Scholarly article on topic 'Early declarative memory predicts productive language: A longitudinal study of deferred imitation and communication at 9 and 16months'

Early declarative memory predicts productive language: A longitudinal study of deferred imitation and communication at 9 and 16months Academic research paper on "Psychology"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Infant development" / "Deferred imitation" / "Declarative memory" / "Productive Language" / "Receptive Language" / "Gestural Communication"}

Abstract of research paper on Psychology, author of scientific article — Annette Sundqvist, Emelie Nordqvist, Felix-Sebastian Koch, Mikael Heimann

Abstract Deferred imitation (DI) may be regarded as an early declarative-like memory ability shaping the infant’s ability to learn about novelties and regularities of the surrounding world. In the current longitudinal study, infants were assessed at 9 and 16months. DI was assessed using five novel objects. Each infant’s communicative development was measured by parental questionnaires. The results indicate stability in DI performance and early communicative development between 9 and 16months. The early achievers at 9months were still advanced at 16months. Results also identified a predictive relationship between the infant’s gestural development at 9months and the infant’s productive and receptive language at 16months. Moreover, the results show that declarative memory, measured with DI, and gestural communication at 9months independently predict productive language at 16months. These findings suggest a connection between the ability to form non-linguistic and linguistic mental representations. These results indicate that the child’s DI ability when predominantly preverbal might be regarded as an early domain-general declarative memory ability underlying early productive language development.

Academic research paper on topic "Early declarative memory predicts productive language: A longitudinal study of deferred imitation and communication at 9 and 16months"

Early declarative memory predicts productive language: A longitudinal study of deferred imitation and communication at 9 and 16 months

Annette Sundqvist *, Emelie Nordqvist, Felix-Sebastian Koch, Mikael Heimann

The Infant and Child Lab, Division of Psychology, Department of Behavioral Sciences and Learning, Linkoping University, SE 581 83 Linkoping, Sweden

ARTICLE INFO ABSTRACT

Deferred imitation (DI) may be regarded as an early declarative-like memory ability shaping the infant's ability to learn about novelties and regularities of the surrounding world. In the current longitudinal study, infants were assessed at 9 and 16 months. DI was assessed using five novel objects. Each infant's communicative development was measured by parental questionnaires. The results indicate stability in DI performance and early communicative development between 9 and 16 months. The early achievers at 9 months were still advanced at 16 months. Results also identified a predictive relationship between the infant's gestural development at 9 months and the infant's productive and receptive language at 16 months. Moreover, the results show that declarative memory, measured with DI, and gestural communication at 9 months independently predict productive language at 16 months. These findings suggest a connection between the ability to form non-linguistic and linguistic mental representations. These results indicate that the child's DI ability when predominantly preverbal might be regarded as an early domain-general declarative memory ability underlying early productive language development.

© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecom-

mons.org/licenses/by-nc-nd/4.0/).

Article history: Available online xxxx

Keywords: Infant development Deferred imitation Declarative memory Productive Language Receptive Language Gestural Communication

* Corresponding author. E-mail address: anett.sundqvist@liu.se (A. Sundqvist).

http://dx.doi.org/10.1016/jjecp.2016.01.015 0022-0965/© 2016 The Authors. Published by Elsevier Inc.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

An infant's ability to imitate actions and communicative behavior (gestures and words) when responding to individual interactions is an important aspect of cognitive development (Meltzoff, Williamson, & Marshall, 2013). During the infant's preverbal development, imitation is a primary path to learning about the world and the actions of others (e.g., Meltzoff, 2005; Nadel, 2014; Ramachandran, 2011). In addition, a critical developmental step is the ability to observe new actions and, later, to retrieve the previously observed action from memory and perform the action (Piaget, 1962). This ability, known as deferred imitation (DI), is commonly described as reflecting early declarative or declarative-like memory (Jones & Herbert, 2006).

As the infant moves from the preverbal stage to the verbal stage of language development, simultaneous changes occur in encoding, consolidation, and storage of declarative memory (Bauer, 2006; Mullally & Maguire, 2014; Rovee-Collier & Cuevas, 2009). Previous studies have found that DI performance during the first year predicts DI performance during the first months of the second year of life (Heimann & Meltzoff, 1996; Heimann et al., 2006; Strid, Tjus, Smith, Meltzoff, & Heimann, 2006). Furthermore, this research showed that the DI measured at 9 months also predicts gestural communication and was positively related to receptive language measured 5 months later. However, no study to date has reported a link between early DI and productive language during the second year of life.

Early memory development

Past research suggests that DI is a reliable pathway for assessing the preverbal infant's developing declarative memory (Heimann & Meltzoff, 1996). DI is an active and goal-directed activity (Marshall & Meltzoff, 2014) that indicates the infant's memory representation of the action to be reproduced.

Research during the last decade has demonstrated evidence of DI well before a child's first birthday (e.g., Barr, Dowden, & Hayne, 1996; Bauer, Wiebe, Carver, Waters, & C. Nelson, 2003; Collie & Hayne, 1999; Heimann & Nilheim, 2004; Meltzoff, 1988a, 1988b; Rovee-Collier & Giles, 2010). A few previous studies have investigated the developmental trajectory of declarative memory measured by DI (Heimann & Meltzoff, 1996; Heimann et al., 2006; Kolling, Goertz, Frahsek, & Knopf, 2009; Strid et al., 2006). Heimann and Meltzoff (1996) found that the variation in observed DI was stable between 9 and 14 months. That is, the children who performed less DI at 9 months were also at the lower end at 14 months. Similarly, the children who displayed more DI at 9 months were to a large extent the same children who performed more DI at 14 months.

It is possible that the ability to reliably make mental representations of novel acts presented through the DI tasks is connected to the ability to make mental representations of novel words during the initial lexical development given that both tasks are reliant on the ability to represent actions and events of the world mentally. This would suggest that a domain-general declarative memory ability acts as a facilitator for creating mental representations of actions and words (Heyes, 2012).

Early language development

The infant's lexical and communicative abilities develop remarkably quickly during the transition from the preverbal period to the verbal period. The preverbal infant's communication is characterized by the use of sounds, body language, and gestures (Bishop, 1997). For instance, a 9-month-old will typically understand simple phrases that represent actions such as ''come here" and ''clap your hands" (Eriksson & Berglund, 1999); this understanding is greatly aided by the body language and gestural communication of the adult (Fenson et al., 1994). The productive communication of a 9-month-old consists primarily of communicative gestures (Bloom, 1993). For example, infants at this age will play peekaboo or extend their arms upward when they want to be picked up (Eriksson & Berglund, 1999; Fenson et al., 1994). The productive language use at this stage is limited to attempts to imitate expressions, sounds, and words produced by the caregivers (K. E. Nelson, Heimann, Abuelhaija, & Wroblewski, 1989).

By 16 months, the infant moves into the verbal stage of development and begins to verbally label objects, people, and actions in his or her proximity (Bloom, 1993; Eriksson & Berglund, 1999; Fenson et al., 1994). Infants at this age actively use approximately 20 words and understand approximately 170 words (Eriksson & Berglund, 1999), but this number rapidly increases with time. Many of the first words are so-called holophrases, mentally representing actions or events that the children wish to communicate about (Tomasello, 2003). The holophrases are typically mental representations of actions or routines that are familiar to the children (K. Nelson, 1996). For example, children may express existence versus non-existence of people and objects (Tomasello, 2003). An example of this would be when a child produces "Daddy"; it could mean ''We said good-bye to Daddy earlier" or, depending on the situation, ''Daddy is coming soon." This production would entail retrieving a previously formed mental representation to the present and expressing the related word (Tomasello, 2003).

Imitation and DI have long been hypothesized to be the foundation of the initial and early lexical development (e.g., K. E. Nelson et al., 1989; Snow, 1981). As described by Meltzoff and Gopnik (1989), ''language depends on deferred imitation" (p. 38). An example of this could be when parents offer an infant water to drink; they will help the child to mentally connect the word ''water" with the object ''sip mug" and with the child's own perceived ''thirstiness." Subsequently, when the child later feels ''thirsty," he or she may eventually be able to deferringly imitate by retrieving the formed mental representation of the word ''water" to receive the sip mug. As the child's oral motor function increases, the child will be able to more closely approximate the adult pronunciation of ''water." The child will implicitly self-monitor his or her own production of the word while comparing it with the stored mental representation of the word. It is evident that this process is supported by the child's ability to remember. Other aspects are also relevant, including a child's interest in social relations (Bishop, 1997).

Declarative memory and language

As the infant's cognitive capacities mature, there is a need to represent the world and the novelties and regularities experienced and, by doing so, linking the past to the present (Bruner, 1964). This is the advent of language and memory. One way to explore the relationship between the two is to examine the association between the early memory capacity, measured with deferred imitation, and different aspects of the infant's developing language.

Because many of the infant's first words often are holophrases expressing mental representations of previously experienced actions or events, there is a probable connection to the retrieval of previously stored mental representations of actions expressed through deferred imitation tasks. A mental representation in language can be characterized as a social event organized in time and space (K. Nelson, 1996). Thus, an infant who is able to form stable and easily retrieved mental representations of events and actions may be better equipped to attach an expression—a word—to these mental representations as the infant's cognitive and oral motor skills develop during the second year.

DI also makes a significant contribution to the older child's language development and foremost lexical development. Kuczaj (1987), for example, taught 2-year-olds novel nonwords in a structured play session. The children did not engage in overt spontaneous imitation to any great degree during the learning phase, but at the test phase the children produced several of the nonwords that had been taught, demonstrating an active use of the memory capacity tapped by DI in producing the nonwords. McEwen and colleagues (2007) further demonstrated the importance of imitation to children's language ability at 2 years of age. Through the use of both parental questionnaires and parents as test leaders, researchers identified a moderate correlation between direct imitation and lexical development.

It is likely that lexical development before 2 years of age is in part reliant on the child's ability to defer imitation—the ability to form a mental representation of the act observed and store it in long-term memory. Heimann and colleagues (2006) showed that an infant's early mental representations of communicative gestural actions were predicted by the child's ability to form mental representations of novel actions in the DI task. Thus, DI may tap cognitive processes essential for language learning, at least with regard to nonverbal communication. In contrast, a parallel analysis of DI at 9 months did not reveal any correlation with early lexical development at 14 months.

There are at least two different ways of conceptualizing the impact of deferred imitation on language development. One way is to view declarative memory as a domain-general ability underlying several abilities across different modalities such as remembering language or actions with objects (Heyes, 2012). Another way is based on the idea that declarative memory develops in a domain-specific manner (Fodor, 1992; Leslie, 1994; Subiaul, Patterson, Schilder, Renner, & Barr, 2014). Visual and verbal imitation would reflect different underlying systems not depending on each other. If both the ability to remember actions with objects and to produce words are reliant on a common early memory factor reflected by DI, these two abilities should be closely related. Recent studies have also shown that retention rates seem to be parallel for visual and auditory modalities (Visscher, Kaplan, Kahana, & Sekuler, 2007). This supports the view of a more general learning capacity rather than one that is domain specific.

Aim and hypotheses

This study investigated early memory performance, measured by deferred imitation, and language development in infants from 9 to 16 months. Specifically, two hypotheses were formulated. First, we expected stability in DI development from 9 to 16 months. Second, we expected that DI at 9 months also would predict language development (gestural, receptive, and productive measures) at 16 months.

Method

Participants

The study included a total of 42 infant participants (22 boys and 20 girls) recruited through open day-care units for parents and their children, baby clinics, and word of mouth. Mean age at the first testing session (T1) was 9.55 months (SD = 0.50, range = 8.12-11.34, N = 42) and at the second testing session (T2) was 15.96 months (SD = 1.86, range = 13.18-20.71, N = 36). Six infants were excluded from the follow-up session due to technical errors (n = 4) or failure to contact the families to schedule a second visit (n = 2). Group total mean gestational age was 40 weeks (range = 36-42), mean birth weight was 3547 g (SD = 612), and mean Apgar score was 9.33 (SD = 1.00). No participant had known medical or developmental problems.

Two groups of infants were included as control groups for the deferred imitation procedure: (a) 19 infants (10 girls and 9 boys) constituting the controls for the first observation (M = 9.72 months, SD = 0.37) and (b) 8 infants (3 girls and 5 boys) constituting the controls for the follow-up session (M =15.43 months, SD = 0.62, range = 14.59-16.14).

General procedure

Participants in the study made two visits to the Infant and Child Lab at Linkoping University accompanied by at least one parent. Each visit lasted approximately 1 h (M = 55.0 min, SD = 6.55). Children sat in a parent's lap during testing; parents were instructed not to interact with their children or comment on anything done during the session. All visits were video-monitored for later coding.

Although several measures were administered over the course of each visit, the current study relies on results from deferred imitation and language development testing. Because DI was the main focus of the study, it always initiated the visit, immediately following the initial warming-up toy play. Other measures focused on parent-child interaction, communication, mind-mindedness, and novelty preference.

Written informed consent was obtained from all caregivers prior to the first session. The regional ethical review board for Linkoping, Sweden, approved this study.

Materials and procedures Deferred imitation

The study relied on an observation-only design (Meltzoff, 1988a). The procedure included a set of three age-appropriate actions-on-objects tasks presented to participants one at a time during a period of 20 s. Infants were not allowed to handle the objects during or after the presentation until the response session began. The mean delay between presentation of the objects and response was 26.75 min (SD = 6.13) for the session at 9 months and 25.87 min (SD = 7.61) for the follow-up session at 16 months. After the delay, participants were presented with the objects one at a time in the same order as they were originally presented. The response period was 20 s from infants' first touch of each object. No verbal instructions specific to the actions presented preceded the response session; the experimenter spoke only to direct infants' attention as needed (e.g., saying ''Look here"). The set of objects used differed between T1 and T2 (Meltzoff, 1988b).

Objects used at 9 months (T1). Three objects were used in a counter-balanced order. The first object was a rectangular black box with a hidden button on the front. The target action was to press the button to make a ringing sound. The second object was an L-shaped wooden object consisting of two wooden blocks that were held together at a 90-degree angle by a hinge, and the target action was to fold them together. The third object was an egg that made a rattling sound if shaken; the target action was to shake the egg three times bidirectionally.

Objects used at 16 months (T2). Three objects were used in a counter-balanced order. The first object was a telescope-shaped cup that could be collapsed flat when pressed downward. The extended cup was presented, and the target action was to press down on top of it with an open palm until it collapsed completely. The second object was a plastic cup and a short string of beads placed next to each other. The target action was to take the string of beads and place it in the cup; no more than a third of the string should hang over the edge of the plastic cup. Finally, the same egg as described for 9 months was used and also the target action was the same—to shake the egg three times bidirectionally.

For deferred imitation, there were two age-matched comparison groups (9 months [n = 19] and 16 months [n = 8]) that were presented with the same objects as described above but without demonstration of the target actions. These infants' spontaneous actions with the objects were observed and recorded.

The Swedish Early Communicative Development Inventories (SECDI) is a Swedish version of the MacArthur Communicative Development Inventories (CDI; Eriksson & Berglund, 1999; Fenson et al., 1994). The CDI is a well-established parental questionnaire providing a reliable estimate of infants' communicative skills. For the current study, the Swedish standardization was used. The Words and Gestures version of the SECDI, aimed at children at 8 to 16 months, was administered to parents at both visits. The inventory documents infants' understanding and production of words and sentences (maximum = 385 words) along with production of communicative and symbolic gestures (maximum = 62). Parents were asked to fill out the SECDI within a week of the visit to the lab and to return the questionnaire promptly in the supplied envelope. The rating scales are the frequency counts of each of the sections described above.

Data analysis Deferred imitation

A score of 1 or 0 was used to indicate whether the child performed the target action described above or not, yielding a range of scores of 0 to 3 for the three DI actions on objects.

All actions were scored by both an experimenter and an independent research assistant blind to the purpose of the study. The overall Cohen's kappa for deferred imitation at 9 and 16 months was k > .90.

A score for each of the parts of the questionnaire was calculated by adding all of the checked-in boxes representing words or actions that the child understands, says, or does. This yielded one total summary score for the number of words the child understands, one score representing the number of words the child says, and one score representing the number of gestures/actions the child uses for communication at that particular age. For this study, the Swedish norms were used (Eriksson & Berglund, 1999).

Statistical analyses

SPSS (Statistical Package for Social Sciences 22.0) was used for all statistical analyses. Two-tailed analyses are used throughout except for the comparison between the experimental and control groups, where clear expectations from previous research existed. For DI, both the overall mean score and the proportion score are presented. For comparisons between measures and groups at different time points, parametric correlations are used. An analysis of linear regression used measures of productive language at 16 months as a dependent variable, and independent variables were DI and gestural communication at T1 as well as gender and age at T2.

Results

Deferred imitation At 9 months

The mean score on DI for the experimental group on DI was 1.45 (SD = 0.92), with a mean proportion score of .53 (SD = .32) (see Table 1). Participants in the experimental group performed a higher proportion of target actions than those in the control group (M = .39, SD = .28, p = .047, d = 0.47). No gender differences were found (p > .10).

At 16 months

The mean score on DI for the experimental group was 2.44 (SD = 0.77), as shown in Table 1. Their mean proportion score was .80 (SD = .27), a significantly higher proportion than that of the control group (M = .46, SD = .18, p < .001, d = 1.48). No gender differences were found (p > .10).

Correlations

There was a significant correlation between performance on DI at 9 and 16 months, indicating stability in the pattern of the ability to remember actions-on-objects. The correlation of the proportional scores between DI at 9 and 16 months was r(35) = .391, p = .01.

Children improved significantly from 9 to 16 months on all of the subscales of the SECDI (see Table 1). The results of the children's communicative development at 9 and 16 months were within standard norms for a Swedish sample (Eriksson & Berglund, 1999). No gender differences were found

Table 1

Descriptive data of deferred imitation and SECDI at 9 and 16 months.

9 months 16 months

Mean SD Mean SD pa

Deferred imitation 1.45 0.92 2.44 0.77 <.001

Gestural communication 11.35 7.04 43.36 8.97 <.001

Productive language 1.23 2.32 17.63 22.33 <.001

Receptive language 19.05 24.59 131.57 73.57 <.001

a Paired samples t-test.

Please cite this article in press as: Sundqvist, A., et al. Early declarative memory predicts productive language: A

longitudinal study of deferred imitation and communication at 9 and 16 months. Journal of Experimental Child

Psychology (2016), http://dx.doi.org/10.1016/j.jecp.2016.01.015

(all ps > .10). The concurrent correlations for the subscales of the SECDI are shown in Table 2. The correlations at 9 months are indicated in the upper right-hand corner. At 9 months, gestural communication correlates with productive and receptive language (p < .01). The correlations at 16 months are shown in the lower left-hand corner of Table 2. All of the SECDI measures correlate (p < .01). DI did not concurrently correlate with measures at 9 or 16 months.

Correlations between subscales of SECDI: Longitudinal relations

As shown in Table 3, receptive language at 9 months was correlated with receptive language at 16 months. Gestures at 9 months correlated with all of the SECDI subscales at 16 months (receptive language, productive language, and gestures). As expected, there was no correlation for the production of words between the preverbal 9-month-olds and the verbal 16-month-olds. All of these correlations suggest that there is stability in the communicative development between 9 and 16 months.

Predictive relationship between memory and language: Longitudinal relations Correlations

There was a significant positive correlation between early declarative memory as measured by DI at 9 months and productive language at 16 months, r(34) = .35, p = .04. No other significant correlations between the SECDI subscales and DI, at any age, were found (see Tables 2 and 3).

Regression analysis

A linear regression analysis was used to analyze the variables with a significant correlation to productive language (see Table 4). To account for the variance in age at T2 and gender, these variables were also included in the model (see Table 3). A significant regression equation was found, F(4, 27) = 4.8, p = .005, with R2 = .41 (adjusted R2 = .33), with both gestural communication at 9 months (b = 1.50, p = .008) and DI at 9 months (b = 8.34, p = .02) as significant predictors to productive language at 16 months. Age at T2 (b = 2.86, ns) and gender (b = 7.13) were not significant.

Discussion

The first hypothesis of the current study was confirmed; there was a significant degree of stability in early declarative memory performance as measured by DI. There was also stability in the development of language in infants from 9 to 16 months. The second hypothesis, which investigated the predictive relationship between DI and language measures, was partly confirmed. Declarative memory, measured with DI, in the preverbal period (9 months) predicted productive language in the early verbal period (16 months).

A significant correlation between DI at 9 and 16 months was observed, consistent with findings of early stability in DI between 9 and 14 months (Heimann & Meltzoff, 1996). The observation-only procedure for DI used in the current study ensured that infants used a mental representation when imitating the target action. Thus, infants' ability to form a stable mental representation, store the observed event in long-term memory, and retrieve the representation when reproducing the act later increases consistently through the transition to the verbal stage of development.

Table 2

Concurrent correlations of SECDI and deferred imitation at 16 months (lower left) and 9 months (upper right).

Receptive language Productive language Gestural communication Deferred imitation

Receptive language - .12 .42** .19

Productive language .51** - .49** .08

Gestural communication .67** .46** - .12

Deferred imitation .21 .16 .008 -

** p <.01.

Table 3

Correlations between SECDI subscales and deferred imitation at 9 and 16 months.

16 months

Receptive Language Productive Language Gestural Communication DI

9 months Receptive language .55

Productive language .15

Gestural communication .50*

Deferred imitation .26

.11 .32 .46** .35*

.34 .25 .64* .02

.17 .14 .06 .37*

p < .05. p <.01.

Table 4

Predictors of productive language at 16 months.

Variable

95% CI

Constant

Gestural communication at T1

Deferred imitation at T1

Age at T2

Gender

-65.71

.41 (adjusted R2 = .33) 4.76**

-132 to 0.58 0.42 to 2.56 -1.43 to 15.24 -1.04 to 6.75 -6.83 to 21.1

Note. CI, confidence interval.

* p < .05.

** p <.01.

The longitudinal correlation found in this study for gestural communication and receptive language suggests individual stability over time irrespective of infants' general language development. The early achievers were still advanced at 16 months. Infants' preverbal gestural communication seemed to be particularly important in that it also predicted receptive and productive language at 16 months.

The second hypothesis of the current study, that DI at 9 months would predict gestural, receptive, and productive language at 16 months, was partly confirmed. DI did not predict gestural communication and receptive language at 16 months. However, a main finding was that infants' deferred imitation at 9 months predicted their productive language at 16 months. Hence, early declarative memory may be an important facilitator for the development of productive language. When infants are 16 months of age, they have entered the verbal stage of development and are producing more words at a rapid rate. They are not as dependent on gestures to understand and express their intent as they were a few months earlier. The words they use are often holophrases that symbolize events, things they have experienced, of which they previously formed mental representations. They use their ability to form mental representations of prior actions and events and their knowledge of gestural communication to produce words of symbolic content. This is apparent in the analysis because DI and gestural communication at 9 months independently predicted productive language at 16 months. Infants' ability to form mental representations (DI) and early communicative behavior (gestural communication) is certainly an integral part of why infants learn new words. Although both factors accounted for a substantial part of the variance (33%), other factors may also play important roles (e.g., communicative environment, temperament, linguistic stimulation).

Heimann and colleagues (2006) previously reported that infants' DI at 9 months predicted infants' gestural communication at 14 months but not their productive language at this age. Although a similar correlation was not observed in the current study, the difference may be accounted for by the experimental design. The longitudinal design in Heimann and colleagues' study was assessing 9-and 14-month-old children, a shorter interval than that used in the current study. At this stage in an infant's verbal development, 2 months may have a significant impact on progress. A 14-month-old is still primarily preverbal; in comparison, 2 months later, most 16-month-olds have entered the verbal stage of development and are actively trying to use words instead of only gestures

whenever possible. The results from Heimann and colleagues' study and the current study show that DI predicts gestural communication measured before the onset of the verbal period and predicts productive language measured after the onset of the verbal period.

Deferred imitation and functions of language did not show any significant concurrent correlations at either 9 or 16 months in the current study. This might be considered contrary to McEwen and colleagues' (2007) study, which found a correlation between the ability to immediately imitate actions and children's vocabulary at 2 years of age. The lack of correlation may indicate that immediate imitation and deferred imitation do measure different abilities (short-term memory vs. long-term memory) and that these different memory abilities are differently allocated and used at different ages. Previous studies regarding preschool children (e.g., Gathercole & Baddeley, 1993) have shown support for a domain-specific account regarding memory. It is possible that the initial domain-general declarative memory functions become more specialized as children's cognitive abilities develop.

Limitations

Attrition is important in longitudinal studies because it might affect the statistical power. In the current study, 6 children (14.3% of the original sample) were lost at T2 (4 due to technical failures at the lab and 2 due to failure to get in contact with the parents). Furthermore, there is always an uncertainty in using a parental questionnaire to estimate children's abilities in that parents might be biased in their estimates of their own children's abilities. It might be easier for parents to assess their children's productive language, which they have heard, rather than receptive language, which needs to be derived from the context; this might account for the lack of correlations between DI and receptive language. However, the questionnaire used in this study has been used extensively in prior international and Swedish research studies and, thus, has been validated as a useful method to estimate young children's language abilities (Eriksson & Berglund, 1999; Fenson et al., 1994). The DI tasks used in the current study used only single-step actions, which could be viewed as an additional limitation. Multiple-step actions might have provided additional information regarding children's early memory capacity. Finally, the reuse of the ''shaking egg procedure" task at T2 may have been affected by infants' exposure to the same target action at T1. It is unlikely that learning occurred, however, because no evidence exists indicating that children remember a single task without reminders over 7 months at this early stage in life.

Conclusions

This study adds to our understanding of declarative memory and language abilities during the time period when the infant develops from a primarily preverbal individual to a verbal individual. The first hypothesis was confirmed; declarative memory development, as measured with DI, revealed both stability and a significant increase over time. Infants who were high achievers at an early age also displayed advanced abilities at 16 months. The second hypothesis was partly confirmed; DI at 9 months predicted infants' productive language at 16 months but not infants' gestural communication and receptive language. The declarative memory and gestural ability of 9-month-olds independently predicted the children's productive language at 16 months. In other words, infants' ability to form mental representations of actions occurring in everyday life and their knowledge of communicative gestures are essential parts of early productive language development. Early declarative memory may be regarded as a domain-general ability required for both remembering and producing actions with objects and for remembering and producing new words.

Acknowledgments

Support for this research was provided by research grants from the Swedish Council for Working Life and Social Research (2008-0875) and the Swedish Research Council (2008-2454 and 20111913) to Mikael Heimann. The authors thank all of the parents and children who participated as well as Anneli Frostell and Emil Holmer for collecting part of the data. Thanks are also due to Gustaf Gre-deback for valuable input on an early draft of the article.

References

Barr, R., Dowden, A., & Hayne, H. (1996). Developmental changes in deferred imitation by 6- to 24-month-old infants. Infant Behavior and Development, 19,159-170.

Bauer, P.J. (2006). Constructing a past in infancy: A neuro-developmental account. Trends in Cognitive Sciences, 10,175-181.

Bauer, P. J., Wiebe, S. A., Carver, L. J., Waters, J. M., & Nelson, C. A. (2003). Developments in long-term explicit memory late in the first year of life: Behavioral and electrophysiological indices. Psychological Science, 14, 629-635.

Bishop, D. V. M. (1997). Uncommon understanding: Development and disorders of language comprehension in children. Hove, UK: Psychology Press.

Bloom, L. (1993). The transition from infancy to language: Acquiring the power of expression.New York: Cambridge University Press.

Bruner, J. S. (1964). The course of cognitive growth. American Psychologist, 19, 1-15.

Collie, R., & Hayne, H. (1999). Deferred imitation by 6- and 9-month-old infants: More evidence for declarative memory. Developmental Psychobiology, 35, 83-90.

Eriksson, M., & Berglund, E. (1999). Swedish early communicative development inventories: Words and gestures. First Language, 19(55), 55-90.

Fenson, L., Dale, P. S., Reznick, J. S., Bates, E., Thal, D. J., & Pethick, S. J. (1994). Variability in early communicative development. Monographs of the Society for Research in Child Development, 59 (5, Serial No. 242).

Fodor, J. A. (1992). A theory of the child's theory of mind. Cognition, 44, 283-296.

Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language. Hove, UK: Psychology Press.

Heimann, M., & Meltzoff, A. N. (1996). Deferred imitation in 9- to 14-month-old infants: A longitudinal study of a Swedish sample. British Journal of Developmental Psychology, 14, 55-64.

Heimann, M., & Nilheim, K. (2004). 6-month-olds and delayed actions: An early sign of an early explicit memory? Cognition, Brain and Behavior, 8, 249-254.

Heimann, M., Strid, K., Smith, L., Tjus, T., Ulvund, S. E., & Meltzoff, A. N. (2006). Exploring the relation between memory, gestural communication, and the emergence of language in infancy: A longitudinal study. Infant and Child Development, 15,233-249.

Heyes, C. (2012). What's social about social learning? Journal of Comparative Psychology, 126,193-202.

Jones, E. J. H., & Herbert, J. S. (2006). Exploring memory in infancy: Deferred imitation and the development of declarative memory. Infant and Child Development, 15, 195-205.

Kolling, T., Goertz, C., Frahsek, S., & Knopf, M. (2009). Stability of deferred imitation in 12- to 18-month-old infants: A closer look into developmental dynamics. European Journal of Developmental Psychology, 6, 615-640.

Kuczaj, S. A. (1987). Deferred imitation and the acquisition of novel lexical items. First Language, 7(21), 177-182.

Leslie, A. (1994). Pretense and representation: The origins of "theory of mind". Psychological Review, 94, 412-426.

Marshall, P. J., & Meltzoff, A. N. (2014). Neural mirroring mechanisms and imitation in human infants. Philosophical Transactions of the Royal Society B: Biological Sciences, 369. http://dx.doi.org/10.1098/rstb.2013.0620.

McEwen, F., Happé, F., Bolton, P., Rijsdijk, F., Ronald, A., Dworzynski, K., et al (2007). Origins of individual differences in imitation: Links with language, pretend play, and socially insightful behavior in two-year-old twins. Child Development, 78, 474-492.

Meltzoff, A. N. (1988a). Infant imitation after one-week delay: Long-term memory for novel acts and multiple stimuli. Developmental Psychology, 24, 4470-4476.

Meltzoff, A. N. (1988b). Infant imitation and memory: Nine-month-olds in immediate and deferred tests. Child Development, 59, 217-225.

Meltzoff, A. N., & Gopnik, A. (1989). On linking nonverbal imitation, representation, and language learning in the first two years of life. In G. E. Speidel & K. E. Nelson (Eds.), The many faces of imitation in language learning (pp. 22-51). New York: SpringerVerlag.

Meltzoff, A. N. (2005). Imitation and other minds: The "like-me" hypothesis. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science. Imitation, human development, and culture (Vol. 2, pp. 55-77). Cambridge, MA: MIT Press.

Meltzoff, A. N., Williamson, R. A., & Marshall, P. J. (2013). Developmental perspectives on action science: Lessons from infant imitation and cognitive neuroscience. In W. Prinz, M. Beisert, & A. Herwig (Eds.), Action science: Foundations of an emerging discipline (pp. 281-306). Cambridge, MA: MIT Press.

Mullally, S. L., & Maguire, E. A. (2014). Counterfactual thinking in patients with amnesia. Hippocampus, 24,1261-1266.

Nadel, J. (2014). How imitation boosts development in infancy and autism spectrum disorders. Oxford, UK: Oxford University Press.

Nelson, K. (1996). Language in cognitive development: The emergence of the mediated mind. Cambridge, UK: Cambridge University Press.

Nelson, K. E., Heimann, M., Abuelhaija, L., & Wroblewski, R. (1989). Implications for language acquisition models of children's and parents' variations in imitation. In G. E. Speidel & K. E. Nelson (Eds.), The many faces of imitation in language learning (pp. 305-323). New York: Springer-Verlag.

Piaget, J. (1962). Play, dreams, and imitation in childhood. New York: Norton.

Ramachandran, V. S. (2011). The tell-tale brain: A neuroscientist's quest for what makes us human. New York: Norton.

Rovee-Collier, C., & Cuevas, K. (2009). Multiple memory systems are unnecessary to account for infant memory development: An ecological model. Developmental Psychology, 45, 160-174.

Rovee-Collier, C., & Giles, A. (2010). Why a neuromaturational model of memory fails: Exuberant learning in early infancy. Behavioural Processes, 83, 197-206.

Snow, C. E. (1981). The use of imitation. Journal of Child Language, 8, 205-212.

Strid, K., Tjus, T., Smith, L., Meltzoff, A. N., & Heimann, M. (2006). Infant recall memory and communication predicts later cognitive development. Infant Behavior and Development, 29, 545-553.

Subiaul, F., Patterson, E. M., Schilder, B., Renner, E., & Barr, E. (2014). Becoming a high-fidelity super-imitator: What are the contributions of social and individual learning? Developmental Science, 18,1025-1035.

Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. Cambridge, MA: Harvard University Press.

Visscher, K. M., Kaplan, E., Kahana, M. J., & Sekuler, R. (2007). Auditory short-term memory behaves like visual short-term memory. PLoS Biology, 5, 662-672.