This is the first paper to model a comprehensive and concurrent set of predictors across health, development, and environment in relation to self-regulatory development of young children, across a two-year period beginning from age 4-5 years. Controlling for a range of background factors, significant predictors of self-regulatory growth included: fewer behavioural sleep problems; higher gross motor and pre-academic skills; lower levels of maternal and paternal angry parenting; lower levels of financial hardship; and marginal effects for home learning environment and child-educator relationships. As predictors were modelled simultaneously, significant findings provide a (likely conservative) estimate of the associations between each variable and self-regulation change, over and above the combined associations of all other variables in the model. While previous studies have provided insight into the transactional mechanisms between some factors known to influence self-regulation (e.g. parenting and sleep), this model better reflects the complexity of children’s lives and the combined impact of a range of factors on self-regulatory change. Thus the study makes an important contribution toward prevention and intervention efforts by identifying the most salient and high-potential factors to target for self-regulation interventionists, taking a holistic approach to supporting self-regulatory growth in young children.
Substantial research and theory supports both acute and persistent associations of self-regulation with learning and academic skills (33) with self-regulation typically positioned as a predictor of academic skills. In a related finding, but with self-regulation as the outcome, in our model pre-academic skills were one of the strongest predictors of self-regulation growth. It is clear why self-regulation would predict learning and academic skills: the ability to direct and sustain attention, tackle new challenges, resist maladaptive impulses, and work collaboratively and pro-socially with others – all hallmarks of high self-regulation – serve to support on-task behaviour, effort and persistence during learning. However, there is comparatively less research focused on the possible reciprocal effects with pre-academic skills predicting self-regulation growth. A number of explanations are feasible. First, it is likely that self-regulation and early literacy and numeracy skills, as represented by our pre-academic skill assessment, develop in a bidirectional manner across early childhood (34, 35). For example, time spent in focussed literacy and numeracy learning activities provide the opportunity to extend and enhance self-regulatory capacities, particularly in attentional and cognitive control aspects. It is likely that had we had an earlier and multiple measures of both self-regulation and early concept comprehension, literacy, and numeracy, we would have established birdirectional and reciprocal associations across time. A second and related explanation is that the pre-academic assessment used here may have tapped children’s visual-motor skills given it was a pencil and paper task requiring the writing of letters. While there was no visual-motor data available for children in this dataset, scores on the pre-academic test did correlate (r = .40) with the fine motor variable in our model (single item of teacher report of fine motor competence). Recent research has suggested that visual-motor skills and cognitive self-regulation, as enabled by executive functions, co-develop in a bidirectional manner (35) and it may be that our findings are reflecting a small portion of this transactional process at this period of development. That is, children who scored more highly on the pre-academic score may have done so due to higher visual-motor skills, which may themselves co-develop with and support self-regulatory growth.
Pre-school gross motor abilities were also significantly, albeit modestly, associated with children’s self-regulation growth. This is consistent with suggestions of common mechanisms (i.e., executive functions) that are implicated in both self-regulation and motor learning (36-38), such that both show common areas of neural activation, are impaired after damage to neural regions for the other, and are often both impaired in cognitive disorders, such as ADHD and dyslexia. Indeed, tasks that are motor-demanding for young children, such as navigating uneven surfaces and/or obstacles, are more cognitively demanding and lead to more cognitive errors than less cognitively demanding motor tasks (39). As such, one possibility is that this finding is indicative of the concomitance between self-regulatory and motor skills. However, that gross motor skills were associated with change in self-regulation may additionally suggest that the acquisition of motor proficiency creates new learning opportunities (40) such as experiences that serve to foster self-regulation (e.g., increased mobility causing children to encounter rules associated with access, involvement in physically active shared play providing opportunities for impulse control and turn-taking, etc.) As such, gross motor skills may open a gateway to important self-regulation-promoting experiences and activities, whereas low levels of gross motor skills might consume much of the cognitive resource that otherwise could be directed toward these same activities.
Another factor that was modestly but significantly and uniquely related to self-regulation growth was sleep problems. This aligns with a large body of existing research that identifies sleep problems as a key contributor to daytime self-regulatory problems in young children both in the short (41) and long term (18, 42). It is possible that behavioural sleep problems in young children reflect an underlying phenotype associated with regulatory problems (43, 44), and/or that early behavioural sleep problems initiate a developmental cascade that disrupts emotional and attentional development over time (15). Either way, brief sleep interventions are known to be safe and effective in improving both sleep behaviours and daytime self-regulatory functioning in young children in both typically-developing (45-47) and clinical populations (48, 49).
Our finding that angry parenting was associated with less growth in self-regulation for children echoes a range of prior studies that have linked aggressive, controlling parenting with poor self-regulation in children (50-54). However, this study extends that work by including not only mothers’ but also fathers’ parenting, a rare inclusion. We suggest that angry parenting as measured here is indicative of dysregulated parenting, and potentially of overall emotional regulation skills of parents. Mechanisms through which this might limit self-regulatory growth in children include heritability pathways in terms of self-regulation capabilities (55), and socialisation pathways in which children learn about self-regulatory behaviours through modelling their parents’ behaviours. It is also important to note that child-driven effects are possible, as reflected in prior studies that show dysregulation in young children is associated with increased parenting stress and more-negative parenting approaches (56, 57). These bidirectional relationships between parenting and children’s self-regulation, which are likely to establish mutual promotion/exacerbation processes over time, were not modelled in this study and should be the focus of future longitudinal work.
A number of socioeconomic variables were associated with enhanced self-regulatory growth including higher household incomes, higher maternal education levels and living in households with lower levels of financial hardship. The experience of significant financial hardships such as those tapped here is likely associated with stressful home environments, which impact on children’s physiology and neurodevelopment in ways that limit their capacity for self-regulation development (58, 59). Indeed, early self-regulation has been identified as one of the foremost mechanisms through which early stressors and socioeconomic disadvantage can lead to poorer academic and wellbeing outcomes (60). For these reasons, much of the prevention and intervention focus to date has been on children from disadvantaged backgrounds in an effort to address socio-economic gradients in achievement likely mediated through early self-regulatory capacity. Our findings suggest this focus is well-placed.
Marginal effects were also found for the association between educator-child relationships and the home learning environment, with self-regulatory change. The finding regarding importance of educator-child relationship in terms of children’s early self-regulation development reflects other similar findings in both Australia (61) and Europe (62). Positive student-teacher relationships likely matter because they set the context within which teachers can enact strategies particularly important for acquiring self-regulation during the preschool developmental period (63) including co-regulation, modelling and coaching (64). Our findings regarding the home learning environment align with a prior American longitudinal study linking parental involvement in home learning activities with children’s self-regulatory development (65).
Although this study included a comprehensive array of predictors of self-regulation growth across a specific period in early childhood, there are a number of limitations related primarily to measurement. Most measures were broad and blunt instruments of their constructs. This reflects the nature of the population dataset, in which a broad spectrum of measures capturing child development and the environment were desired, rather than an in-depth measurement of any particular constructs. In addition, our self-regulation composite was only available at two time points in this dataset, meaning that more sophisticated growth curve modelling, which requires a minimum of three time points, could not be undertaken. It is also important to note that although we included a wide array of predictors, nearly 60% of the variance in our self-regulation composite at 6-7 years was still unexplained by the model. This suggests that even large-scale studies such as these are missing key ingredients related to self-regulatory growth. Our understandings could be enhanced through studies which capture potential variables that are not often measured, including chronic stress (e.g. cortisol), psychophysiological arousal and regulation, sensory processing, and more detailed understandings of the nature of home learning and early education and care activities. Finally, it is important to note that participants in this study were recruited in 2004. While it is anticipated that there has been limited change in most lifestyle factors investigated (e.g., parenting), new cohort studies are required to better understand the influence of more prominent societal change such as increased access and use of digital devices.