Executive functions (EFs) strategies, i.e., strategies that help students learn new content or solve problems are vital in developing lifelong learning skills. However, up to now, most educational research has focused on self-regulated learning (SRL) to explain successful learning and study success (e.g., Hayat et al., 2020; Moghadari-Koosha et al., 2020; Sun et al., 2023). In contrast, EF is mainly approached from a neuropsychological and clinical perspective, focusing on EF dysfunction and related educational problems (e.g., Meltzer et al., 2018, pp. 109–141; Dijkhuis et al., 2020), and it is hardly studied in ecological settings (i.e., standard learning settings). The studies conducted on EF in the educational context have primarily focused on children and adolescents (e.g., Diamond and Ling, 2016; Pascual et al., 2019; Zelazo and Carlson, 2020), not on young adults in higher education. To our best knowledge, EF and SRL have yet to be examined in combination within the context of higher education. Therefore, this study explores the relationship between EF and SRL and the extent to which they impact students’ study success in higher education.

Executive function and SRL originate from two paradigms, respectively founded in neuropsychology and based on educational research. Both have their methods, tests, and language. Researchers generally base their research on one of two perspectives. Nonetheless, both concepts have been associated with successful studying and study success in young adults (e.g., Garner, 2009; Musso et al., 2019; Pinochet-Quiroz et al., 2022) and are essential to a broader understanding of student’s ability to learn. The following section describes the definitions, similarities and differences between EF and SRL.

Self-regulated learning is about students becoming masters of their learning process which implies being able to adopt the most appropriate strategy for a learning task to be developed (Zimmerman, 2015; Dent and Koenka, 2016). SRL is generally considered a dynamic, cyclical process consisting of different phases and sub-processes of learning (Panadero, 2017). One of the most used and well-operationalized SRL models states that the cyclical process contains the following phases: (1) forethought, planning, and activation; (2) monitoring; (3) control; and (4) reaction and reflection (Pintrich, 2004). Each phase has four different areas for regulation: cognition, motivation/affect, behavior, and context.

Executive functions can be defined as a set of cognitive processes, partially independent and involved in top-down control of behavior, emotion, and cognition (Baggetta and Alexander, 2016; Nigg, 2017). EF refer to the most basic level of behavioral analyses or the neuropsychological level (De la Fuente et al., 2022). EF are effortful and invoked when automatic responses and routines do not work. This mainly happens in novel, complex, or otherwise challenging situations (Miller and Cohen, 2001; Barkley, 2012; Diamond, 2013). Therefore, EF are critical in learning, study success, and flexible behavior (Denckla and Mahone, 2018, p. 6).

Executive functions are a multidimensional concept; the literature describes several classifications of EF. Baggetta and Alexander (2016) found 39 different components or processes of EF in their review, with three core EF being the most commonly mentioned in the 106 studies they examined, i.e., inhibition (68%), working memory (35%), and cognitive flexibility (31%):

Combining these core “lower-order” processes creates “higher-order” or complex processes such as planning, reasoning, and problem-solving (Diamond, 2013).

One of the classifications that describes both the core and complex EF and is often used in research (in both academic and clinical contexts) is that of Gioia et al. (2000). We chose this classification because of its well-operationalized EF components and its emphasis on assessing behavioral manifestations of EF in an individual’s daily life (Baggetta and Alexander, 2016). Additionally, based on this classification Gioia et al. (2000) developed the Behavioral Rating Inventory Executive Functions (BRIEF), a self-reported questionnaire that has been translated into Dutch and standardized for children (Huizinga and Schmidts, 2012) and adults (Scholte and Noens, 2011).

This classification – based on factor analyses of EF behavioral descriptions – comprises nine EF, including the three core EF described before, next to the more complex or higher-order EF “self-monitor,” “emotional control,” “initiate,” “task monitor,” “plan/organize,” and “organization of materials.”

Executive function can be conceptualized on two levels: the core EF on a cognitive level (i.e., how the brain thinks) and the core EF and complex EF on a behavioral level (i.e., how the brain thinks expressed in behavior). Both levels refer to EF; however, in studies, they are operationalized and measured differently and refer to different underlying mechanisms of EF (e.g., Barkley and Fischer, 2011; Toplak et al., 2013). Researchers hypothesize that this is why directly or task-based measured core EF hardly overlap with the indirectly or self-reported measured EF (e.g., Barkley and Fischer, 2011; Toplak et al., 2013).

An advantage of directly assessing EF is that these task-based tests better test the actual performance of a specific EF. However, these results provide information about how well the student functions in an optimal and highly structured environment and, therefore, are not easily generalized across settings (Naglieri and Otero, 2014, pp. 191–208). The advantage of a self-reported EF is its higher ecological validity because it provides information about how well the student functions in a less structured environment, such as a school or home setting (Barkley and Fischer, 2011). An assumption with self-reports is that they measure behaviors related to the cognitive processes measured by task-based measures of EF. Because we are interested in how students experience their EF in their daily settings, we use self-report questionnaires in this study to assess EF.

The same reason applies to SRL; we are interested in the students’ perceptions of their SRL in general. SRL self-reports fall under the category of “offline measures,” referring to the timing of the measurements, in this case, that the self-reports are taken before or after the task and not during the task (i.e., “online measures,” such as think-aloud protocols or systematic observations) (Veenman, 2005; Schellings, 2011). When taking an SRL self-report, the student reflects on how they usually approach the learning task, so it provides more general information than specific information about a task at that moment. Thus, it depends on the research question of which measurement instrument is most appropriate (Rovers et al., 2019).

Self-reports also have drawbacks. Paradoxically, being able to complete the self-reports requires EF of the student to reflect on past and future behaviors (Garner, 2009), suggesting that a student with weak EF will be less able to self-report.

Another issue might be that students over- or underestimate themselves and whether there is a discrepancy between their intentional behavior and what they actually demonstrate [as demonstrated for SRL by Broadbent and Poon (2015)]. Students who overstated their performance on EF performance measures also achieved significantly higher scores on self-reports (Follmer, 2021).

Rovers et al. (2019) showed that students can report – via questionnaires – relatively accurately what their general self-regulatory functioning is, while at a detailed level, they have difficulty pinpointing exact SRL strategies. They argue that the level of granularity is of influence and that different types of measurement are valuable, depending on the research question. The same kind of reasoning could apply to the practical use of these measurements. For example, the benefit of self-reporting is that students become aware of their SRL strategies, which is an intervention in itself (Panadero et al., 2016). So, if self-monitoring is the objective, self-reports are an excellent option.

Conceptually, both EF and SRL refer to higher-order (top-down) cognitive processes. However, they differ in the context in which they are applied. EF are essential for navigating everyday life and engaging in social interactions (Barkley, 2012). EF become active when a student faces new, complex, or challenging daily life problems, including but not limited to problems encountered in the learning environment. In those moments, the student must make decisions, resolve issues, learn from mistakes, mentally play with ideas (be creative), think before acting, resist temptations, and stay focused (Diamond, 2013). In contrast, SRL occurs specifically and exclusively in the learning context and focuses on acquiring knowledge, automating skills, and achieving learning results. SRL involves both conscious and unconscious deep processing of information, or the repetition of facts, to eventually consolidate this information in long-term memory (Wirth et al., 2020).

Another difference is that SRL strongly emphasizes motivation or the “why” someone does something and the willingness to put effort into it (Schunk and Greene, 2018), in contrast to EF, which focus more on the “how,” i.e., “how do I solve this problem or adapt to the situation?”

Studies on the relationship between EF and SRL suggest a partial overlap between the constructs (e.g., Garner, 2009; Effeney et al., 2013; Follmer and Sperling, 2016). In these studies, EF and SRL are – indirectly – measured via self-reports demonstrating that EF expressed on a behavioral level overlap partially with SRL, also expressed on a behavioral level. Moreover, it seems that, in particular, the metacognitive dimensions of SRL are associated with or coincide with EF, for instance, planning (Effeney et al., 2013; Pinochet-Quiroz et al., 2022). The ability to plan allows a student to set and achieve goals in everyday life (context of EF) and focus explicitly on prioritizing learning tasks (context of SRL). However, self-reported EF and SRL are not the same in learning environments, and when overlapping, EF appear to contribute to variability in SRL processes, and the other way around, SRL processes implicate EFs (Garner, 2009).

However, in contrast, EF appear to be more unidirectionally related to SRL when measured directly through neuropsychological tasks, i.e., meaning task-based EF mediate through SRL on academic achievement and not the other way around (e.g., Rutherford et al., 2018; Musso et al., 2019). Only the core EF (i.e., working memory, inhibition, or cognitive flexibility) are task-based measured in these cases. In other words, SRL strategies seem to employ core EF – on a cognitive level – to achieve learning results, which makes sense because to sustain a learning strategy, the student must focus, keep information online, avoid distractions, and be cognitively flexible in disregarding old information in favor of new information.

In summary, EF can be conceptualized and measured at a cognitive and behavioral level (typically task-based and self-reported). Self-reported EF are most likely to have a partially overlapping relationship with self-reported SRL. In contrast, task-based EF are more likely to support self-reported SRL in achieving study success, thus showing a mediating role.

Studies about EF, SRL, and study success are scarce, particularly in young adult students. In this population, we identified only the study by Musso et al. (2019), who investigated the coherence between EF, SRL, and study success in a group of first-year university students. They found mediating effects of EF via SRL on math performance. Musso et al. (2019) measured EF directly (i.e., measured with neuropsychological tasks) and focused solely on working memory and executive attention.

To our knowledge, no study has focused on self-reported EF and SRL together as predictors of study success in higher education. There are a few studies that have investigated the relationship between self-reported EF and study success in the context of higher education. These studies indicate that self-reported EF problems negatively affect study success (e.g., Knouse et al., 2014; Baars et al., 2015; Ramos-Galarza et al., 2020). On the other hand, numerous studies have shown a positive relationship between SRL and study success in higher education (e.g., Honicke and Broadbent, 2016; Virtanen et al., 2017; Sun et al., 2018). The added value of EF in conjunction with SRL and their explanatory value for study success still needs to be determined to investigate if the concepts combined have the potential power to improve study success. Therefore, this study investigates the following research question:

Do self-reported EF and SRL combined explain variations in study success among higher education students better than either separately? Notably, since there appears to be a reciprocal non-mediating relationship between self-reported EF and self-reported SRL, we assume two independent variables that can directly or combined affect study success.

Following the research model depicted in Figure 1, we will investigate (1) the relationships between EF and SRL (measured 2 months after the start of the academic year), and study success (measured at the end of the academic year, and (2) the combined effect of EF and SRL on study success.

We hypothesize that EF and SRL combined explain statistically significantly more variance in the number of credits obtained at the end of the academic year than each construct separately.

In addition, the aim is not to examine the different dimensions of EF and SRL and their relationships. Given the inter- and intra-individual differences due to the developmental trajectory of both EF and SRL, we expect these specific dimensions to have little expressive power when looking at individual students. We expect the results to provide insight into the group. However, this picture may differ if students have been developing for 6 months or if a different group of students is involved. Therefore, we explore the concepts of EF and SRL without identifying the specific dimensions.

The COVID-19 pandemic made studying and life more challenging for students due to lockdowns and regulations (e.g., Copeland et al., 2021; Ihm et al., 2021). During this time, students may have faced a constant stream of new and complex issues, which could have impacted their EF. While not the main focus of our research, we also wanted to understand how these circumstances affected students’ self-reported EF and SRL. As such, we asked students if the lockdowns and regulations influenced how they completed our questionnaire.

The study was conducted as planned with no significant details in implementation. The results will answer the research question of the combined value of EF added to SRL to better understand the differences in study success. We hypothesize that EF and SRL combined explain statistically significantly more variance in the number of credits obtained at the end of the academic year than each construct separately.

First, we provide descriptive data, then discuss the correlations between the different constructs (EF and SS, SRL and SS, and EF and SRL) and finally, test the hypothesis.

This study sought to investigate the added value of including EFs and SRL in predicting study success after one academic year among higher education students. We explored (1) the relationship between EF and SRL the relationship between EF and study success and SRL and study success, and we hypothesized that (2) the combination of self-reported EF and SRL would explain differences in study success better than separately. First, Regarding the relationships between the constructs our findings show that EF and SRL are correlated. This is consistent with previous studies demonstrating associations between EF and SRL among high school students (Effeney et al., 2013; Rutherford et al., 2018) and university students (Garner, 2009; Follmer and Sperling, 2016). The weak to moderate correlations between EF and SRL makes it clear that these concepts partially overlap but are not the same (Garner, 2009). In addition, EF and SRL both correlate – partially and weakly – with study success, indicating a trend of more EF problems or less SRL comes with fewer credits earned after one school year, in line with studies such as those by Baars et al. (2015) and Ramos-Galarza et al. (2020) for EF, and Li et al. (2018) and Moghadari-Koosha et al. (2020) for SRL.

Second, to better understand the influence of EF and SRL combined on the differences in study success, we compared the imposed models separately and combined. In line with our hypothesis, EF and SRL combined explained the variance in study success after one academic year better than EF and SRL separately. This indicates that a student who performs poorly on EF will likely demonstrate less effective SRL and likely have less study success. Similar results were found by Musso et al. (2019), although they used task-based EF measurements, whereas we used self-reported EF. Even though more research is needed, Musso’s and our findings indicate that combining EF and SRL is vital for the learning processes. A student with more developed EF strategies can reflect on, choose from, or integrate rules where appropriate (Moran and Gardner, 2018, pp. 29–56) and thus be better able to self-regulate their learning and achieve more success.

The current study has some important strengths, such as the empirical confirmation of the need for integration of two theoretical models relevant to education and study success, with the use of proven valid and reliable instruments and the use of SEM to test the models while better accounting for measurement error (Tomarken and Waller, 2005).

On the other hand, this study has a few limitations. First, a non-probabilistic sample was used, namely students of Applied Psychology, which limits the generalization of the results to other groups of students or young adults. Future research could include students from different studies and levels as a more representative sample of young adult learners.

Second, self-reporting measurements were used, which have apparent advantages, such as surveying a large population without much effort and high ecological validity (Barkley and Fischer, 2011). However, a known pitfall with self-reporting is that students may (un)consciously fill out the questionnaires differently than they would show in observed behavior (e.g., McDonald, 2008; Demetriou et al., 2015).

Particularly, self-reports of cognitive abilities are sensitive to response bias and psychological factors, such as depression, anxiety, or chronic pain. For instance, Schwartz et al. (2020) found that self-reporting EF with the BRIEF-A probably measured emotional distress over executive dysfunction. However, they argued that this could be the case in specific samples such as theirs, namely middle-aged veterans who all showed intact EF and experienced heightened psychiatric distress. Abeare et al. (2021) demonstrated inconsistent results to the conclusion of Schwartz et al. (2020), suggesting a more plausible explanation that “non-credible presentation manifests as extreme levels of symptoms on the BRIEF-A-SR- and self-report inventories in general” (Abeare et al., 2021, p. 9). Additionally, a reasonable number of studies have shown that the BRIEF-A can validly measure EF in various target groups such as deaf and hearing students (Hauser et al., 2013), students and procrastination (Rabin et al., 2011), and depression within students (Mohammadnia et al., 2022).

Nevertheless, both Schwartz et al. (2020) and Abeare et al. (2021) suggest that on the validity scale “negativity” a cut-off score of 4 (instead of ≥6) is more representative (i.e., essentially a frequency count of the extreme self-ratings on 10 items of the BRIEF-A). In our study, this would imply that 10 more students should have been disregarded as outliers. However, considering this small number of students, we do not expect different outcomes. To gain insight into the impact of assessment mode on outcomes, research is needed that includes both self-report measures of EF and SRL and objective measures, such as neuropsychological tests for EF or learning analytics for SRL (Yamada et al., 2017). Additionally, measurement instruments that can support screening for non-credible symptom reports can be used (Abeare et al., 2021), such as the MMPI-2 (Schwartz et al., 2020).

Subsequently, a CFI (just) below 0.9 indicates a reasonable but not good fit of the model with the dependent variable. That is, as argued, if the CFI scale is considered as a continuum and not, as is often incorrect, as a dichotomous scale. If our model had had a higher CFI, it would be easier to make statements about the relationship between the variables in the model and the outcome measure. However, this does not alter the fact that the correlations between the various SRL scales and the EF scales with study success have been reliably established. The lower CFI mainly concerns correlations between these (sub)scales, making it more difficult to see what their unique contribution is to study success. Further research will be required to investigate this further.

A final limitation might have been that this study was conducted during the COVID-19 pandemic. Research demonstrates that the lockdowns and other restrictive regulations impacted students’ lives considerably (e.g., Copeland et al., 2021; Ihm et al., 2021), and therefore we investigated the self-reported impact of COVID on how students completed the questionnaires. We found that this period negatively related to how students experience their EF and SRL. This was especially true for students reporting severe EF problems, which implies that the results might be colored because assessments were conducted during the pandemic. Appelhans et al. (2021) found a similar result: young adults with preexisting EF deficits have shown increased unhealthy behavior since COVID-19. We think this period especially challenged students’ EF because of the constant flow of new and complex issues they encountered. Nevertheless, although students’ response patterns might have deviated a bit due to COVID-19, we think that, in light of previous research, patterns would have been the same when assessed in regular times. However, it might be valuable to repeat the study in non-pandemic times.

Future research could further explore how EF and SRL impact study success in theory and practice. One aspect is that a large part of the variance in study success is still unaccounted for, and future research could focus on finding additional answers, for example, in personal and contextual regulatory factors, such as studied by De la Fuente et al. (2022) and Pachón-Basallo et al. (2022).

Another aspect is that EF and SRL have different yet complementary conceptual lenses on how students learn and achieve success [such as Dinsmore et al. (2008) suggest for metacognition and SRL]. Although our study is not about the conceptual lens of EF and SRL, further research into how we can learn from both ways of looking at things to understand student study behavior is desirable since the results confirm that, taken together, they better predict study success even though they do not measure the same thing. The findings of this study can be used to motivate improving learning environments in higher education. Since EF and SRL combined better explain the differences in study success, it makes sense to look at the available EF tools to expand the educational professional’s toolbox beyond the already available SRL tools (e.g., Theobald, 2021). Providing education of EF in addition to SRL probably increases the levels of success in students. Furthermore, metacognitive knowledge about SRL and EF leads to more motivation to use the learned strategies correctly (e.g., Veenman, 2011; Follmer, 2021). Additionally, in the design of (blended) learning environments, educational professionals can build a certain degree of adaptivity when considering different levels of students’ EF, knowing that individual differences are significant. For example, regarding problems with task initiation, one can think of a lesson structure with more intermediate moments during which a student can ask a supervisor for help, more formative tests, or additional (warm-up) assignments in a module that support and encourage the start-up. This way of working is not new and also falls under the intersection of educational science, psychology, and neuroscience, also called neuroeducation (e.g., Jolles and Jolles, 2021; Willingham, 2023).

In conclusion, this study highlights that while EF and SRL cohere and are related to study success, they do not measure the same. This is also reflected in that each separately explains less variance in study success than taken together. Nonetheless, combined they provide more information about how student achieve study success. Generally, a student reporting EF deficits will likely report less effective SRL and achieve less study success in the long-term. This suggests that attention to EF alongside SRL in education is justified and valuable. Future theoretical research on both the working mechanisms of EF and SRL is needed, as well as the more practical application of the knowledge associated with EF and SRL.

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

The studies involving humans were approved by Ellis van Dooren-Wissebom, chairperson of the LLM Saxion Ethical Advice Committee (SEAC). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants.

DM secured funding for the study, conceptualized the study, oversaw, and conducted the data collection, analyzed the data, and led the writing of the manuscript. MS-D managed the data collection in SPSS and contributed to drafting the method and result sections of the manuscript. JD supervised the design, data collection and analyses, and manuscript writing. JF supervised the writing of the manuscript. AJ provided feedback on the latest versions of the article. All authors contributed to the article and approved the submitted version.