Sources Of Developmental Changes In Interval Timing

7.5.1 Long-Term Memory Sources of Developmental Changes in Timing

Our developmental models of interval timing suggest that the increase with age in the sensitivity to time in the temporal bisection and generalization tasks is due to a decrease in the variability of the long-term memory representations of the standard durations. In short, young children have a "fuzzier" memory of durations. The issue is: Why do they have a fuzzier memory representation of time? A first working hypothesis is to consider that the standard durations have been correctly encoded and stored in long-term memory, in the form of a distribution with means equal to the standard values and some given coefficient of variation. In this case, the higher variability of time representation in memory obtained from young children could be attributed to a greater degradation over time, in other words, to a larger degree of memory decay for durations. This decay would erase memory traces of the standard durations and would increase the coefficient of variation of their representation, without alteration of their mean representation. In sum, with this sort of decay or forgetting, the long-term memory of standard durations would get fuzzier, which would flatten the psychophysical functions.

In order to test whether the degradation over time of the long-term memory representation of the standard durations is greater in younger children, we submitted children aged 5 and 8 years old to a temporal bisection task with a 2-sec short standard and an 8-sec long standard. In addition, the children were given an interfering task lasting for 15 min, either between the training and the testing phase, for

5-year-olds 8-year-olds

Stimulus duration (seconds) Stimulus duration (seconds)

FIGURE 7.8 Psychophysical functions for the 5- and 8-year-olds in a bisection task before an interfering task (immediate test) and after (deferred test). (From Rattat, A.-C. and Droit-Volet, S., Behav. Process., 55, 81-91, 2001.)

one experiment, or between two testing phases, for a second experiment (Rattat, in preparation). The results from Rattat and Droit-Volet (2001) showing the temporal bisection functions for the 5- and 8-year-olds in the testing phase before and after the interfering task (immediate vs. deferred test) are shown in Figure 7.8. In these two age groups, the psychophysical functions were flatter and the Weber ratio higher after the interfering task than before, but more particularly in the 5-year-olds. Our developmental version of the scalar timing model applied to these bisection data attributed the lower temporal sensitivity with the interfering task to an increase in the coefficient of variation for the remembered time. Furthermore, the magnitude of this increase was greater for the 5-year-olds (before, 0.65; after, 0.95) than for the 8-year-olds (before, 0.30; after, 0.50). Thus, the forgetting of the standard durations was more significant in the younger children, which produced a greater variability in the memory representation of durations. In the field of psychology, the development of long-term memory in children is still not well understood (for reviews, see Cowan, 1997; Gathercole, 1998). However, some studies suggest that the memory of events in children is influenced by the repeated experience of events and how well the mental representation of these events is organized. Indeed, a well-organized representation of events in memory allows a better recall of event memories and their preservation over a longer period of time. In this case, the variability of the memory representation of the standard durations at the outset of the temporal bisection testing phase can explain, in part, the greater effect on the bisection performance of the youngest children.

In any event, with regard to familiar events or actions for which the temporal characteristics have been reactivated sufficiently in memory to strengthen their memory traces, the observed forgetting is greatly reduced. Friedman (1990b) showed that children aged 3.6 years old are able to correctly recall, on a verbal estimation scale, the relative duration of familiar activities, such as drinking a glass of milk and viewing a TV cartoon. In a recent study, we showed that 5-year-olds are able to keep in memory the duration of a learned action for up to 6 weeks. More precisely, the 3-

and 5-year-old children were trained to produce a given action duration of 5 sec (button press) with the procedural learning methods used by Droit-Volet (1998) and Droit-Volet and Rattat (1999). In this method, the children were forced to repetitively produce a 5-sec action duration by simultaneously imitating the experimenter's action. Each correct response duration (between 4 and 6 sec) resulted in positive feedback (smiling clown), and incorrect response duration (shorter than 4 sec or longer than 6 sec) in negative feedback (frowning clown). Then, after an immediate test of the learning of the required response duration, the 3- and 5-year-olds were randomly assigned to a retention interval condition given 15 min, 1 h, 24 h, or 48 h later. The results showed that the proportion of correct responses decreased for the 3-year-olds as soon as the 15-min retention interval. However, the 5-year-olds maintained their level of performance until 48 h. Increasing the duration of the retention interval up to 6 days or even 6 weeks did not alter their performance (Rattat, in preparation). Therefore, with a repeated experience (i.e., procedural learning) the memory for duration is not or slightly affected by forgetting. Finally, the most critical process in children's abilities to time events is probably the encoding of duration.

7.5.2 Attentional Sources of Developmental Changes in Timing

Difficulties in the encoding of duration can also account for the high variability in the memory representation of the standard durations in young children. Indeed, the memory representation is the result of how it has been encoded. Among the cognitive processes involved in the encoding of duration, we have specifically investigated the role of attention. Indeed, the inefficiency of attentional processes contributes to increases in the trial-by-trial variance and, as a consequence, makes the memory of event durations fuzzier. Furthermore, it is well known that the prefrontal cortex plays a central role in attention and that its maturation is not complete until the end of adolescence. More precisely, its maturation is relatively fast until 2 years, but it continues after, with an important evolution between 4 and 7 years, followed by a more progressive evolution up to adulthood (Luria, 1961, 1973; see Pang and McAuley, this volume). Thus, the maturation of the prefrontal cortex influences the efficiency of different cognitive operations relative to attention, such as the mobilization of attentional resources in a dual task, the resistance to distraction, or the selective orientation of attention for the simultaneous processing of information.

Within a developmental framework for the mobilization of attentional resources, there is a consensus according to which the amount of attentional resources available to process a given task increases with age. Indeed, most psychologists believe that the total capacity of the pool of attentional resources remains constant throughout development, or at least after the age of 2 years. However, thanks to cognitive development, the processing demands of any particular task decrease, thus releasing attentional resources. For example, the more the processing of information becomes automatic, the less it consumes attentional resources and the more these resources are available for the processing of other information. However, psychologists do not agree on the causes of this releasing of resources through cognitive development. For Case (1985), this release of resources is caused by a better chunking of infor mation related to the development of knowledge; for Pascual-Leone and Baillargeon (1994), it is caused by the increase in automatic cognitive operations as a result of practice; and for Bjorklund and Harnishfeger (1995), it comes about from an improvement in inhibitory efficiency. Recently, others have suggested that an increase with age of the speed of general cognitive processing is involved: the faster the processing of information, the more resources are available for the processing of other information. In any event, all these psychologists agree that the amount of attentional resources increases with age during early development.

According to attentional models of time perception in human adults (e.g., Thomas and Weaver, 1975; Zakay, 1989), subjective time is directly related to the amount of attentional resources allocated to the processing of temporal information: the fewer the attentional resources that are allocated to time, the shorter the time estimate. This shortening effect is interpreted as the consequence of a loss of pulses by the flickering of the switch connecting the pacemaker to the accumulator (for reviews, see Buhusi, this volume; Fortin; this volume). The predictions of the attentional models have been extensively validated in human adults by studies using the dual-task paradigm (see Fortin, this volume). In contrast, few studies have investigated children's timing performance with this paradigm (Arlin, 1986a, 1986b; Gautier and Droit-Volet, 2002b). If the amount of attentional resources is less in younger children, then we might expect a greater interference effect of a concurrent nontemporal task on duration estimates, in terms of underestimation. This is exactly what was found by Gautier and Droit-Volet (2002b). In their study, the 5- and 8-year-olds were required to reproduce a given duration (6 or 12 sec) in both a single-and dual-task condition with a concurrent nontemporal task. The concurrent nontemporal task was to name the pictures presented in the center of the stimulus to be timed. These pictures were easy to name (low attentional demand) or difficult to name (high attentional demand). The results showed that the 5- and 8-year-olds, as well as adults, reproduced shorter durations in the dual task than in the single task (Figure 7.9). This provides support to the predictions of the attentional models and extends them to children as young as 5 years old, and even as young as 3 years, as reported by other studies (Gautier, 2002). This shortening effect was also greater in the 5-year-olds than in the 8-year-olds. Moreover, in the 5-year-olds, temporal reproductions were significantly shorter in both dual tasks (low and high attentional demands) than in the single task, whereas in the 8-year-olds, differences reached significance only between the higher attentional demand dual task and the single task. These findings indicated a greater interference effect on duration judgments in the 5-year-olds than in the 8-year-olds. The amount of available attentional resources is therefore a potential contributor to the developmental changes we have observed in the timing behavior of children.

Nevertheless, as we can easily imagine on the basis of our results, the overall explanation of the age-related changes in the development of interval timing is particularly complex. There is not one, but several sources of developmental changes. Probably related to the limited pool of attentional resources in young children, we found that children also have difficulties in resisting distraction or are susceptible to interference from nontemporal information. Using an attentional distracter in a temporal bisection task, we showed that a distracter disrupted the bisection performance

Single-task LA dual-task HA dual-task task conditions

FIGURE 7.9 Mean durations (in seconds) reproduced by the 5- and 8-year-old children for the 6- and 12-sec stimulus durations in the single task and the low (LA) and high (HA) attentional dual tasks. (From Gautier, T. and Droit-Volet, S., Behav. Process., 58, 57-66, 2002b.)

Single-task LA dual-task HA dual-task task conditions

FIGURE 7.9 Mean durations (in seconds) reproduced by the 5- and 8-year-old children for the 6- and 12-sec stimulus durations in the single task and the low (LA) and high (HA) attentional dual tasks. (From Gautier, T. and Droit-Volet, S., Behav. Process., 58, 57-66, 2002b.)

more in the 5-year-olds than in the 8-year-olds (Gautier and Droit-Volet, 2002a). In the same vein, when the subjects were instructed to process the duration of a sequence of stimuli and to ignore the varying number of stimuli in this sequence, the number of stimuli interfered more with temporal bisection performance in the 5-year-olds than in the 8-year-olds and the adults (Droit-Volet, Clément, and Fayol, in press). In contrast, the duration dimension did not interfere with the numerical bisection performance when the subjects were instructed to process the number of stimuli in the sequence while ignoring the varying duration. The younger children therefore had more difficulty in inhibiting the processing of nontemporal information (i.e., number) that altered their duration judgments and masked their fundamental competence to time events. On the other hand, it was relatively easy for them to selectively ignore time, for which the processing was more attentionally demanding (for a discussion of counting and timing in infants, see Brannon and Roitman, this volume).

0 0

Post a comment