Motor Skill Interventions to Improve Fundamental Movement ...€¦ · FMS Interventions 211 A key developmental aim among preschool-aged children (ages 3–5) is the development of

210

Adapted Physical Activity Quarterly, 2011, 28, 210-232© 2011 Human Kinetics, Inc.

Megan A. Kirk and Ryan E. Rhodes are with the University of Victoria, School of Exercise Science, Physical and Health Education in Victoria, British Columbia, Canada.

Motor Skill Interventions to Improve Fundamental Movement Skills

of Preschoolers With Developmental Delay

Megan A. Kirk and Ryan E. RhodesUniversity of Victoria, Canada

Preschoolers with developmental delay (DD) are at risk for poor fundamental movement skills (FMS), but a paucity of early FMS interventions exist. The pur-pose of this review was to critically appraise the existing interventions to establish direction for future trials targeting preschoolers with DD. A total of 11 studies met the inclusion criteria. Major findings were summarized based on common subtopics of overall intervention effect, locomotor skill outcomes, object-control outcomes, and gender differences. Trials ranged from 8 to 24 weeks and offered 540–1700 min of instruction. The majority of trials (n = 9) significantly improved FMS of preschoolers with DD, with a large intervention effect (η2 = 0.57–0.85). This review supports the utility of interventions to improve FMS of preschoolers with DD. Future researchers are encouraged to include more robust designs, a theoretical framework, and involvement of parents and teachers in the delivery of the intervention.

Keywords: preschoolers, developmental delay, motor skill interventions, funda-mental movement skills

The early childhood years represent an important time period where funda-mental physical, social, and cognitive skills are formed. Children identified with developmental delay (DD) do not form these fundamental skills at the same pace as typically developing children and, consequently, are at an even greater risk of compromised health and further delays in social, emotional, and cognitive develop-ment across the lifespan (Centers for Disease Control and Prevention, CDC, 2009; Majnemer, 1998). DD refers to a nonpermanent chronological delay (e.g., physi-cal, intellectual, social) in achieving typical milestones expected at a certain age (Statistics Canada, 2001). In Canada, DD has been identified as the most common disabling condition among young children (ages 3–5). Among the nearly 20,000 children identified with a disability in Canada, 68% of those cases were because of a DD diagnosis (CDC, 2009; Statistics Canada, 2001). Thus, research efforts focused on improving the social, cognitive, and physical development of young children with DD is a potential public health priority.

FMS Interventions 211

A key developmental aim among preschool-aged children (ages 3–5) is the development of proficient fundamental movement skills (FMS). FMS are con-sidered the building blocks for future complex motor skills, e.g., sport-specific participation, habitual physical activity (PA), where greater focus can be placed on the health outcomes of PA (Temple, Naylor, Rhodes, & Wharf Higgins, 2009; Timmons, Naylor, & Pfeiffer, 2007). FMS are comprised of locomotor and object control skills. Locomotor skills require the movement of the body through space and include running, hopping, jumping, skipping, galloping, sliding, and leaping (Stodden et al., 2008). Object control skills are those that require the manipulation of tangible objects and include throwing, catching, bouncing, striking, kicking, and rolling (Stodden et al., 2008). The development of proficient FMS is not a naturally occurring process and requires sufficient time, instruction, and reinforcement by educators, parents, and health professionals to ensure children are appropriately engaging in movement skills that build FMS (Stodden et al., 2008).

Children identified with DD lack the motor skill competency necessary to perform FMS and are at risk for poor physical, social, and emotional functioning later in life (Majnemer, 1998). From a PA perspective, motor skill competence during the early childhood years (ages 3–5) has demonstrated to track and predict future physical activity (PA) and sport participation (Telama, Yang, Laakso, & Viikari, 1997; Telama et al., 2005). Thus, children with DD who have not developed proficiency in FMS are at risk for having poor physical functioning and inactiv-ity extending across the lifespan. Therefore, children with DD especially need remediation, intervention, and instruction to develop competence in FMS to help promote overall daily functioning and future PA and sport participation (Stodden et al., 2008; Strong et al., 2005).

From a public health standpoint, early intervention strategies that target FMS development among children with DD can (a) help minimize and remediate exist-ing DD disorders, (b) improve overall daily functioning, and (c) promote lifelong participation in health promoting behaviors (e.g., PA; Majnemer, 1998). Prior intervention efforts have focused heavily on typically developing preschool-aged children or integrated samples that do not differentiate between those with and without DD (Riethmuller, Jones, & Okely, 2009). Thus, our understanding of the efficacy of interventions in improving motor skills of preschool children with DD is limited and supports the need for a review. To our knowledge, no known systematic review explicitly examining the effect of motor skill interventions on motor skill development of preschool-aged children with DD has been conducted. Therefore, the purpose of this study is to examine and critically appraise the existing litera-ture investigating the impact of a motor skill intervention on FMS development of preschool-aged children with DD to help establish a guiding platform for future research and identify the potential research limitations and potential targets for future PA interventions among this at-risk population.

Method

Inclusion Criteria

Studies investigating the effect of a motor skill intervention on FMS scores of children identified as having DD were initially assessed for eligibility based on

212 Kirk and Rhodes

preestablished criteria set by both reviewers (MK, RR). The population of inter-est was preschool-aged children (ages 3–5) identified as having DD. Eligible studies were selected from published English peer-reviewed journal articles that examined an experimental or quasi-experimental trial and reported the effects of the intervention. Because of a limited literature, studies that also examined single case-reports that described the motor skill outcomes of motor interventions were critically appraised and included in this review.

Exclusion Criteria

Studies were excluded from this review based on preestablished criteria set by both reviewers (MK, RR). A study was excluded if it examined infant (ages < 3) or youth (ages > 6) populations since motor skill development and competency differs from that of preschool-aged children. Excluded studies were also those that (a) did not exclusively describe the gross motor outcomes of a sample of children defined as having a developmental delay (e.g., integrated sample of children with and without developmental delay) because the results may not exclusively generalize to the target population and (b) examined gross motor skills among preschoolers identified as having a specific developmental disability (e.g., cerebral palsy, Autism, Down Syndrome, ADHD, visual impairments), since the gross motor development of children with permanent disabilities may require unique intervention procedures (e.g., intensive physical therapy, unique instruction, adapted equipment) that differ from those used for children with DD.

Search Strategy

Based on previous recommendations (Egger, Smith, & Altman, 2001), database searches were conducted from March 2010 to August 2010 in 6 online databases: Academic Search Premier, ERIC, ISI Web of Knowledge, MEDLINE, SPORTDis-cus, and CINAHL. The literature search strategy was developed by both authors (MK, RR) and was not restricted to year of publication or study design. Various combinations of key words were used including physical activity intervention, gross motor skills, fundamental movement skills, AND intervention studies AND preschoolers AND developmental delay. To ensure saturation of the literature, manual cross-referencing of reference lists of relevant articles was also completed.

Screening

Citations, including the title and abstract, were independently screened by both authors (MK, RR). Potentially relevant and ambiguous studies were retrieved in full and judged against the predefined inclusion criteria. Potential studies for inclusion were compared, discussed, and agreed upon by both reviewers (RR, MK). Discrepancies were resolved through discussion until 100% consensus in all cases was reached.

Study Quality Assessment

The study quality and risk of bias of all relevant studies was assessed by both authors using a modified version of (Downs & Black, 1998) validated quality assessment


tool. The assessment tool evaluates 5 areas of study quality: reporting, external validity, internal validity—risk of bias, internal validity—confounders, and power. A 27-item checklist that includes a yes/no response assesses the quality of each category: reporting = 10 items, external validity = 3 items, internal validity—bias = 7 items, internal validity—confounding = 6 items, power = 1 item (Downs & Black, 1998). All studies were initially scored out of a maximum of 27 (see Appendix A). Based on predetermined criteria set by both reviewers (MK, RR), the overall quality of the study was then determined based on the maximum score out of 5 to coincide with the 5 quality categories. For reporting quality, studies that scored a minimum of 8 out of 10 were coded as 1; studies scoring less than 8 were coded as 0. For external validity quality, a minimum score of 2 out of 3 was coded as 1; less than 2 was coded as 0. For Internal Validity—Bias Quality, studies that scored a minimum of 5 out of 7 were coded as 1; studies that scored less than 5 were coded as 0. For Internal Validity—Confounding Quality, studies that scored a minimum of 5 out of 6 were coded as 1; studies that scored less than 5 were coded as 0. For Power, studies that scored 1 out of 1 were coded as 1; studies that scored 0 were coded as 0. Both authors independently coded and assessed the overall study quality. Discrepancies were resolved through discussion until 100% agreement in all cases was reached. High quality studies were those that scored five, moderate quality studies scored three to four, and low quality studies were those that scored two or less (see Appen-dix A). The overall quality of the studies was reported to describe the general state of the research on the topic. Because of the limited research available, all studies, regardless of quality rating, were included in this review to help establish current limitations and future methodological recommendations for future interventions.

Data Abstraction and AnalysisA 10-item data abstraction form that included the authors, year, study design, par-ticipants, developmental delay measurement tool, intervention length, intervention approach, intervention setting, gross motor outcome measure, results, and study quality was used to abstract the data (see Table 1). Subtopics and themes were identified and categorized based preestablished criteria set by both reviewers that followed previous recommendations (Sallis, Prochaska, & Taylor, 2000). Due to the limited number of studies that passed the inclusion criteria, the major findings were highlighted if they were present in a minimum of two independent studies. Each study was initially scanned to identify what themes were present across all studies. The study characteristics, intervention design characteristics, overall intervention effect, locomotor skill outcomes, object control skill outcomes, and gender differences were discussed.

The major findings from each study including the subtopics and emerging themes are discussed and synthesized in a narrative review with particular focus on whether the intervention was effective in changing gross motor scores. The qualita-tive appraisal included the overall intervention effect, locomotor skill outcomes, object control outcomes, and gender differences. The quantitative appraisal of studies included discussing and summarizing the necessary statistical information that was available. Reporting decisions were based on (a) significant/null findings (p < .05) and (b) at least a small effect size (η2 = 0.009) using standardized prede-termined criteria (Cohen, 1992). Homogeneity across study methods and measures was inadequate to perform a meta-analysis (Hunter & Schmidt, 2004).

214

Table 1 Summary of the 11 Articles Reviewed

Authors Study Design Participants DD MeasureIntervention

LengthIntervention Approach

Robinson, L.E. & Good-way, J.D. (2009)

USA

RCT, multiple interven-tions, with a reten-tion test

N = 117;

63 male, 54 female

TGMD2

At or below the 30th per-centile

9-weeks

2×/wk for 30-min ses-sions; 24-min motor-instruc-tion

Researcher admin-istered

TARGET

(a) Low Autonomy: instructor directed, task specific

(b) Mastery Moti-vational Climate: child-directed, instructor –facili-tated

Apache, R.R.G. (2005)

USA

Experi-mental, 2 treatment trial

N = 28; 13 male, 15 female, ages 3–6

U.S. State Criteria

Placement in a special edu-cation pre-school based on classified DD

15-weeks

3×/wk for 30 min

Researcher administered

(a) Activity-Based program: child-initiated, teacher-facilitated

(b) Direct Instruc-tion program: task-specific, teacher directed

Valentini, N., & Rudisill, M.E. (2004)

USA

Experimental, 2 treatment trial

N = 39

(25 female, 14 male)

TGMD


12-weeks

2×/wk, 35 min

GM Specialist administered

TARGET

1) Mastery Moti-vational Climate: child-directed, instructor -facili-tated

2) Low Autonomy: instructor directed, task specific

215

SettingOutcome Measure Results

QR(/5)

School TGMD2

6 object con-trol skills

Retention Test 9 weeks fol-lowing post

intervention testing

Object Control: Significant effect of time, F(2, 228) = 647.96, p .001, η2 = 0.85); treatment, F(2, 114) = 76.83, p = 0.001, η2 = 0.57); and treatment × time interaction, F(4, 228) = 156.13, p = .001, η2 = 0.73). The LA and MMC participants showed significant improve-ments in OC scores from pretest to posttest, pretest to retention, and significant decreases in OC scores from posttest to retention (p = 0.001). No significant change in OC scores over time among the con-trol group (p > 0.05). At posttest, statistically significant differences found in OC scores with the LA and MMC not differing significantly (p = 0.60), but both had OC scores significantly higher than control group (p = 0.001). Overall, the LA group improved from the 19th percentile to the 77th percentile, and the MMC group improved from the 18th percentile to the 84th percentile.

5

School TGMD2

12-items measured

(7 locomotor, 5 object con-trol)

Locomotor: Significant instruction × time interaction for locomotor skills, F(2,26) = 15.5, p .01, η2 = 0.89). Activity-based intervention significantly improved locomotor skills from pre- to posttest, t(27) = 14.33, p < .001, with no significant increase in skills found in the direct-instruction program. Locomotor skills were significantly higher in the Activity-based intervention group compared with the direct instruction group, t(27) = 7.12, p < .001) at posttest.

Object Control: Significant instruction × time interaction for object control, F(2,26) = 6.54, p < 0.001, η2 = 0.87. Those in the activity-based intervention significantly improved object control skills from pre to posttest, t(27) = 16.31, p < .001. No significant improvement in the direct-instruction program (p > 0.05). Object control scores were significantly higher in the activity-based intervention that direct-instruction, t(27) = 6.97, p < .001.

Intervention Effect: Significant instruction effect, F(2, 23) = 53.18, p < .001, η2 = 0.85 found.

Gender: No significant effects or interactions found.

2

School TGMD

12-items measured (7 locomotor, 5 object con-trol)

Locomotor: Significant group effect (F(1,37) = 3.95, p = .05, η2 = .096), group × time interaction (F(1,37) = 4.41, p = .04, η2 = 0.11), and time effect (F(1,37) = 415.75, p = .0001, η2 = 0.91). Locomotor skills improved significantly for both MMC (t(18) = 22.05) and LA group (t(19) = 10.81, p = .0001). Significant differences (p = .001) in locomotor skills postintervention found, with the MMC group (M = 13.55, SD = 1.50) performing better than LA group (M = 12.30, SD = 1.92).

Object Control: Significant time effect (F(1,37) = 237.10, p = .0001, η2 = 0.86). Group effect and interaction were n/s (p > .05). OC increased from pre (M = 8.88, SD = 1.93) to postintervention (M = 13.11, SD = 2.34). MMC and LA groups performed similarly.

3

(continued)

216



Valentini, N. & Rudisill, M.E. (2004)

USA

NRCT N = 56;

30 male, 26 female

TGMD


12-weeks

2×/wk, 35 min

GM Specialist administered

TARGET

(a) Mastery Moti-vational Climate: child-directed, instructor - facilitated

Buccieri, K.M. (2003)

USA

Case-Report N = 1;

Female age 2.5

PDGMS

2-SD below the mean. Motor skills in 1st percen-tile.

6-months

1×/wk, 60 min

Therapist administered

1:1 instruction by PT; Physical therapy exercise program

Goodway, J.D., et al. (2003)

USA

NRCT N = 63; 29 male, 34 female, ages 3–5, 84.7% Hispanic

TGMD2

At or below the 7th per-centile for locomotor and below the 13th percen-tile for object control.

9-weeks

2×/wk, 35 min


Direct instruction

Curriculum

Table 1 (continued)

217


QR(/5)

School TGMD

12-items measured (7 locomotor, 5 object con-trol)

Retention Test

6-months fol-lowing post-intervention testing

Locomotor: Significant group × time interaction, F(2, 53) = 11.34, p = .0001, η2 = 0.30; group effect, F(1, 54) = 7.21, p = 0.01, η2 = 0.12; and time effect, F(2, 53) = 284.10, p = 0.0001, η2 = 0.92. Locomotor skills increased significantly from pre to postintervention for both the mastery and control group. Locomotor scores were sig. higher in the intervention group compared with the control group.Object Control: Significant group × time interaction, F(2, 53) = 12.31, p = 0.0001, η2 = 0.32; time effect, F(2, 53) = 49.31, p = 0.000, η2 = 0.65. The group effect was non/sig, F(1, 54) = 11.34, p = 0.12, η2 = .002. OC skills increased significantly from pre- to postinter-vention for the mastery and control groups.Follow-up: From postintervention to follow-up, the master group (postintervention M = 12.19, SD = 1.57 vs. follow-up M = 11.70, SD = 2.28) maintained their skill development and the control group significantly decreased (postintervention M = 11.14, SD = 1.24 vs. follow-up M = 9.50, SD = 1.84).From postintervention to follow-up—no sig. change in OC skills for mastery group, but significant decreases in OC skills were found for the control group. Significant difference was found at follow-up intervention, F(1, 54) = 11.59, p = 0.001. The mastery group (M = 13.37, SD = 2.07) performed better than the control group (M = 11.35, SD = 2.36).

Home PDGMS

5-items; reflexes, bal-ance, non-locomotor, locomotor, receipt and propulsion

PDGMS: Total score improved 12-months age equivalency following intervention. Gross motor score advanced from z = -2.33 to z = -.74 (standard deviation below the mean).Improvement in all skill categories improved with z-scores < 1 stan-dard deviation from the mean for children of similar age. Participant improved from 1st percentile to 23 percentile in PDGMS.

0

School TGMD2

12-items measured

(7 Locomo-tor, 5 Object Control)

Locomotor: Significant group × time interaction for locomotor skills, F(1,61) = 101.04, p < .001, η2 = 0.63. Intervention group signifi-cantly improved locomotor skills, t(32) = -19.71, p < .001, but the comparison group did not (p > .05). The intervention group had significantly better posttest locomotor scores than the control group, t(61) = 8.49, p < .001.Object Control: Significant group × time interaction for object con-trol skills, F(1, 61) = 99.05, p < .001, η2 = 0.63. Intervention group improved significantly in object control, t(32) = -13.60, p < .001, control group did not. The intervention group performed significantly better than the control group on object control skills at posttest, t(61) = 6.98, p < .001.Intervention Effect: The effect of the intervention was strong, η2 = 0.75, with the intervention group improving from the 7th to 50th percentile in locomotor and 11th to 60th percentile in object control.Gender: No significant effects of interactions found.

3

(continued)

218



Goodway, J.D. & Branta, C.F. (2003)

USA

NRCT N = 59;

29 male, 30 female, 4-years old, African American

US State Criteria/ Self-Report

Low scores on preschool readiness test

Mental and motor skill delays

12-weeks

2×/wk, 45 min


Direct Instruction Curriculum

Hamilton, M. et al. (1999)

USA

NRCT N = 27;

16 male, 11 female, ages 3–5, African American

TGMD

Below 20th percentile of object-control scales

8-weeks

2×/wk, 45 min

Researcher/ Parent administered

Direct Instruction Curriculum

Young, H.J. & Lewis, C.L. (1998)

USA

Case-report N = 1;

Male, 3yrs/10m

PDGMS

4th percentile for motor skills

6-weeks

3×/wk for each component; duration based on completion of tasks

Therapist/ Parent administered

Physical therapy

1) Aerobic

2) Strength

Zittlel, L.L., & McCubbin, J.A. (1996)

USA

Experimental Single-subject reversal design

N = 4;

3 male, 1 female, ages 3–5

TGMD

1.5 standard deviations below the mean

24-weeks

1×/wk, 40 min

Researcher/ Teacher administered

I-CAN program:

child-directed

Table 1 (continued)

219


QR(/5)

School TGMD2

12-items measured

(7 locomotor, 5 object con-trol)

Locomotor: Significant group × prepostintervention interaction for locomotor skills with a large intervention effect, F(1, 57) = 134.23, p < .001, η2 = 0.70. MSI group, t(30) = -21.69, p < .001 (2-tailed) and control group, t(28) = -3.83, p = 0.001, (2-tailed), both improved sig-nificantly in locomotor skills from pre- to postintervention. The MSI group had significantly better postintervention locomotor scores than the control group, t(57) = 13.11, p < .001 (2-tailed). 3% of MSI at or below 50% for locomotor skills compared with 93% of control Ps.

Object Control: Significant group × prepostintervention interac-tion with a large intervention effect, F(1,57) = 161.55, p < .001, η2 = 0.74. MSI group, t(30) = -20.49, p < .001, (2-tailed), and control group, t(28) = -3.40, p = 0.002, (2-tailed) significantly improved object control skills from pre- to postintervention. The MSI groups had significantly higher postintervention object control scores, t(57) = 11.74, p < .001, (2-tailed).

4

School TGMD

5-items measured

(5 Object control skills)

Object Control: Experimental group advanced to the 67th percen-tile; control group remained at the 15th percentile postintervention. Significant group effect, F(1, 26) = 12.55, p < .002, η2 = 0.97; test effect, F(1, 26) = 13.18, p < .001, η2 = 1.14; and group by test inter-action, F(1, 26) = 24.18, p < .002, η2 = 1.85. Significant gains in object-control scores among the experimental group with little/no change in the control group. 60% of experimental group performed at or above the 75th percentile following intervention. Only 8% of the control group performed at or above the 50th percentile following the intervention.

1

Home PDGMS

5-items:

reflexes, bal-ance, non-locomotor, locomotor, receipt and propulsion

PDGMS: Total motor score improved 20-months age equivalency following intervention. Gross motor score advanced from z = -1.75 to z = -.77 (standard deviation below the mean). Improvement in all skill categories improved with z-scores < 1 standard deviation from the mean for children of similar age. Participant improved from 4th percentile to 22nd percentile in PDGMS.

0

School Study-Created

4-items measured

(2 locomotor, 2 Object con-trol)

Locomotor: No significant improvements in skill level in the inte-grated condition, p < .05. Skill level remained stable.

Object Control: Skills remained stable. No significant differences based on segregated or integrated condition, p < .05.

0

(continued)

220 Kirk and Rhodes

Table 1 (continued)



DeGangi, G.A. et al. (1993)USA

Experi-men-tal, repeated intervention (A-B cross-over)

N = 12;

36-71 months; 10 male, 2 female

PDGMS

1 standard deviation below the mean

16-weeks

8-weeks per intervention

1x/wk, 60 min

Therapist administered

(a) Child-centered activity: child-initiated, instructor facilitated

(b) Structured Sen-sorimotor Therapy: instructor-directed, task-orientated

Jenkins, J.R. et al. (1983)USA

Experi- mental, 2 treatments

N = 44;ages 3–5; 33 male, 11 female

PDGMS

High need—motor quo-tient < 80; Low need—motor quo-tient > 80

17-weeks

Sensory Motor Integration = 2×/wk, 25 min

Motor Lessons = 4×/wk 25 min

Therapist/ Assis-tant administered

(a) ST: individu-alize programs, direct-instruction, 1:1

(b) Motor Lessons: direct-instruction, class setting

Note. Bold text = valid/reliable tool; TGMD – Test of Gross Motor Development, PDGMS – Peabody Developmental Gross Motor Scale, RCT—randomized control trial, NRCT—nonrandomized control trial, MMC—mastery motivational climate, LA—low autonomy, TARGET – Task, Authority, Recognition, Grouping, Evaluation, Time, OC—object control,

ResultsOverall, 30 potentially relevant records were retrieved from the database search. Based on the inclusion criteria, 24 potentially relevant abstracts and full-text articles were obtained and reviewed. Overall, 11 articles examining 11 independent samples passed the eligibility criteria and were included in this review (see Figure 1).

Study Characteristics

Table 2 presents a detailed summary of the intervention studies. In terms of qual-ity rating, only one study was identified as being of high quality (Robinson & Goodway, 2009), three studies were identified as moderate quality (Goodway & Branta, 2003; Goodway, Crowe, & Ward, 2003; Valentini & Rudisill, 2004), and the remaining seven studies were scored as low quality (Apache, 2005; Buccieri, 2003; DeGangi, Wietlisbach, & Scheiner, 1993; Hamilton, Goodway, & Hauben-stricker, 1999; Jenkins, Fewell, & Harris, 1983; Young & Lewis, 1998; Zittle & McCubbin, 1996). Overall, the studies represented a total of 395 participants (211 male and 184 female). Sample sizes ranged from 1 to 117 preschool children with a mean sample size of 36. Developmental delays were most commonly assessed by prevalence of low test scores (e.g., at or below 30th percentile) on the Test of Gross Motor Development (TGMD; n = 5), followed by prevalence of low test



QR(/5)

Therapy Facility

PDGMS PDGMS: Significant improvements in gross motor skills after ST, t = -2.966, p =.016, found for all subjects. SST was more useful than CCA in improving motor skills.

Therapy Facility

PDGMS PDGMS: No significant main effects or interactions found for both groups.

1

T1—time one, T2—time two, MSI—motor skill intervention, QR—quality rating, ST—structured sensorimotor therapy, CCA—child-centered activity.

scores (e.g., 2 standard deviations below the mean, 1st percentile) on the Pea-body Developmental Gross Motor Scale (PDGMS; n = 4) and predefined U.S. state specific criteria (n = 2). Motor skill outcomes of preschool children were measured using the Test of Gross Motor Development (TGMD; n = 6), Peabody Developmental Gross Motor Scale (PDGMS; n = 4), and a study-created measure (n = 1).

Intervention Characteristics

Research Design. Of the eleven studies, the majority conducted nonrandomized control trials (n = 3), followed by a quasi-experimental two-treatment intervention design (n = 3), a single case-report design (n = 2), and other experimental designs (n = 2). Only one study conducted a randomized control trial. Authors reported that true randomized control trials were difficult to conduct in the majority of the studies due to the educational structure and nature of the preschool setting.

For the five control trials, all studies indicated that the intervention and control groups were comprised of children identified with DD that attended a community-based preschool. Two of the studies reported that the preschool setting used for recruitment was part of the U.S. Head Start program (Hamilton et al., 1999; Rob-inson & Goodway, 2009). None of the five studies indicated that the control group would be receiving the motor skill intervention following the study, but all studies

222

Fig

ure

1 —

Sys

tem

atic

flow

-cha

rt o

f th

e lit

erat

ure

sear

ch.

223

Table 2 Characteristics of Included Studies (N = 11)

Characteristic Value

Study Design

RCT, N (%) 1 (9.1)

NRCT, N (%) 4 (36.4)

Two-treatment Trial, N (%) 3 (27.3)

Single Case-Report, N (%) 2 (18.2)

Other Experimental, N (%) 2 (18.2)

Independent samples, N (%) 11 (100)

Sample size, M 35.9

Sample size, min 1

Sample size, max 117

Population Characteristics

Study Location (country), N (%)

USA 11 (100)

Outcome Measures, N (%)

TGMD 7 (63.6)

PGMDS 3 (2.3)

Study-created measure 1 (9.1)

Intervention Approach, N (%)

Therapy-based program 2 (18.2)

Direct-Instruction 3 (27.3)

Child-Initiated/Teacher-Facilitated 1 (9.1)

Multiple Approaches 5 (45.5)

Intervention Setting, N (%)

Preschool Classroom 7 (63.6)

In-home Setting 2 (18.2)

Therapy Facility Setting 2 (18.2)

Community-based Setting 0 (0)

Intervention Implementation, N (%)

Primary Researcher/Author 4 (36.4)

Therapist/Specialist 3 (27.3)

Mixed (e.g., parents and therapist) 4 (36.4)

Intervention Length (weeks), M 13.7

Intervention Length (weeks), min 6

Intervention Length (weeks), max 24

Total Instruction Duration (minutes), min 540

Total Instruction Duration (minutes), max 1700

Note. RCT – randomized control trial, NRCT – nonrandomized control trial, TGMD –Test of Gross Motor Development, PDGMS –Peabody Developmental Gross Motor Scale.

224 Kirk and Rhodes

reported that the control group would be receiving the typical play/recess program they were accustomed to.

Intervention Approach. The type of intervention approach used in the studies varied. Eight of the eleven interventions (72.7%) implemented an instructor-directed approach that was task specific. Five studies (45.5%) implemented a child-directed approach were the child initiated learning, and three studies (27.3%) followed a physical/occupational therapy based approach. Overall, five studies implemented both instructor-directed and child-directed interventions to compare gross motor outcomes (Apache, 2005; DeGangi et al., 1993; Jenkins et al., 1983; Robinson & Goodway, 2009; Valentini & Rudisill, 2004). Intervention objectives were most commonly developed based on preestablished guidelines and procedures: physical therapy (n = 4), education curriculum guidelines (n = 3), TARGET (n = 2), I-CAN (n = 1). One study followed guidelines developed by the primary researcher.

Intervention Implementation. All studies described the intervention implemen-tation procedures. The intervention trials were most commonly delivered in the school setting (n = 7); two studies examined an in-home intervention trial and two studies were conducted in a therapy setting. The primary researcher (n = 4) and therapist/specialist (n = 3) delivered the intervention in the majority of studies. For the remaining four studies, the researcher/therapist worked with parents, research assistants, and preschool teachers to implement the intervention. The intervention length and hours of instruction provided to participants varied. Intervention trials ranged from 8 to 24 weeks with an average duration of 13.7 weeks. The weekly frequency of instruction ranged from one to four times per week, with only one study administering the intervention program four times per week. Overall, the intervention trials offered between 540–1700 min of total instruction time.

Overall Efficacy of Intervention Trials

Overall, nine of the eleven (81.8%) studies reported improvements in gross motor skills among preschoolers with DD at postintervention testing, with six studies reporting significance at the p < .05 level (Apache, 2005; Buccieri, 2003; DeGangi et al., 1993; Goodway & Branta, 2003; Goodway et al., 2003; Jenkins et al., 1983; Robinson & Goodway, 2009; Valentini & Rudisill, 2004; Young & Lewis, 1998). Two studies found no significant difference (p > 0.05) in motor skill scores based on the intervention condition (Jenkins et al., 1983; Zittle & McCubbin, 1996); the results may be attributed to the poor study quality. The magnitude of the interven-tion effect was reported in three studies and ranged from η2 = 0.57 to η2 = 0.85, indicating a strong/large effect of the motor skill intervention in improving gross motor skills (Apache, 2005; Goodway et al., 2003; Robinson & Goodway, 2009).

In terms of motor skill retention over time, only two studies investigated par-ticipants’ motor skill competence following postintervention measures (Robinson & Goodway, 2009; Valentini & Rudisill, 2004). The follow-up measures were conducted at nine weeks (Robinson & Goodway, 2009) or six months (Valentini & Rudisill, 2004) after postintervention testing. From postintervention to retention, one study found significant decreases in scores from postintervention to retention among intervention groups (Robinson & Goodway, 2009), and one study found no


significant change in motor skill scores at follow-up (Valentini & Rudisill, 2004). Both studies reported that the intervention groups had significantly higher scores than the control group at retention, suggesting that the intervention is efficacious in maintaining and developing motor skills. Despite these promising findings, however, more research is needed to substantiate the findings.

Locomotor Skill Outcomes

Locomotor skill outcomes were assessed in nine of the eleven studies (Apache, 2005; Buccieri, 2003; DeGangi et al., 1993; Goodway & Branta, 2003; Goodway et al., 2003; Jenkins et al., 1983; Valentini & Rudisill, 2004; Young & Lewis, 1998; Zittle & McCubbin, 1996). Seven of the nine studies reported significant improvements in locomotor skills following the intervention among the intervention group and treat-ment effects ranged from η2 = 0.10 to 0.85 (Apache, 2005; Buccieri, 2003; DeGangi et al., 1993; Goodway & Branta, 2003; Goodway et al., 2003; Valentini & Rudisill, 2004; Young & Lewis, 1998). For control trials, all three studies reported that the intervention group performed significantly better on locomotor tests compared with the control group at postintervention testing (Goodway & Branta, 2003; Goodway et al., 2003; Valentini & Rudisill, 2004). For studies comparing two treatments, both studies reported significantly higher locomotor scores for the child-initiated/directed approach compared with the instructor-directed approach (Apache, 2005; Valentini & Rudisill, 2004). Running and jumping emerged as the most common locomotor skills that saw the greatest improvement. Among the studies that reported locomotor performance norms pre- and postintervention, locomotor performance was found to be at or above the 50th percentile following the intervention, indicat-ing that the intervention helped preschool children with developmental delays to perform at a level similar to typically developing preschoolers (Goodway & Branta, 2003; Goodway et al., 2003; Hamilton et al., 1999).

Object Control Skills

Overall, all eleven intervention trials assessed and reported object control skill outcomes. Nine of the studies reported significant improvements in object-control skills following the intervention among treatment groups with intervention effects ranging from η2 = 0.002 to 0.97 (Apache, 2005; Buccieri, 2003; DeGangi et al., 1993; Goodway & Branta, 2003; Goodway et al., 2003; Hamilton et al., 1999; Rob-inson & Goodway, 2009; Valentini & Rudisill, 2004; Young & Lewis, 1998). The interaction effect of intervention and time for object-control skills was reported in six studies and ranged from η2 = 0.32–1.85, indicating a strong/large effect of the intervention on posttest scores (Apache, 2005; Goodway & Branta, 2003; Goodway et al., 2003; Hamilton et al., 1999; Robinson & Goodway, 2009; Valentini & Rud-isill, 2004). For control trials, four of the five studies reported that the intervention group performed significantly better on locomotor tests compared with the control group at postintervention testing (Goodway & Branta, 2003; Goodway et al., 2003; Hamilton et al., 1999; Robinson & Goodway, 2009; Valentini & Rudisill, 2004). For studies comparing two treatments, one study found the child-initiated/directed intervention to produce significantly better object control skills than the instructor-directed intervention (Apache, 2005), and one study found no significant differences

226 Kirk and Rhodes

b/n treatments (Valentini & Rudisill, 2004). The strike and throw emerged as the most common object control skills that showed the greatest improvement in performance among the participants. Object-control skill performance among the intervention groups improved to a level at or above the 50th percentile following the intervention (Goodway & Branta, 2003; Goodway et al., 2003; Hamilton et al., 1999; Robinson & Goodway, 2009).

Gender Differences in Motor Skill Outcomes

Only two of the eleven studies reported and tested for differences in gross motor scores among male and female participants following the intervention (Apache, 2005; Goodway & Branta, 2003). No significant main effect or interactions were found based on gender.

DiscussionThe purpose of this review was to summarize and critically appraise the existing literature pertaining to the efficacy of motor skill interventions in improving FMS of preschool-aged children identified with DD. A total of eleven studies were identified as meeting the eligibility criteria and were critically appraised in the review. Only one study met the criteria for “high quality.” None of the remaining ten studies included random assignment of participants to the intervention group or adequately controlled for potential confounders and, therefore, were rated as “moderate” or “low” quality. Overall, 81% of the studies identified significant improvements in motor skill scores following an intervention with locomotor skills showing the largest improvements. Significant gains in both locomotor (e.g., run-ning, jumping) and object-control (e.g., throwing, catching) skills to a performance level similar to typically developing preschoolers was found among the majority of controlled intervention trials. Thus, this review provides preliminary support of the effectiveness of early motor skill interventions in improving gross motor skills among preschoolers with DD.

The findings from this review are in congruence with previous research inves-tigating the effectiveness of motor skill interventions on gross motor development of typically developing preschool-aged children (Riethmuller et al., 2009; van Sluijs, McMinn, & Griffin, 2008). In support of our findings, van Sluijs et al. (2008) identified three PA interventions that targeted children from low SES backgrounds. Of these, all three reported significant positive intervention effects. Of particular importance, however, our review highlights that the intervention effect of motor skill programs may be even larger and more consistent among preschoolers with DD than those studies exclusively examining motor outcomes among typically developing children (Riethmuller et al., 2009; van Sluijs, et al., 2008). Future researchers are encouraged to implement early motor skill interventions.

The findings from this review highlight that the setting and type of intervention approach are important considerations for future FMS interventions. It appeared that the school setting was the most opportune location to help improve FMS compared with home-based and therapy-based settings. The school environment may provide greater opportunities for motor skill development as a result of greater access to equipment and space than perhaps the child’s home environment. In addition, inter-


vention approaches that emphasize child-initiated learning appeared to produce the most significant gains in FMS among children identified with DD compared with instructor-directed approaches. More specifically, the intervention strategies that focused developmentally appropriate progressions in motor skill acquisition and tailored the session toward how the child best learns and processes information appeared to have promising results. Further research investigating the effective-ness of a child-centered approach toward improving and maintaining FMS among children with DD is encouraged to help build the preliminary evidence. Overall, use of a motor skill intervention to assist preschoolers with DD to become profi-cient in FMS is supported by this review. Thus, targeted interventions are urgently needed to prevent habitual inactivity and adverse health consequences among this at-risk population.

Of particular interest, only two of the studies conducted follow-up retention tests after postintervention measures. While both studies concluded that motor skill competences remained significantly higher among the intervention group compared with the control group, the lack of research examining retention effects is concerning since the overall effectiveness of an intervention is determined by the long-term follow-up effects. The findings from this review provide initial support for the efficacy of FMS interventions among children with DD, but further retention measures are urgently needed to establish robust support in this area.

Despite the strengths of motor skill interventions in improving motor skills among preschool children with developmental delays, there are inherent limita-tions to the studies reviewed that must be considered. First, the lack of high-quality studies (e.g., small samples, few RCT trials, poor internal validity) limits our conclusions concerning the efficacy of motor skill interventions in improving FMS among preschoolers with developmental delays. Second, despite increased emphasis on conducting theoretically sound research, few studies included in this review used previously validated intervention approaches to guide the inter-vention trials. Successful interventions are thought to be best implemented if they follow a theoretical framework to help identify the critical determinants of behavior (Baranowski, Anderson, & Carmack, 1998; Rhodes & Pfaeffli, 2009). Therefore, it is expected that even greater improvements in motor skills among preschool children with DD are possible if future trials use theoretically sound frameworks to guide their intervention (e.g., Self-Determination Theory). Third, only one of the studies followed a randomized controlled trial design, and few studies attempted to conduct multicomponent interventions. Prior research has established that parents, teachers, and community organizations play an impor-tant role in the physical development of young children (Stodden et al., 2008; Timmons et al., 2007; Tucker, 2008). Future intervention trials should involve parents and teachers in the direct delivery and design of the intervention as well as attempt to include various settings (e.g., school, home, community). Overall, this review supports the use of more robust designs and longitudinal RCT trials to further strengthen the findings.

It is important that this review also be interpreted within the context of its limitations. First, the literature reviewed was limited to English written journal articles, which prevents the results from being generalized to studies conducted in other languages. Second, this review is limited to the search terms and databases referred to in the methods section. Thus, potentially relevant articles that were

228 Kirk and Rhodes

not identified with the key words used are missing from this review. Based on the evidence included in this review, it appears that motor-skill interventions aimed at increasing FMS of preschoolers with DD, particularly child-directed approaches, are effective. Continued research investigating the long-term outcomes (e.g., health, physical functioning) are needed to confirm the efficacy of early FMS interven-tions in improving overall health and daily functioning from the preschool years to adolescence. Future studies aimed at improving the motor skills of this at-risk population are needed to minimize and remediate their DD and maximize their overall physical, social, and emotional development across the lifespan.

Acknowledgments

Megan A. Kirk is supported by the Canadian Institutes of Health Research Frederick Banting and Charles Best Canada Graduate Scholarship and a University of Victoria President’s Research Scholarship. Ryan E. Rhodes is supported by a scholar award from the Michael Smith Foundation for Health Research, a new investigator award from the Canadian Insti-tutes of Health Research, and with funds from the Social Sciences and Humanities Research Council of Canada and the Human Early Learning Partnership.

ReferencesApache, R.R.G. (2005). Activity-based intervention in motor skill development. Perceptual

and Motor Skills, 100, 1011–1020.Baranowski, T., Anderson, C., & Carmack, C. (1998). Mediating variable framework in

physical activity interventions: How are we doing? How might we do better? American Journal of Preventive Medicine, 15(4), 266–297.

Buccieri, K.M. (2003). Use of orthoses and early intervention physical therapy to minimize hyperpronation and promote functional skills in a child with gross motor delays: A case report. Physical & Occupational Therapy in Pediatrics, 23(1), 5–20.

Centers for Disease Control and Prevention [CDC]. (2009). Child Development. Content from the National Center on Birth Defects and Developmental Disabilities. Retrieved from http://www.cdc.gov/ncbddd/child/default.htm

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155–159. DeGangi, G.A., Wietlisbach, M.G., & Scheiner, N. (1993). A comparison of structured

sensorimotor therapy and child-centered activity in the treatment of preschool children with sensorimotor problems. The American Journal of Occupational Therapy, 47(9), 777–786.

Downs, S.H., & Black, N. (1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology and Community Health, 52, 377–384.

Egger, M., Smith, D.S., & Altman, D.G. (Eds.). (2001). Systematic reviews in health care: Meta-analysis in context (2nd ed.). London: BMJ Books.

Goodway, J.D., & Branta, C.F. (2003). Influence of a motor skill intervention on fundamental motor skill development of disadvantaged preschool children. Research Quarterly for Exercise and Sport, 74(1), 36–46.

Goodway, J.D., Crowe, H., & Ward, P. (2003). Effects of motor skill instruction on funda-mental motor skill development. Adapted Physical Activity Quarterly, 20, 298–314.

Hamilton, M., Goodway, J.D., & Haubenstricker, J. (1999). Parent-assisted instruction in a motor skill program for at-risk preschool children. Adapted Physical Activity Quar-terly, 16, 415–426.


Hunter, J.E., & Schmidt, F.L. (Eds.). (2004). Methods of meta-analysis: Correcting for error and bias in research findings (2nd ed.). Thousand Oaks, CA: Sage.

Jenkins, J.R., Fewell, R., & Harris, S.R. (1983). Comparison of sensory integrative therapy and motor programming. American Journal of Mental Deficiency, 88(2), 221–224.

Majnemer, A. (1998). Benefits of early intervention for children with developmental dis-abilities. Seminars in Pediatric Neurology, 5(1), 62–69.

Rhodes, R.E., & Pfaeffli, L. (2009). Mediators of behaviour change among adult non-clinical populations: A review update. Annals of Behavioral Medicine, 37, s85.

Riethmuller, A.M., Jones, R.A., & Okely, A.D. (2009). Efficacy of intervention to improve motor development in young children: A systematic review. Pediatrics, 124, e782–e792.

Robinson, L.E., & Goodway, J.D. (2009). Instructional climates in preschool children who are at-risk: Part I: Object-control skill development. Research Quarterly for Exercise and Sport, 80(3), 533–542.

Sallis, J.F., Prochaska, J.J., & Taylor, W.C. (2000). A review of correlates of physical activ-ity of children and adolescents. Medicine and Science in Sports and Exercise, 32(5), 963–975.

Statistics Canada. (2001). A profile of disability in Canada: Participation and activity limita-tion survey, 2001. Statistics Canada, Health Reports, Catalogue 89-577-XIE. Retrieved from http://www.statcan.gc.ca/pub/89-577-x/pdf/4228016-eng.pdf

Stodden, D.F., Goodway, J.D., Langendorfer, S.J., Roberton, M.A., Rudisill, M.E., Garcia, C., et al. (2008). A developmental perspective on the role of motor skill competence in physical activity: An emergent relationship. Quest, 60, 290–306.

Strong, W.B., Malina, R.M., Blimkie, C.J.R., Danils, S.R., Dishman, R.K., Gutin, B., et al. (2005). Evidence based physical activity for school-age youth. The Journal of Pediatrics, 146(6), 732–737.

Telama, R., Yang, X., Laakso, L., & Viikari, J. (1997). Physical activity in childhood and adolescence as predictor of physical activity in young adulthood. American Journal of Preventive Medicine, 13(4), 317–323.

Telama, R., Yang, X., Viikari, J., Valimaki, I., Wanne, O., & Raitakari, O. (2005). Physical activity from childhood to adulthood: A 21-year tracking study. American Journal of Preventive Medicine, 28(3), 267–273.

Temple, V.A., Naylor, P.J., Rhodes, R.E., & Wharf Higgins, J. (2009). Physical activity of children in family care. Applied Physiology, Nutrition, and Metabolism, 34, 794–798.

Timmons, B.W., Naylor, P.J., & Pfeiffer, K.A. (2007). Physical activity for preschool chil-dren - How much and how? Applied Physiology, Nutrition, and Metabolism, 32(Suppl. 2E), S122–S134.

Tucker, P. (2008). The physical activity levels of preschool-aged children: A systematic review. Early Childhood Research Quarterly, 23, 547–558.

Valentini, N.C., & Rudisill, M.E. (2004). Motivational climate, motor-skill development, and perceived competence: Two studies of developmentally delayed kindergarten children. Journal of Teaching in Physical Education, 23(3), 216–234.

van Sluijs, E.M.F., McMinn, A.M., & Griffin, S.J. (2008). Effectiveness of interventions to promote physical activity in children and adolescents: Systematic review of controlled trials. British Medical Journal, 42(8): 653-657.

Young, H.J., & Lewis, C.L. (1998). Effects of a home fitness program for a preschool-aged child with gross motor delay. Physical Therapy Case Reports, 1(2), 84–93.

Zittle, L.L., & McCubbin, J.A. (1996). Effect of an integrated physical education setting on motor performance of preschool children with developmental delays. Adapted Physical Activity Quarterly, 13, 316–333.

230 Kirk and Rhodes

Appendix A: Quality of Studies (N = 11)Overall Study Quality

Author ReportingExternal Validity

Internal Validity—Bias

Internal Validity -Confounding Power QR /5

Robinson & Goodway 1 1 1 1 1 5

Apache (2005) 1 0 0 0 1 2

Valentini & Rudisill (2004)

1 0 1 0 1 3

Buccieri (2003) 0 0 0 0 0 0

Goodway & Branta (2003)

1 1 1 0 1 4

Goodway et al. (2003) 0 1 1 0 1 3

Hamilton et al. (1999) 0 1 0 0 0 1

Young & Lewis (1998)

0 0 0 0 0 0

Zittle & McCubbin (1996)

0 0 0 0 0 0

DeGangi et al. (1993) 0 0 0 0 0 0

Jenkins et al. (1983) 0 0 0 0 1 1

Reporting

Author 1 2 3 4 5 6 7 8 9 10

Robinson & Goodway (2009) 1 1 1 1 1 1 1 0 1 1

Apache (2005) 1 1 1 1 1 1 1 0 0 1

Valentini & Rudisill (2004) 1 1 1 1 1 1 1 0 1 1

Buccieri (2003) 1 1 0 1 0 1 0 0 0 0

Goodway & Branta (2003) 1 1 1 1 0 1 1 0 1 1

Goodway et al. (2003) 1 1 1 1 0 1 1 0 0 1

Hamilton et al. (1999) 0 1 0 1 1 1 1 0 0 0

Young & Lewis (1998) 0 0 1 1 0 1 0 0 0 0

Zittle & McCubbin (1996) 0 0 1 1 0 0 0 0 0 0

DeGangi et al. (1993) 0 1 1 1 0 1 0 0 0 1

Jenkins et al. (1983) 0 1 1 1 0 1 0 0 0 0

External Validity

Author 11 12 13

Robinson & Goodway (2009) 1 1 1

Apache (2005) 0 0 1

Valentini & Rudisill (2004) 0 0 1

Buccieri (2003) 0 0 0

Goodway & Branta 1 1 1


External Validity (continued)

Author 11 12 13

Goodway et al. (2003) 1 0 1

Hamilton et al. (1999) 1 0 1

Young & Lewis (1998) 0 0 0

Zittle & McCubbin (1996) 0 0 1

DeGangi et al. (1993) 0 0 0

Jenkins et al. (1983) 1 0 0

Internal Validity—Bias

Author 14 15 16 17 18 19 20

Robinson & Goodway (2009) 0 0 1 1 1 1 1

Apache (2005) 0 0 1 1 0 1 1

Valentini & Rudisill (2004) 1 0 1 1 1 1 1

Buccieri (2003) 0 0 1 0 1 0 1

Goodway & Branta (2003) 0 0 1 1 1 1

Goodway et al. (2003) 0 0 1 1 1 1 1

Hamilton et al. (1999) 0 0 0 1 1 0 1

Young & Lewis (1998) 0 0 0 0 1 0 1

Zittle & McCubbin (1996) 0 0 1 1 0 0 1

DeGangi et al. (1993) 1 0 0 1 1 0 1

Jenkins et al. (1983) 0 0 1 0 1 0 1

Internal Validity—Confounding

Authors 21 22 23 24 25 26

Robinson & Goodway (2009) 1 0 1 1 1 1

Apache (2005) 1 1 0 0 0 1

Valentini & Rudisill (2004) 1 0 0 0 1 1

Buccieri (2003) 0 0 0 0 0 0

Goodway & Branta (2003) 1 1 0 0 1 1

Goodway et al. (2003) 1 0 0 0 0 0

Hamilton et al. (1999) 1 0 0 0 0 0

Young & Lewis (1998) 0 0 0 0 0 0

Zittle & McCubbin (1996) 0 0 0 0 0 0

DeGangi et al. (1993) 0 0 1 0 0 0

Jenkins et al. (1983) 0 0 0 0 0 0

(continued)

232 Kirk and Rhodes

Power

Authors 1

Robinson & Goodway (2009) 1

Apache (2005) 1

Valentini & Rudisill (2004) 1

Buccieri (2003) 0

Goodway & Branta (2003) 1

Goodway et al. (2003) 1

Hamilton et al. (1999) 0

Young & Lewis (1998) 0

Zittle & McCubbin (1996) 0

DeGangi et al. (1993) 0

Jenkins et al. (1983) 1

Documents

Motor Skill Interventions to Improve Fundamental Movement ...€¦ · FMS Interventions 211 A key developmental aim among preschool-aged children (ages 3–5) is the development of