BRIEF RESEARCH REPORT

published: 10 January 2022

doi: 10.3389/fpsyg.2021.773492

Parent Provision of Choice Is a Key

Component of Autonomy Support in
Predicting Child Executive Function
Skills
Romulus J. Castelo 1 , Alyssa S. Meuwissen 1 , Rebecca Distefano 1,2 ,
Megan M. McClelland 3 , Ellen Galinsky 4 , Philip David Zelazo 1 and Stephanie M. Carlson 1*
1
Institute of Child Development, University of Minnesota, Minneapolis, MN, United States, 2 Teachers College, Columbia
University, New York, NY, United States, 3 Human Development and Family Sciences, Oregon State University, Corvallis, OR,
United States, 4 Bezos Family Foundation, Seattle, WA, United States

Although previous work has linked parent autonomy support to the development
of children’s executive function (EF) skills, the role of specific autonomy-supportive
behaviors has not been thoroughly investigated. We compiled data from four preschool-
age samples in the Midwestern United States (N = 366; M age = 44.26 months;
Edited by:
Jin Sun, 72% non-Hispanic White, 19% Black/African American, 5% Multiracial) to examine
The Education University three relevant autonomy-supportive behaviors (supporting competence, positive
of Hong Kong, Hong Kong SAR,
China
verbalizations, and offering choice) and their associations with child EF. We coded parent
Reviewed by:
autonomy-supportive behaviors from a 10-min interaction between parent and child
Ruth Ford, dyads working on challenging jigsaw puzzles together. Children completed a battery of
Anglia Ruskin University,
EF. Overall, child EF was most consistently correlated with the offering choice subscale.
United Kingdom
Carolyn Palmquist, Additionally, only the offering choice subscale predicted child EF while controlling for
Amherst College, United States the other autonomy support subscales and child age. These results suggest that
*Correspondence: parent provision of choice is an especially relevant aspect of autonomy-supportive
Stephanie M. Carlson
smc@umn.edu
parenting and may be important to the development of EF in early childhood. Future
research should directly measure children’s experience with choice and how it relates to
Specialty section: emerging EF.
This article was submitted to
Developmental Psychology, Keywords: autonomy support, parenting, early childhood, choice, executive function (EF)
a section of the journal
Frontiers in Psychology
Received: 10 September 2021 INTRODUCTION
Accepted: 13 December 2021
Published: 10 January 2022 There is a wealth of evidence for the predictive importance of early executive function (EF) skills
Citation: for social and moral competence, emotion regulation, and academic achievement (e.g., Kochanska
Castelo RJ, Meuwissen AS, et al., 2000; Carlson and Wang, 2007; McClelland et al., 2013; Allan et al., 2014; Jacob and Parkinson,
Distefano R, McClelland MM, 2015; Willoughby et al., 2017). EF refers to a set of higher-order neurocognitive skills that are
Galinsky E, Zelazo PD and critical for goal-directed behaviors and self-regulation and that are thought to be comprised of
Carlson SM (2022) Parent Provision
working memory (the ability to hold relevant information in mind), inhibitory control (the ability
of Choice Is a Key Component
of Autonomy Support in Predicting
to suppress prepotent behavioral responses), and cognitive flexibility (the ability to think about
Child Executive Function Skills. different aspects of a stimulus or situation) (Miyake et al., 2000; Diamond, 2013). EF skills undergo
Front. Psychol. 12:773492. exceptionally rapid development during the preschool period, peaking in late adolescence, before
doi: 10.3389/fpsyg.2021.773492 declining into later adulthood (Zelazo et al., 2013). Importantly, evidence suggests that EF skills

Frontiers in Psychology | www.frontiersin.org 1 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

are highly malleable and are sensitive to training (Diamond and EF in children is providing children with choices. Theoretical
Ling, 2019). Given the relevance of EF skills for a range of child accounts on the development of autonomy highlight the
outcomes and their plasticity in early childhood, researchers have importance of choice and perceived control – the child’s sense
become increasingly interested in identifying factors that might of agency (Ryan and Deci, 2006). As put forth by James
support the early development of these important skills. (1890/1950) and Baldwin (1892), a sense of volition enables
Ecological models of development suggest that EF emerges one to attend to things on purpose and exercise the will as a
in the context of multi-level biological and contextual processes conscious choice. This is relevant to EF because before children
(Blair and Raver, 2012; Carlson et al., 2013). Considering that can exert conscious control over their behaviors, they must
caregivers make up most of children’s early social interactions, first realize that they have a choice in how to act, think, or
an emerging area of research has concentrated on how parenting feel. For example, imagine a scenario where a toddler is given
behaviors might serve as potential facilitators for EF skill a choice of which cereal to eat for breakfast. The ability to
development in children. Researchers have examined how make a conscious choice prompts the child to develop a sense
transactional parent-child dynamics contribute to the emergence of control over their environment, take ownership over their
of EF. For example, Carlson (2003) proposed three dimensions decision, and resist an emotional meltdown. It is important to
of parenting that might promote EF development in children: recognize that provision of choice in the context of autonomy-
sensitivity, autonomy support, and mind-mindedness. One supportive parenting has reasonable limits. In the example
common conceptualization of autonomy-supportive parenting above, parents might provide their child with limited options
defines it as behaviors that serve to instill children with a sense of for which cereal to eat for breakfast, as opposed to allowing
agency over their own actions by scaffolding difficult tasks, taking their child to choose whatever they want to eat. The latter
the child’s perspective, and offering choices (Deci and Ryan, 2000; characterizes a laissez-faire type of parenting. Furthermore, as
Grolnick and Farkas, 2002). However there is not full agreement children use their working memory and inhibitory control skills,
on what behaviors constitute autonomy-supportive parenting in they actively select goal-relevant information to hold in mind
the literature (McCurdy et al., 2020). In the current study, we while rejecting non-relevant information and associated actions,
conceptualize autonomy support based on four dimensions: the however, tempting they might be. When building a tower, for
extent to which the parent adapts the task according to the child’s example, they make choices that require attending to the size
needs, encourages and provides the child with suggestions using of the base and ignoring the fancy top piece with a turret
a positive tone of voice, shows flexibility in their attempts to keep until the end. If someone else were to tell them which piece
the child on task, and provides choices to ensure that the child to place each time, or do it for them, or determine the pace
plays an active role in the task while following the child’s pace at which they proceeded, they would not have the opportunity
(Whipple et al., 2011). to make choices and exercise these skills. In contrast, early
There is robust evidence to suggest that autonomy-supportive and repeated experiences with choice could increase children’s
parenting is most consistently predictive of EF skills in children perceived control (i.e., their sense of agency) and strengthen
(Bernier et al., 2010, 2012; Fay-Stammbach et al., 2014; Matte- their EF skills through autonomous action. In line with this
Gagné et al., 2015; Valcan et al., 2018). For example, Bernier et al. theoretical framework, we propose that parent behaviors that
(2010) found that maternal autonomy support was the strongest consistently provide children with opportunities to make choices
predictor of EF in young toddlers, independent of general are especially important to the development of child EF skills,
cognitive ability and maternal education, over and above parent above and beyond some other autonomy-supportive behaviors.
sensitivity and mind-mindedness. Hence, subsequent work has We tested this prediction in the current study by examining the
focused on autonomy support. In studies by our team, Distefano extent to which different types of autonomy-supportive behaviors
et al. (2018) found that parent autonomy support positively during a brief parent-child interaction were associated with
predicted a composite measure of child EF in a diverse sample children’s EF performance.
over and above child age and academic knowledge. These results
also held true in a homeless and highly mobile sample (Distefano,
2019). Meuwissen and Carlson (2015) reported similar findings METHOD
in fathers such that father autonomy support was positively
associated with child EF controlling for family income and child Participants
verbal abilities. In a later study training study, we found that Participant data across several study samples from one lab were
a brief intervention increased parent autonomy support and compiled to analyze a total of 366 children (52% male) and their
child self-regulation (Meuwissen and Carlson, 2019). Still, it caregivers (49% fathers and 51% mothers). Child age ranged
remains unclear whether different aspects of autonomy support from 35 to 73 months (M = 44.26, SD = 9.03) and caregiver age
equally predict child EF. To address this gap, we compiled ranged from 19 to 70 years (M = 33.67, SD = 5.94). A majority
samples from prior published studies in our lab to examine these of the children were non-Hispanic White (72%), 19% were
associations more closely. Black/African American, and 5% were Multiracial. The average
Most of the existing literature has examined this association family income in the last year ranged from a bracket of less than
using an overall autonomy support score based on observations $25,000 to $200,000 or more with the mean corresponding to
of a brief parent-child interaction. One aspect of autonomy $75,000–$99,999. Sixty percent of parents had an education level
support that we predict is particularly important to developing of a bachelor’s degree or higher.

Frontiers in Psychology | www.frontiersin.org 2 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

Procedure when the child digressed from the task for more than 5 s); (4)
Samples from four independent studies were combined for the Provides the child with opportunities to make choices on their
current study. All children were pre-screened for documented own, and follows the child’s pace while ensuring that the child
developmental delays or physical or language barriers to takes an active role in the completion of the task. Each subscale
participating. The first sample was recruited from a university was rated from 1 (not autonomy-supportive) to 5 (extremely
participant database (Meuwissen and Carlson, 2015). Fathers autonomy-supportive). The Control and Laissez-faire scales were
and their children (N = 101) worked on a puzzle task together coded as well but were not the focus of the present study (further
for 10 min in the laboratory. Children completed a battery of details can be found in the original publications). Two trained
EF tasks. The second sample consisted of parent-child dyads research assistants coded 25–33% of the videos to establish inter-
(N = 71) recruited through Pre-K programs that were considered rater reliability. Inter-rater reliability for each subscale across the
Title I or served Title I eligible children (i.e., schools with a full sample was excellent (Supporting Competence ICC = 0.781,
large low-income population) (Distefano et al., 2018). Dyads Positive Verbalizations = 0.807, Offering Choice ICC = 0.806).
completed the puzzle task together during a parent night at Inter-rater reliability for the overall autonomy support scores was
the child’s school. Children completed a battery of EF tasks also excellent (ICC = 0.869).
at their school with a researcher. The third sample consisted
of parent-child dyads (N = 73) that resided at a shelter
Child Executive Function
for families experiencing homelessness. Parents and children The EF measures varied across samples and each sample
completed the puzzle task together in a research room at determined that combining the EF measures into a composite was
the shelter. Children completed a battery of EF tasks with warranted by the inter-correlations among EF tasks. Standardized
a research assistant (Distefano, 2019). The remaining parent- scores on each task were averaged to create a composite EF
child participants (N = 121) were recruited from a university variable. We list the EF measures that made up the composite for
participant pool including both mothers and fathers. Dyads each study sample in the Supplementary Table 6. The following
completed all study activities in the laboratory with a research tasks were used.
assistant (Meuwissen and Carlson, 2019). Given this was a Bear/Dragon
training study, only the first timepoint was included in our The Bear/Dragon task (Kochanska et al., 1996) is a simplified
analyses. In each study, the procedures were video recorded and version of the Simon Says game that was adapted by Meuwissen
took approximately 60 min to complete. Data collection took and Carlson (2015) to include a range of difficulty. The task began
place in Minneapolis, Minnesota, with the exception of the Title with Level 1 followed by subsequent levels if children completed
1 school district, which was located in Evansville, Indiana. In at least 8 out of 10 items correctly. Children were instructed to
each of the samples, all study procedures were reviewed and do various actions (e.g., touch your tummy) by the “nice bear”
approved by the University of Minnesota’s Institutional Review puppet but not to do the actions directed by the “naughty dragon”
Board. Informed consent was obtained for all adult participants. puppet. The puppets were voiced by the experimenter in distinct
voices. In Level 1, five commands were first given from the bear
and then five more commands were given from the dragon, with
Measures
the experimenter holding the child’s hands on the table during the
Dyadic Puzzle Task
dragon trials. Level 2 progressed identically except that children
An experimenter asked parents and children to complete puzzles were directed to sit on their hands during the dragon trials. In
together for 10 min. The instructions and procedure were Level 3, 10 commands were given, alternating from the bear and
identical across all studies. The sets of puzzles were selected dragon. In Level 4, the first five commands were given, alternating
to be slightly too difficult for children to complete on their from the bear and dragon, followed by five more commands
own, such that some adult assistance would be necessary. The where the rules were switched (children were instructed to do
instructions given to parents were purposefully vague to evoke what the dragon said but not what the bear said). Each child was
more natural interactions between the parent-child dyads: “We given a score of 0–4 to describe the highest level passed.
would like to see what your child can do by him or herself, but
feel free to provide him or her with any help that you would Delay of Gratification
like.” The experimenter left the room after providing instructions In the Delay of Gratification task (Mischel et al., 1989), children
and returned after 10 min. The interactions were video and selected their favorite treat from three options. The experimenter
audio recorded and were later coded for four types of autonomy- placed a small number of treats on one plate and a larger number
supportive behavior using a well-established autonomy support on another. The experimenter then explained that she was going
coding scheme (Whipple et al., 2011). Parent behavior was coded to do some work in the corner and that if children waited until she
on four subscales characterizing the extent to which the parent (1) came back, they would get the large pile of treats. Alternatively,
Intervenes according to the child’s needs and adapts accordingly if they rang a bell, the experimenter would return immediately,
to create a more optimal challenge; (2) Provides appropriate hints and children would be given the smaller pile of treats. The
and suggestions, and uses a positive tone of voice to communicate experimenter left the table and sat in a chair behind the children,
to the child that they were there to help; (3) Recognizes and takes pretending to work. The experimenter returned after 10 min or
the child’s perspective and shows flexibility in their attempts to when the children rang the bell, ate the treats, or left the table. The
keep the child engaged on the task (this scale was only coded primary dependent variable from this task was the time until the

Frontiers in Psychology | www.frontiersin.org 3 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

child’s first transgression (touched or rang the bell, touched the children by adding a developmental extension (Anderson et al.,
plates or the treats, ate the treats, or left the table). A score of 600 s 2021). Using a tablet, children were shown five fish in a line
were given to children who never transgressed during the 10 min. and were told by the experimenter that the fish in the middle
is hungry. Children were told that to feed the hungry fish,
Gift Delay they must touch the button on the screen that points the same
In the Gift Delay task (Kochanska et al., 1996), the experimenter way that the middle fish is swimming. After the experimenter
explained to children that they would receive a present but that it demonstrated how to touch the button facing the way the middle
was going to be a surprise. Accordingly, children were told not to fish is pointing, children were given several practice trials with
peek while the present was being wrapped. Children sat down in a feedback. If children passed the practice trials, they proceeded
chair with their backs to the experimenter while the experimenter to the test trials. If children did not pass the practice trials, they
noisily wrapped the present for 1 min. Children were scored on moved down to the developmental extension version of the task.
the level of their transgression: 0 = turned body around, 1 = turned In this version, the experimenter introduced children to one fish
head, 2 = did not peek. and a bowl of food on the screen. Children were instructed to feed
the fish by touching the button pointing in the same direction as
Head-Toes-Knees-Shoulders
the fish. The levels then became increasingly more challenging,
In the Head-Toes-Knees-Shoulders task (McClelland et al., 2014),
scaffolding children’s understanding of the task. The task ended if
the experimenter described two commands to children: “touch
children answered less than 80% of trials correctly or when they
your head” and “touch your toes.” The experimenter then told
completed the final level of the developmental extension. Total
children about the “silly” game where they should do the opposite
scores were computed using a scoring algorithm.
of what the experimenter says. For example, if the command was
to “touch your head,” they should instead touch their toes. There Minnesota Executive Function Scale
were four practice trials with feedback followed by 10 test trials All children across samples were assessed using the Minnesota
without feedback. Children scored a 0 on any given trial if they Executive Function Scale, a standardized application-based
performed an incorrect response. Children received a score of 1 measure delivered via tablet (MEFS; Carlson and Zelazo, 2014).
if they self-corrected and a score of 2 for a fully correct response The MEFS is normed on over 50,000 typically developing
on any given trial. If children earned 4 or more points on the children in the United States (Carlson, 2021). The MEFS is a
10 test trials, they continued to part II where the experimenter broad measure of EF, requiring the use of working memory,
introduced two new rules: “touch your shoulders” and “touch inhibitory control, and cognitive flexibility to be successful. In
your knees.” Part II also included four practice trials followed by the game, children were asked to sort cards into boxes based
10 test trials that included all the rules from Part I and Part II. If on specific dimensions (e.g., by shape or color). For example,
children scored 4 or more points on 10 test trials, they continued children were shown two boxes, one with a green rabbit and
to Part III. In Part III, all the rules that children had learned were another with a purple pig. In the shape game, children were asked
mixed such that if the experimenter instructed them to “touch to put the target card in the box that corresponds with the correct
your head,” they should touch their knees. Points were summed shape (i.e., rabbit card into the rabbit box), disregarding the color.
from all three parts for a possible of 60 points total. Conversely, in the color game, children were asked to put the
target card into the box that matched the color (i.e., green card
Peg-Tapping into the green box), disregarding the shape. Children completed
In the Peg-Tapping task (Diamond and Taylor, 1996), the up to two practice trials with feedback before proceeding to the
experimenter presented children with a wooden dowel used test trials. There were seven levels increasing in complexity and
in the game. The experimenter first demonstrated to children difficulty, with starting level dependent on age. Children were
that when the experimenter taps the dowel once, the children required to correctly sort at least 80% of the trials to move to the
should tap the dowel twice. Children were given one opportunity next level. If they did not pass a certain level, they moved to an
to practice the first rule. The experimenter then introduced easier level until they were able to meet the passing threshold.
the second rule such that when the experimenter taps twice, The task took 4 min to complete on average. Total scores ranging
the children should tap just once. Children were given an from 0 to 100 were computed using an algorithm that accounts
opportunity to practice the second rule. Children were then given for both accuracy and response time. Given that the MEFS was
two pre-test trials. If children answered both trials correctly, the the only EF task that was common across all studies, we report
experimenter provided positive feedback and proceeded with 14 the results for it separately in the analyses that follow.
test trials without feedback. If children answered one or both
trials incorrectly, the experimenter reminded the children of
the rules and completed two additional trials before proceeding RESULTS
with 14 test trials without feedback. The total possible score that
children could earn was 16. Averaging across subscales, overall autonomy support ranged
from 1.0 to 5.0, with a mean of 3.68 (SD = 0.96). Mean scores
National Institutes of Health Toolbox Developmental for each autonomy support subscale are shown in Table 1.
Extension Flanker Given the substantial association between age and EF skills
The National Institutes of Health (NIH) Toolbox Flanker task (see Table 2), we calculated partial correlations among autonomy
(Zelazo et al., 2013) was adapted for younger and less advantaged support and child EF, controlling for child age. Overall parent

Frontiers in Psychology | www.frontiersin.org 4 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

TABLE 1 | Means, standard deviations, and ranges for autonomy support and Table 3). Although model 2 shows that Supporting Competence
child EF measures.
significantly predicted child EF while controlling for Positive
Measure M SD Observed range Verbalizations, this effect disappeared once Offering Choice
was included in the model (see Table 3). Importantly, only
Supporting Competence 3.46 1.11 1–5
the Offering Choice subscale predicted children’s EF over and
Positive Verbalizations 4.03 1.05 1–5
above child age and the other autonomy support subscales. The
Offering Choice 3.62 1.14 1–5
Offering Choice subscale accounted for an additional 4.1 and
Total Autonomy Support 3.68 0.96 1–5
4.9% of the variance in the EF Composite and MEFS, respectively.
EF Composite −0.090 0.73 −1.73 to 2.04
We then conducted the same regression analyses to examine
MEFS 32.20 14.31 0–76
whether these associations persisted in each subsample (see
MEFS, Minnesota Executive Function Scale. online Supplementary Material). A similar pattern of results
emerged in the mothers, fathers, and lab parents subsamples. In
TABLE 2 | Correlations between autonomy support and child EF in multiple contrast, however, the associations between the Offering Choice
samples. subscale and child EF were not statistically significant in the
Title 1 and Homeless subsamples when controlling for the other
Supporting Positive Offering Choice Autonomy
Competence Verbalizations Support subscales, with one exception: Offering Choice predicted MEFS
performance in the homeless and highly mobile sample over
Overall (N = 366, M = 44 mos)
and above child age and the other aspects of autonomy support
EF Comp. 0.27*** 0.14** 0.34*** 0.31***
(p = 0.05, see Supplementary Material).
MEFS 0.28*** 0.15** 0.36*** 0.32***
Mothers (N = 182, M = 48 mos)
EF Comp. 0.39*** 0.21** 0.42*** 0.42***
MEFS 0.36*** 0.20** 0.43*** 0.39*** DISCUSSION
Fathers (N = 178, M = 39 mos)
EF Comp. 0.18* 0.12 0.24*** 0.23** There has been growing evidence for the association between
MEFS 0.21** 0.15 0.28*** 0.25*** autonomy-supportive parenting behaviors and the development
Lab Parents (N = 222, M = 38 mos) of children’s EF, critical neurocognitive skills important for life
EF Comp. 0.13 0.06 0.22*** 0.17* success. Employing secondary data analysis, the current study
MEFS 0.18** 0.01 0.26*** 0.21** examined this association more closely by considering three
Title 1 District (N = 72, M = 53 mos) different types of autonomy-supportive behavior. Specifically, we
EF Comp. 0.43*** 0.31** 0.40*** 0.42*** predicted that giving children opportunities to make choices is
MEFS 0.34** 0.22 0.33** 0.32** an especially important type of autonomy support that helps
Homeless (N = 73, M = 53 mos)
promote the development of children’s EF. Consistent with
EF Comp. 0.29* 0.32** 0.31** 0.37***
this hypothesis, we found that across samples, the Offering
MEFS 0.22 0.26* 0.33** 0.30*
Choice subscale predicted child EF skills independent of child
Partial correlations controlling for child age in months. MEFS, Minnesota Executive age and other autonomy support behaviors. This finding
Function Scale. Data from Meuwissen and Carlson (2015, 2019), Distefano et al.
(2018), and Distefano (2019). Summing across the subsamples does not equate
suggests that the provision of choice in early childhood may
to overall sample size due to some participants being counted in more than one be a particularly effective parenting approach that contributes
category. Flexibility was excluded because it was coded only when the child leaves to building children’s capacity for autonomy, which in turn
the task, which was rare.
promotes their EF skills by helping them reflect on their options
*p < 0.05, **p < 0.01, ***p < 0.001.
and exert self-control over their chosen behavior. Having more
opportunities to practice these skills strengthens them. In turn,
autonomy support was significantly correlated with the child EF growing EF skills would further support children’s agency and
composite and MEFS, controlling for child age, rs = 0.31 and autonomy, in a mutually reinforcing dynamic system. More
0.32, respectively, ps < 0.001. We then assessed the extent to research on how children behave and respond when presented
which each autonomy support subscale correlated with child EF, with choices across early childhood is warranted to test these
controlling for child age. All three subscales were significantly ideas developmentally.
correlated with child EF, indexed by both the EF composite and Our study had several strengths. First, our sample was highly
the MEFS. In line with our hypothesis, we found that child EF diverse. We were able to include both mothers and fathers as
was most strongly correlated with the Offering Choice subscale. well as families from a diverse range of socioeconomic and
This effect was consistent across mothers and fathers as well racial/ethnic backgrounds. The inclusion of a diverse sample
as for diverse socioeconomic and race samples. The Supporting facilitates the generalizability of our findings. Consistent with the
Competence subscale was the next most highly correlated with literature, we found that parent autonomy-supportive behaviors
child EF, followed by the Positive Verbalizations subscale. positively predicted child EF across different subsamples. This
To glean more information about how these three autonomy suggests that autonomy support is important for EF development
support subscales and child EF were associated, we conducted in a broad range of families. Second, we examined different
hierarchical linear regressions, pitting the subscales against one components of autonomy support and its relation to child EF
another in predicting children’s EF over and above child age (see separately. Most of the research linking the two has focused on an

Frontiers in Psychology | www.frontiersin.org 5 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

TABLE 3 | Hierarchical regression analyses of autonomy support subscales predicting child EF.

EF Composite MEFS

Beta t p R2 (p) Beta t p R2 (p)

Model 1
(Constant) −2.82 0.005 0.023 (0.004) −4.17 0.000 0.044 (0.000)
Age (months) 0.15 2.90 0.004 0.21 4.05 0.000
Model 2
(Constant) −3.66 0.000 0.095 (0.000) −5.08 0.000 0.118 (0.000)
Age (months) 0.19 3.72 0.000 0.25 4.94 0.000
Competence 0.29 4.58 0.000 0.29 4.60 0.000
Verbalizations −0.03 −0.45 0.654 −0.017 −0.28 0.779
Model 3
(Constant) −3.69 0.000 0.136 (0.000) −5.16 0.000 0.167 (0.000)
Age (months) 0.18 3.73 0.000 0.24 4.99 0.000
Competence 0.09 1.08 0.279 0.06 0.83 0.405
Verbalizations −0.06 −0.96 0.337 −0.05 −0.85 0.396
Offering Choice 0.30 4.14 0.000 0.33 4.60 0.000
MEFS, Minnesota Executive Function Scale. N = 366 for EF Composite and N = 362 for MEFS.

overall autonomy support score as a predictor of EF (e.g., Bernier completed. It is plausible that levels of autonomy-supportive
et al., 2010, 2012). However, as our findings suggest, various behaviors could change as a function of the type of interaction
forms of autonomy-supportive behaviors could be differentially between parent and child. For example, getting ready to go
important in facilitating EF in early childhood. When examining to preschool in the morning might create more time pressure
each subsample separately, we found that the evidence for choice for parents, resulting in less autonomy support and fewer
as a unique predictor of children’s EF was weaker in our lower- opportunities for choice. Additionally, social desirability bias
income samples (Title 1 and Homeless). This suggests that further might have come into play as parents were aware that they were
research is needed to better understand parents’ beliefs about being observed. Nonetheless, there was sufficient variability in
choice, parent provision of choice, and child choice preference parents’ behavior to detect the associations we reported. Third,
in higher risk environments. the theoretical mechanism underlying the association between
autonomy support (including choice) and EF development is
Limitations and Future Directions children’s enhanced sense of agency and autonomy over their
This study also has some notable limitations. First, there are own actions; however, autonomy itself was not measured in
several ways of conceptualizing the construct of autonomy these studies, given that it is usually assessed using self-report
support (McCurdy et al., 2020). We aimed to capture the questionnaires in older children (e.g., Soenens et al., 2007).
broader array including supporting the child’s competence to Finally, as with most of the research on this topic, our study
solve problems on their own, giving encouragement along was correlational. As a result, we are limited in our capacity
the lines of “you can do it,” and offering choices (Whipple to make causal inferences regarding parent provision of choice
et al., 2011). It is possible that other aspects of parenting, and children’s EF skills. For example, one could argue that
such as providing structure, would prove to be important parents offer children more opportunities to make choices
for EF skills as well (e.g., Landry and Smith, 2010). Also, because they believe that their children are ready to manage
our focal autonomy support subscale was designed to be these responsibilities given higher EF skills. It is also possible
coded as a combination of parent behaviors of providing that children’s experience with choice and their emerging EF
opportunities to children for making choices and following skills are associated in a bidirectional, mutually reinforcing
the child’s pace to ensure that the child had an active role in system developing over time. Future research should incorporate
completion of the task. Offering choice and following the child’s experimental designs with random assignment where parent
pace are adjacent ways of making the child feel a sense of provision of choice is manipulated to better understand the
agency and ownership in the activity. In fact, it is difficult to direction of these effects. There is evidence that parents can
imagine how they could be orthogonal. Nonetheless, although change their autonomy-supportive behaviors during a brief
our prediction was grounded in theory, we are not able to interaction with their child when presented with written and
conclusively claim that parent provision of choice alone is verbal instructions by an experimenter (Meuwissen and Carlson,
driving the association with EF. Future studies should directly 2019). Accordingly, it may be possible to implement an
measure provision of choice by caregivers and examine its intervention where parents are encouraged to provide children
relations with children’s emerging EF. Second, although we with opportunities to make choices. Finally, extending this
were able to observe parent behavior in a largely naturalistic research to teachers will be beneficial in better understanding
setting, the coded interactions between parent and child were how autonomy-supportive behaviors promote positive child
relatively brief and were limited to the puzzle task the dyads outcomes across different types of interactions and relationships

Frontiers in Psychology | www.frontiersin.org 6 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

between adults and children. Meta-analytic evidence suggests of Minnesota, Twin Cities. Written informed consent to
that autonomy-supportive teaching is malleable and predicts a participate in this study was provided by the participants’ legal
wide range of student outcomes (Reeve and Cheon, 2021). Thus, guardian/next of kin.
it will be important to closely consider how teacher autonomy
support facilitates children’s EF skills as they enter school.
AUTHOR CONTRIBUTIONS
CONCLUSION SC conceptualized the study. AM and RD collected the data.
In this study, we demonstrated that parent behaviors that provide RC and SC analyzed and interpreted the data. RC drafted the
children with opportunities to make choices is an especially manuscript with support from all co-authors. All authors gave
important aspect of autonomy-supportive parenting across a their final approval of the manuscript.
diverse range of families. Our findings suggest that children’s
experience with choice, particularly in early childhood, could
be a potential antecedent to the development of EF skills. FUNDING
Additionally, this is a promising area of research for parent and
teacher interventions that aim to improve child well-being. This research was funded in part by a grant to EG from the W.K.
Kellogg Foundation and through a subcontract from this funding
to SC and PZ via the Families and Work Institute. AM and RD
DATA AVAILABILITY STATEMENT also received small grants totaling $2250 from the University of
Minnesota, Twin Cities, Institute of Child Development.
The raw data supporting the conclusions of this article will be
made available by the authors, without undue reservation.
SUPPLEMENTARY MATERIAL
ETHICS STATEMENT
The Supplementary Material for this article can be found
The studies involving human participants were reviewed and online at: https://www.frontiersin.org/articles/10.3389/fpsyg.
approved by the Institutional Review Board at the University 2021.773492/full#supplementary-material

REFERENCES Deci, E. L., and Ryan, R. M. (2000). The “what” and “why” of goal pursuits:
human needs and the self-determination of behavior. Psychol. Inq. 11, 227–268.
Allan, N. P., Hume, L. E., Allan, D. M., Farrington, A. L., and Lonigan, C. J. (2014). doi: 10.1207/S15327965PLI1104_01
Relations between inhibitory control and the development of academic skills Diamond, A. (2013). Executive functions. Annu. Rev. Psychol. 64, 135–168.
in preschool and kindergarten: a meta-analysis. Dev. Psychol. 50, 2368–2379. Diamond, A., and Ling, D. S. (2019). “Review of the evidence on, and fundamental
doi: 10.1037/a0037493 questions about, efforts to improve executive functions, including working
Anderson, J. E., Kalstabakken, A. W., Carlson, S. M., Zelazo, P. D., Distefano, R., memory,” in Cognitive and Working Memory Training: Perspectives From
and Masten, A. S. (2021). Technical Report on the Developmental Extension Psychology, Neuroscience, and Human Development, eds J. M. Novick, M. F.
of the NH Toolbox Flanker Task: E-Prime Version. Minnesota: University of Bunting, R. W. Engle, and M. R. Dougherty (Oxford: Oxford University Press),
Minnesota. 145–389.
Baldwin, J. M. (1892). Origin of volition in childhood. Science 20, 286–287. doi: Diamond, A., and Taylor, C. (1996). Development of an aspect of executive control:
10.1126/science.ns-20.511.286 development of the abilities to remember what I said and to “Do as I say, not
Bernier, A., Carlson, S. M., Deschênes, M., and Matte-Gagné, C. (2012). Social as I do. Dev. Psychobiol. 29, 315–334. doi: 10.1002/(SICI)1098-2302(199605)29:
factors in the development of early executive functioning: a closer look at the 4<315::AID-DEV2<3.0.CO;2-T
caregiving environment. Dev. Sci. 15, 12–24. doi: 10.1111/j.1467-7687.2011. Distefano, R. (2019). Autonomy Support in Parents and Young Children
01093.x Experiencing Homelessness: a Mixed Method Approach. Ph.D. thesis. Ann Arbor:
Bernier, A., Carlson, S. M., and Whipple, N. (2010). From external regulation ProQuest Dissertations Publishing.
to self-regulation: early parenting precursors of young children’s executive Distefano, R., Galinsky, E., McClelland, M. M., Zelazo, P. D., and Carlson, S. M.
functioning. Child Dev. 81, 326–339. doi: 10.1111/j.1467-8624.2009.01397.x (2018). Autonomy-supportive parenting and associations with child and parent
Blair, C., and Raver, C. C. (2012). Child development in the context of adversity: executive function. J. Appl. Dev. Psychol. 58, 77–85. doi: 10.1016/j.appdev.2018.
experiential canalization of brain and behavior. Am. Psychol. 67, 309–318. doi: 04.007
10.1037/a0027493 Fay-Stammbach, T., Hawes, D. J., and Meredith, P. (2014). Parenting influences on
Carlson, S. M. (2003). Executive function in context: development, measurement, executive function in early childhood: a review. Child Dev. Perspect. 8, 258–264.
theory, and experience. Monogr. Soc. Res. Child Dev. 68, 138–151. doi: 10.1111/cdep.12095
Carlson, S. M. (2021). Minnesota Executive Function Scale: technical Report. Saint Grolnick, W. S., and Farkas, M. (2002). “Parenting and the development of
Paul, Minnesota: Reflection Sciences, Inc. children’s self-regulation,” in Handbook of Parenting, ed. M. H. Bornstein
Carlson, S. M., and Wang, T. S. (2007). Inhibitory control and emotion regulation (New Jersey: Lawrence Erlbaum Associates), 89–110.
in preschool children. Cogn. Dev. 22, 489–510. doi: 10.1016/j.cogdev.2007.08. Jacob, R., and Parkinson, J. (2015). The potential for school-based interventions
002 that target executive function to improve academic achievement: a review. Rev.
Carlson, S. M., and Zelazo, P. D. (2014). Minnesota Executive Function Scale:Test Educ. Res. 85, 512–552. doi: 10.1002/14651858.CD009728.pub2
Manual. Saint Paul, Minnesota: Reflection Sciences, Inc. James, W. (1890/1950). The Principles of Psychology (Vol. 1). New York:
Carlson, S. M., Zelazo, P. D., and Faja, S. (2013). “Executive function,” in The Dover.
Oxford Handbook of Developmental Psychology (Vol 1): Body and Mind, ed. P. D. Kochanska, G., Murray, K., Jacques, T. Y., Koenig, A. L., and Vandegeest,
Zelazo (Oxford: Oxford University Press), 706–743. K. A. (1996). Inhibitory control in young children and its role in emerging

Frontiers in Psychology | www.frontiersin.org 7 January 2022 | Volume 12 | Article 773492

Castelo et al. Choice and Child Executive Function

internalization. Child Dev. 67, 490–507. doi: 10.1111/j.1467-8624.1996.tb01 Ryan, R. M., and Deci, E. L. (2006). Self-regulation and the problem of human
747.x autonomy: does psychology need choice, self-determination, and will? J. Pers.
Kochanska, G., Murray, K. T., and Harlan, E. T. (2000). Effortful control in early 74, 1557–1586. doi: 10.1111/j.1467-6494.2006.00420.x
childhood: continuity and change, antecedents, and implications for social Soenens, B., Vansteenkiste, M., Lens, W., Luyckx, K., Goossens, L., Beyers, W., et al.
development. Dev. Psychol. 36, 220–232. doi: 10.1037/0012-1649.36.2.220 (2007). Conceptualizing parental autonomy support: adolescent perceptions of
Landry, S. H., and Smith, K. E. (2010). “Early social and cognitive precursors and promotion of independence versus promotion of volitional functioning. Dev.
parental support for self-regulation and executive function: relations from early Psychol. 43, 633–646. doi: 10.1037/0012-1649.43.3.633
childhood into adolescence,” in Self and Social Regulation: social Interaction and Valcan, D. S., Davis, H., and Pino-Pasternak, D. (2018). Parental behaviours
The Development of Social Understanding and Executive Functions, eds B. W. predicting early childhood executive functions: a meta-analysis. Educ. Psychol.
Sokol, U. Müller, J. I. M. Carpendale, A. R. Young, and G. Iarocci (Oxford: Rev. 30, 607–649. doi: 10.1007/s10648-017-9411-9
Oxford University Press), 386–417. doi: 10.1093/acprof:oso/9780195327694. Whipple, N., Bernier, A., and Mageau, G. A. (2011). Broadening the study of infant
003.0016 security of attachment: maternal autonomy-support in the context of infant
Matte-Gagné, C., Bernier, A., and Lalonde, G. (2015). Stability in maternal exploration. Soc. Dev. 20, 17–32. doi: 10.1111/j.1467-9507.2010.00574.x
autonomy support and child executive functioning. J. Child Fam. Stud. 24, Willoughby, M. T., Magnus, B., Vernon-Feagans, L., Blair, C. B., and Family Life
2610–2619. doi: 10.1007/s10826-014-0063-9 Project Investigators (2017). Developmental delays in executive function from
McClelland, M. M., Acock, A. C., Piccinin, A., Rhea, S. A., and Stallings, M. C. 3 to 5 years of age predict kindergarten academic readiness. J. Learn. Disabil. 50,
(2013). Relations between preschool attention span-persistence and age 25 359–372. doi: 10.1177/0022219415619754
educational outcomes. Early Child. Res. Q. 28, 314–324. doi: 10.1016/j.ecresq. Zelazo, P. D., Anderson, J. E., Richler, J., Wallner-Allen, K., Beaumont, J. L., and
2012.07.008 Weintraub, S. (2013). II. NIH Toolbox Cognition Battery (CB): measuring
McClelland, M. M., Cameron, C. E., Duncan, R., Bowles, R. P., Acock, A. C., executive function and attention. Monogr. Soc. Res. Child Dev. 78, 16–33. doi:
Miao, A., et al. (2014). Predictors of early growth in academic achievement: the 10.1111/mono.12032
head-toes-knees-shoulders task. Front. Psychol. 5:599. doi: 10.3389/fpsyg.2014.
00599 Conflict of Interest: SC and PZ are Co-founders and hold equity in Reflection
McCurdy, A. L., Williams, K. N., Lee, G. Y., Benito-Gomez, M., and Fletcher, A. Sciences, Inc., which has licensed the Minnesota Executive Function Scale (MEFS)
C. (2020). Measurement of parental autonomy support: a review of theoretical from the University of Minnesota, Twin Cities. These interests have been reviewed
concerns and developmental considerations. J. Fam. Theory Rev. 12, 382–397. and managed by the University of Minnesota, Twin Cities in accordance with its
doi: 10.1111/jftr.12389 Conflict of Interest policies.
Meuwissen, A. S., and Carlson, S. M. (2015). Fathers matter: the role of father
parenting in preschoolers’ executive function development. J. Exp. Child The remaining authors declare that the research was conducted in the absence of
Psychol. 140, 1–15. doi: 10.1016/j.jecp.2015.06.010 any commercial or financial relationships that could be construed as a potential
Meuwissen, A. S., and Carlson, S. M. (2019). An experimental study of the effects conflict of interest.
of autonomy support on preschoolers’ self-regulation. J. Appl. Dev. Psychol. 60,
11–23. doi: 10.1016/j.appdev.2018.10.001 Publisher’s Note: All claims expressed in this article are solely those of the authors
Mischel, W., Shoda, Y., and Rodriguez, M. I. (1989). Delay of and do not necessarily represent those of their affiliated organizations, or those of
gratification in children. Science 244, 933–938. doi: 10.1126/science.265 the publisher, the editors and the reviewers. Any product that may be evaluated in
8056 this article, or claim that may be made by its manufacturer, is not guaranteed or
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, endorsed by the publisher.
A., and Wager, T. D. (2000). The unity and diversity of executive
functions and their contributions to complex “frontal lobe” tasks: a Copyright © 2022 Castelo, Meuwissen, Distefano, McClelland, Galinsky, Zelazo and
latent variable analysis. Cogn. Psychol. 41, 49–100. doi: 10.1006/cogp.1999. Carlson. This is an open-access article distributed under the terms of the Creative
0734 Commons Attribution License (CC BY). The use, distribution or reproduction in
Reeve, J., and Cheon, S. H. (2021). Autonomy-supportive teaching: other forums is permitted, provided the original author(s) and the copyright owner(s)
its malleability, benefits, and potential to improve educational are credited and that the original publication in this journal is cited, in accordance
practice. Educ. Psychol. 56, 54–77. doi: 10.1080/00461520.2020.186 with accepted academic practice. No use, distribution or reproduction is permitted
2657 which does not comply with these terms.

Frontiers in Psychology | www.frontiersin.org 8 January 2022 | Volume 12 | Article 773492