Maternal scaffolding of analogy and metacognition in the early pretence of gifted children.
This study investigated whether mothers of children assessed as having gifted/high IQ at 5 years were more likely to scaffold their children in analogical and metacognitive thinking during the infant/toddler period than mothers of children with more typical IQs. The researcher videotaped 21 children in monthly play sessions with their mothers, from the time that the children were 8 months old until they were 17 months old, and coded the mothers' verbalizations for scaffolding of analogical and metacognitive thinking. A psychologist assessed these children on the Stanford-Binet IV (Thorndike, 1986) and found ability levels ranging from average to high. Analysis showed that mothers of the children with high IQs introduced analogical and metacognitive scaffolding earlier than mothers of children with average IQs. The findings are consistent with a bidirectional model of gifted development in which mothers respond to support advanced development from infancy.

Gifted children (Reports)
Gifted children (Research)
Metacognition (Research)
Infants (Development)
Infants (Research)
Morrissey, Anne-Marie
Pub Date:
Name: Exceptional Children Publisher: Council for Exceptional Children Audience: Academic; Professional Format: Magazine/Journal Subject: Education; Family and marriage Copyright: COPYRIGHT 2011 Council for Exceptional Children ISSN: 0014-4029
Date: Spring, 2011 Source Volume: 77 Source Issue: 3
Event Code: 310 Science & research
Geographic Scope: United Kingdom; New York Geographic Code: 4EUUK United Kingdom; 1U2NY New York
Accession Number:
Full Text:
Educators generally understand intellectual giftedness in childhood as advanced capacities for cognitive functioning that are significantly ahead of age norms, including the early development of a capacity for abstract and higher order thinking (N. M. Robinson, 2008). Although heredity is significant in giftedness, environment also plays a role; and both are necessary and complementary factors in the development of giftedness (Sternberg, Grigorenko, & Bundy, 2001; Tannenbaum, 2003). Limited evidence, including evidence on the early childhood years, indicates that the parents of gifted children provide stimulating home environments and interactions that can promote gifted development (Fowler, 1981; Gottfried, Gottfried, Bathurst, & Guerin, 1994; Perleth, Lehwald, & Browder, 1993; N. M. Robinson, 1993; N. M. Robinson, Lanzi, Weinberg, Ramey, & Ramey, 2002). Fowler argued that advanced development that is accompanied by intensive caregiver stimulation in early childhood can lead to a cognitive "critical mass," so that by school entry these children "reach a cognitive threshold ... not ordinarily attained at any age by most children" (p. 360).

Some authors have proposed a bidirectional process of mutual responsiveness, whereby young gifted children elicit high levels of stimulation from their parents (Gottftied et al., 1994; Moss, 1990; N. M. Robinson, 1993), and some case study research supports that idea (Harrison, 2004; Lewis & Michalson, 1985). We know little, however, about the specific nature of parental interactions with their young gifted children, particularly in the infant and toddler periods, and how these interactions may promote advanced thinking. This gap in our knowledge is a significant one, because the first years of life are a crucial period for laying the foundations of subsequent intellectual development.

The study reported here was part of a larger longitudinal study investigating relationships between mother and child play and interactions, during the infant/toddler period and subsequent child IQ. The researcher followed a reversed contingency analysis method, similar to the one used in the Fullerton Longitudinal Study (Gottfried et al., 1994), which collected data before children were identified as gifted/high IQ rather than after. The basis for the findings presented here is an analysis of mothers' verbal scaffolding of analogical and metacognitive thinking in infant/toddler pretend play. The aim was to investigate whether the researcher could associate high child IQ at 5 years with earlier and/or more frequent maternal scaffolding of analogical and metacognitive thinking in the infant/toddler period.

Extensive research evidence exists on the role of caregiver interactions in the early cognitive development of typically developing young children, and there is a growing body of literature on caregiver-child interactions involving children with disabilities. Many of these studies draw on the sociocultural theory of Vygotsky and in particular, his concept of the zone of proximal development (ZPD; Vygotsky, 1978). The ZPD represents the difference between what children can accomplish unaided and what they can achieve with the assistance of adults or expert peers, thereby providing a lens for studying development as it unfolds within the context of dyadic interaction. A related concept is that of scaffolding, a term used to describe the range of responsive tutoring strategies--such as modeling, simplifying, maintaining interest and motivation, and marking features and discrepancies--used by adults to assist children's learning within the ZPD (Wood, Bruner, & Ross, 1976).

Researchers have shown that adult scaffolding involving the use of such specific strategies as modeling, the use of referential language, and both verbal and nonverbal assistance--as well as more-global authoritative, warm, and responsive interaction styles--have led to higher child outcomes in areas including play, language, and problem-solving (Berk & Spuhl, 1995; Conner, Knight, & Cross, 1997; Damast, TamisLeMonda, & Bornstein, 1996; Dilworth-Bart, Poehlmann, Hi|gendorf, Miller, K., & Lambert, 2009; Pratt, Kerig, Cowan, & Pape Cowan, 1988; Tamis-LeMonda & Bornstein, 1991; Tamis-LeMonda, Bornstein, & Baumwell, 2001; Vibbert & Bornstein, 1989). The interactions of both mothers and fathers with their young children have been the focus of study; and some evidence indicates that although there may be differences in style and learning outcomes, both can lead to positive effects (Conner et al., 1997; Pratt et al., 1988). More extensive research has dealt with the positive effects of mothers' scaffolding on young children's learning and development, and this area has developed a solid body of evidence and well-tested methodologies for investigating the nature of mother-child interactions (Berk & Spuhl, 1995; Clarke-Stewart & Beck, 1999; Smith, Landry, & Swank, 2000; Tamis-LeMonda & Bornstein, 1991, 1994; Tamis-LeMonda et al., 2001).

Since Bell and colleagues demonstrated that infants influence the sequence of activity with their caregivers (Bell, 1979; Bell & Harper, 1977; Kuczynski, 2003), researchers have widely acknowledged the bidirectional nature of child-caregiver interaction. Many researchers have recently come to regard informal teaching interactions, such as that between parent and child, as representing integrated processes of teaching/learning, in which the dyad forms the unit of analysis and the actions of dyad members are not independent of one another (Kenny, Kashy, & Cook, 2006; Scrimsher & Tudge, 2003). According to these views, the teaching or tutoring behavior of the caregiver within the dyad shows something about the learning or development of the child and about the dynamics of the teaching/learning relationship. The literature on mother-child dyads engaged in play or problem-solving tasks shows that child characteristics such as age and competence influence maternal interactions (Heckhausen, 1987; McCune, Dipane, Fireoved, & Fleck, 1994; J. B. Robinson, Burns, & Davis, 2009). Research shows, for example, that mothers of younger children provide more physical assistance (Heckhausen, 1987), whereas mothers of children with disabilities use more concrete and simple language and are more directive and active within the dyad (McCune et al., 1994; Power, Wood, Wood, & McDougall, 1990). A study that looked at maternal scaffolding of preschoolers living in poverty found that the mothers of children with poor attention-regulation skills were more likely to engage in intensive verbal scaffolding during a puzzle task (J. B. Robinson et al., 2009). Some researchers have interpreted such strategies as mothers' adaptive responses to children's specific needs (Jamieson, 1994; McCune et al., 1987; Power et al., 1990).

Mothers of gifted children, however, appear to offer higher levels of challenge in response to their children's advanced development. For example, Moss (1990, 1992a) found that mothers of gifted preschoolers were more likely to model metacognitive strategies than mothers of nongifted children. Morrissey and Brown (2009) found that toddlers subsequently identified with high IQ showed faster learning and earlier independence within the ZPD for pretend play. Their mothers responded by reducing the frequency of their own play modeling and increasingly transferring responsibility for dyadic play to the child. Similarly, Moss and Strayer (1990) found that mothers of gifted children responded to their children's higher level of metacognitive verbalizations by decreasing support to encourage autonomy. These findings indicate that mothers of gifted children engage them in stimulating and challenging interactions, possibly on the basis of cues that they are receiving about their children's advanced development (Moss, 1990, 1992a).

The aim of the present study was to investigate whether mothers of very young gifted children were more likely to engage them in intellectually stimulating interactions by scaffoling advanced forms of higher order thinking than mothers of more typically developing children. The researcher chose analogical and metacognitive thinking as the scaffolding variables for study. Both are core components of intelligence and have been linked to giftedness in older children (Brown & Kane, 1988; Gentner & Toupin, 1986; Holyoak, Junn, & Billman, 1984; Kanevsky, 1992; Klavir & Gorodestky, 2001; Sternberg, 1985; Vosniadou, 1989).

Traditional measures of giftedness, such as standardized IQ tests, draw directly on analogical reasoning (Fagan, 1984; Holyoak et al., 1984; Sattler, 1992, 2001); and gifted children demonstrate advancement in this area (Davidson, 1986; Davidson & Sternberg, 1984; Klavir & Gorodestky, 2001). Since the 1980s, researchers have developed age-appropriate methods for assessing analogical thinking in even very young children, including infants and toddlers (DeLoache, Simcock, & Marzolf, 2004; Goswami, 1992; Holyoak & Thasgard, 1997; Schafer, 2005); and Goswami argues that the ability to recognize relational or structural similarity, the basis of analogical reasoning, is present from the first months of life. Some research has also suggested links between mothers' analogical verbalizations and young children's cognitive development. Morelock, Brown, and Morrissey (2003) found that mothers of toddlers showing advanced development in pretend play engaged in more verbalizations containing analogies and world links involving connections between past and current experiences and knowledge. Higher levels of maternal use of world links with 2-year-olds have been linked to more extensive pretend play and advanced deductive reasoning in children (Mills & Funnell, 1983).

We know less about the early development of metacognition and to what extent this form of thinking develops differently in gifted children. Some evidence indicates that older gifted children and adolescents show advanced metacognitive thinking in comparison with age-typical peers (Borkowski & Peck, 1986; Cho & Ahn, 2003; Hoh, 2008; Jackson & Butterfield, 1986; Kanevsky, 1992; Moss, 1992a; Schwanenflugel, Stevens, & Carr, 1997; Steiner, 2006). Despite the lack of evidence on metacognitive development in the earliest years, some argue that development in metacognition is present during the first 5 years; or as Flavell (1987) states "probably ... almost from the beginning" (p. 25). Flavell regards a child's developing sense of self as an "active cognitive agent" (p. 26), along with parental modeling or teaching of metacognitive activity, as potentially important factors in the promotion of metacognitive thinking.

Other authors also propose that the early years are a period during which basic processes essential for metacognitive development are being laid down (Borkowski & Peck, 1986; Moss, 1992b; Schwanenflugel et al., 1997). These processes include conceptual and strategic knowledge, self-regulation, and evaluative functions. A number of authors point to the role of interactions with caregivers in promoting higher levels of metacognition in young gifted children. In particular, they argue that caregivers notice their gifted preschooler's advanced information-processing abilities and respond by supporting and encouraging higher level executive functions (Borkowski & Peck, 1986; Moss, 1992a, 1992b; Perleth et al., 1993; Schwanenflugel et al., 1997).

In a study on mother-child interactions involving infants, toddlers, and preschoolers, Moss (1992b) found that dyadic interactions in the first 3 years focused on such basic skills as labeling and joint task orientation. It was not until the fourth year that mothers began working to help children acquire simple metacognitive skills, including predicting consequences, checking results, reality testing, and activity monitoring. She argues that the beginning of metacognitive activity within the dyads depends on the establishment of the earlier, more basic skills in the first 3 years. Interestingly though, Moss found that 20% of mothers in her research verbalized and modeled metacognitive strategies when their children were as young as 24 months old, suggesting that some mothers of even very young children may engage in scaffolding metacognitive thinking.

Research on maternal scaffolding of very young children's learning and development has largely been located in informal or play-based contexts, sometimes involving measures of pretend play development (Damast et al., 1996; Fiese, 1990; Jamieson, 1994; McCune et al., 1994; Rome-Flanders, Cronk, & Gourde, 1995; Tamis-LeMonda & Bornstein, 1991, 1994; Tamis-LeMonda et al., 2001; Vandermaas-Peeler et al., 2002). During the time that pretend play develops, it demands higher forms of thinking in the child, including early forms of analogical and metacognitive thinking. For example, the creation of pretend scenarios requires recognizing and creating relational similarities--the structural mapping that underlies analogical reasoning (Goswami, 1991; Harris, 2000; Lillard, 2001). While their pretend play abilities develop, young children also demonstrate increasing skills in planning their play. These skills represent emerging forms of metacognitive thinking. They include identifying play goals and subgoals; selecting appropriate strategies and materials for use in pretense; predicting consequences of play actions; and performing such evaluative functions as checking results, reality testing, and activity monitoring (Moss, 1992b). Mother-child interactions involving pretend play can thus provide opportunities for maternal scaffolding of analogical and metacognitive thinking.

This study reports findings produced by an analysis of data from a larger longitudinal study investigating the early development of high-ability children and the role of mother-child interactions in supporting that development. The basis of the analysis was data on mothers' verbal scaffolding of analogical and metacognitive thinking during pretend play interactions with their very young children. This study investigated potential relationships between mothers' use of verbal scaffolding with their children during the infant/toddler period and measures of the children's IQ taken when they were approximately 5 years old. The researcher expected that she would be able to associate higher child IQ at 5 years old with earlier and/or more frequent use of analogical and metacognitive verbalizations by mothers during the infant/toddler period.



The participants were 21 mother-child dyads, involving 12 male children and nine female children. Recruitment procedures attempted to maximize the chances that at least half the child participants would achieve high levels on an IQ assessment when they were approximately 5 years old. Drawing on findings in the literature, initial recruitment procedures focused on the younger siblings of children formally or informally identified as gifted (Silverman, 1988). This study recruited 14 children on that basis and recruited another two children because they showed unusual alertness and responsiveness (Dalzell, 1998; Gross, 1993; N. M. Robinson, 1993). The study recruited the remaining five dyads through leaflets distributed through Maternal and Child Health Centres in suburban Melbourne. Child participants were all second- or later-born children, except for two first-born girls. Family backgrounds were largely professional and middle-class, and the families were predominantly of Anglo-Celtic or European cultural background. All families were English-speaking. One child was part Indonesian, one was part Mauritian, and one was an adopted Korean child. One child came from a sole-parent family. The children's parents generally had high levels of education, with 27 holding university degrees. They included 13 parents with postgraduate qualifications and five enrolled in doctoral programs or with completed doctoral degrees. The education levels of five of the fathers were unknown. At the time of taping, two mothers worked full-time, with the remainder either full-time homemakers or working or studying part-time. The parents' occupational backgrounds were largely professional or managerial, with other occupations including a builder, a carpenter, and a laborer. One father was unemployed at the time of the study.


The researcher coded the content of mothers' utterances according to categories for scaffolding children's analogical and metacognitive thinking. Utterances did not need to be a complete sentence, and the researcher coded them according to all elements present. They could therefore code one utterance as both analogical and metacognitive thinking (see Table 1). As noted in Moss and Strayer (1990), interpreting utterances within a dyadic context requires situational evidence. Coding for this study required interpreting maternal intentions by using such contextual evidence as the sequence of events, verbal inflection, gesture and facial expression, and references to play materials. Table 1 presents examples that illustrate how the categorization of utterances could vary according to their purpose or focus, as well as the type of play materials to which the mother was referring. Following are the definitions and criteria used for coding the analogical and metacognitive categories.

Analogical The basis for this category was a broad definition of analogical reference and included verbalizations that focused on pretend play transformations and world links. Examples included maternal references to analogical similarity, such as pointing to a small basket and remarking "That looks like a bowl," as well as world links to the child's previous experiences such as "It's like the truck we saw this morning." Also coded as analogical were verbal transformations that were associated with pretend play substitutions (Morelock et al., 2003). A typical example involved a mother announcing "Here's Teddy's bed" while indicating a box and then acting out a sequence of pretend transformations by putting the teddy bear to bed in the box and covering it with a piece of fabric substituted for a blanket. The researcher also coded verbal transformations involving imaginary substances as analogical, such as asking a child to wipe up an imaginary spill or asking him or her whether a pretend cup of tea was hot.

Metacognitive. The researcher designed the measures for metacognitive utterances to capture maternal verbalizations of planning and self-regulatory processes. She largely based these measures on a coding scheme developed by Moss (1990, 1992a) and Moss and Strayer (1990) but also drew on the wider literature on metacognition. Moss (1990, 1992a) categorized mother and child verbalizations in relation to problem-solving tasks according to level of cognitive distancing. The relevant categories for this study were Moss's two highest levels of structuring and metacognitive regulatory tactics. Structuring tactics include planning references such as identifying goals and subgoals and statements of rules. Metacognitive regulatory tactics include predicting consequences, checking results, monitoring activity, and reality testing.

Although both categories involved components of planning, monitoring, and evaluation, Moss considered only the processes in the metacognitive regulatory category to be truly metacognitive. Unlike the preschoolers in the Moss study (1992a), however, the researcher expected the infants and toddlers in the current study to demonstrate, at most, only an emergent metacognitive awareness. Therefore, the researcher decided to adapt Moss's coding categories to capture the ways in which a mother might introduce metacognitive ideas and scaffold any emerging capacities in a very young child, including through verbalization of both structuring and metacognitive regulatory tactics. In addition, the wider literature offers justification for a broader definition of metacognition that includes such planning activities as deciding on use of actions, space, or materials (Borkowski & Peck, 1986; Flavell, 1987; Schraw & Graham, 1997; Schwanenflugel et al., 1997). For this study, therefore, the researcher decided to regard subgoal identification as maternal scaffolding of metacognitive activity; for example, when a mother said that she needed to make a bed before she could put a doll to sleep. Thus, the researcher coded maternal utterances as metacognitive if they contained any of the following elements:

* Predicting the effectiveness or consequences of future actions or use of materials ("This would make a good bed"; "Will that balance?").

* Planning that involved verbalizing strategies or steps for achieving a goal, including identifying subgoals ("We could use this as a table for our tea party"; "First, we need to mix the batter").

* Monitoring the effectiveness or appropriateness of maternal or child actions ("We don't drink tea from a teapot--we use a cup"; "Yes, that's how you do it"; "We built a strong tower").

* Reflection on the self's capacities and/or experiences in relation to the task ("I don't think I can make an elephant noise"; "Do you know how to ...?"; "We used that as a hat last time").

The researcher defined metacognitive statements as being about the play but made outside it (Hacker, n.d.; Livingstone, 1997). These statements therefore needed to consider the relationship of the utterance to the play context. For example, if a child pretended to make a cup of tea and his or her mother pointed out that the child also needed to provide (pretend) milk and sugar, the researcher coded this utterance as metacognitive. If, however, the mother took on the play role of recipient of the cup of tea and within that role requested milk ("I need some milk"), the researcher considered this response to be maternal participation in the play and did not code it. The researcher coded maternal questions or statements such as "Can you brush Dolly's hair?" or "Can you balance it., as metacognitive when they referred to a specific challenge or capacity required of the child. They did not usually code general positive comments--such as "good," "good boy," or "clever girl"--as metacognitive. They considered positive statements that involved explicit evaluation of child or maternal actions--such as "That's right," "Yes, that's it," or "Almost"--as metacognitive,

Child IQ. The psychologist assessed 20 of the children by using the Stanford-Binet IV (Thorndike, 1986). This test is appropriate for assessing higher levels of intellectual ability in young children (Davis & Rimm, 1998; Sattler, 1992). The parents of one child chose to have him tested privately using the Stanford-Binet LM (Terman, 1973).


The play materials consisted of four rotating sets, with three age-based levels; each set dealt with one of three themes: tea party, dolls, or farm. The sets included realistic and replica materials, such as a teddy bear, a doll, plastic cups and plates, a hairbrush, a toy truck, and animal figures. They also included more abstract materials, for example, wooden blocks, wooden Popsicle sticks, squares of material, and aerosol lids, with an increasing proportion of abstract materials as children got older. For further details of the play sets, see Morrissey and Brown (2009).


Most of the participating mother-child dyads entered the study by the time the children were 8 months old. Thirteen of the children commenced videotaped play sessions "early," between the ages of 5 and 7 months; and six commenced videotaping at 8 months. Two of the dyads entered late: one when the child was 10 months old, and one when the child was 11 months old. To investigate any possible effects of these variations in starting age, the researcher conducted a Spearman's correlation (two-way) between starting age and all relevant variables and found no significant correlations.

The participants came to a university laboratory for monthly play sessions until children were 17 months old. The researcher drew data for this study from 10 videotaped sessions when children were 8 to 17 months old, although some data were missing. One dyad left the study when the child reached 14 months of age because of family relocation, and this dyad had also missed one previous session. In addition to the two dyads that started late, four dyads missed one session each, one dyad missed two sessions, and one dyad missed four sessions.

The researcher videotaped the play sessions in a special-purpose two-room suite. One room had two cameras, mats on which mothers and children could sit, and a box containing the relevant set of play materials for that session. The re searcher sat in the other room, behind one-way glass, controlling the videotaping equipment. The researcher informed the mothers that she was investigating children's play and language development and asked the mothers to play with the child as they normally would at home. They received no other instructions. The researcher videotaped the dyads for 15 min, except for occasions when a child was distracted or unwell. Three children had one shorter taping (ranging from 10.32 to 13.06 min), one child had two shorter tapings (11.48 and 14.12 min), and four children had three shorter tapings (ranging from 10.19 to 13.57 min). All but three of these shorter tapings occurred when children were between 8 and 14 months old.

When children turned 4 1/2 to 5 years old, the researcher again contacted the mothers and gave them the name of the psychologist who would conduct the IQ assessment of their child. A university psychologist tested 14 of the children and supervised the testing of another four children by postgraduate psychology students. The remaining three children received testing through arrangements with private psychologists. Because of unavoidable circumstances, only seven of the children received testing within the target age range of 4 to 5 years. Of the remainder, 10 received testing when they were 5 to 6 years old, and four of the children were not assessed until they were 6 years old. Although this age range was not ideal, the literature indicates that IQ scores are fairly stable within individuals during development (Sternberg et al., 2001). Sternberg and colleagues point out that the "best predictor of IQ in a given year is the IQ from the previous year" (p. 15). They point to long-term data showing, for example, Stanford-Binet scores at age 5 correlating (r = .87) with scores at age 6. Scores across a 3-year age range from ages 3 to 6 years also correlated (r = .73; Sternberg et al., 2001). On the basis of that evidence, the researcher decided to use all the child IQ assessments for the analysis.


The researcher based data for this study on three 5-min samples containing the child's highest play levels for each of three age ranges. The researcher had selected these 5-min samples as the database for a previous analysis of child and mother play as part of the larger longitudinal study. She used the samples from the available sessions within three age ranges: 8 to 9 months (Session 1), 11 to 13 months (Session 2), and 16 to 17 months (Session 3). These age ranges represented discrete periods in which the child-participant group demonstrated clear developmental progression through one or more play levels. In the other periods--at the 10-, 15-, and 16-months sessions--children appeared to plateau in their pretend play development. The selection of samples from within three age ranges also minimized the amount of missing data. The researcher digitally transferred these samples to CD and coded them by using the Observer (1996) by Noldus behavior recording program supplemented with the support package for video analysis.

Informal observation had indicated that the rates of maternal verbal scaffolding tended to increase around children's play activity. This verbal scaffolding also tended to be more challenging when mothers were scaffolding their children to higher levels of play. This study explored the potential for maternal scaffolding of children's analogical and metacognitive thinking during play. The researcher believed that sampling maternal utterances around children's highest play levels, rather than a random sampling across sessions, would be most likely to pick up the sort of maternal utterances that were of interest. Because of practical considerations and methodological requirements, the researcher based the analysis on mothers' verbal scaffolding on the first 3 min of these 5-min samples from the larger study. The video-coding process was long and time-consuming, particularly in regard to the verbal scaffolding. The researcher examined each maternal scaffolding utterance for the presence of elements of analogical and/or metacognitive scaffolding. The rate of mother-and-child play actions required three 5-min samples to provide sufficient data for the proposed analyses. The rate of maternal verbal utterances was higher, and the three 3-min samples yielded 582 codings of maternal verbal scaffolding, which provided sufficient representative data for the proposed analyses.

The researcher calculated intercoder agreement for a representative one third of the sample and initially coded the 3-min samples that made up the verbal scaffolding data. To establish interobserver reliability, the researcher and a second coder went through one third of the sample together, discussing the researcher's codings and establishing whether the second coder agreed or disagreed with each one. Intercoder agreement was therefore based on the level of agreement and disagreement between the researcher and the second coder as to the occurrence or nonoccurrence of analogical and metacognitive elements in maternal verbalizations. Percentage agreement was 99% (Cohen's [kappa] = .94) for analogical utterances and 98% (Cohen's [kappa] = .91) for metacognitive utterances.


The results of the IQ assessments showed that the children's IQ scores represented an average to high range of 96 to 150 (M = 122.62, SD = 13.37). Coding of maternal utterances for the whole group across the three sessions yielded totals of 328 analogical codings and 254 metacognitive codings. Analysis of the frequency of the two types of maternal utterance demonstrated that mothers increased their use of analogical and metacognitive verbal scaffolding across the sessions, particularly between Sessions 1 and 2. The frequency of analogical utterances increased from 33 in Session 1 (Mdn = 0.00, range = 0-8), to 132 in Session 2 (Mdn = 4.00, range = 0-31), and 170 in Session 3 (Mdn = 4.50, range = 0-22). Metacognitive utterances increased from 34 in Session 1 (Mdn = 0.00, range = 0-10), to 100 in Session 2 (Mdn = 2.00, range = 0-14), and 120 (Mdn = 2.50, range = 0-24) in Session 3.

To investigate whether mothers' use of analogical and metacognitive scaffolding linked with children's subsequently tested IQ level, the researcher divided dyads into higher and lower IQ groups on the basis of a natural split in child IQ scores. The higher IQ group (HIQ) comprised 11 children and their mothers, with scores ranging from 123 to 150; whereas the lower IQ group (LIQ) comprised 10 children and their mothers, with scores in an average to high-average range of 96 to 119. Recent definitions of giftedness allow the children in the HIQ group to be designated as gifted (with the lowest score of 123 considered borderline; Porter, 1999). The child who had received testing on the Stanford-Binet LM (Terman, 1973) had an IQ score of 150. To compensate for the out-of-date norms of the LM, the researcher deducted 10 points from this child's IQ score (Silverman & Kearney, 1992); and he and his mother remained in the HIQ group.

Because two children started in the study late and one child moved overseas before completing all sessions, the actual numbers of dyads varied across the sessions, with 19 dyads in Session 1 (HIQ = 9, LIQ = 10); 21 in Session 2 (HIQ = 11, LIQ = 10); and 20 in Session 3 (HIQ = 10, LIQ = 10). Table 2 gives the median frequencies and the ranges of each type of maternal utterance for the two groups across the three sessions. Although both groups showed low frequencies of both types of utterance in Session 1, the median and upper range scores for mothers' use of analogical utterances is marginally higher in the HIQ group in this first session. Inspection of data on individual dyads confirmed that mothers in the HIQ group were more likely to use analogical utterances in that session. Two mothers in the HIQ group used no analogical utterances, and three of the mothers used this type of utterance once. The remaining four mothers in the HIQ group used analogical utterances five to eight times. In the LIQ group, however, seven of the mothers used no analogical utterances in this session; and the remaining three mothers used one to two utterances.

Table 2 also shows that although mothers in both groups increased their use of analogical and metacognitive utterances from Session 1 to Session 2, median scores were higher for mothers in the HIQ group in Session 2. In the HIQ group, two of the mothers used no analogical utterances, two used between 2 and 4, six used between 7 and 14, and one used 31. In the LIQgroup, three of the mothers used no analogical utterances, two mothers used 1, three mothers used 3 to 5, and two mothers used 10. Also in that session, two mothers in the HIQ group used 0 to 1 metacognitive utterances, one used 2 utterances, three used 6 to 9, and five used 10 to 14. In the LIQ group, seven mothers used 0 to 1 metacognitive utterances, one used 2 utterances, and two used 5 to 7.

In Session 3, mothers in the LIQ group continued to increase their use of these utterances. Although the use of both types of utterance was still more frequent in the HIQ group in Session 3, these mothers' use of analogical utterances declined slightly from the level of Session 2, but their median score for metacognitive utterances halved.

To test whether these group differences were statistically significant, the researcher conducted a Mann-Whitney U Test (Tabachnik, 2001) on

each measure, using an alpha level of .05. Because the researcher expected mothers in the HIQ group to be more likely to use analogical and metacognitive utterances with their children, she used a one-tailed test of significance. Table 2 presents the results. Mothers in the HIQ group used significantly more analogical utterances with their children in Session 1, as well as significantly more analogical and metacognitive utterances in Session 2. No significant differences occurred between the groups in maternal use of utterances in Session 3.


The findings of this study show that mothers of children assessed as having a high IQ at 5 years old commenced scaffolding analogical and metacognitive thinking earlier in the infant/toddler period than mothers of children subsequently assessed as having an IQ that was closer to average. The results suggest that children subsequently identified as gifted were more likely to experience maternal scaffolding of analogical and metacognitive thinking from a younger age--and consequently for a longer period--than children subsequently identified with more typical development.

In interpreting these results, examining the findings from the larger longitudinal study on mother and child play (Morrissey & Brown, 2009) is helpful. The results from that study indicated no differences between the two IQ groups in levels of child play in any of the three sessions, with children in both groups demonstrating emerging pretend play behaviors from 8 to 9 months. Children in the HIQ group did not attain higher levels of play earlier than children in the LIQ group, but they did demonstrate earlier independence in their play by Session 2, indicating faster learning of pretend play skills within the ZPD. The mothers of the children in the HIQ group responded to this independence by transferring joint responsibility for play to their children in that same session. This outcome compared with mothers in the LIQ group, who transferred joint responsibility to their children by Session 3.

Those findings, in conjunction with the present findings on mothers' verbal scaffolding, show that by the time that the children subsequently assessed as having high IQ reached 16 to 17 months of age, their experience of the play sessions with their mothers had differed in several ways from that of children subsequently assessed as having an IQ that was closer to average. Not only did the children in the higher IQ group experience scaffolding of analogical thinking for longer, but they were also more likely to experience it during the period of the earliest emergence of their pretend play development. Almost from the very beginning, then, abstract verbal commentary around analogical thinking accompanied their development of pretense. Children in the lower IQ group, however, did not experience equivalent levels of analogical scaffolding until 16 to 17 months of age, at a time when their pretend play was well established. The findings suggest that by introducing verbal elements of abstraction, complexity, and challenge into their play interactions, the mothers of the high IQ children were providing a qualitatively different play experience for their children from an early age. This outcome is in accord with Fowler's (1981) proposition that intensive caregiver stimulation from an early age plays a significant role in the development of giftedness.

The more independent play skills of children in the HIQ group in Session 2 (Morrissey & Brown, 2009) could also have elicited the maternal metacognitive scaffolding that appeared in that same session. The mothers in the higher IQ group may have considered that their children had grasped the basics of engaging in pretend play and no longer needed basic modeling support. They may have seen their children as ready to move to the next level, in which they could begin to learn how to plan and organize their own pretend play activity. This process is similar to the process in which children progress from coming to understand the basics of pretense per se and move to the level of understanding that allows them to become active collaborators in pretense (Walker-Andrews & Kahana-Kalman, 1999).

The introduction of metacognitive scaffolding in Session 2 by mothers in the HIQ group suggests that the HIQ dyads had moved from acting within a ZPD for learning to engage in various forms of pretend play to a new ZPD that was based on the children's emerging higher order skills such as planning, organizing, monitoring, and evaluating their own play activity. Findings both from the previous analysis on mother and child play (Morrissey & Brown, 2009) and the present analysis support this interpretation. Morrissey and Brown found that by Session 3, children in the LIQ group had caught up with children in the HIQ group in their play skills and that LIQ dyads also demonstrated maternal transfer of responsibility for play activity to the children. This outcome parallels the findings in the current analysis that no significant difference existed between the two groups in the frequency of maternal metacognitive scaffolding by Session 3. Taken together, these findings support the contention that mothers' introduction of metacognitive scaffolding links with their children's development of independent play skills. The findings also provide support for those who argue that metacognitive development has its roots in the first years of life and that it is essentially a social skill originating in interaction, particularly interaction between parents and children (Moss, 1990).

Although both groups of mothers promoted analogical and metacognitive thinking at what appears to be an early age, the interactions of the mothers of children with higher IQ seem particularly challenging, especially with regard to metacognitive thinking. As previously noted, Moss's (1992b) research indicated that mothers do not typically engage in scaffolding metacognitive thinking until their children are 4 years old. Moss did find that one fifth of her mothers introduced metacognition to their children at 24 months; however, this age is still 11 to 13 months later than the age at which mothers in the HIQ group introduced metacognition to their children and 7 to 8 months later than mothers in the LIQ group. The developmentally appropriate context of play may have provided unusually good opportunities for mothers in both groups in this study to introduce and promote these forms of thinking in their children. This possibility suggests that pretend play can be a useful context for investigating the ways in which caregivers can promote higher order thinking in toddlers.

The small sample size and the range of ages at which children were assessed for IQ are obvious limitations to this study. A small sample relates to low power and can lead to Type II errors. Therefore, a larger sample size may have led to the discovery of continued group differences in maternal scaffolding at Session 3. The study may have also found group differences in Session 3 if the coding scheme had retained more of Moss's distinctions in the metacognitive coding category. The basis for the metacognitive coding category for this study was the literature, but it represented a less rigorous adaptation of Moss's metacognitive category (1990, 1992a). The criteria used in the current study produced findings that distinguished between the IQ groups in Session 2 but not in Session 3. The less rigorous version of the category may have created a ceiling that some dyads reached by Session 3. In devising the category for this study, the researcher expected that only a few dyads involving such young children would exhibit metacognitive activity. This theory was the reason for the less rigorous adaptation, and the resulting prevalence of metacognitive verbalizations by Session 3 in both groups was surprising. Discrimination of levels within this category could be useful, in a manner closer to Moss's original formulation, as illustrated by the following example involving a dyad from the HIQ group.

The mother of C1 (M1) was the study's most frequent user of metacognitive verbalizations. A comparison of interactions between C1 and his mother at Session 2 and Session 3 provides an illustration of how metacognitive interactions can evolve over time. The first example is from Session 2, when C1 (designated as J.), was 12 months old. He had picked up a baby's hairbrush and brushed the floor with it, as if using a dustpan and a small floor brush. His mother acknowledged this use of the brush ("Oh, a 'brush-and-shovelly-brush.' Yes, might be a 'brush-and-shovelly-brush.'") and then asked whether it was a hairbrush. She then made the following suggestion: "J., might be a toothbrush. Is it a toothbrush (laughs)? Is it a great big toothbrush?" The researcher coded these utterances as metacognitive on the basis that they represented planning and considering possible ways of using the brush in pretend activity (the child appeared to take up his mother's suggestion, later making a brushing motion with the hairbrush across his mouth).

The second example comes from Session 3, when C1 was 16 months of age. At one point in this session, M1 demonstrated to C1 how to set out a number of Popsicle sticks on the floor so that they could count them one by one. This process involved C1 pointing to each stick while his mother verbalized the appropriate numeral in the sequence. After C 1 had successfully accomplished this activity once, he attempted it again but this time missed pointing to one of the sticks. His mother responded with the following comments:

"One, two, three, four ... missed one ... missed the middle one. Yeah, where was the middle one? Let's try again. Shall we try again? Here we go" (arranges the sticks in a row). "OK, J., count" (C1 points in turn to three sticks while his mother counts them, and then he stops. M1 then completes the count.). "One, two, three ... four, five, mister! ... It's easier if we start at one end and move through."

This latter example of metacognitive interaction between C1 and his mother represents more complexity and challenge than the previous example from Session 2. In addition to the relatively simple considerations of planning demonstrated in the example from Session 2, the example from Session 3 also contains elements of evaluation and monitoring, tactics that belong to Moss's highest category of metacognition (Moss, 1990, 1992a). Thus, a limitation of the metacognitive coding category used in this study was that although it could measure the sheer quantity of metacognitive utterances, it was not able to distinguish the sort of qualitative differences exemplified by the two interactions described.

Although the present study focused on the role of maternal scaffolding in the cognitive development of young gifted children, the findings also have implications for future research into the development of higher order thinking in the wider population of young children. Research using play contexts and involving larger samples could further explore the ways in which caregivers respond to promote infant and toddler thinking, including analogical and metacognitive thought. Because the early development of metacognitive thinking is little understood, the topic calls for further investigation. Further longitudinal research could also contribute to our understanding of potential mechanisms through which early caregiver interactions may influence children's long-term intellectual development. The implications of the findings on maternal scaffolding by mothers in both groups also add weight to the argument that research into the influences on early gifted development can also contribute to our understanding of influences on early intellectual development in general.

The results of this analysis support other findings that parents of young gifted children engage them in stimulating and challenging interactions that are likely to support their advanced development from the first years of life. Together with findings from the larger longitudinal study, the results reported here also support the proposition that in providing these sorts of interaction, parents are responding to cues about their children's advanced development, an outcome that supports the bidirectional model of gifted development proposed by researchers (Gottfried et al., 1994; Moss, 1990; N. M. Robinson, 1993).

These findings have several implications for professionals. First, the findings suggest that pretend play may provide a favorable context for interactions that can promote young children's thinking. Further investigation of the possible advantages of naturalistic and formal interventions through pretend play could potentially affect the practice of both early childhood education and early intervention professionals. Second, the findings suggest that professionals need to become aware of the individual development and needs of infants and toddlers in their care and ensure that they provide sufficient stimulation and challenge for those children who have the potential for more advanced development. Third, the findings can guide professional attitudes toward the important role of parents in responding to and promoting gifted potential in their children, thereby helping to counteract negative stereotypes of pushy parents who "hothouse" their children. As N. M. Robinson states:

When parents of [gifted children] behave differently from other parents, should one attribute the children's giftedness to those differences, or should one assume that the parents are responding differently because their children are different? Obviously the answer is that both are true, and trying to disentangle them is usually fruitless. (N. M. Robinson, 1993, p. 513)


Bell, R. Q. (1979). Parent, child, and reciprocal influences. American Psychologist, 34, 821-826.

Bell, R. Q., & Harper, L. V. (1977). Child effects on adults. Hillsdale, NJ,: Erlbaum.

Berk, L. E., & Spuhl, S. T. (1995). Maternal interaction, private speech, and task performance in preschool children. Early Childhood Research Quarterly, 10, 145-169.

Borkowski, J. G., & Peck, V. A. (1986). Causes and consequences of metamemory in gifted children. In R. J. Sternberg & V. A. Peck (Eds.), Conceptions of giftedness (pp. 182-200). Cambridge, NY: Cambridge University Press.

Brown, A. L., & Kane, M. J. (1988). Preschool children can learn to transfer: Learning to learn and learning from example. Cognitive Psychology, 20, 493-523.

Cho, S., & Ahn, D. (2003). Strategy acquisition and maintenance of gifted and nongifted young children. Exceptional Children, 69, 497-505.

Clarke-Stewart, K. A., & Beck, R. J. (1999). Maternal scaffolding and children's narrative retelling of a movie story. Early Childhood Research Quarterly, 14, 409-434.

Conner, D. B., Knight, D. K., & Cross, D. R. (1997). Mothers and fathers scaffolding of their two-year-olds during problem-solving and literacy interactions. British Journal of Developmental Psychology, 15, 323-338.

Dalzell, H. J. (1998). Giftedness: Infancy to adolescence--a developmental perspective. Roeper Review, 20, 259-265.

Damast, A. M., Tamis-LeMonda, C. S., & Bornstein, M. H. (1996). Mother-child play: Sequential interactions and the relation between maternal beliefs and behaviours. Child Development, 67, 1752-1766.

Davidson, J. E. (1986). The role of insight in intellectual giftedness. In R. J. Sternberg & J. E. Davidson (Eds.), Conceptions of giftedness (pp. 201-222). Cambridge, NY: Cambridge University Press.

Davidson, J. E., & Sternberg, R. J. (1984). The role of insight in intellectual giftedness. Gifted Child Quarterly, 28, 58-64.

Davis, G. A., & Rimm, S. B. (1998). Education of the gifted and talented (5th ed.). Boston, MA: Allyn & Bacon.

DeLoache, J. S., Simcock, G., & Marzolf, D. P. (2004). Transfer by very young children in the symbolic retrieval task. Child Development, 75, 1708-1718.

Dilworth-Bart, J., Poehlmann, J., Hilgendorf, A. E., Miller, K., & Lambert, H. (2009). Do mothers' play behaviours moderate the associations between socioeconomic status and 24-month neurocognitive outcomes of toddlers born pre-term or with low birth weight? Journal of Pediatric Psychology. Retrieved from http:// jpepsy.jsq064.full

Fagan, J. (1984). The intelligent infant: Theoretical implications. Intelligence, 8, 1-9.

Fiese, B. H. (1990). Playful relationships: A contextual analysis of mother-toddler interaction and symbolic play. Child Development, 61, 1648-1656.

Flavell, J. H. (1987). Speculations about the nature and development of metacognition. In F. E. Weinert & R. H. Kluwe (Eds.), Metacognition, motivation and understanding (pp. 21-29). Hillsdale, NJ: Erlbaum.

Fowler, W. (1981). Case studies of cognitive precocity: The role of exogenous and endogenous stimulation in early mental development. Journal of Applied Developmental Psychology, 2, 319-367.

Genmer, D., & Toupin, C. (1986). Systematicity and surface similarity in the development of analogy. Cognitive Science, 10, 277-300.

Goswami, U. (1991). Analogical reasoning: What develops? A review of research and theory. Child Development, 62, 1-22.

Goswami, U. (1992). Analogical reasoning in children. Hove, UK: Erlbaum.

Gottfried, A. W., Gottfried, A. E., Bathurst, K., & Guerin, D. W (1994). Gifted IQ: Early developmental aspects--The Fullerton Longitudinal Study. New York, NY: Plenum Press.

Gross, M. (1993). Exceptionally gifted children. London, UK: Routledge.

Hacker, D. J. (n.d.). Metacognition: Definitions and empirical foundations. Retrieved from http://www.psyc.

Harris, P. L. (2000). The work of the imagination. Oxford, UK: Blackwell.

Harrison, C. (2004). Giftedness in early childhood: The search for complexity and connection. Roeper Review, 26, 78-84.

Heckhausen, J. (1987). Balancing for weaknesses and challenging developmental potential: A longitudinal study of mother-infant dyads in apprenticeship. Developmental Psychology, 23, 762-770.

Hoh, P.-S. (2008). Cognitive characteristics of the gifted. In J. A. Plucker & C. M. Callaghan (Eds.), Critical issues and practices in gifted education (pp. 57-83). Waco, TX: Prufrock Press.

Holyoak, K. J., Junn, E. N., & Billman, D. O. (1984). Development of analogical problem-solving skill. Child Development, 55, 2042-2055.

Holyoak, K. J., & Thasgard, P. (1997). The analogical mind. American Psychologist, 52(1), 35-44.

Jackson, N. E., & Butterfield, E. C. (1986). A conception of giftedness designed to promote research. In R. J. Sternberg & J. E. Davidson (Eds.), Conceptions of giftedness (pp. 151-181). Cambridge, NY: Cambridge University Press.

Jamieson, J. R. (1994). Teaching as transaction: Vygotskian perspectives on deafness and mother-child interaction. Exceptional Children, 60, 434-450.

Kanevsky, L. (1992). The learning game. In P. S. Klein & A. J. Tannenbaum (Eds.), To be young and gifted (pp. 204-241). Norwood, NJ: Ablex.

Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. New York, NY: Guilford.

Klavir, R., & Gorodestky, M. (2001). The processing of analogous problems in the verbal and visual-humorous (cartoons) modalities by gifted/average children. Gifted Child Quarterly, 45, 205-215.

Kuczynski, L. (2003). Beyond bi-directionality: Bilateral conceptual frameworks for understanding dynamics in parent-child relations. In L. Kuczynski (Ed.), Handbook of dynamics in parent-child relations (pp. 3-24). Thousand Oaks, CA: Sage.

Lewis, M., & Michalson, L. (1985). The gifted infant. In J. Freeman (Ed.), The psychology of gifted children: Perspectives on development and education (pp. 35-57). New York, NY: John Wiley & Sons.

Lillard, A. (2001). Pretend play as twin earth: A social-cognitive analysis. Developmental Review, 21, 495-531.

Livingstone, J. A. (1997). Metacognition: An overview. Retrieved from cep564/Metacog.htm

Tabachnik, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham Heights, MA: Allyn & Bacon.

McCune, L., Dipane, D., Fireoved, R., & Fleck, M. (1994). Play: A context for mutual regulation within mother-child interaction. In A. Slade & D. P. Wolf (Eds.), Children at play: Clinical and developmental approaches to meaning and representation (pp. 148-166). Oxford, UK: Oxford University Press.

Mills, M., & Funnell, E. (1983). Experience and cognitive processing. In S. Meadows (Ed.), Developing thinking (pp. 161-187). London, UK: Methuen.

Morelock, M. J., Brown, P. M., & Morrissey, A. M. (2003). Pretend play and maternal scaffolding: Comparisons of toddlers with advanced development, typical development, and hearing impairment. Roeper Review, 26, 41-51.

Morrissey, A. M., & Brown, P. M. (2009). Mother and toddler activity in the zone of proximal development for pretend play as a predictor of higher child IQ. Gifted Child Quarterly, 53, 106-120.

Moss, E. (1990). Social interaction and metacognitive development in gifted preschoolers. Gifted Child Quarterly 34, 16-20.

Moss, E. (1992a). Early interactions and metacognitive development of gifted pre-schoolers. In P. S. Klein & A. J. Tannenbaum (Eds.), To be young and gifted (pp. 278-318). Norwood, NJ: Ablex.

Moss, E. (1992b). The socioaffective context of joint cognitive activity. In L. T. Winegar & J. Valsiner (Eds.), Children's development within social context: Vol. 1.

Metatheory and theory (pp. 117-154). Hillsdale, NJ: Erlbaum.

Moss, E., & Strayer, F. F. (1990). Interactive problem-solving of gifted and non-gifted preschoolers with their mothers. International Journal of Behavioural Development, 13, 177-197.

The Observer (Version 3.0 for Windows) [Computer software] Support package for video analysis. (1996). Wagingen, The Netherlands: Noldus Information Technology.

Perleth, C., Lehwald, G., & Browder, C. S. (1993). Indicators of high ability in young children. In K. A. Heller, E J. Monks, & A. H. Passow (Eds.), International handbook of research and development of giftedness and talent (pp. 283-310). Oxford, UK: Pergamon Press.

Porter, L. (1999). Gifted young children.. A guide for parents and teachers. St Leonard's, New South Wales, Australia: Allen & Unwin.

Power, D. J., Wood, D. J., Wood, H. A., & McDougall, J. (1990). Maternal control over conversations with hearing and deaf infants and young children. First Language, 10, 19-35.

Pratt, M. W., Kerig, P., Cowan, P. A., & Pape Cowan, C. (1988). Mothers and fathers teaching three-year-olds: Authoritative parenting and adult scaffolding of young children's learning. Developmental Psychology, 24, 832-839.

Robinson, J. B., Burns, B. M., & Davis, D. W. (2009). Maternal scaffolding and attention regulation in children living in poverty. Journal of Applied Developmental Psychology 30, 82-91.

Robinson, N. M. (1993). Identifying and nurturing gifted, very young children. In K. A. Heller, F. J. Monks, & A. H. Passow (Eds.), International handbook of research and development of giftedness and talent (pp. 507-524). Oxford, UK: Pergamon Press.

Robinson, N. M. (2008). Early childhood. In J. A. Plucker & C. M. Callahan (Eds.), Critical issues and practices in gifted education: What the research says (pp. 179-194). Waco, TX: Prufrock Press.

Robinson, N. M., Lanzi, R. G., Weinberg, R. A., Ramey, S. L., & Ramey, C. T. (2002). Family factors associated with high academic competence in former Head Start children at third grade. Gifted Child Quarterly, 46, 278-290.

Rome-Flanders, T., Cronk, C., & Gourde, C. (1995). Maternal scaffolding in mother-infant games and its relation to language development: A longitudinal study. First Language, 15, 339-355.

Sattler, J. M. (1992). Assessment of children (Revised and updated 3rd ed.). San Diego, CA: Jerome M. Sattler.

Sattler, J. M. (2001). Assessment of children: Cognitive applications (4th ed.). San Diego, CA: Jerome M. Sattler.

Schafer, G. (2005). Infants can learn decontextualized words before their first birthday. Child Development, 76, 87-96.

Schraw, G., & Graham, T. (1997). Helping gifted students develop metacognitive awareness. Roeper Review, 20, 4-8.

Schwanenflugel, P. J., Stevens, T. P. M., & Carr, M. (1997). Metacognitive knowledge of gifted children and nonidentified children in elementary school. Gifted Child Quarterly, 41, 23-35.

Scrimsher, S., & Tudge, J. (2003). The teaching/learning relationship in the first years of school: Some revolutionary implications of Vygotsky's theory. Early Education and Development, 14, 293-312.

Silverman, L. K. (1988). The second child syndrome. Mensa Bulletin, 320, 18-20.

Silverman, L. K., & Kearney, K. (1992). The case for the Stanford-Binet LM as a supplemental test. Roeper Review, 15, 34-37.

Smith, K. E., Landry, S. H., & Swank, E R. (2000). Does the content of mothers' verbal stimulation explain differences in children's development of verbal and nonverbal cognitive skills. Journal of School Psychology, 38, 27-49.

Steiner, H. H. (2006). A micro genetic analysis of strategic variability in gifted and average-ability children. Gifted Child Quarterly, 50, 62-74.

Sternberg, R. J. (1985). Beyond IQ: A triarchic theory of intelligence. New York, NY: Cambridge University Press.

Sternberg, R. J., Grigorenko, E. L., & Bundy, D. A. (2001). The predictive value of IQ. Merrill Palmer Quarterly, 47, 1-41.

Tamis-LeMonda, C. S., & Bornstein, M. H. (1991). Individual variation, correspondence, stability, and change in mother and toddler play. Infant Behaviour and Development, 14, 143-162.

Tamis-LeMonda, C. S., & Bornstein, M. H. (1994). Specificity in mother-toddler language-play relations across the second year. Developmental Psychology, 30, 283-292.

Tamis-LeMonda, C. S., Bornstein, M. H., & Baumwell, L. (2001). Maternal responsiveness and children's achievement of language milestones. Child Development, 72, 748-767.

Tannenbaum, A. J. (2003). Nature and nurture of giftedness. In N. Colangelo & G. A. Davis (Eds.), Handbook of gifted education (pp. 45-59). Boston, MA: Allyn & Bacon.

Terman, L. M., & Merrill, M. A. (1973). Stanford-Binet Intelligence Scale: Manual for the Third Revision Form LM (1972 Norm Tables by R. L. Thorndike). Boston, MA: Houghton Mifflin.

Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). The Stanford-Binet Intelligence Scale: Fourth Edition. Itasca, IL: Riverside Publishing.

Vandermaas-Peeler, M,, King, C., Clayton, A., Holt, M., Kurtz, K., Maestri, L., ... Woody, E. (2002). Parental scaffolding during joint play with preschoolers. In J. R. Roopnarine (Ed.), Conceptual, social-cognitive, and contextual issues in the field of play (pp. 165-181). Westport, CT: Ablex.

Vibbert, M., & Bornstein, M. H. (1989). Specific associations between domains of mother-child interaction and toddler referential language and pretense play. Infant Behaviour and Development, 12, 163-184.

Vosniadou, S. (1989). Analogical reasoning as a mechanism in knowledge acquisition: A developmental perspective. In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning (pp. 413-437). Cambridge, UK: Cambridge University Press.

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Walker-Andrews, A. S., & Kahana-Kalman, R. (1999). The understanding of pretence across the second year of life. British Journal of Developmental Psychology, 17, 523-536.

Wood, D. J., Bruner, J., & Ross, S. (1976). The role of tutoring in problem-solving. Journal of ChiM Psychology and Psychiatry, 17, 89-100.


University of Melbourne

ANNE-MARIE MORRISSEY, Lecturer, School of Education, Deakin University (Burwood), Victoria 3217, Australia (formerly, Lecturer, Melbourne Graduate School of Education, University of Melbourne).

Correspondence concerning this article should be addressed to Anne-Marie Morrissey, School of Education, Deakin University, Victoria 3217, Australia. (e-mail:

This research was supported by a Postgraduate Scholarship for Women With Career Interruptions, awarded to the author by the School of Postgraduate Studies at the University of Melbourne.

The author would particularly like to thank the mothers and children who volunteered to participate in this study and my supervisor, Associate Professor Margaret Brown. I am also grateful to Romana Morda, who undertook most of the IQ assessments, and to Mary Lawson, Vicki McKenzie, Stuart Dansinger, Fiona Marcum, Rebecca Rossi, and Linda Silverman for additional support with assessments. The CHIP Foundation, Glen Alsop, and Melbourne Municipal Maternal and Child Health Centre staff assisted with recruitment. I would also like to thank staff at the Statistical Consulting Centre at the University of Melbourne for support in data coding and analysis and Helen Skouteris for her valuable suggestions for revisions.

This research draws on data from a longitudinal study, portions of which were presented in a paper at the 17th Annual Conference of the European Early Childhood Education and Research Association on Exploring Vygotksy's Ideas: Crossing Borders, held in Prague, August-September, 2007.

Manuscript received October 2009; accepted July 2010.
Examples of Categories of Maternal Analogical and Metacognitive Verbal
Scaffolding as Applied to a Plastic Cup or Aerosol Lid

Scaffolding            Plastic Cup                  Aerosol Lid

Analogical      Is the tea hot?              Pour some tea in my cup.
                "This upside-down cup        "You're drinking the tea."
                  looks like a rocket."
                "This is just like the       "This is like a cup."
                  cups we had on our
                                             "It's like the lid on
                                               Daddy's shaving cream."
Metacognitive   "We will need a cup."        "We could use this cup" *
                "Yes, that's the right       "What a good idea to use
                  place for the cup."          this as a cup"*

Note. * These utterances are also codable as analogical.

Mann-Whitney Table With Comparisons forAnalogical and Metacognitive
Verbal Scaffolding Frequencies in Sessions 1, 2, and 3

                     HIQ Group       LIQ Group

Scaffolding       Median   Range   Median   Range     U      p

Session 1
  Analogical       1       0-8      0       0-2     19.00   0.01
  Metacognitive    0       0-10     0       0-5     39.00   0.29
Session 2
  Analogical       8       0-31     2       0-10    31.50   0.05
  Metacognitive    9       0-11     0.5     0-7     15.00   0.00
Session 3
  Analogical       7.5     0-22     5       0-22    43.00   0.31
  Metacognitive    4.5     0-16     2.5     0-16    39.50   0.21

Note. Tabachnik, B. G., & Fidell, L. S. (2001). Using multivariate
statistics (4th ed.). Needham Heights, MA: Allyn & Bacon.
Gale Copyright:
Copyright 2011 Gale, Cengage Learning. All rights reserved.