Executive functions (EFs) are the cognitive processes that control and regulate goal-directed behaviors (Barkley, 2012; Miyake and Friedman, 2012; Diamond, 2013). Executive functions have been investigated at a basic, core-process, level and at a higher cognitive-control level. For example, working memory, inhibition, and set-shifting are generally considered to be core components of executive functions (Miyake et al., 2000; Diamond, 2013). Working memory is the ability to hold information in mind for a short period of time and, if necessary, to mentally manipulate this information (Miyake and Friedman, 2012; Diamond, 2013). Evidence suggests that this ability to keep representations in mind develops in the first 6 months of life (Pelphrey et al., 2004; Reznick et al., 2004) and improves across preschool years (Gathercole, 1998, 1999; Espy and Bull, 2005). Inhibition is the ability to control one's attention, thoughts, behaviors, and/or emotions to override an external temptation or a strong internal predisposition (Miyake and Friedman, 2012; Diamond, 2013). Response inhibition, the ability to suppress a dominant response, particularly develops in the 1st year of life, continuing to improve over toddlerhood and the preschool years (Kochanska and Aksan, 1995; Garon et al., 2008; Kochanska et al., 2008; Friedman et al., 2011). Lastly, set-shifting is the ability to shift attention between different tasks and mental sets (Miyake and Friedman, 2012; Diamond, 2013); in addition, both response shifting and attention shifting develop over preschool years (Zelazo et al., 1996, 2003; Espy et al., 1999; Garon et al., 2008).

It is hypothesized that the foundation of these core components lies in infancy or toddlerhood (Garon et al., 2008; Anderson and Reidy, 2012). However, there are also higher-level and more complex EF skills such as planning, reasoning, and problem-solving. It has been suggested that these higher EF skills are likely to be dependent on lower-level core aspects of EF (including working memory, inhibition, and set-shifting) (Figure 1, McCormack and Atance, 2011; Miyake and Friedman, 2012; Diamond, 2013). Developmental research into EF has largely been focused on the development of these lower-level core aspects of EF; however, research into the development of higher-level EF aspects has been relatively neglected. In the current study, we focus on the development of a naturalistic measurement of the higher-level EF component of planning. Planning is a complex set of mental and behavioral operations that brings together cognitive, emotional, and motivational resources to achieve the desired goals (Shallice, 1982; Friedman and Schonick, 2014). In past studies, planning of simple actions and action sequences in infancy and childhood has been linked to improvements in the core aspects of executive functions (Pennequin et al., 2010; Gottwald et al., 2016; Yanaoka and Saito, 2019, 2020; Schröer et al., 2021), and planning abilities on standard tasks such as the Tower of London have argued to be related to inhibition and set-shifting (Baughman and Cooper, 2007; Cooper and Marsh, 2016), suggesting a tight link between lower- and higher-level components of EF.

There is substantial evidence that the core aspects of EF are related to school readiness and academic achievement in early childhood years (e.g., Bull and Scerif, 2001; Blair and Raza, 2007; McClelland et al., 2007; Espy et al., 2010; Welsh et al., 2010; Fuhs et al., 2014; Blair et al., 2015; Blair, 2016). For example, Fitzpatrick and Pagani (2012) found that improved working memory skills in toddlerhood were related to better classroom engagement, number knowledge, and receptive vocabulary in kindergarten, while Welsh et al. (2010) found that working memory at the beginning of kindergarten was related to improved emergent literacy and numeracy skills. Furthermore, working memory efficacy has been found to contribute to variance in school achievement at the end of Grade 1 (Monette et al., 2011). Better working memory scores have also been found to provide children with a head start in mathematics and reading achievement (Bull et al., 2008). However, while both working memory and inhibition have been found to predict early arithmetic competence in children between 2 and 5 years of age, Espy et al. (2010) found that only inhibition explained unique variance in mathematical skills after controlling for other core aspects of EF. Moreover, Blair and Raza (2007) found that inhibition was positively related to mathematical ability and knowledge of letters within emerging literacy in 3- to 5-year-olds.

Similarly, many studies have found a combination of several core aspects of EF that are related to school readiness and academic achievement, suggesting that those higher EFs that are dependent on multiple core aspects of EFs might also relate to school readiness and school success. For example, the combination of inhibition and set-shifting was found by Shaul and Schwartz (2013) to be related to emergent literacy and mathematical knowledge in preschool children, while Bull and Scerif (2001) found that mathematical abilities in children of 7 years of age were associated with measures of set-shifting, inhibition, and working memory. Furthermore, the head-to-toes task, which assesses inhibition, attention, and working memory, significantly predicted emergent literacy, vocabulary, and math skills (McClelland et al., 2007). Similarly, set-shifting and inhibition at age 4 were related to mathematical achievements at age 6 (Clark et al., 2010). Lastly, low-level EF components such as improved working memory are related to more appropriate classroom behavior, such as a larger attention span, lower levels of distractibility, fewer problems in monitoring the quality of working, and better ability to generate new solutions (Gathercole et al., 2008), suggesting that EFs contribute to school success.

While it is well-known that the development of lower-level EF skills is related to improved school readiness in later life, much less is known about how the development of higher-level EF skills such as planning is related to school readiness and school success. This stands in contrast to the fact that, in later life, for example in late adolescence, higher-level EFs such as planning are known to be important for academic success (Baars et al., 2015). Furthermore, problems in classroom behavior that are associated with poor working memory are also associated with lower scores on measures of planning (St. Clair-Thompson, 2011). Only one study has shed some light on the relationship between planning and school readiness; in the study, performance on the Tower of London task during 4 to 5 years of age was predictive of early math abilities (Bull et al., 2008), but how planning in very early childhood is related to later school success remains unknown. This is an important question because the first formal school systems begin during 3 to 4 years of age in many countries. However, most developmental studies exploring EFs and school readiness focus on children 4 to 5 years of age (e.g., Bull et al., 2008; Clark et al., 2010; Welsh et al., 2010; Monette et al., 2011).

There are various tasks to provide measures of the core aspects of executive functions in early childhood (e.g., Garon et al., 2008; Howard and Melhuish, 2017). However, naturalistic tasks that can be used in everyday nursery contexts are rare. This is particularly true for children in their 3rd year of life when school readiness is of critical concern. The main reason for this is that 2- to 3-year-olds are an especially difficult age group to study as they are highly mobile and have low concentration, making them difficult to assess using rigid standardized tasks. One exception is Schröer et al. (2022) coin sorting task, which is designed to measure the ability of 2- to 3-year-olds to plan extended alternating action sequences. This task is especially simple and only requires some plastic coins and two boxes. The experimenter demonstrates sorting the coins between the two boxes in a left–right–left–right (alternating) pattern. Toddlers are then instructed to continue putting the coins in the boxes in this alternating way. Schröer et al. (2022) found that the ability to plan and execute these alternating action sequences improved between 2 and 3 years of age and, critically, was predicted by working memory capacity.

However, real-world action planning often involves multiple goals, constraints, or goal hierarchies. For example, something as simple as making a cup of tea in the morning involves a goal hierarchy of setting an overarching goal and then establishing the necessary subgoals to accomplish the task (Cooper et al., 2014). Freier et al. (2017) used a coloring task in which children had to color six farm animals following the direction of an arrow and use each coloring pencil equally often as a measure of hierarchical planning abilities in 3- to 5-year-olds. In this study, action planning involved taking into account a higher- and a lower-level goal simultaneously. While this task revealed improvements in action-planning abilities between 3 to 5 years of age (Freier et al., 2017) and could be used within a school environment, it is very difficult for 2- to 3-year-olds. Hence, our current task uses Duplo blocks with simpler instructions to investigate planning development in 2- to 3-year-olds. Toddlers were instructed to build a tall tower with a height constraint and to make it striped by alternating between two colors. Performance on these two constraints was used to assess planning improvements. This simple task is short, does not require complex materials, and is not dependent on verbal expressive abilities that are known to be poor in this age group (Colson and Dworkin, 1997).

The aim of the current study was to develop a measure of planning based on a naturally occurring behavior suitable for use with children on the cusp of entering the school system. To this end, children were instructed to build a tall tower (height constraint) that was striped—i.e., alternating between colors (striped constraint)—with Duplo blocks. In this particular study, we aimed to evaluate this task cross-sectionally in 2-year-olds and investigate whether performance on this task is related to performance on standardized measures of executive functions. It was hypothesized that planning actions according to multiple goal constraints would improve over the 3rd year of life and that children's planning ability would be related to their performance on standard tests of the core components of executive functions. If true, then the quality of tower building could be used as a proxy measure of executive function development.

All analyses were carried out using RStudio (version 1.2.1335). The data demonstrating that performance on both the working memory task as well as the inhibition task improves with age in this sample can be found in Schröer et al. (2022). There were no significant differences in the age of children who were excluded from the sample as compared to those who remained in the analyses for both the 24- and 30-month-old groups (24 months: t(17) = 1.10, p =.332; 30 months: t(17) = 0.87, p =.398). We first discuss performance on the tower building task before turning to an examination of the association between performance on that task and traditional EF measures.

This study presents a simple and cost-effective way of assessing planning and executive functions in very young children in naturalistic environments. Children participated in a task in which they were instructed to build a Duplo tower taking into account two goal constraints: the height and the striped constraint. The task is simple and naturalistic, allowing its use in nurseries and school settings and making it particularly well suited for use in future to investigate the relationship between planning and school readiness. It is a coarse measure of EF level of development which can be used as a screening instrument. One advantage of the Duplo tower task is that it does not require a language response. It is also simple to operationalize with materials commonly available in nurseries or at home. Most 2-year-olds showed evidence of acting in line with the goal, that is, they built a tower of at least three blocks, demonstrating that they at least understood the instructions and were motivated to act in line with the instructions. Children between 2 and 3 years of age were equally able to achieve the height constraint; however, the ability to achieve the striped constraint improved with age. Older children were more likely to coordinate the two constraints effectively and build a striped tower.

Importantly, inhibition scores measured using a standard task could be related to performance on the height constraint. Children who were successful on the height constraint were also more likely to be better at the inhibition task. Moreover, working memory was predicted by performance on the striped constraint, and better performance on the planning task was associated with a higher working memory score. Thus, while the tower task does not have the resolution of many of the standard EF tasks, it provides an easy-to-use, fun, and natural screening tool for EF-related school readiness factors, such as planning, working memory, and inhibition. This might be especially important given that, in late adolescence, planning is known to be an important factor related to academic success (Baars et al., 2015), and many studies have already demonstrated a relationship between working memory and inhibition and between school readiness and school success in childhood (Bull and Scerif, 2001; Blair and Raza, 2007; Bull et al., 2008; Gathercole et al., 2008; Clark et al., 2010; Espy et al., 2010; Welsh et al., 2010; Monette et al., 2011; Fitzpatrick and Pagani, 2012; Shaul and Schwartz, 2013).

Our results are also consistent with the hypothesis that the development of executive functions and complex action planning are related (McCormack and Atance, 2011), specifically that planning is a higher-level EF that draws upon the lower-level core EF components (Diamond, 2013). First, working memory could be predicted by planning action sequences (i.e., striped constraint). This is consistent with a previous study in which working memory was related to planning simple alternating actions in toddlerhood (Schröer et al., 2022). Second, inhibition was predicted by age as well as the height constraint in the action sequence planning task. In a previous study, inhibition was also found to be related to avoiding actions irrelevant to the goal hierarchy in the preschool period (Schröer et al., 2021). Thus, it might be that inhibition is essential in avoiding actions that do not achieve the goal and focusing on actions that accomplish the goal. In short, our results provide further evidence that planning is dependent on lower-level EF components such as working memory and inhibition (Diamond, 2013).

Interestingly, the working memory score was not predicted by the height constraint, while inhibition was not predicted by the striped constraint in the Duplo tower task. Furthermore, working memory did not predict the inhibition score. In adulthood, the three core components of EF are correlated but functionally separated (Miyake and Friedman, 2012). This distinction into separate but related core aspects of EF is not clear in infancy (Hendry et al., 2016), but several studies have detected separable components in 2- to 3-year-old children (Garon et al., 2014; Skogan et al., 2016). In the current study, working memory and inhibition scores were not associated, suggesting separable functions. Similarly, Gottwald et al. (2016) did not find a correlation between simple inhibition and working memory in 18-month-olds. Thus, our study suggests that both working memory and inhibition contribute independently to the higher-level EF of planning. Further research could investigate this link between the three lower EF components and how these components relate to higher-level EFs such as planning, reasoning, and problem-solving.

One limitation of the current study is that a cross-sectional design was used to investigate age-related effects on the Duplo planning task. Future studies could investigate the improvement of planning over preschool age by using a longitudinal design to examine whether this task can be used to predict EF and school readiness prospectively. A second potential limitation is that we did not take into account the children's height. All children were seated at the same table and the same chair, and the picture was located at exactly the same location on the wall. However, as there were no age effects in the height construction constraints, it is unlikely that age-related physical height of the children influenced their tower building performance.

Future studies could also use this Duplo tower task to assess the relationship of planning (higher-level EF) in very early childhood with school readiness and academic success in later childhood. This research area is of special interest since many studies have assessed EFs to relate to school readiness and success among children of 4–5 years of age (e.g., Bull et al., 2008; Clark et al., 2010; Welsh et al., 2010; Monette et al., 2011), while, in many countries, the first formal school systems begin during 3 to 4 years of age. For example, prior research has shown that working memory in toddlerhood is related to better classroom engagement and school success in mathematics and literacy in kindergarten (Fitzpatrick and Pagani, 2012), suggesting that assessing EF abilities in toddlerhood is important to understand the relationship between EFs skills and school readiness and school success later.

The aim of the current study was to develop a naturalistic measure of planning that could be used in the real world. A key challenge in psychology and neuroscience is to move away from traditional laboratory studies and investigate the development and brain function in the real world (Dahl, 2017; Pinti et al., 2018; Matusz et al., 2019). Studying development in controlled lab contexts removes the richness of the real world (Cantlon, 2020). The key challenge of moving away from lab-based tasks to naturalistic and easily administered behavioral tasks is also present in educational neuroscience (Janssen et al., 2021). A first step in addressing this issue is to create easily administered behavioral tasks that can be used in a nursery or school setting to assess cognitive and neurocognitive processes such as planning or other executive functions that are relevant for understanding the relationship between cognition and school readiness and school success. The Duplo tower task is a good example of this; it is an easily administrated naturalistic behavioral task that can assess very young children's planning, working memory, and inhibition skills. Furthermore, it is simple and does not require any complex material meaning that it could be used to assess these abilities in everyday nursery or school settings. These simple naturalistic behavioral tasks could be used in future to assess the relationship between cognition and education in nurseries, schools, or anywhere outside of the traditional laboratory, for example, the relation between planning abilities in toddlers and school readiness in the nursery environment.

The next step for educational neuroscience is to adopt simple naturalistic tasks such as the Duplo tower task to assess brain development using wireless neuroimaging methods. This would allow the investigation of the functional brain development associated with EF development that is critical for school readiness. In cognitive neuroscience, research into real-world neuroscience has already demonstrated that wireless fNIRS can be used to examine brain patterns during real-world behavior in adults (Pinti et al., 2015; Balardin et al., 2017), making it a particularly interesting method to apply to educational neuroscience with naturalistic tasks (Janssen et al., 2021). Similarly, mobile EEG can be used to assess differences in processes in the classroom, such as attention (Xu et al., 2022). In the same way, as adult cognitive neuroscience must be taken out of the laboratory and into the real world, educational neuroscience research needs to be taken out of the laboratory and into real classrooms.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Birkbeck College Ethics Committee. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

DM, RC, and LS conceived and designed the study and wrote the manuscript. LS collected the data. DM and LS analysed the data. All authors contributed to the article and approved the submitted version.