Training Caregivers to Intervene in the Exploratory Play of Young Children with Developmental Delays

Play, particularly exploratory play, can be viewed as the arena in which a series of interactions between a child and the environment occur. However, children with developmental delays appear to be less able to fully engage in exploratory play and utilize the environmental information that such play provides (Krakow & Kopp, 1983). There is also preliminary evidence that caregivers can be trained to use behaviors that facilitate exploratory play when engaged with their children (Belsky, Goode, & Most, 1981). The purpose of the present study was to explore the potential of training caregivers to facilitate the exploratory play of developmentally delayed children using an experim~ntal design. Additionally, relationships between caregiver and child exploratory play behavior are considered. Using preand post intervention videotaped data of 29 mothers and their children, caregiver and child exploratory play was examined across two groups. Fifteen children and caregivers were assigned to a group that provided caregivers with specific training on strategies to enhance exploratory play. The remaining fourteen ·children and caregivers comprised a group that received general information on child development. Results did not indicate any significant differences between these groups in maternal or child behavior, however significant differences in the quality of children's play were noted over time. Additionally changes in the patterns of relationships between maternal behavior and child exploratory behavior are noted over time. Current findings are discussed in the context of previous findings on exploratory play. David Caruso


Introduction
Table 5 Table 6 In the past, play has been called a window into cognitive development. Play, particularly exploratory play, can be viewed as the arena in which a series of interactions between a child and the environment occur. Exploratory play, as defined by Caruso (1985), is behavior which generates or maximizes informational feedback from objects in the environment (Caruso, 1985).
In exploratory play the focus is on initial interactions with relatively novel objects. These interactions provide a child with learning opportunities which contribute to cognitive growth (Belsky & Most, 1980). The motivation to explore the environment through play is considered to be "an integral part of development for all children," (Hauser-Cram, 1996, p. 236).
However, children with developmental delays appear to be less able to fully engage in exploratory play and utilize the environmental information that such play provides (Krakow & Kopp, 1983). Research has suggested that the actions of caregivers who play a primary role in their children's environment can facilitate the exploratory ; play of typically -developing children (Anzalone, 1994).
There is also preliminary evidence that caregivers can be trained to use facilitory behaviors when playing with their children (Belsky, Goode & Most, 1981). The purpose of the present study is to explore the potential of training caregivers to facilitate the exploratory play of developmentally delayed children using an experimental design.
It is hypothesized that such training teaches caregivers to structure 1 child play appropriately, to be responsive to child cues, and to provide useful environmental feedback to children. Children were · · expected to gain exploratory play skills and use those skills more effectively.

Justification For The Study
The present study is designed to contribute to current understanding of exploratory play in a number of ways. First, the study uses an experimental design, rather than the non-experimental designs used primarily in the past, to test hypotheses regarding exploratory play and its relationship to caregiver behavior. Second, the study examines the relationship between several exploratory play variables used by various researchers in the past. Finally, the training used in the study may provide additional support for the widespread trend m early intervention programs to train parents to provide services to their children. It may also provide a model of training and training evaluation that early intervention centers can use for programming.

Review Of Relevant Literature
Play -has long been recognized as an important human endeavor. For example, it is reported that Plato encouraged young boys to play with apples in the belief that the experience would later be beneficial in learning math concepts (Rubin, Fein, & Vandenburg, 1983). Although there is general agreement that play is important, the reason for its importance has not always been agreed upon. Four primary theories have been advanced over the years to explain the 2 importance of play.
The eighteenth century philosopher, Schiller,( as cited in Rubin, Fein, & Vandenburg, 1983) proposed play as a means · of releasing surplus energy. Larazus (as cited in Rubin, Fein, & Vandenburg, 1983) hypothesized that play serves to relieve stress and promote relaxation. G. Stanley Hall stated that play was part of the recapitulation of evolution that all individua,s undergo (as cited in Rubin, Fein, & Vandenburg, 1983). Piaget (1952) proposed that play is a means of consolidating, or integrating, cognitive skills for current and future use.
The theories of Schiller, Larazus, and Hall may be of limited use in explaining actual observed play. For example, none of these theories can account for why tired children still play (Weisler & McCall, 1976;Rubin, Fein, & Vandenburg, 1983). Piaget's work, on the other hand, has provided a valuable framework for conceptualizing play. Piaget was the first to emphasize the importance of self-initiated exploratory play. The basis for Piaget's writings on play are formed within a theory of interactionism.
Piaget's work is based on the belief that the coordination of mind and body is achieved through a series of exchanges between organisms and their environment. Thus, a person gains information about the environment, on how to affect the environment, and applies that information to the environment, which produces more information, which is then used to affect the environment and so on. Cognitive changes in the organism are mediated by the processes of assimilation and accommodation. According to Piaget, many of the early exchanges between an infant and the environment occur within the context of play. Bruner (1973) expanded upon Piaget's conceptualization of play to conclude that play is a forum for the development and the practice of behaviors that are later used in goal directed activity. Belsky and Most (1980) call play a "window" into cognitive development.
A number of other theorists have filled out Piaget's conceptual framework of play.
Drawing information from such diverse sources as animal learning studies and psychoanalytic theory, White (1959) synthesized a innovative view of play. Basically, White postulated children control their environment through play, gaining mastery over it, thereby developing a sense of efficacy or competence. This sense of competence provides the motivation for continued play. White (1959) suggested that the motivation to play is evolutionarily adaptive, as play provides information about the environment and how to affect the environment, which increases the chance of survival. Hunt (1965) and Yarrow, Rubenstein, Pederson and Jankowski ( 1972) extended Piaget and White's con~eptualization of play stating that it is intrinsically motivated, not externally imposed, nor a primary biological drive. In summary, the usefulness of play is theorized to be, in part, in its impact on cognitive development, its relationship to feelings of competence, and its ability to produce the motivation to continue exploration, which may enhance an organism's likelihood of survival. Gaiter and Yarrow (1979), Yarrow, Rubenstein and Pederson (1975) and Hcnir, Speller and West (1985) Caruso (1984Caruso ( , 1986 found that the degree of sophistication of infant play was related to the d~gree of sophistication in problem solving tasks at one and two years of age. Thus, it appears that cognitive development and competence are related to exploratory play behavior.  Krakow & Kopp, 1982;Vietze et al., 1983) began to look at the 6 differences in the play of delayed and non-delayed children.
Early research indicated that both delayed and non-delayed children · obtained the same repertoires of behaviors in the same sequence, with the delayed children progressing at a slower rate of development, as might have been predicted by the work of Zigler (1968). However, upon closer examination, a number of qualitative differences could be seen in the play of delayed and non-delayed .
children. Kopp (1982, 1983) found that delayed Down Syndrome children were less socially oriented than non-delayed children, spent more time unoccupied, and showed limited play repertoire, and less monitoring of environment. Vietze, McCarthy, McQuiston, MacTurk and Yarrow (1983) showed that delayed children engaged in less mastery behavior over the environment.
Berry and Gwinn (1984) found that normally -developing children were more organized and showed less perseveration in behavior than delayed children. In general, delayed children were less likely to be able to extract information from all sources in environment, including other people. These differences occurred even when children were matched for developmental age (Krakow & Kopp, 1982). This research supports the theory that play is important in extracting information from environment and that a limitation in this ability may be related to delays in cognitive development.
The research in the relationship of exploratory play to other cognitive competencies, the developmental sequence of play, and the differences in the play of delayed and non-delayed children all support the theoretical importance of play. However, this research supports only that exploratory play is related to cognitive 7 development not that it is a causal agent in cognitive development.
There may be an underlying factor or a third variable which relates · to both exploratory play and cognitive development. This third variable may be rearing practices.
Researchers such as Yarrow et al. ( 1975) have found significant relationship between rearing practices and IQ.
More recent research has focused on the relationship between exploratory play and rearing practices. Caruso (1986) found that children with responsive mothers were more skilled at play. Berry and Gwinn (1984) indicated that the delayed children with mothers responsive to them were less likely to display aimless behavior. Again, these studies all provide correlational evidence for the link between rearing practices, exploratory play, and cognitive development.
Studies, such as those conducted by Caruso (1986), Berry and Gwinn (1984), and Anzalone (1994) have begun to determine specifically which rearing practices, or in other words, which caregiver behaviors might impact on exploratory play and cognitive development.
As Anzalone ( 1994) states, exploratory play functions as a dynamic action system in which features of the individual (attention span, mastery motivation, cognitive ability) interact with features of , the environment including social context. Interactions with a primary caregiver form the basis of that social context. Anzalone ( 1994) found that maternal behaviors such as attention directing and expansion of children's play repertoire were associated with high developmental levels of play, while maternal intrusiveness was negatively correlated with breadth and sophistication of exploratory play. In general, accurate caregiver response and 8 sensitivity to infant cues during exploratory play appears to be related to a variety of measures that assess the quality as well as the · quantity of infant exploratory play.
An experimental study by Belsky et al. ( 1980) attempted to demonstrate that a maternal rearing practice that is associated with exploratory play could be altered and that altering this practice would in turn produce a change in infant exploratory play. This study divided children into two groups. One group, the control group, simply received three visits from an observer. In the experimental group, the "observer" would comment ("It's interesting, how you ... ") when the mother focused her child's attention on a toy in the course of daily interactions. The authors concluded that this intervention increased maternal attention focusing behavior.
Additionally, children in the experimental group were more likely to demonstrate sustained attention than control children at the end of the intervention. The general conclusion was that it was possible to affect the maternal behavior related to one skill used in play -sustained attention -and that further this change in maternal behavior affected infant behavior. The authors speculated that maternal attention focusing may help children learn to sustain their own attention.
The Belsky et al. study (1980) provides stronger evidence for a causal link between rearing condition and caregiver behavior to exploratory play.
A related line of research (Mahoney, 1988) which demonstrated the reciprocal nature of the exploratory play action system, indicated that mothers of delayed children attempt to modify the interaction pattern with their children (e.g., more 9 directive, more physical cues). These modifications can facilitate infant exploratory play if they are appropriately mediated by · responsiveness to infant cues, as is the case with typically developing children (Anzalone, 1994). Krakow and Kopp (1982) also indicated that delayed children responded to simplified maternal cues at a higher rate than non-simplified cues.
The research evidence presented thus far demonstrates there is a relationship between 1) exploratory play and cognitive development, 2) between rearing conditions and cognitive development, 3) between rearing condition and exploratory play.
There is also preliminary evidence for a causal link between some aspects of rearing conditions and exploratory play.  Kopp (1982, 1983) point out as delayed children attend less to the social information available in their environment, it might be productive for parents to take increased responsibility for initiating and structuring social interaction.
The current study is experimental. The parents of one group of children are trained to initiate and structure social interaction with their delayed children while the other group receives general information on parenting.
The impact of training on parent behavior and child behavior, particularly exploratory play, is then examined.
This study attempts to provide information on a casual link between parent behavior and exploratory play. It also attempts to provide indirect evidence for the importance of the role exploratory play has in cognitive development. This type of study may also be useful m that it is compatible with the current emphasis on training parents to be the provider of services to children needing early intervention.

Research Questions
This study focuses on the design of an intervention conducted with developmentally delayed children and their parents. It is hypothesized that providing parents with training on strategies that 1 1 are related to exploratory play competence would lead to an increase in use of those strategies by parents which in turn, leads to higher developmental levels of play and exploratory competence in children.
Specifically it was predicted that children who participate in the training group would have a higher developmental level of play than children whose caregivers do not receive training on facilitating exploratory play.
Additionally it was expected that the children in the training group would display a higher quality of exploratory play in terms of greater breadth, increased sophistication, and a faster behavioral tempo. Also children in the training group were predicted to have a greater quantity of exploratory play as shown by increased total exploration and increased percentage of time spent in exploration post-training. It was also predicted that caregivers who participated in the training group would display more behaviors that facilitate infant exploratory play (i.e., attention directing, expanding, organizing, supporting infant) and less behaviors that inhibit exploratory play (i.e., intruding, initiating, and removing toys) at the conclusion of the intervention. . The relationship between caregiver behavior and infant exploratory play was also examined pre-and postin terven tion.

Participants
The participants in this study were 29 young children, 18 males and 11 females, as well as their primary caregivers. Prior research (Anzalone,1994) has determined that an infant can initiate and execute a play action in 15 seconds, thus 15 seconds was the unit of time sampling for all coding. Caregiver actions were also analyzed using the same approach (five minutes of play with child, 20 intervals).
The coding system used to derive these variables were based on methods described by Anzalone ( 1994) and Caruso ( 1990Caruso ( , 1993. Acceptable levels of interrater reliability (K= 0.86 for caregiver behavior, K= 0.88 for infant behavior), as defined in Bordens and Abbott (1992), were obtained using pilot children prior to coding data tapes. Although the behavior coding strategies used in this study have been used successfully in previous research (i.e., Anzalone,1994;Belsky & Most, 1981;Caruso, 1990;1993), no additional reliability and validity data are available at this point.
Samples of coding sheets used are available in Appendixes B and C.

Infant Exploratory
Play. Developmental Level Coding. Developmental level of play was measured by both a Spontaneous Mastery Score which is the highest 1 4 level of play that an infant is able to sustain (Hrncir et al., 1985) and Visually guided manipulation (2).
Visually guided manipulation that is appropriate for a particular object (3).
Place two objects in an inappropriate relationship ( 4).
Place two objects in an appropriate relationship ( 5).
Approximate pretense activity without confirming evidence of pretense behavior (6).  Feel texture of object surface with finger(s).
Move or turn object with fingers.
Look at object(s) without touching.
Look carefully at object while holding.
Move object back and forth in air while holding.
Repeatedly hit object against table.
Move object back and forth across table.
Purposefully release grasp of object and drop.
Toss object using arm or wrist movement.
Using two hands, pull or twist objects in two directions.
Place two objects in a relationship.
Place or hold object in mouth. Total of all exploratory behaviors.
Percentage of time available spent in exploration.
The Sophistication Ratio was a ratio of the frequency of sophisticated exploratory behaviors divided by the frequency of unsophisticated exploratory behaviors.
The determination of a play behavior as sophisticated or unsophisticated and the use of a sophistication ratio was based on previous findings of Caruso (1990Caruso ( , 1993 and Anzalone (1994). Behavioral Tempo was defined as ratio of total exploratory play behaviors to time spent in exploration.
Exploratory Quantity Coding. Exploratory quantity again utilized the exploratory behaviors that are presented in Table 2.
These behaviors were used to calculate two quantity variables: the 1 8 total number of exploratory behaviors an infant used and percentage of the time available that the infant engaged in exploration.

Caregiver Behavior Coding.
Caregiver coding attempted to capture information on caregiver behaviors that have previously been shown to either facilitate or inhibit infant exploratory play.
Using Anzalone' s ( 1994) coding scheme, which was based on the work of Fiese (1990), these behaviors were classified into nine categories including the following: Social, Follow/Imitate, Support, Expand, Intrude, Initiate, Remove Toy, Attention Directing, Organize.
Operational definitions of these behaviors are available in Table 3. Additionally any consent forms typically used by individual regional programs were completed during the initial session. After caregivers were contacted and the number of caregivers interested in group participation was known, group assignment was determined.
With six or more potential child/caregiver pairs in one early intervention program, random assignment of pairs to one of two groups (with a minimum of three participants) occurred without replacement. When less than six · potential participants were available at one site, they were, as a group, assigned to a condition (via coin toss). When possible a subsequent group at that site was assigned to the other condition.
One type of group focused on teaching caregivers specific skills to facilitate their children's play. The other type of group, a comparison group, focused on providing caregivers with general information on different areas of development (i.e. social, physical, cognitive).

Sequence of Eight Week Intervention.
This study was conducted at the early intervention centers that served as the base for recruitment of participants. The study took place over an eight week period for each group of participants. An interpreter was present during all sessions for those caregivers whose primary language was not English.
In the first week, the purpose of the study was reviewed and introductions were made in both groups. During the second week each child's developmental age was individually assessed using the Bayley Scales Of Infant Development-2nd Edition, Mental Scale (Psychological Corporation, 1993).
In the third week, pre-test data on the play behavior of the children and on the child/caregiver interaction were collected using a standardized set of toys via video camera for all groups. First each child was videotaped playing independently for five minutes. The child and his/her caregiver were then videotaped together for an additional five minute period.
The instructions used to introduce the different play segments are available in Appendix F.

1
The set of standardized toys selected for use in the present study has been used in previous research using similar measures of exploratory play and has been shown to provide the potential for sustained play and diverse exploration (Anzalone, 1994;Belsky & Most, 1981;Caruso, 1993). Additionally these toys are commercially available and are safe for children who fall within the developmental range used in this study. A list of the toys used is available in Appendix G.
In the following four weeks the primary caregivers of children in the exploratory play group were taught to facilitate their child's play. First, they were taught to recognize when their child has lost interest in the age appropriate toys provided (non-attention). Then they were taught strategies, such as verbal and physical cues, to reengage their child's interest.
In the last two weeks of the intervention phase of the study, caregiver's learned how to expand their infant's play repertoires using skills including verbal descriptions and praise. This phase of the intervention was based in part on the Nursing Child Assessment Teaching Scale (Barnard, 1988 In the general information group, caregivers heard how their children's performance reflects developmental skills. In the last ten minutes of the sessions, a summary of the session was given and caregivers were encouraged to use the information they gained at home. During the pre-test and post-test weeks (weeks 3 and 8) the set of standardized toys were presented for videotaping · as an additional play station.

Results
Parallel sets of analyses were used to examine data obtained from the caregivers and children involved in the study. In other words, the dependent variables based on caregiver behavior in the two group across time (Social, Follow/Imitate, Support, Expand, Intrude, Initiate, Remove Toy, Attention Directing, Organize) were summarized, compared and analyzed for statistically significant differences in one set of analyses.
In the other set of analyses, the differences between the children of the two groups across time on the dependent variables related to infant behavior (i.e., SMS, AOL, Sophistication Ratio, Breadth, Behavioral Tempo, Total Exploring, Exploration Time) were summarized, compared and analyzed for statistical differences. Child behavior was examined both when children were playing independently and when children were interacting with their caregivers.
In the case of significant differences, ANOVAs were used to assess each dependent variable's contribution to the effect. Additionally the relationship between caregiver behavior and infant behavior was examined pre-and postintervention using Pearson product moment correlations.
To provide background information, demographic and descriptive data will first be presented. Remaining results will be presented to address the major research questions under exploration.
These questions include the effects of group participation on caregiver and child behavior as well as the relationship between caregiver behavior and child play behavior.

Demographic and Descriptive Data
All caregivers and their children were involved in regional early intervention programs in Rhode Island and Massachusetts.
English was the primary language for 24 of the caregivers, with four caregivers using primarily Spanish and one caregiver using primarily Portuguese.
The majority of the mothers participating in the study were married (65%).  As wide ranges of caregiver education (less than high school to graduate degrees) and employment (unemployed to professional occupations) were represented within the current sample and as Anzalone ( 1994) found no significant effects of demographic variables (child gender, birth order, maternal education, work status) • on play and developmental variables, current analyses did not control for demographic effects.
The first step in the analyses was to summarize the data using descriptive statistics .
In   Developmental Level of all children ranged between 2 and 6 both before and after the intervention.

Multivariate Analyses of Variance
The second step in the analyses of the data was to test for significant group differences using two 2x2 MANOVAs (Group: 1) Specific Training 2) General Information x Time: 1) Pre-intervention 2) Post-intervention).
One MANOVA was used with the caregiver variables, while a second MANOV A was used with the child 30 variables.
The MANOV A is the analysis of choice in the present study because there are several dependent variables for both · caregivers and children and because there is evidence (e.g., Anzalone, 1994) that these variables may be correlated.
Prior to using the MANOV A the data were checked to ensure that assumptions underlying multivariate statistics were met. This included examining frequency distributions for outliers, skewness, and kurtosis.
Given the presence of two outliers (as defined in Tabachnick & Fidell, 1989) in the distribution of variable "sophistication ratio", this variable was eliminated from subsequent analyses.

Relationship Between Child and Caregiver Behavior
Pearson product-moment correlations were calculated to explore the degree of relationship between caregiver and child behavior before and after the intervention period. The Pearson product-moment correlation is used as all variables are continuous.
All correlation coefficients presented are statistically significant.
Prior to the intervention, the children's exploratory play behavior when playing independently was not, in general, related to the patterns of behavior that the mothers demonstrated when playing with their children.
For example, mother's attention directing behavior was not related to any of the variables that measured independent child play. An exception to this was the relationship of the variable Expand to child behavior.
The degree to which mothers expanded upon their children's play was positively related to measures of play in both groups prior to the intervention.
In the group (Group 1) that would later receive training specifically related to exploratory play, the mother's expansion of play was related to increased sophistication in the child's play, r=.7, Q_<.05. In the group (Group 2) that later received general information, expansion of play was related to increased time spent in exploration, play his way, not forcing him." A mother in the general information group indicated that she had learned "what to do to help motor skills." It appears that while at least some parents learned strategies specifically related to their group assignment, these strategies were not as valued as opportunities to interact with others. Additionally more parents reported increased understanding of the importance of play in general as opposed to increased understanding of specific strategies to promote play, regardless of group assignment.

Discussion
The finding of this study are discussed in relation to two central research questions: (a) the effect of group participation on caregiver and child behavior as well as (b) the relationship between caregiver and child exploratory play behavior before and after the intervention period. In addition, the exploratory behavior of the children in the present study is compared to prior findings regarding exploratory play in developmentally delayed children. The Effect Of Group Participation On Caregiver Behavior And Child Exploratory Play

36
The hypothesis that providing parents with training on strategies that are related to exploratory play competence in children · leads to an increased use of those strategies was not supported by the results of the current study. Caregiver behavior on variables previously shown to be related to child exploratory play competence (Anzalone, 1994) did not change over the course of the current intervention. This lack of change was noted in both the group that received general information on development and in the group that received specific training ~n these behaviors. Caregivers who participated in the exploratory play training group did not display more behaviors that facilitate infant exploratory play (i.e., attention directing, expanding, organizing, supporting) or less behaviors that inhibit exploratory play (i.e., intruding, initiating, and removing toys) at the conclusion of the intervention.
In addition, the hypothesis that changes in the behavior of caregivers who received specific training regarding the facilitation of exploratory play leads to higher developmental levels of play and exploratory competence in their children · was not supported.
Children who participated in the specific training group did not demonstrate a higher developmental level of play, an increased quantity of exploration, or a greater breadth of exploration than children whose caregivers received training on general development.
Children in the training group and children in the general information group di(j not differ significantly in terms of quality of exploratory play or in quantity of exploratory play subsequent to intervention.
Although the exploratory behavior of children in the general information group did not differ from that of children in the specific  (Krakow & Kopp, 1982;.
The present study suggests that it may be possible to increase the quality of infant exploratory play with a group intervention, however that intervention may not need to incorporate strategies that are specifically geared toward facilitating exploratory play.
Eight weekly session that focus on the importance of play and provide opportunities for caregivers to play with their children using 3 8 developmentally appropriate toys may lead to an increase in the quality of child play whether specific information regarding exploratory play is provided or not. Additionally, the information mothers gained by watching other caregivers playing with their children and the support obtained from interacting with others may also be related to change. In other words, participation in either group of the present study may have led to changes in child behavior. If so, these changes may have been mediated by changes in maternal behavior that were not measured in the present research design.
In their previous experimental study, Belsky  This possibility is given support by the description feedback caregivers provided on group participation. Caregivers, regardless of group assignment, indicated that the opportunity for their children to interact with other children and the opportunity to interact with other parents were the most valued aspect of group participation. In 39 addition, several parents · indicated increased understanding of the importance of play in general. So while some parents noted that · they gained specific information related to group assignment, this information was not considered to be the most advantageous or salient aspect of group participation by parents.
Although it may be plausible to ask whether participation in an eight week group focused on play can affect the quality of children's exploratory play regardless of the specific information provided to caregivers, that question can not be answered by the present study.
Alternative hypotheses for the change in child behavior in the present study must also be . considered. One possibility is that this change is the result of testing itself. Participants may have become accustomed to the setting that the videotaping occurred in or to the process of being videotaped. Another possibility is that maturational changes in the children led to a change in their exploratory play behavior. While possible, neither of these explanations fully accounts for the changes in quality of exploratory play when no change in quantity of exploratory play was demonstrated.

Relationships Between Caregiver and Child Exploratory Play Behaviors Before and After the Intervention Period
Previous research (Fiese, 1990;Anzalone, 1994) -Cram, 1996). Additionally, Anzalone ( 1994) suggests that relationship of maternal behavior to child behavior may change over time. In part, this study attempts to add to the literature available in these areas.
To an extent, the relationships between caregiver and child behavior in the present study follow the general rules regarding maternal behavior and child exploratory play. Prior to intervention, when children were engaged in play with their caregiver, support and expansion by caregiver was positively associated with quality and quantity of play. Patterns of expanding upon children's play were associated with increased sophistication (Group 1) and increased time spent in exploration (Group 2) even when children were playing independently. Additionally, a pattern of removal of toy by caregivers (an intrusive behavior) was related to a decrease in exploratory breadth in children playing alone (Group 1).
Interestingly, however, these relationships were not stable across time for either group. Subsequent to intervention, when playing together, caregiver social behavior (i.e., comments unrelated to child's play) was related to the amount of time children spent exploring and to behavioral tempo in the group that received general information on development. No significant relationships between caregiver behavior and child independent play were noted for this group at the conclusion of the intervention.
The relationship between maternal and child behavior was not stable for the participants in the exploratory play training group either.
In fact, several caregiver variables generally considered to facilitate mature play were negatively related to measures that tap quantity of exploration post -intervention. In addition, the caregiver behavior of initiating a play sequence, often considered to impede exploratory play, was in fact related to a decrease in the sophistication of play. However, Anzalone (1994) postulates that in some cases, maternal initiation could increase the novelty of the play situation and facilitate exploratory play.
Looking at the patterns of the relationships between maternal behavior and child exploratory play from the present study in conjunction with previous literature, two major conclusions can be drawn.
The first is that these relationships are not stable over time.
The maternal behaviors that facilitate exploratory play for an individual may change in relation to that child's developmental level.
For example, younger children may profit from increased maternal structuring of the environment, while older children may respond to more control over the play interaction. Second, the patterns of behavior that facilitate exploratory play in atypically developing children may not be the same patterns found between typically developing children and their caregivers. In other words, facilitation of exploratory play may not be as straightforward as was suggested by previous research.

Exploratory Play Behavior In Developmentally Delayed Children
In previous studies (e.g., Jennings et al., 1979), cognitive development has been shown to be related to exploratory play.
Additionally it has been shown that developmentally delayed children progress through Belsky and Most' s play sequence more slowly than typically developing children (Krakow & Kopp, 1982; 42 1983). It has also been noted that the quality of delayed and non delayed children's play may differ (Krakow & Kopp, 1982;· MacTurk et al., 1985). Many of these studies, while providing valuable information on exploratory play, have focused on relatively homogeneous populations (Hauser-Cram, 1996).
The present study contributes to the existing body of literature by expanding the definition of developmental delay to that used in practice by early intervention programs rather than focusing on a specific population of children with developmental disabilities (i.e., children with Down's Syndrome).
Using a more heterogeneous sample, the results of the current study provide additional support for previous findings.
In the present study developmentally delayed children, ranging in age from one to two years, were primarily using play behaviors that are typically associated with 9 to 14 month old children who do not have developmental delays (Belsky & Most, 1981 (Anzalone, 1994). There is also a growing interest in the possibility that mothers with developmental delayed children differ from mothers of typically developing children during interactions focused on play. Mahoney ( 1988) indicated that mothers of children with delays are more directive. Blasco, Hrncir and Blasco (1990) found that quality of play in infants with cerebral palsy was related to the support mothers provided.
In the present study the most frequently occurring maternal behaviors included expanding upon a child's play and introducing a new object to a child. Mothers rarely followed the lead of their children as demonstrated by the low incidence of caregivers imitating their children.
The pattern of maternal behavior found across groups in this study appears to support Anzalone' s contention that parents of children with disabilities may try to teach their children rather than "play" with them. It is of note that this pattern of maternal behavior persisted post intervention in the group that was encouraged to be more responsive to child cues and consequently less directive.
Overall there is a growing body of literature that suggests that the maternal behavior of mothers of children with developmental delays differs from that of mothers of children without delays.
Moreover, these differences tend to be in the direction of being more directive. Current findings lend additional support to previous research in this area.

Conclusion
In summary, the purpose of this study was to determine if caregivers could be trained to facilitate the play of their 44 developmentally delayed children.
In addition, the relationships between caregiver behavior and child exploratory play pre-and post-intervention were examined. The present study also provided support for prior findings regarding exploratory play in developmentally delayed children.
Providing parents with training on strategies that are related to exploratory play competence in children did not lead to an increase in use of those strategies in the current study. However, children who participated in the study demonstrated increased quality of exploratory play at the conclusion of the study. Several hypotheses must be considered to account for this change.
These include the possibility that group participation did affect change in parent and child behavior, although not along the predicted path. Other possibilities include the effects of testing or maturation. Given the importance ascribed to and the widespread use of parent training in early intervention programs, this study provides a valuable model of evaluating, rather than assuming the effectiveness of parent training programs.
Previous research has provided general guidelines about the types of maternal behavior that enhance or inhibit child exploratory play. Maternal behavior such as expanding upon and supporting children's play is associated with more mature exploratory play, while maternal initiation and intrusiveness has been associated with less mature play.
The current study provides additional support for the growing body of evidence that these guidelines are, at best, just guidelines. The relationship between maternal behaviors and child exploratory play may not be stable across time. Changes in the relationship between maternal behavior and exploratory play may be associated with a child's developmental level as well as individual differences.
The present study also contributes to the existing body of literature on exploratory play by expanding the definition of developmental delay to that used in practice by early intervention programs rather than focusing on a specific population of children with developmental disabilities.
Using a more heterogeneous sample , the results of the current study provide additional support for previous findings. That is the level and quality of exploratory play of developmentally delayed children differs from the exploratory play of children without developmental delays. Additionally the pattern of maternal behavior found across groups in this study appears to support Anzalone's contention that parents of children with disabilities may try to teach their children rather than "play" with them.

Limitations and Recommendations for Further Study
The conclusions of this study are limited by several factors.
These factors include sample size as well as threats to internal and external validity.
First, · although this study attempted to consider ecological validity by holding groups in the early intervention center that parents were recruited from , the behavior of both parents and children may have been affected by coming to a potentially less familiar setting (than home) and interacting with unfamiliar adults and children. However, this situation does typically occur in early intervention programs, and thus results might be generalized to similar groups held at early intervention sites.
Second, the process of being videotaped might be considered to be a reactive procedure. This procedure may have differentially affected participants at the pre-and post-intervention videotaping sessions. Subsequent studies might decrease the effect of the novelty of videotaping by videotaping the mother and child over time, prior to the initiation of the intervention. I have been asked to take part in a study which will look at how caregivers of developmentally delayed infants might best help their children take advantage of play situations to practice and develop thinking skills.
If I decide to participate with my child, this is what will happen: my child and I will participate in 8 ninety minute sessions at our regional early intervention center with up to 4 other caregivers and their children. During one session I will meet with the researcher and have the opportunity to discuss any questions I have.
In another session a brief developmental evaluation (the Bayley Scales of Infant Development -2nd edition) will be completed.
During the remaining sessions I will be asked to play with my infant or watch my infant play using a variety of safe and age appropriate toys, while the researcher provides me with information on play or early childhood development. Researcher : For the next ten minutes or so, you and your baby will be playing with some toys. We'll do this in three different segments.

Segment 1
Researcher: First we'll let play with a "warm-up" toy for a few minutes, until s/he is in a relaxed and playful mood.

Segment 2
Researcher: At the end of the warm-up period I will give of toys. During the first five minutes I want you to just watch.

a set
Please respond normally if ___ tries to get your attention but do not start to play with the toys or your baby.

Segment 3
Researcher: I will signal you when the next five minute period begins.
In this segment I would like you to play with your baby and the toys as you typically might at home.

At conclusion
That was great! Thank you for your help. At first, they may not even be very good at touching a toy but eventually they will be able use a rough grasp (demonstrate), that grasp becomes better and better developed until it is much cleaner and they can even start to pick up and hold more then one thing at once. As a child gains more control over his/her hands, s/he will also start to gain control over their fingers, like when they pick up a

You can really
It is especially nice to take some time to play with your child and practice these skills, because the you will be working on things and skills your child 8 1 is interested in and motivated to learn.
Over the next week, practice imitating, supporting and praising your child at home.