function of context. Thus, an interesting aspect of this study was that half of the children were tested in the laboratory by a trained psychology undergraduate who was unknown to them and the other half were tested at home by their older siblings (also psychology undergraduates).
The results showed entirely different clock-monitoring patterns in these two contexts. In the more anxiety-provoking environment of the psychology laboratory, in which the prospective memory task was probably perceived as quite important, the number of clock checks linearly increased, with the highest number being made in the last 5 minutes of the delay interval. Although this strategy was not most parsimonious, given the large number of clock checks that children had to make, it paid off in that all but one child remembered to take out the cupcakes (or recharge the batteries) on time. In contrast, in the more relaxing and familiar environment (i.e., at home), those children who remembered to take out the cupcakes on time demonstrated strategic monitoring that resembled the U-shaped pattern. Thus, children checked the clock quite frequently in the first 10 minutes of the delay interval as if trying to synchronize or calibrate their internal clock with the elapsed time shown by the external clock. After this, the number of clock checks dropped markedly for some time until it again sharply increased in the last 5 minutes of the delay period. Interestingly, there were no marked age effects in clock monitoring. Although overall younger children made more clock checks than older children, the pattern of clock checks was similar across age groups in both contexts.
These findings are remarkable for two reasons. First, they show that children can use different clock monitoring strategies as a function of context. In the laboratory, where the consequences of forgetting the prospective memory task were probably perceived as less acceptable, children chose to use the less parsimonious but safer strategy of linearly increasing monitoring. At home, however, where children were more relaxed and probably deemed the forgetting of the prospective memory task as more acceptable, they chose to use a completely different U-shaped pattern of monitoring. Second, this U-shaped pattern of monitoring indicates that 10- and 14-year-old children, if necessary, can engage in fairly complex strategic behavior that involves temporal calibration of internal clocks at the beginning of the delay interval. As a result of this calibration, the overall number of clock checks is substantially reduced, allowing children to deploy their attentional resources elsewhere (in this case on playing the computer game).
To study this temporal calibration strategy in more detail, Ceci et al. (1988) conducted a follow-up study in which 10-year-old children had to carry out the same prospective memory task at home as before, but the speed of clocks was manipulated (accelerated or decelerated by 10%, 33%, or 50%). The results showed that 10-year-olds managed to successfully use the temporal calibration strategy (reflected in the U-shaped pattern of monitoring) in conditions in which the time on the external clocks was accelerated or decelerated by 10% and 33%. However, when the time was accelerated or decelerated by as much as 50%, children chose a linearly increasing pattern of monitoring instead, as if realizing that they could no longer trust their internal estimation of time that did not seem to match the one shown by the external clock.
The findings of Ceci and his colleagues were recently replicated by Kerns (2000) in 7- to 12-year-old children who, in the course of the computer game Cyber Cruiser, had to periodically check the levels of a gas tank and refuel it whenever it reached a certain critical level. The gas tank ran out of the fuel five times in the course of this computer game (approximately once every 60 seconds). The findings showed that there was no age effect in the pattern of strategic monitoring. All children, irrespective of age, displayed the J-shaped pattern of monitoring originally reported by Harris and Wilkins (1982) on young adult participants. Mantyla and Carelli (2005) also reported the J-shaped pattern when they studied time monitoring and time estimation across the life span in children (8- to 12-year-olds), young adults, and older adults. Moreover, children were as good as young adults at a time estimation task in which they had to reproduce short time intervals of 4 to 32 seconds.2
Conclusions Taken together, the results of Ceci and colleagues and Kerns (2000) show that young children can use fairly complex monitoring strategies in time-based prospective memory tasks. However, an intriguing aspect of these findings is that children may be using these strategies fairly automatically without much conscious knowledge of what they are doing. For example, when Ceci and Bronfenbrenner (1985) probed their participants at the end of the session, children seemed to be unaware or unable to verbally formulate the temporal calibration strategy that they were exhibiting in their behavior (i.e., they seemed to be unaware of the fact that they checked the clock more frequently at the beginning than in the middle of the delay period). Ceci and Bronfenbrenner (1985) argued that if the temporal calibration strategy is indeed deployed automatically, then "this would help explain why young children appear to be adept at its use, as automatic processing has been shown to be age-invariant" (p. 162).3
Was this article helpful?
It seems like you hear it all the time from nearly every one you know I'm SO stressed out!? Pressures abound in this world today. Those pressures cause stress and anxiety, and often we are ill-equipped to deal with those stressors that trigger anxiety and other feelings that can make us sick. Literally, sick.