The selection procedure of the articles followed the guidelines of the PRISMA method in order to ensure the quality of this work [21, 22]. As proposed by the methodology of Guirao Goris (2015) for reviews, a systematic review of the literature published over the last 5 years (between 2015 and 2020) was carried out [23]. The protocol can be found on http://neurocics.udd.cl/data/Larrain-Valenzuela_et_al_2021_Protocolo_registro.pdf

The search for published studies was carried out using the PUBMED database (https://www.ncbi.nlm.nih.gov/) because it includes more than 32 million citations for biomedical literature from MEDLINE, life science journals, and online books. The search terms were “self-regulation”, “cognition”, “neuroscience” and “brain”. These words were chosen with the aim of addressing the different dimensions of the word self-regulation as a psychological construct and as a whole, they were chosen to delimit the construct of this mental capacity and the neurobiological findings associated with it. Because the individual words were broad, the following combinations of words were used in the PUBMED database search for Title/Abstract: “self-regulation AND cognition”, “self-regulation AND neuroscience” and “self-regulation AND brain”. Words that were possibly linked to self-regulation such as “self-control or inhibitory control or emotional regulation” were not used because we needed to identify the dimensions and conceptualizations in a broad way and, from there, understand their particularities. Only articles from 1st of January of 2015 until the 31st of december of 2020 were included.

The inclusion criteria were: a) Self-regulation as an object of study. b) Studies of a quantitative nature with the possibility of replicating the methods. c) Studies in humans with or without neuropsychiatric diagnosis. d) Only the age range of childhood and adolescence was accepted. Specifically, late adolescence up to the age of 21 [24]. Articles that presented results from infants and included an adult population as a control group within the experimental design were accepted. e) Only articles indexed in Web of Science or Scopus were included as a criterion of academic quality.

The exclusion criteria were: a) Systematic review articles and / or meta-analysis, b) Articles referring to self-regulation in terms of neuronal electrical activity (e.g., The concept used in the context of neurofeedback techniques). c) Articles that used emotion regulation, and not self-regulation of a higher order of complexity. d) Articles where the word self-regulation was only mentioned in the abstract and was not a variable within the article.

(1) The search was conducted by two independent researchers following inclusion and exclusion criteria.

(2) Collection of articles: three hundred thirty-eight articles were retrieved from the PUBMED database according to our criteria for the years considered and word search.

(3) Elimination of repeated articles (n = 35).

(4) Elimination of reviews or meta-analysis (n = 113).

(5) Elimination because the age requirement was not met or the age of the sample was not specified. Furthermore, articles were excluded when self-regulation was not a variable within the study. Information extracted only by abstract (eliminated n = 146). Remaining articles (n = 44).

(6) Elimination of articles due to exclusion criteria when analyzing the study in greater depth, in which there were inconsistencies associated with the approach to the concept of self-regulation (n = 15).

(7) Finally, 29 articles were selected for the extraction of information according to the previously established inclusion criteria.

To ensure a greater organization of the systematic review, a review matrix was created in an Excel spreadsheet. A different sheet was created for each of the steps in the article selection and data collection process of this review, which can be reviewed in the supplemental material. Fig. (1) shows the summary of the information acquisition process and Table 2 shows all the selected articles with their general identification for subsequent extraction linked to the analysis categories.

The selected articles were downloaded and stored digitally in PDF format. A database was built for the systematic analysis of each of the selected articles, establishing 18 categories of analysis. These were: year, journal, country of study, concept of self-regulation, dimensions of self-regulation, hypothesis, sample size, characteristics of the sample, inclusion and exclusion criteria, behavioral instruments, experimental task, use of therapeutic intervention programs and the main limitations of the study. Table 1 defines each of the categories of analysis.

Fig. (1). PRISMA flow diagram of the articles included in this analysis. Source: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71

The productions of research in self-regulation presented 338 articles published in the last 5 years. An example was the combination of words “self-regulation [Title / Abstract] AND Brain [Title / Abstract]” which registered 228 articles compared to the combination of words self-regulation [Title / Abstract] AND cognition [Title / Abstract] which registered 83 articles. This could suggest that the concept of self-regulation is predominantly intertwined with a neurobiological substrate. However, the combination self-regulation [Title / Abstract] AND neuroscience [Title / Abstract] only yielded 29 articles. In relation to the analysis of the 29 articles selected for this review, a greater number of publications was detected during 2018. See Graph. 1 to see the annual distribution of the publication of the 29 selected articles during the last 5 years.

Graph. (1). Annual distribution of the 29 selected Publications.

When grouped by continent, the largest proportion of the reviewed research was from North America with 22 articles. Four studies were conducted in Europe, two in Oceania and one in Asia. The country with the highest number of publications included in this study was the United States with 18 articles. The specific age ranges of the participants were: Early childhood 10.3% (0 to 5 years), childhood 20.6% (6 to 11 years), adolescence 20.6% (12 to 18 years) and youth 6.8% (18 years or more). The remaining percentages were part of the articles with broader age ranges (e.g., articles [26, 27, 29, 32, 35, 36, 39, 42, 43, 49, 50, 53]).

For this section of the results, a descriptive Table 3 was prepared regarding the participants, interventions, comparisons, results and study designs (PICOS). Likewise, information was included on the classification of the articles with respect to the clinical diagnosis of the sample associated with typical or atypical development. Articles that explicitly presented the clinical diagnosis were considered within the atypical development category. Articles that presented a population at “risk” were incorporated in the typical development category.

There are only four articles that explicitly define the study sample as healthy (articles) [29, 45, 51, 53]. However, there are six other articles with difficulties in the inclusion and exclusion criteria. Given this, they were classified after reviewing the article in general, according to the variables they presented associated with health, cognitive and / or behavioral characteristics. Likewise, there are two articles that describe a population at “risk” with emphasis on the predisposition to suffer from physical and / or mental health problems. One article had a population associated with Early Adverse Experiences and the other article had a sample that presented a low socioeconomic level.

Also, obstacles to categorizing studies as being of typical development appeared because only four articles were explicit about the study population being healthy or having typical development. Eight articles are ambiguous regarding this classification. This allows hypothesizing possible predisposing factors that influence a broad spectrum of typical child development. Without a doubt, it reveals certain gray areas with respect to this concept of self-regulation. In none of the selected articles, the self-regulatory capacity appears as pathognomonic for the understanding of a certain neuropsychiatric difficulty, although it does influence the modulation of behavior sensitive to contextual changes.

The descriptive analysis PICOS detected therapeutic interventions in only seven articles, some of which involved children and others parents. Only seven articles carried out intervention designs with a small sample and with short intervention periods. There were no findings on prevention or promotion of self-regulation as a protective capacity in the field of mental health.

Another production condition was the variability of the sample within the atypical development. Article 12 was the only one that conducted research in infants with the aim of evaluating the development of self-regulatory capacity after moderate brain injury. This reveals a weakness associated with the complexity of the sample, and the increase in studies after 24 months of age (e.g., article) [45]. Perhaps the studies of self-regulation increase because it is understood as part of a maturation process, and that sometimes it can be altered.

With regard to developmental disorders, articles of children with a diagnosis of autism spectrum disorder appeared (e.g., articles [30, 32, 38, 49]), children with brain injury (e.g., articles [28, 36, 50]), children with behavior problems (e.g., articles [27, 34]), children with hyperactivity disorder (e.g., articles [31, 51]), children with eating disorders (e.g. articles [33, 39, 40]), children with a diagnosis of epilepsy (e.g. article [46]), infants with early adverse experiences (e.g. article [37]) and intellectually gifted adolescents (e.g. article [52]). There were articles related to relationships with parents and the incorporation of parenting styles (e.g., articles [29, 39, 44]), two related to the abuse of substances such as alcohol (e.g., articles [47, 48]). Two others who performed a construct validation with structural equations and different psychological evaluations related to self-regulation (e.g., article ([42, 51]). Finally, articles were detected that focused on the effects of an intervention that aims to develop self-regulatory capacity (e.g., articles [30, 31]).

Not all articles presented an explicit definition of self-regulation or of the theoretical constructs that support its conceptualization. In this regard, two articles were essential to guide our results because they delve into the conceptualization of self-regulation (articles [42, 51]).

One was carried out by the group of Malanchini, Engenlhardl, Grotzinger, Harden & Tucker-Drob (2018) [42] and the other was carried out by the researchers Tiego, Bellgrove, Whittle, Pantelis and Testa (2018) [51]. The interpretation as a whole strengthened at least two theoretical currents from which self-regulation can be measured: Cognitive and Personality dimensions [51].

The cognitive dimension of self-regulation contemplates executive functions, which are cognitive processes that supervise, monitor and control the most basic processes related to thinking, reasoning and decision-making. Furthermore, Malanchini et al. (2018) and Tiego et al. (2018) state that inhibitory control would be a part of executive functions, which refers to the ability to control predominant responses [42, 51]. The concept of inhibitory control was studied as a central component within the autoregulatory capacity [25, 34, 42, 52]. Specifically, article 28 of Urben et al. (2018) studied 39 intellectually gifted adolescents. They studied the influence of emotions on inhibitory control evaluated using a Stop Signal task, with the aim of evaluating the relationship between cognitive control and emotional abilities [52]. The results showed that the main effect was obtained in the higher performance of the cognitive task of the STOP stimulus and in a shorter reaction time in the emotional responses associated with sadness in intellectually gifted adolescents, which would imply a different inhibitory capacity decrease to adolescents with typical development [52].

Malanchini et al. (2018) incorporate working memory (memory system of temporary limited capacity that maintains and stores information [54], switching (ability to change attention in the face of a different stimulus) and updating (ability to replace the previous information with the incoming one) as dimensions of self-regulation [9, 42].

Likewise, Tiego et al. (2018) incorporate the capacity of effortful control into self-regulation. This includes the efficiency of top-down executive attention in children and adolescents that gives them the ability to control cognition, emotion, and behavior in a goal-oriented manner [51].

A common aspect of the selected articles presenting a cognitive dimension of self-regulation is that they repeatedly mention cutting-edge research associated with temperament, development of attention and executive functions, among others [17, 55, 56]. A clear example was Tiego et al. (2018) who investigated the relationship between the behavioral measures of executive function and effortful control (e.g., BRIEF), and the relationship with performance in cognitive executive attention tasks, and problems associated with developmental psychopathology, based on different measurements. A factor analysis was performed, which reported that cognitive self-regulation processes underlie a unitary construct [51]. Likewise, it reported that executive attention as part of self-regulation significantly explained the variance in the self-regulation construct measured by behavioral measures of executive function and effortful control. Undoubtedly, the results show a strong overlap between the concepts of effortful control and executive functions, which is consistent with previous studies [4, 57].

As for self-regulation in its dimension associated with personality, Malanchini et al. (2018) [42] incorporate conceptualizations from Diamond (2013), who proposes that self-regulation allows motivation and interest to achieve one’s goals [9]. This ability changes during development and is related to motivation [58]. Bridgett, Oddi, Laake, Murdock & Bacjmann (2013) also refer to the self-regulatory capacity as a complex process that allows guiding activities aimed at objectives over time through changing circumstances [4]. Article [35] by researchers Hanson, Gillmore, Yu, Holmes, Hallowell, Barton, Beach, Galván, MacKillop, Windle, Chen, Miller, Sweet & Brody (2019) associated low self-regulatory capacity with externalizing behavioral problems.

This personality dimension of self-regulation was measured in many of the selected articles through adult ratings of children, observing their behavior within different environments such as at home or school, incorporating more ecological measures. Likewise, in the selected articles [26, 27, 31, 45], different observations were made of the behavior of the children. Levels of irritability, anxiety, impulsivity, among others, were evaluated in different interventions or contexts. The factorial model of personality (Big Five; Costa and McCrae, 1992) that contemplates openness, conscientiousness, extraversion, kindness, and neuroticism, are also reported within the selected articles [42].

For both dimensions of self-regulation, cognitive and personality, concepts within the articles were sought that could contribute to a better understanding of their underlying theoretical constructs. The definitions of self-regulatory capacity did not appear in all the articles. In some, the concept was explicitly defined and in others, the definition could only be deduced once the entire article had been analyzed.

In order to contribute to the comprehension of the concept of self-regulation, all of the selected articles were classified according to their level of description and deepening of the concept of self-regulation (Table 4).

In this section, the numerous instruments used to measure self-regulation are organized. Only the instruments that the authors report as a measure of self-regulation were analyzed, which allowed for the verification of objectives or hypothesis linked to self-regulation. The instruments used to measure self-regulation are classified in Table 5 according to alphabetical order. In addition, the type of measurement, the potential dimension of self-regulation that the instrument targets (Cognitive / Personality) and the aspects evaluated by the instruments associated with self-regulation are described.

This last section shows the main neurobiological correlates associated with the tasks used to evaluate self-regulation. Only 11 of the articles selected for this review presented results related to neurobiology. All reviewed articles used non-invasive techniques, widely used in neurosciences, such as functional magnetic resonance imaging (fMRI) and electroencephalography (EEG).

Regarding the use of functional magnetic resonance imaging techniques, the randomized controlled study by Hanson et al. (2019) evaluated the effect of an intervention program focused on the mitigation of stressors associated with poverty. This study considered, among others, infants with an average age of 11 years 2 months (n = 93). There were significant differences between the group that received the intervention compared to the control group, mainly in the area of ​​the left dentate gyrus and in the right ventromedial prefrontal cortex (vmPFC) [35].

Another article that presented neurobiological findings was Lim et al. (2015), who used a decision-making task associated with healthy eating preferences in infants and how they incorporate the recommendations of their own mothers when choosing the type of diet (n = 25). They reported that vmPFC further encodes children's own preference and that an increase in the left dorsolateral prefrontal cortex (dlPFC) was associated with the children’s responses when they incorporate their mother’s projected choices. Also, an increase in activity of the dlPFC was shown when compared to inhibitory control-type functions when children established their own preferences about their food [39].

Also, Musarak et al. (2018) studied the relationship between parental control over children and neuronal and behavioral responses to socio-emotionally challenging situations in 27 children. Greater parental control was associated with children with lower self-esteem and lower self-regulatory capacity, when responding to a face conflict categorization task, which asked to indicate whether the faces showed fear or happiness. Greater parental control and faster but less precise behavioral responses were associated with less activation in the left anterior insula. This is complemented by article 29, where a greater activation of the anterior cingulate cortex (ACC) and the dlPFC was reported in a delay discounting task in 228 adolescents [44].

Article [40] by researchers Lopez, Chen, Huckins, Hofmann, Kelley & Heatherton (2017) conceptualizes self-regulation based on the equilibrium theory proposed by Heatherton and Wagner (2011) [19], which supports the idea of ​​a balance between the activity of ascending subcortical neuronal regions (from amygdala and nucleus accumbens towards cortex) and descending prefrontal neural regions (lateral prefrontal cortex towards subcortical). Failure of self-regulation occurs when the balance tips in favor of the subcortical regions. This translates into an inverse relationship between the descending cognitive control of the prefrontal cortex over the affective impulses from below - above the subcortical regions involved in the generation and reward of emotions.

As for EEG techniques, Grabell et al. (2017) studied infants with behavioral problems (n = 20) and control subjects (n = 30) during a Go / No-go task adapted to evoke certain event-related potentials (ERP). They observed that there is a positivity related to the error (Pe is a positivity of the signal approximately 200-400ms after the onset of a commission error and the maximum amplitude of the Pe has been reported in frontal, frontocentral and parietal locations. Initially framed as a component related to conscious processing or evaluating errors after they have occurred.) and a No-go N2, which is presented as a negative deviation with Fronto-central electrode distribution that occurs in adults between 200 and 400ms after the presentation of a No-go stimulus. Both ERPs identified preschoolers with high levels of disruptive behavior. Specifically, it showed that the children with better behavior presented a greater amplitude in Pe and No-go N2. Ultimately, Grabell et al. (2017) report that the increase in Pe is associated with a better processing of the negative effect after an error and a greater response of the autonomic nervous system to errors, as well as the increase in the amplitude of No-go N2, which would reflect a greater cognitive control required to inhibit a response or a greater level of conflict between competitive responses, that is, execution versus inhibition [34].

Finally, the reviewed articles that present neurobiological findings suggest that the generalizability of their results needs to be evaluated with caution. In addition, most of the studies emphasize limitations associated with the sample size, which affects the robustness of the findings.

For research articles with several authors, a short paragraph specifying their individual contributions must be provided. The following statements should be used “Conceptualization, J.L.; methodology, J.L and F.M.; formal analysis, J.L and F.M.; writing—original draft preparation, J.L.; writing—review and editing, E.A., L.K.; visualization, J.L and F.M.; project administration, J.L. All authors have read and agreed to the published version of the manuscript