The interrelation between children’s play behaviour, temperament, and special educational needs (SEN) in early childhood special education (ECSE)

ABSTRACT

The aim of this study is to identify if children's play behaviour, temperament, and special educational needs (SEN) are interrelated. It is important to increase the awareness of the joint effects of temperament and SEN on behaviour so that early childhood special education (ECSE) teachers can provide the appropriate support for children with SEN and optimally guide their development.

This study is a part of an early childhood special education (ECSE) project at the University of Helsinki. Children's (N = 332) educational needs were divided into four status groups: language disorders, self-regulatory difficulties, severe disabilities, and children without SEN. The temperament of each child was evaluated by their parents and play behaviour observed by ECSE teachers.

The results show that children without SEN had more effortful control and social play than children with SEN. In addition, there were significant differences in temperament and play behaviour between the different SEN groups.

KEYWORDS:

• Early childhood education (ECE)
• special educational needs (SEN)
• play behaviour
• temperament

Introduction

Building a safe social learning environment is the most effective way to prevent social exclusion and comprehensively boost a child’s development. In principle, inclusive settings are the optimal means of enabling social interactions for all children (Allen & Cowdery, 2015; Odom & Diamond, 1998). However, it is widely known that children with special educational needs (SEN) are frequently less accepted and may experience greater loneliness than their peers without SEN (Avramidis, 2010; Lackaye & Margalit, 2006). It is therefore essential to enhance peer interactions and encourage participation from the moment a child enters ECE (Pesonen, 2016; Syrjämäki, Pihlaja, & Sajaniemi, 2018). During a child’s early years, play is the most natural arena for learning the socio-emotional and cognitive skills that are needed for cooperation with peers (Lynch & Simpson, 2010; O’Connor & Stagnitti, 2011). For some children, however, their opportunities to join play are limited because of their compromised social, cognitive, or verbal development (Kilburn & Mills, 2018). In addition, a range of temperament characteristics may also increase the barriers to joining play (Rothbart Mary, 2011; Suhonen, Nislin, Alijoki, & Sajaniemi, 2015). Thus, teachers should be aware of even small obstacles that may increase the risk of social isolation (Hermanfors, 2017; Lynch & Simpson, 2010). By focusing on the goal of actual inclusion (not just physical, but also social inclusion and full participation), we investigated whether certain temperament traits in SEN children have an impact on play.

In Finnish ECE, the percentage of children with SEN is approximately 7.4% (Finnish National Agency for Education, 2018). According to statistics, most SEN children have speech and language difficulties; other common challenges include developmental delays, socio-emotional difficulties, and attention difficulties. Approximately 30% of SEN children do not received a formal diagnosis.

The importance of play

Play is essential for children’s development, learning, and self-regulation (Pellis, Pellis, & Bell, & C, 2010). Children also acquire other important abilities through play, such as sharing, social communication, empathy, intentional learning, problem solving, and creativity (Whitebread, Coltman, Jameson, & Lander, 2009). Play is rewarding, as it encourages children to pursue challenges and learn from more capable peers (Bodrova, Germeroth, & Leong, 2013; Vygotsky, 1978). While every child should have the right to participate in play, the process can be more difficult for children with SEN because of physical, communicative, or self-regulative challenges (Barron et al., 2017). These challenges frequently prevent SEN children from experiencing the joy of interacting with peers, and thus they can miss out on the multiple advantages that are gained through play (Hoogsteen & Woodgate, 2010). Therefore, it is extremely important that ECE professionals identify children who are not actively involved in play and understand how to increase their participation (Syrjämäki, Sajaniemi, Suhonen, Alijoki, & Nislin, 2017).

Children by nature are different from each other, and their differences are manifested in many domains, including play. Children with shyness prefer being alone, and they need encouragement to join social play. In contrast, extraverted children enjoy playing with peers and may occasionally require help to calm down (Coplan & Bullock, 2012). Temperament also affects a child’s ability to cope and adapt to new situations (Kagan, 2019; Rothbart Mary, 2011). Most importantly, every child has the right to be accepted and access support when needed (Convention on the rights of the child, 1989). Children thrive in learning environments that ensure physical, social, and psychological safety. Safe environments should be responsive to individual needs and to the diverse behaviours that originate from individual temperament. Understanding different temperaments and their connections to behaviour, and adapting interactions accordingly are key elements in sensitive interactions (Kristal, 2005). It is fundamentally important that teachers learn to recognize these components of sensitive interactions and understand how they impact on behaviour (Eisenberg, Eggum, Sallquist, & Edwards, 2010; Rothbart & Jones, 1998).

Children's temperament in early childhood has been widely researched (Bagner, Rodríguez, Blake, Linares, & Carter, 2012; Brock et al., 2012; Hepburn, 2003; Kang & Kwack, 2018; Lukowski & Milojevich, 2017; Northerner, Trentacosta, & McLear, 2016; Rothbart, 2007; Rothbart, Ahadi, Hershey, & Fisher, 2001; Talge, Donzella, & Gunnar, 2008). Several of these studies have found that certain traits hinder social play, for example, shyness and high reactivity (Coplan, 2020; Coplan & Weeks, 2009; Glover Gagnon et al., 2014 Kalutskaya, Archbell, Moritz Rudasill, & Coplan, 2015). Relationships between temperament and different SEN statuses have also been examined, with the research revealing diverse findings, for example, in studies on autism (Brock et al., 2012; Konstantareas & Stewart, 2006), Down Syndrome (Gartstein, Marmion, & Swanson, 2006; Lukowski & Milojevich, 2017), and attention deficits (Kang & Kwack, 2018; Karalunas, Gustafsson, Fair, Musser, & Nigg, 2019). However, only a limited number of studies have considered the combined effect of SEN, temperament, and play behaviour.

The concept of temperament

This study utilizes Rothbart's three dimensions of temperament (Rothbart et al., 2001). Rothbart defined temperament as the individual differences in reactivity and self-regulation that affect emotional expressions, activity, and attention. Rothbart employed factor analysis to interpret data that was based on questionnaires of 3 -12-month-old children: three broad factors were identified and labelled as surgency/extraversion, negative affect, and effortful control (descriptions located in the method section).

Previous studies have shown that children with SEN have less effortful control and more negative affectivity than their peers (Konstantareas & Stewart, 2006; Suhonen, Sajaniemi, Alijoki, & Nislin, 2018). Effortful control is related to self-regulation, which is a core ability in learning, executive functions, and social competence (Robson, Allen, & Howard, 2020). Effortful control develops in early childhood and can be learnt through social play (Blair & Diamond, 2008; Timmons, Pelletier, & Corter, 2016). In addition, Broekhuizen, Slot, van Aken, and Dubas (2017) found that children with inadequate self-regulation can benefit greatly from emotional and behavioural support provided by ECE teachers. Surgency/extraversion refers to novelty seeking, cheerfulness, spontaneity, and sociability. It may also appear as high activity level and impulsivity when it can cause challenges in interaction. As a temperament dimension, surgency has been described as an accelerator in behaviour, while effortful control acts as a brake (Rothbart Mary, 2011). Thus, children with self-regulation difficulties and low effortful control require adult support in order to learn how to inhibit their immediate reactions, i.e. employing ‘brakes’ when necessary. Likewise, children with extreme surgent tendencies require sensitive guidance to decelerate difficult behaviour. In the other words, every child should have access to adjusted and sensitive guidance – or co-regulation – when needed (e.g. Sajaniemi, Suhonen, Nislin, & Mäkelä, 2015). Research findings in psychology, neuroscience, and genetics have indicated that individual differences in reactivity and self-regulation are based on temperament (Eisenberg et al., 2010). Optimal development in self-regulation boosts resilience while, in the long term, compromised self-regulation jeopardizes social, emotional, and cognitive development (Calkins & Howse, 2004; Keenan & Shaw, 2003). Hence, it is crucial that self-regulation is enhanced through co-regulation, particularly in children with low effortful control or high surgency. This process is important for the development of every child and vital for children with SEN.

A previous study by the current authors (Kesäläinen, Suhonen, Alijoki, & Sajaniemi, 2022) found that over the course of one year in ECE, children with language disorders had better cognitive progression than children with self-regulation difficulties. However, social play improved in both groups. In the present study, we assume that children with self-regulation difficulties have lower effortful control than their peers, which could explain the cognitive findings. Improvements in social play could precede the emerging cognitive abilities that may be detected in later measurements. In addition, we assume that children without SEN have more effortful control and social play than children with SEN.

Research implications

Within the context of ECSE, this study aims to examine the connections between children's play behaviour, temperament, and SEN to increase the awareness of the joint effects of temperament and SEN on behaviour. ECSE professionals can use this knowledge to provide a learning environment that meets the needs of every child. This research includes data from two follow-up studies (Suhonen et al., 2016; Suhonen et al., 2019).

Hypotheses:

• - Children with self-regulation difficulties have lower effortful control than their peers.

• - Children without SEN have more effortful control and social play than children with SEN.

Ethical considerations

This study has been conducted according ethical principles of University of Helsinki Ethical Review Board in the Humanities and Social and Behavioural Sciences. Research permission was requested from the municipal authorities. The parents and the children were well informed and could withdraw from the study at any time. Written consent from the children’s parents was requested.

Method

Participants

This study is a part of the early childhood special education (ECSE) research at the University of Helsinki. The data was collected between 2012 and 2017 during two separate projects. The first project studied the development and learning of 257 children (3-6 years) in integrated ECSE groups (N = 24) in Helsinki, Finland; the integrated groups usually included five children with SEN, seven children without SEN, two ECSE teachers, and one nurse. Some of the children also received personalized therapies, such as speech therapy, physiotherapy, or occupational therapy. The second project studied 75 children with SEN who were supported by an ECSE teacher in regular ECE settings. Thus, a total of 332 children participated in the current study (182 with SEN and 150 without SEN). The parents of each child gave written consent for the research. The gender distribution of the children was not balanced, as the majority of children with SEN were boys. In total, 206 children were boys (138 with SEN and 68 without SEN) and 126 girls (44 with SEN and 82 without SEN).

Parents completed the Children's Behaviour Questionnaire (CBQ) when their child entered the study. ECSE teachers evaluated the children’s play behaviour three times in total, once per year over three consecutive years (pre-measure, repeat measure, and post-measure), using the Preschool Play Behaviour Scale (PPBS).

Children's SEN was divided into status groups at the beginning of the project. When the status groups were established, each child’s primary need for support was carefully considered. In addition, the overlap between the status groups was almost non-existent. The distribution was based on both medical documents and the teachers’ descriptions of the children’s general difficulties and behaviour in the ECE environment. Based on this information, the status groups were developed by a specialist in clinical neuropsychology. First, children's diagnoses were divided into eight different categories and then later narrowed down to three main categories: language disorders, self-regulation difficulties, and severe disabilities. The severe disabilities category included children with intellectual disabilities, children with cerebral palsy, and children with autism spectrum disorders (ASD). Although some children had been diagnosed with more than one disability, their primary need was clearly prominent. Table 1 shows the children’s status groups and the number of participants in each group.

Table 1. Children's status groups.

Status groups	Children without SEN	Children with language disorders	Children with self-regulation difficulties	Children with severe disabilities
Boys	68	48	68	22
Girls	82	18	19	7
Total	150	66	87	29

Measurements

Background information

A family background questionnaire was distributed to the parents at the beginning of the study. The questionnaire concerned family composition, the occurrence of learning difficulties among family members, and the familýs socio-economic status (SES), with the latter based on the parents’ education level and their yearly income. The questionnaire also requested information on birth-related factors of the participating children as well as current medications and developmental concerns.

Play behaviour

The children’s play behaviour was evaluated by ECSE teachers using the Preschool Play Behaviour Scale (PPBS) by Coplan and Rubin (1998), an assessment tool that is widely used in ECE settings. The PPBS includes 18 items that are scored on a 5-point Likert scale from ‘never’ to ‘very often’. PPBS divides play into five categories: reticent behaviour, solitary play, outward directed solitary play, social play, and rough and tumble play. Reticent behaviour is expressed as shy or anxious avoidance in early childhood (Jarcho et al., 2016). Solitary play can reflect a child’s conscious choice and their preference for solitude. However, research has shown that children have an increased risk of adjustment difficulties if they are willing to play with others but are too shy or dońt have the skills necessary for social play (Ooi, Baldwin, Coplan, & Rose-Krasnor, 2018). Social play requires various abilities, such as sharing and cooperation, but it also teaches many important skills and buffers learning (Palagi, 2018). Young children in particular use their bodies for communication and learning; therefore, rough and tumble is a vigorous and intense form of play that many young children enjoy. Because it sometimes resembles actual fighting, adults may find it challenging and thus prohibit this behaviour. However, rough and tumble is generally not aggressive, and children can use this form of play to learn physical skills as well as cooperation (Carlson, 2011a; Flanders, Herman, & Paquette, 2013).

Temperament

Children's temperament was evaluated using the Children’s Behaviour Questionnaire (CBQ) developed by Rothbart et al. (2001); the questionnaire was completed by the parents of each child. CBQ has 94 items scored on an 8-point Likert scale from ‘very untrue’ to ‘very true’ and includes the option ‘Ím not sure’. Individual differences are assessed with 15 primary temperament characteristics: positive anticipation, smiling/laughter, high intensity pleasure, activity level, impulsivity, shyness, discomfort, fear, anger/frustration, sadness, soothability, inhibitory control, attentional focusing, low intensity pleasure, and perceptual sensitivity. In this study, these 15 dimensions were condensed into the following three broad groups (see Rothbart et al., 2001): negative affectivity, effortful control, and extraversion/surgency.

Negative affectivity has an influence on behaviour problems that manifest early in children. These problems can appear as both internalizing (fearful, withdrawn) and externalizing (aggression, rule-breaking) behaviour (Bagner et al., 2012; Northerner et al., 2016). In addition, children who have high levels of negative affectivity are less likely to form social play relationships (Neal, Durbin, Gornik, & Lo, 2017). In turn, low negative affectivity appears to protect children who have other risk factors (Northerner et al., 2016).

Effortful control is visible, for example, in a child’s attempt to accept delays, slow down their motor activity, provide effortful attention, and lower their voice when needed (Kochanska, Murray, & Harlan, 2000). Effortful control involves the ability to regulate one’s own actions, emotions, and attention, which allows a child to pursue a coherent plan of action. These behaviours and abilities are all important for learning and social competence (Eisenberg et al., 2010); thus, effortful control is also related to peer acceptance (Hernández et al., 2017).

Extraversion/surgency temperament reflects a child’s propensity towards positive affect, approach, sociability, high-intensity pleasure, reward seeking, and high activity levels (Holmboe, 2017; Rothbart Mary, 2011). Children with high extraversion/surgency are more likely to experience negative peer encounters than children with low extraversion/surgency who behave with more prudence when interacting with their peers (Dollar & Stifter, 2012).

Analysis

The analysis of the data was statistically conducted with IBM SPSS 25. Correlations were tested using the Pearson correlation coefficient. Gender differences were tested with a t-test. Analysis of variance (ANOVA) was used to assess the relationships between the children's status groups and the temperament dimensions and play behaviour categories; gender was covariated. The estimated relationships between the variables were tested with regression analysis.

When conducting the temperament sub-dimensions, a number of low Cronbach alpha values were observed: approach/positive anticipation (.092), discomfort (.441), high intensity pleasure (.583), impulsivity (.489), sadness (.447), and fear (.473). Approach/positive anticipation was excluded from the surgency/extraversion dimension and from further analysis because of its extremely low alpha value. The remaining sub-dimensions had acceptable alpha values (<.06). The three broad temperament dimensions (negative affect, effortful control, and surgency/extraversion) also had adequate Cronbach's alphas (<.06.).

Results

Background information

There was no connection between SES, SEN, and temperament. The parents’ level of education was generally high: 35% of mothers and 28% of fathers had a university degree; 29% of mothers and 28% of fathers had a college or polytechnic degree; and 25% of mothers and 30% of fathers had a vocational school degree. Only 11% of mothers and 18% of fathers lacked further education.

The children's gender variable was skewed because of the higher number of male participants (206 boys; 126 girls). Gender was highly correlated with effortful control (r = .242, p < .001); girls had more effortful control than boys (t −4.528 (330), p < .001).

Key results

Table 2 shows the key results divided by the status groups. Targeted results for play behaviour and temperament are presented in Table 3 and Table 4, respectively.

Table 2. Key results.

Status groups	Children without SEN	Children with language disorders	Children with self-regulation difficulties	Children with severe disabilities
	high effortful control	solitary play	high surgency / extraversion	low effortful control
	social play		high negative affect	low positive anticipation
			increased activity	low negative affect
			rough and tumble play	solitary play

Table 3. The results of play behaviour (PPBS) and SEN.

PPBS	Pre-measure N = 320	Repeat measure N = 248	Post-measure N =  62
RETICENT BEHAVIOUR	Children without SEN	Children without SEN	Children without SEN
	N =  147	N =  107	N =  20
	Mean 2.31	Mean 2.00	Mean 1.84
	Std. deviation .80	Std. deviation .77	Std. deviation .75
	Language disorders	Language disorders	Language disorders
	N =  62	N =  51	N =  17
	Mean 2.79	Mean 2.61	Mean 2.22
	Std. deviation .96	Std. deviation .75	Std. deviation .87
	Self-regulatory	Self-regulatory	Self-regulatory
	Difficulties	Difficulties	Difficulties
	N =  84	N =  71	N =  19
	Mean 2.78	Mean 2.48	Mean 2.17
	Std. deviation .75	Std. deviation .82	Std. deviation .67
	Severe Disabilities	Severe Disabilities	Severe Disabilities
	N = 27	N =  18	N =  6
	Mean 3.19	Mean 3.13	Mean 3.00
	Std. deviation .95	Std. deviation 1.11	Std. deviation .76
SOLITARY PLAY	Children without SEN	Children without SEN	Children without SEN
	Mean 2.65	Mean 2.39	Mean 2.33
	Std. deviation .65	Std. deviation .67	Std. deviation .50
	Language disorders	Language disorders	Language disorders
	Mean 3.00	Mean 2.74	Mean 2.57
	Std. deviation .71	Std. deviation .68	Std. deviation .81
	Self-regulatory	Self-regulatory	Self-regulatory
	Difficulties	Difficulties	Difficulties
	Mean 2.83	Mean 2.65	Mean 2.43
	Std. deviation .68	Std. deviation .79	Std. deviation .75
	Severe Disabilities	Severe Disabilities	Severe Disabilities
	Mean 2.94	Mean 3.25	Mean 2.71
	Std. deviation .85	Std. deviation 1.11	Std. deviation .40
OUTWARDLY-DIRECTED SOLITARY PLAY	Children without SEN	Children without SEN	Children without SEN
	Mean 2.62	Mean 2.39	Mean 2.53
	Std. deviation .90	Std. deviation .87	Std. deviation .89
	Language disorders	Language disorders	Language disorders
	Mean 2.55	Mean 2.49	Mean 2.47
	Std. deviation 1.11	Std. deviation 1.05	Std. deviation 1.02
	Self-regulatory	Self-regulatory	Self-regulatory
	Difficulties	Difficulties	Difficulties
	Mean 2.62	Mean 2.44	Mean 2.26
	Std. deviation 1.07	Std. deviation 1.10	Std. deviation 1.13
	Severe Disabilities	Severe Disabilities	Severe Disabilities
	Mean 2.06	Mean 2.36	Mean 2.25
	Std. deviation .95	Std. deviation 1.10	Std. deviation 1.41
SOCIAL PLAY	Children without SEN	Children without SEN	Children without SEN
	Mean 4.14	Mean 4.55	Mean 4.45
	Std. deviation .94	Std. deviation .72	Std. deviation .76
	Language disorders	Language disorders	Language disorders
	Mean 3.19	Mean 3.68	Mean 4.15
	Std. deviation 1.00	Std. deviation .87	Std. deviation .74
	Self-regulatory	Self-regulatory	Self-regulatory
	Difficulties	Difficulties	Difficulties
	Mean 3.53	Mean 3.84	Mean 4.39
	Std. deviation .99	Std. deviation .87	Std. deviation .89
	Severe Disabilities	Severe Disabilities	Severe Disabilities
	Mean 2.61	Mean 2.84	Mean 2.50
	Std. deviation 1.26	Std. deviation 1.25	Std. deviation 1.14
ROUGH AND TUMBLE PLAY	Children without SEN	Children without SEN	Children without SEN
	Mean 2.76	Mean 2.95	Mean 2.48
	Std. deviation .99	Std. deviation .84	Std. deviation .85
	Language disorders	Language disorders	Language disorders
	Mean 2.42	Mean 2.48	Mean 2.80
	Std. deviation .98	Std. deviation .86	Std. deviation .87
	Self-regulatory	Self-regulatory	Self-regulatory
	Difficulties	Difficulties	Difficulties
	Mean 3.16	Mean 3.22	Mean 3.08
	Std. deviation 1.17	Std. deviation 1.04	Std. deviation .96
	Severe Disabilities	Severe Disabilities	Severe Disabilities
	Mean 2.44	Mean 2.59	Mean 2.67
	Std. deviation .99	Std. deviation 1.02	Std. deviation 1.40

Table 4. The results of temperament and SEN.

	Negative affect	Effortful control	Surgency/extraversion
CHILDREN WITHOUT SEN (N =  150)	Mean 3.8923 Std. deviation .56940 DIFFERENCE WITH GROUPS: Self-regulation difficulties -.2771 p = .021 Severe disabilities .4342 p =  .014	Mean 5.2417 Std. deviation .64362 DIFFERENCE WITH GROUPS: Language disorders .3753 p = .004 Self-regulation difficulties .4407 p<.001 Severe disabilities .9835 p<.001	Mean 4.7959 Std. deviation .54157 DIFFERENCE WITH GROUPS: Self-regulation difficulties -.3098 p = .012 Severe disabilities .4070 p = .040
CHILDREN WITH LANGUAGE DISORDERS (N =  66)	Mean 3.9664 Std. deviation .69945 DIFFERENCE WITH GROUP: Severe disabilities .50823 p = .007	Mean 4.8665 Std. deviation .75743 DIFFERENCES WITH GROUPS: Without SEN -.3753 p = 004 Severe disabilities .6082 p = .001	Mean 4.7965 Std. deviation .82228
CHILDREN WITH SELF-REGULATION DIFFICULTIES (N =  87)	Mean 4.1694 Std. deviation .77946 DIFFERENCES WITH GROUPS Without SEN .2771 p = .021 Severe disabilities .71128 p<.001	Mean 4.8010 Std. deviation .73215 DIFFERENCES WITH GROUPS: Without SEN -.4407 p<.001 Severe disabilities .5427 p = .004	Mean 5.1057 Std. deviation .85834 DIFFERENCE WITH GROUPS: Without SEN .3098 p = .012 Severe disabilities .71168 p<.001
CHILDREN WITH SEVERE DISABILITIES (N = 29)	Mean 3.4581 Std deviation .99216 DIFFERENCES WITH GROUPS: Without SEN -.4342 p = .014 Language disorders -.5082 p = .007 Self-regulation difficulties -.7113 p<.001	Mean 4.2583 Std. deviation 1.05354 DIFFERENCES WITH GROUPS: Without SEN -.9835 p<.001 Language disorders -.6082 p = .001 Self-regulation difficulties -.5427 p = .004	Mean 4.3889 Std. deviation .97089 DIFFERENCE WITH GROUPS: Without SEN -.4070 p = .040 Self-regulation difficulties -.7168 p<.001

Play behaviour

Children without SEN participated in more social play than all the other status groups (p < .001) in the pre- and the repeat measure. The repeated measures across all the measurement points showed that children with severe disabilities were slower to develop social play than the other groups (F 2,1 = 6 p < .01, children without SEN p < .001, language disorders p = .58, self-regulation difficulties p < .01)

In the pre-measure, children with SEN demonstrated significantly more reticent behaviour than children without SEN (language disorders p < .01, self-regulation difficulties p < .001, severe disabilities p < .001); this was also observed in the repeat measure (language disorders p < .001, severe disabilities p < .001, self-regulation difficulties p < .01). However, the reticent behaviour gradually decreased over the three-year period in every SEN group except severe disabilities (p < .05).

Children with language disorders spent more time in solitary play than children without SEN in the pre-measure (p < .01). In the repeat measure, children with language disorders (p < .05) and children with severe disabilities (p < .001) participated in more solitary play than children without SEN.

In the pre-measure, children with severe disabilities had more outwardly-directed solitary play than children without SEN (p < .05).

Children with self-regulation difficulties took part in more rough and tumble play than children without SEN in the pre-measure. In the repeat measure, children with language disorders had less rough and tumble play than children with self-regulation difficulties (p < .001) and children without SEN (p < .05). The results of play behaviour are presented in more detailed in Table 3; the table shows the number of participants in each status group and the mean scores and standard deviations from the PPBS play categories (scale: 1 ‘never’ to 5 ‘very often’) evaluated by the ECSE teachers.

Temperament

The results show that there were significant differences between the status groups in every temperament dimension. The results are presented in more detail in Table 4.

Children without SEN had the highest effortful control scores (mean 5.2). The difference in self-regulation was very significant in comparison to the severe disabilities’ status group (p < .001) and significant in comparison to the group with language disorders (p < .01). Effortful control scores were similar for children with language disorders and children with self-regulation difficulties.

Children with self-regulation difficulties scored higher in negative affect than children without SEN (p < .05) and children with severe disabilities (p < .001). They also had the highest values in the temperament dimension of surgency/extraversion (mean 5.1). The surgency/extraversion score for children with self-regulation difficulties differed from the scores for children without SEN (p < .05) and children with severe disabilities (p < .001).

Children with severe disabilities had less effortful control than all the other status groups: children without SEN (p < .001), children with language disorders (p < .01), and children with self-regulation difficulties (p < .01). They also had less negative affect than all the other status groups: children without SEN (p < .01), children with self-regulation difficulties (p < .001), and children with language disorders (p < .01).

The surgency/extraversion dimension was carefully reviewed across the sub-dimensions. Positive anticipation and increased levels of activity correlated with SEN (r =  -.180, p < .001 and r =  .201, p < .001). The results show that children with severe disabilities had significantly less (p < .001) positive anticipation than the other status groups. Children with self-regulation difficulties had higher scores for increased activity level than all the other status groups: children without SEN (p < .001), children with language disorders (p < .01), and children with severe disabilities (p < .05). The results for temperament for each status group are presented in more detail in Table 4; the table shows the number of participants in each status group and the mean scores and standard deviations from the CBQ dimensions (scale: 1 ‘very untrue’ to 7‘very true’) assessed by the parents.

Relationships between play behaviour, temperament, and SEN

The correlations were examined between background information, play behaviour, temperament and SEN. The correlations were found only between SEN and effortful control (r = -.329 p < .001) and SEN and social play (r = -.406 p < .001). Therefore, these variables were further examined with linear regression analysis. However, coefficients of determination were weak in both cases; SEN and social play (R^2 = .11⁾, SEN and effortful control (R^{2  =}.16⁾.

Discussion

This study added understanding that in addition to SEN, teachers should observe children's individual temperament and its affect in joint play. With this knowledge, they can support the child in the best possible way and improve interaction between children. The results were significant and supported earlier research; in addition, the results confirmed several of our assumptions. In accordance with our hypothesis, children without SEN had more effortful control than children with SEN. It is possible that effortful control helps children adjust to the requirements of ECE environments; effortful control enhances one’s ability to adapt to new situations and inhibit reactions that may create difficulties (Gal-Szabo, Spinrad, Eisenberg, & Sulik, 2019; Posner & Rothbart, 1998). High effortful control in children without SEN indicates adequate self-regulation, which can result in increased social competence (Liew, 2012). The current study did not focus on gender differences because of the unbalanced gender distribution. However, the girls (female participants) clearly had more effortful control than the boys (male participants).

In this study, there was no significant difference in effortful control between children with language disorders and children with self-regulation difficulties. This finding was against our hypothesis and does not explain the cognitive findings of our previous study. Therefore, more research is needed in this regard.

Children with SEN had more reticent behaviour in the pre-measure than children without SEN. This difference reduced over the study phase, which could indicate that ECE environments support children’s participation in joint activities. It is also likely that ECE teachers actively notice reticent behaviour and encourage children to be more social when needed.

Children with language disorders were involved in solitary play more often than children without SEN. This finding should be highlighted when planning pedagogy, as it is also known that language difficulties can contribute to the development of social withdrawal and social phobia (Simms, 2007). In addition, children with language difficulties have nonverbal or faint initiatives that can easily go unnoticed. More sensitivity in regard to communication is required to recognize faint initiatives, and it is important that ECE professionals develop this awareness. Supporting social play and peer-interactions in children’s early years can prevent cumulative social, emotional, and cognitive difficulties in later life (Syrjämäki, Pihlaja, & Sajaniemi, 2019).

As we expected, children with self-regulation difficulties had higher surgency/extraversion. They also exhibited more negative affectivity than the other status groups. In addition, children with self-regulation difficulties demonstrated a preference for rough and tumble play more than the other status groups. This finding was expected based on the teachers’ and parents’ descriptions of restless behaviour. However, it is also important to recognize the positive features of this type of active play behaviour. It might; play that is physical and not aggressive should not necessarily be limited, as it can enhance the development of self-regulation (Carlson, 2011b; Pellis & Pellis, 2007). However, accelerated behaviour and a drive to ‘run wild’ often becomes a problem in group situations. Pedagogical environments require the ability to wait, listen, and delay reactions; in other words, self-regulation is challenged in these settings. Self-regulation skills have been found to moderate the extremes in surgency temperaments (Dollar & Stifter, 2012). Specifically, children with a high surgent temperament who actively seek co-regulation (e.g. need support from adult in play situation), seem to have less extreme behaviour (Dollar & Stifter, 2012). Negative affectivity has been found to predict both internalizing and externalizing problems, and together with high extraversion/surgency, it can cause a wide range of challenges in regard to children's behaviour and learning (Rothbart Mary, 2011).

Children with severe disabilities had significantly less effortful control, positive anticipation, and social play than other children. If they do not have a complete understanding of a situation, children with severe disabilities may not feel excited about an activity that is positively anticipated by their peers. Therefore, it is important that teachers pay attention to these difficulties and help guide children’s positive anticipation. In addition, every child needs a sense of belonging, and they should feel included, accepted, and supported by others (Pesonen, 2016); these feelings, along with positive anticipation, are likely to increase a child’s sense of togetherness. This study did however observe that children with severe disabilities had less negative affect than the other status groups. While limited negative effect could support the development of their social abilities, this finding may be due to an interpretation error: it can be difficult to elicit the thoughts and feelings of children with severe disabilities if their ability to express themselves is limited.

Identifying the strengths and challenges in every child is a prerequisite for supporting development and well-being. Moreover, it is important to create moments that result in positive emotions and expressions of joy for children’s successes. An environment that accommodates the temperament of a child can enhance their developmental potential and boost learning (Kumpulainen, Mikkola, Rajala, Hilppo, & Lipponen, 2015). A teacher–child relationship has a moderating effect on the association between temperament and peer play interaction (Sin, 2019); therefore, it is essential to support ECE professionals and their development of proficiency in sensitive pedagogical guidance and co-regulation (Heritage, 2016).

High-quality ECE should pay attention to children's individual skills, acquirements, and ways of learning (Pihlaja & Viitala, 2018). Therefore, this study has focused on increasing the knowledge of children’s play behaviour and how it is related to temperament and SEN. The results emphasize the importance of examining these issues and indicate that further research is needed.

Limitations and future directions

Temperament research has limitations in regard to the measurability of the phenomenon. In this study, temperament was evaluated by parents; however, the accuracy of this method has been challenged. For example, Kagan (1998) argued that questionnaires based on parents’ reporting of their child’s behaviour are not reliable in child development research, as parents may want to portray their child primarily in a positive light. Despite this concern, Rothbart Mary (2011) has supported the use of parent-filled questionnaires, stating that they offer repeated observations across many different situations. Questionnaires are also an effective method for collecting data from a large number of informants, such as in our study. The reliability of the research could be improved by including observational data; however, this was not an option in the current study because of resource constraints. The tools (CBQ, PPBS) used in this study are relatively old, but widely used and also translated into Finnish.

The distribution for status groups was based on both medical documents and the teachers’ descriptions of the children’s general difficulties and behaviour in the ECE environment. This distribution was based on careful assessment by a clinical neuropsychologist. According to which, overlap between status groups was practically non-existent. However, this possibility of overlapping should be considered.

In addition, there was a limitation in this study concerning the slightly low alpha coefficient in the reliability analysis when compiling the sum variables of the temperament sub-dimensions. Nonetheless, Rothbart and Bates (2006) completed careful and accurate factor analyses when they originally examined the three main temperament dimensions. Hence, the current study utilized the same dimensions and approved the limitation of the low alphas; however, this should be taken into consideration when examining the results.

Future studies should investigate how co-regulation is visible in ECSE using observational data and interviews that consider the children's perspectives. It would give an opportunity to study the impact of co-regulation on support for children's play when they have SEN.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was supported by the Jenny ja Antti Wihurin Rahasto and Antti Wihuri Foundation under grant 170155.

Notes on contributors

Jonna Kesäläinen

Jonna Kesäläinen is a PhD student at the University of Helsinki. Her research is in the field of early childhood special education focusing on children’s learning, temperament and stress regulation.

Eira Suhonen

Eira Suhonen is an adjunct professor at the University of Helsinki. Her main research interest is in early childhood special education. She has focused in her research on play and early intervention. She is also specialized in analyses of biomarkers of stress response regulation in children.

Alisa Alijoki

Alisa Alijoki is University Lecturer at the University of Helsinki. Her main research interest is in early childhood special education. She has focused in her research on early intervention.

Nina Sajaniemi

Nina Sajaniemi is a professor of early education and a clinical neuropsychologist. Her areas of expertise are in developmental sciences and in educational psychology. Her research interests are in stress regulation, interaction, social behaviour and pedagogical sensitivity.