Developmental Trajectories of Executive and Semantic Flexibility Using Task-Switching

Developmental Trajectories of Executive and Semantic Flexibility Using Task-Switching | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Developmental Trajectories of Executive and Semantic Flexibility Using Task-Switching Giada Viviani, Ettore Ambrosini, Annamaria Porru, Silvia Benavides-Varela, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7965304/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Cognitive control supports adaptive behavior through stability and flexibility, with executive flexibility typically assessed through task-switching paradigms. However, while executive flexibility is a well-studied construct, it is unclear whether it relies on the same mechanisms as semantic flexibility – the ability to switch between meanings based on context. An ideal approach to arbitrate this debate is to compare their developmental trajectories, a method hampered by the fact that semantic flexibility’s development remains largely uncharted. Here, 4- to 10-year-old children performed parallel task-switching paradigms: a classic visuospatial paradigm assessing executive flexibility and a novel semantic task assessing semantic flexibility by requiring them to alternate between semantic judgments of meaningful concepts. Results revealed a strong correlation between executive and semantic switch costs, suggesting shared control mechanisms, alongside domain-specific differences and age-related modulations influenced by semantic distance, revealing a growing interplay between semantic knowledge and control as children's conceptual systems mature. These findings provide novel insights into the maturation of cognitive control components in childhood, highlighting the interplay between domain-general executive processes and semantic control mechanisms in flexible cognition. Cognitive Control Semantic Flexibility Task-Switching Paradigms Executive Flexibility Controlled Semantic Cognition Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction To effectively navigate a constantly changing environment, we rely on cognitive control – a family of top-down mechanisms that organize thoughts and actions to align with our goals 1 . A fundamental component of cognitive control is stability, which contributes to maintaining focus on a relevant "task set" (the rules for a task) 2 while shielding it from distraction, as typically assessed by conflict paradigms such as the Stroop task 3 . While crucial, stability alone is insufficient. Our dynamic world often requires cognitive flexibility, a complementary component to rapidly switch between task sets as demands change 4 . Thus, the crucial role of cognitive control is to balance stability and flexibility to effectively achieve our goals. For instance, while driving, we must maintain the stable goal of following a route, yet be able to flexibly react to an unexpected road closure by quickly planning a new one. Given its central role in adaptive cognition, flexibility is extensively investigated using task-switching paradigms 2 , whereby participants alternate between two (or more) tasks. The performance difference between "Switch" trials (where the task changes) and "Repeat" trials (where the task remains the same) is known as the "switch cost". This cost serves as a key behavioral measure of flexibility, with smaller switch costs indicating greater efficiency in shifting between tasks. While the task-switching paradigm provides a robust method for measuring cognitive flexibility, many versions typically rely on stimuli that lack real-world meaning, such as simple geometric shapes or digits. This approach reduces the ecological validity of the findings, as it does not fully capture how we flexibly manipulate meaningful information in everyday life, where we frequently need to adapt our use of semantic knowledge based on context. For example, we may be required to shift from the meaning of the word 'jam' referring to a fruit preserve to its alternative meaning referring to a traffic obstruction. Flexibly interpreting such meanings requires a form of cognitive control known as semantic control 5 – 7 – a mechanism that flexibly prioritizes some of a concept’s features or meanings while actively suppressing others that are irrelevant to the current goal. The concept of semantic control being distinct from domain-general cognitive control (hereafter called executive control) is a central tenet of the Controlled Semantic Cognition (CSC) framework 8 . This framework posits that semantic control is “a set of executive control processes that regulate the activation and deployment of semantic knowledge. These allow flexible, context- and task-appropriate responses by ensuring that only relevant aspects of semantic representations are used to direct thought and behavior.” 9 (p. 259). A dilemma exists, however, regarding the independence of these two control systems. While semantic control is supported by specific brain regions (such as the left posterior middle temporal gyrus and the inferior frontal gyrus) 10 , 11 , neuroimaging evidence also shows partial overlap with the multiple-demand network that underpins executive control 12 . The distinction primarily originates from research on patients with semantic aphasia, who show deficits in using conceptual knowledge despite intact semantic representations 13 – 15 . However, these patients also exhibit deficits in executive control, challenging the assumption that the two mechanisms are entirely independent 13 – 15 . To better untangle this relationship, it is crucial to consider that much like its executive counterpart, semantic control comprises components of both stability (maintaining focus on relevant meanings) and flexibility (shifting between meanings as the context demands). To date, research exploring the link between the two mechanisms has primarily focused on the stability dimension, whereas the interplay between executive flexibility and semantic flexibility remains largely under-explored. An exception is our recent work with healthy adults, which showed a moderate-to-strong correlation between switch costs in executive and semantic tasks, suggesting shared underlying mechanisms 16 . Studying the developmental trajectories of executive and semantic control from childhood offers a unique window into the architecture of cognitive control, allowing researchers to determine whether the observed associations in adulthood reflect shared developmental origins or later functional convergence. This developmental perspective moves beyond correlational evidence to uncover the mechanisms by which cognitive systems differentiate and interact over time. Given that the link between the two control systems is particularly under-explored for the flexibility component, in both adults 16 and children, charting the developmental course of both executive and semantic flexibility is a critical step forward. In children, executive flexibility is a core component of cognitive control that is crucial for school readiness and later academic success 17 . Its developmental trajectory is not a simple, linear improvement. Although the behavioral ability to switch between tasks matures relatively early, showing significant gains between ages 7 and 11 and reaching adult-like performance around age 12, the underlying neural systems supporting this skill undergo a far more prolonged and complex maturation into adolescence 18 . Neuroimaging research has sought to pinpoint the specific brain regions that drive this maturation. One influential line of evidence proposes that the sub-processes of task switching are supported by distinct prefrontal regions that mature at different rates 19 . This model suggests the (pre-)supplementary motor area, a medial frontal region critical for suppressing the previous task set, shows an adult-like activation pattern by early adolescence. In contrast, the ventrolateral prefrontal cortex, a lateral region essential for retrieving and maintaining the current task rule, continues to show immature activation patterns throughout the teenage years 19 . However, this region-specific dissociation is debated. An alternative, network-based perspective suggests that even young children recruit a broad frontoparietal “multiple-demand network” that is spatially similar to the one used by adults 20 . From this viewpoint, development is characterized not by different regions coming online at different times, but by a process of network-wide refinement, where activation becomes more efficient and neurally selective with age and improving ability 21 . In summary, while there is a clear consensus that the neural substrates of executive flexibility undergo a prolonged, non-linear development, the literature presents an active debate. It remains unresolved whether this maturation is best described by the staggered development of specific prefrontal regions handling distinct sub-processes, or by the holistic refinement of a pre-existing, domain-general cognitive control network. Research 22 , 23 has also explored the development of semantic flexibility using stimuli with semantic content. Studies have shown that between the ages of 3 and 6, children improve in switching between verbal rules for sorting cards and in flexibly using changing semantic cues to infer the meanings of new words 22 , 23 . For example, preschoolers can learn to switch between sorting objects by a perceptual feature like shape and a more semantic feature like function 22 . Similarly, word-learning tasks have demonstrated that 4- to 6-year-olds can flexibly use different verbal cues to decide whether a new word refers to an object's shape, material, or another part, a skill that is less developed in 3-year-olds 22 . Verbal fluency tasks, which measure how many words a child can generate from a specific semantic category, have also been used as an indicator of semantic flexibility, with performance improving with age 24 – 27 . These paradigms, while providing valuable insights, often differ from the trial-by-trial cued task-switching methodology typically used to measure executive flexibility. This methodological divergence makes it difficult to determine whether observed improvements are driven by the same underlying control mechanisms isolated by classic switch-cost designs. The development of semantic flexibility is further complicated by its interaction with other cognitive domains. For instance, a child’s growing ability to flexibly use concepts likely reflect the combined contributions of both executive control and the acquisition of domain-specific semantic knowledge 28 . Despite this existing body of work, a significant gap remains. No experimental study has explicitly charted the developmental trajectory of semantic flexibility in children using a cued, trial-by-trial task-switching paradigm, nor has one directly compared it to domain-general executive flexibility using a parallel design. This methodological gap leaves a fundamental developmental question unanswered: are semantic and executive flexibility distinct processes that mature independently, or do they rely on a shared cognitive process that follows a common developmental timeline? Addressing this question has broader theoretical implications, as it informs our understanding of how domain-general and domain-specific control mechanisms interact during development to produce the mature forms observed in adulthood. To address this question, the present study investigates the developmental trajectories of both executive and semantic flexibility. As previously validated with adult participants 16 , we employed a task-switching methodology with two parallel paradigms: a classical visuospatial task with minimal semantic content to measure executive flexibility, and a novel task requiring semantic judgments about pictures of meaningful concepts to measure semantic flexibility. Our study has two primary aims. First, we aim to elucidate the relationship between these two core cognitive abilities across development, assessing their developmental interplay. By directly comparing performance on these parallel tasks, we can provide crucial evidence as to whether they are supported by distinct or shared control mechanisms. Second, we aim to gain a deeper understanding of semantic flexibility itself, characterizing its unique developmental features. We will examine how its development is modulated by semantic distance between consecutive concepts, a factor intrinsic to the organization of semantic information. Methods In line with the journal’s guidelines for transparency and openness, this section details the sample size determination, all criteria for data and participant inclusion/exclusion, and all experimental manipulations and measures. Exclusion criteria were established a priori, before the start of data analysis. While the study was not preregistered, we provide full access to the research materials. The anonymized raw data, materials and codes for the experimental task and analyses are publicly available in our project repository on the Open Science Framework (OSF) platform at https://osf.io/zr4dn/overview for reuse by other researchers. Participants A total of 655 children participated in the first session of a larger two-session study, which included the experimental paradigms reported here. From these, we then excluded 37 participants (see Data analysis). The final experimental sample for the present study consisted of 618 children (290 females, 328 males, see Table 1 ) aged between 3 years and 10 months and 10 years and 10 months. Within this sample, 531 children provided valid data for both the executive and semantic tasks, 76 for the executive task only, and 11 for the semantic task only. Table 1 Distribution of the final sample by grade level and gender . Grade F M Total Preschool 47 69 116 1^ grade 35 42 77 2^ grade 76 75 151 3^ grade 62 65 127 4^ grade 51 55 106 5^ grade 19 22 41 Total 290 328 618 Notes : F, females; M, Males Participants were recruited from several preschools and primary schools in the Northern part of Italy, within the province of Padova. All children were Italian speakers with typical development, as reported by their parents, no history of neurological, neurodevelopmental, or learning disorders, nor any uncorrected sensory impairments (vision or hearing). Written informed consent was obtained from the parents or legal guardians of all participants, and verbal assent was obtained from each child. The study was conducted in accordance with the ethical standards of the 2013 Declaration of Helsinki for human studies of the World Medical Association and received approval from the Ethical Committee for the Psychological Research of the University of Padova (protocol number: 5272). General procedure Participants performed two versions of a two-choice cued task-switching paradigm: an executive and a semantic version. While targeting different domains, both paradigms were designed with an identical structure. These experimental paradigms were part of a broader battery, designed for a larger study aimed at comparing the developmental trajectories of different executive and semantic control processes. The task-switching paradigms were both administered during the first of two experimental sessions; data not pertinent to the current research question will be presented elsewhere. To avoid participants by task order interactions, the order of presentation of the two task-switching paradigms was kept fixed, with all participants performing the executive one before the semantic one. Furthermore, the sequence of trials within each paradigm was also fixed, ensuring all children completed the same trial list. The experiment was administered using the online version of Labvanced 29 , displayed in full-screen mode on tablets with a 1280 x 800 pixel resolution (landscape orientation). Data collection took place in a quiet room within the children’s schools. In each session, the battery of tasks was administered collectively to the entire class, with each child working on an individual tablet under the supervision of a team of at least three experimenters. Children responded by tapping on one of two designated response areas on the left or right side of the tablet screen (see details below). The procedure for each domain (executive and semantic) followed a structured sequence. Before each task-switching paradigm, participants first completed two single-task blocks (20 trials each). In these blocks, they performed two discrimination tasks in isolation. These blocks were designed to assess baseline visuospatial and semantic abilities of each child. While the results from these single-task blocks are not reported directly, performance metrics derived from them were used as predictors in our main analyses to control for baseline abilities and obtain more precise estimates of the experimental effects of interest. Once the two single-task blocks were completed, the task-switching was administered in a single, continuous block of 48 trials. Specifically, the two simple discrimination tasks were mixed, and participants had to perform the relevant one based on a trial-by-trial visual cue (see below). To ensure sustained engagement from our young participants, a gamified approach was adopted throughout the procedure. Each experimental block (executive and semantic single-task and task-switching ones) was preceded by an instructional and training procedure. The lead experimenter (always the same, the author GV) provided standardized instructions to the entire group, adapting the language to the age of the class and using visual aids (slides). To capture the children’s attention and enhance compliance and motivation, the instructions were framed within a short, engaging narrative. This was followed by several practical examples where the experimenter collectively guided the children, asking different children to indicate the correct response to ensure the rules were fully understood. After the group instructions, each child completed a brief, individual training phase (4 trials for the single-task blocks, 6 trials for the task-switching blocks) with immediate, visually intuitive on-screen feedback (a happy face for correct answers; a sad face for incorrect/slow responses). During the experimental blocks, motivational images (e.g., a superhero character) were displayed every eight trials to maintain focus and compliance. Finally, both paradigms used child-friendly stimuli, as detailed in the following sections. Executive Task-Switching To assess executive flexibility, we adapted a classic color/shape task-switching paradigm 30 , implementing a gamified version with child-friendly stimuli. The task required children to sort stimuli representing cartoon aliens based on one of two visuo-spatial features: their color (red, RGB: 192, 0, 0; or blue, RGB: 0, 112, 192) or the shape of their head (squared or rounded) (see Fig. 1 ). Across all stimuli, the body and facial elements were held constant. To promote the use of abstract rules and enhance comparability with the semantic paradigm, four unique variations of each stimulus were created. These variations were generated by combining two internal color-fill patterns (zigzag vs. striped) and two head contours (e.g., square vs. rectangle for the squared-head aliens; circular vs. oval for the rounded-head aliens). The target stimuli (180 x 255 pixels) were presented at the center of the screen. The experiment was presented on a background depicting a lunar landscape with two extraterrestrial houses, one on the left and one on the right side of the screen, which served as the response areas. Crucially, signs outside each house indicated the correct feature-response mapping. The sign on the left house displayed a red paintbrush and a black square, while the sign on the right house showed a blue paintbrush and a black circle (see Fig. 1 ). The color task therefore required children to sort the alien by color, tapping the left house for red and the right house for blue aliens. The shape task required them to sort it by the shape of its head, tapping the left house for squared-head aliens and the right house for rounded-head aliens. Therefore, the employed stimuli were bivalent, that is, they afforded both a color and a shape judgment, requiring children to flexibly select the relevant stimulus feature based on the currently relevant task, which was signaled by a visual task cue (255 x 106 pixels) presented at the top-center of the screen: an image of two paintbrushes for the color task or two geometric shapes for the shape task (see Fig. 1 ). Moreover, we used only the two stimulus types that were task-incongruent: the red alien with the rounded head, requiring a left response for the color task but a right response for the shape task, and the blue alien with the squared head, requiring a right response for the color task but a left response for the shape task. This was critical to maximize flexibility demands through cross-task interference, while ensuring that correct responses on Switch trials required an actual task-set shift. The entire procedure was framed within a narrative where children had to help the lost aliens return to the correct house. They were told that to select the correct house first they had to follow a clue (paintbrushes or shapes). Each trial began with the presentation of the lunar landscape for 1000 ms. The task cue then appeared for 750 ms, after which the target alien appeared in the center while the cue remained visible. Both the cue and stimulus stayed on screen until a response was made or for a maximum of 5000 ms. Trials could either be Repeat (same task as the previous trial) or Switch trials (different task) (see Fig. 1 ). The executive task-switching paradigm consisted of a single block of 48 experimental trials, preceded by one buffer trial. Repeat and Switch trials were interspersed and equally probable (50% switch probability). The trial sequence was pseudorandomized to prevent immediate repetitions of the exact same stimulus image and to avoid more than four consecutive repetitions of the same task or response. Semantic Task-Switching Semantic flexibility was assessed using a semantic task-switching paradigm adapted for children from a version previously validated with adults 16 . While maintaining the same underlying structure as the executive task, this paradigm used stimuli with semantic meaning (instead of visuo-spatial information) to tax flexible control within the semantic domain, requiring participants to flexibly switch between different meaning aspects of a concept 31 . Children were required to categorize cartoon images based on one of two conceptual dimensions: whether the object depicted was living or non-living, or whether it was capable of moving or non-moving. The target stimuli were colorful, child-friendly cartoon images (255 x 255 pixels) presented at the center of the screen (see Fig. 1 ). These images depicted concepts belonging to two categories, chosen to be familiar to children across the entire age range of the sample: Plants (living and non-moving concepts) or Vehicles (non-living and moving concepts). Four exemplars were selected for each category (Plants: tree, sunflower, cactus, potted plant; Vehicles: bicycle, tractor, boat, helicopter. Each image was created ex-novo to be engaging and colorful, and all shared a consistent visual style. As in the executive task-switching paradigm, the selected categories made target stimuli bivalent and task-incongruent (i.e., stimuli that both tasks can be performed on and that required different responses based on the task cue). The experiment was presented on a rainbow-themed background with two response areas on the left and right. Each response area displayed images representing the possible conceptual features. The left area showed both a cartoon child (representing living concepts) and a spinning star (representing moving concepts). The right area showed both a cartoon robot (representing non-living concepts) and a stationary star (representing non-moving concepts). Therefore, for the living/non-living task, children had to tap the left area for living images (i.e., Plants) and the right area for not-living images (i.e., Vehicles), while for the moving/non-moving task, they had to tap the left area for moving images (i.e., Vehicles) and the right area for non-moving images (i.e., Plants) (see Fig. 1 ). The task to be performed was signaled by a cue stimulus (300 x 160 pixels) presented at the top-center of the screen. The cue for the living/non-living task was an image showing both the child and the robot side-by-side, while the cue for the moving/non-moving task showed the two stars. The narrative instructed the children to sort the pictures based on the specific cue they saw on each trial. All other methodological details, including trial timing and the pseudo-randomization of the 48-trial sequence, were kept identical to the executive task-switching paradigm. Data analysis Statistical analyses were performed using R ( https://www.r-project.org/ , version 4.4.1 32 ) on inverse-transformed RTs (iRTs), computed as 1000/RTs. This transformation was used to mitigate the heavy right skewness common in RT distributions and normalize their residuals. Crucially, iRTs serve as a measure of processing speed or performance, where higher values correspond to better performance (i.e., faster responses, interpreted as responses-per-second). Thirty-seven participants (5.65% of the recruited participants) were excluded from the analyses because they did not complete the two single-task and the two task-switching blocks or showed poor compliance/performance (i.e., failed to provide a response in more than 80% of trials, provided the same response in more than 90% of trials, and/or provided the incorrect response in more than 90% of trials). Practice trials and the first trial of each block were excluded from analyses. We then excluded experimental trials with incorrect responses (n = 10472, 18.98% of trials), missed responses (n = 3058, 5.54% of trials) and anticipations (i.e., trials with RTs 3; n = 195, 0.035% of trials). To address both our aims, we used linear mixed-effects model (LMM) analyses 33 . Aim 1 - Executive vs Semantic Flexibility First, to compare the effects of executive and semantic flexibility on behavioral performance and their interplay across development, we performed a within-subject analysis testing an a-priori defined model (M_EvsS_iRT) on the performance data (iRT) from both the executive and semantic paradigms. The model's Wilkinson-notation formula was: iRT ~ iRTpre + Trial*Age + postERR*Age + Dom*Cond*Age + Dom*Cond*PerfBase + (1|SS) + (1|Stim) . Control variables. The fixed part of the model included several confounding predictors: i) a continuous predictor for the iRT of the preceding trial (iRTpre) to account for temporal dependency in iRTs 34 ; ii) a continuous predictor for trial number (Trial) to account for time-on-task effects (e.g., learning and/or fatigue); iii) a dichotomous predictor for an error in the preceding trial (postERR) to account for post-error slowing 35 . The latter two predictors were tested in interaction with Age to capture any age-related modulation of these effects. Fixed effects. To test how executive-semantic flexibility interplay modulated the switch cost across development, we included predictors for Domain (Dom: Executive vs. Semantic), Condition (Cond: Repeat vs. Switch) and the continuous predictor for Age, modeling them in a three-way interaction (Dom*Cond*Age). To control for how the Dom*Cond interaction was modulated by children's baseline abilities, we also included the three-way interaction between Dom, Cond and baseline performance (PerfBase), where PerfBase is a continuous predictor representing each participant's mean iRT in executive and semantic domain when no flexibility was required (i.e., in single-task blocks). Random effects. The random effect structure included a by-participant (SS) random intercept to account for individual differences in overall performance, and a by-stimulus (Stim) random intercept to account for variability attributable to specific stimuli. Variable coding. To facilitate model convergence and interpretation, the continuous confounding predictors (Trial and iRTpre) were centered and scaled (from − .5 to .5) and z-scored, respectively, at the participant level. The categorical predictor postErr was factor-coded (0 = after correct, 1 = after error), with the 0 acting as the reference level. Age was z-scored across the entire sample and PerfBase by-domain. The predictors of interest, Dom and Cond, were treated as continuous numerical variables and centered around zero (-0.5 for Executive/Repeat, 0.5 for Semantic/Switch), so that the estimated main effects and interactions refer to the average across conditions. Outliers. Finally, after fitting the initial model, we examined the residuals to check for the presence of extreme outliers and re-fitted a trimmed version, excluding data points with absolute standardized residuals greater than 3. For the trimmed model, we reported the estimated coefficients ( b ), standard error ( SE ), t and p values for each fixed effect. Satterthwaite's approximation of degrees of freedom was used to calculate p -values and to derive effect size estimates ( d S ) for the experimental effects of interest, which were also calculated using the Westfall’s approach ( d W ) 36 . We additionally calculated the conditional R 2 values using the MuMIn package 37 . An alpha level of .05 was set as the cut-off for statistical significance. Switch cost correlation. Then, to directly compare executive and semantic flexibility effects on performance and further assess their commonalities, we extracted the individual effects of interest (executive and semantic switch costs) as estimated by the LMM model. Covariates other than Age were set to their mean value or reference level, while Age was kept at the child’s observed value. We then computed the Kendall’s τ correlation coefficient between the by-participant executive and semantic switch costs using the package pcor.test 38 . Aim 2 - Semantic Flexibility Our second aim was to characterize the development of semantic flexibility by examining how it is modulated by the organization of semantic information. To this end, we conducted a focused analysis on iRT during the semantic paradigm alone. We tested an a priori defined model (M_Sem_iRT) based on our theoretical assumptions and previous findings 16 . The model's Wilkinson-notation formula was: iRT ~ iRTpre + Trial*Age + postERR*Age + Cond*Age + Cond*PerfBase + Age*SemDist + (1|SS) + (1|Stim) . Control variables. As in the previous model (M_EvsS_iRT), we included confounding predictors (iRTpre, Trial*Age, postERR*Age) and two key interactions: Cond*Age, to test how age modulates the semantic switch cost, and Cond*PerfBase, to assess the role of baseline abilities. Fixed Effects. Crucially, building on our previous evidence showing that the representational distance between concepts influences performance 15 , 16 , 39 , 40 , we also included a continuous predictor for semantic distance (SemDist). SemDist reflects the cosine distance between the concepts of the current and previous stimulus (trial n vs. n–1), derived from fastText, a distributional semantic model trained on Italian corpora 41 – 43 . Our previous results revealed that when stimuli change on every trial (like in our paradigm), participants are required to flexibly navigate through the semantic space. When stimuli are semantically more distant from each other, they require greater flexibility – a mechanism we termed stimulus-level flexibility because it operates regardless of the task condition. Therefore, we tested SemDist in interaction with Age to explore whether this stimulus-level flexibility changes across development. However, in addition to this stimulus-level effect, semantic distance might also modulate the degree of the type of flexibility required to switch between tasks, reflected in the switch cost. Therefore, to test whether SemDist modulates the semantic switch cost and whether this modulation changes with age, we specified a second, more complex model (M_SemFull_iRT), which added a three-way interaction (Cond*Age*SemDist). The formula was: iRT ~ iRTpre + Trial*Age + postERR*Age + Cond*Age + Cond*PerfBase + Cond*Age*SemDist + (1|SS) + (1|Stim) . We then formally compared the fit of M_Sem_iRT and M_SemFull_iRT using a log-likelihood ratio test to determine if the inclusion of this critical interaction was statistically justified. For this analysis, the SemDist predictor was centered, and the rest of the model fitting procedure was identical to that described above. Sensitivity Analysis on Accuracy To control for possible speed-accuracy trade-offs, the analyses detailed above were also performed on accuracy as the dependent variable, but in this case using a generalized LMM (with the binomial family) and considering the mean accuracy in the two single-task blocks for the PerfBase predictor. Therefore, the M_EvsS_Acc model compared the effects of executive and semantic flexibility on children’s accuracy and their interplay across development, while the M_SemFull_Acc model assessed SemDist modulations of the semantic flexibility. Note that in these analyses, Satterthwaite's approach could not be used to compute the effect sizes because this method is not applicable to generalized mixed models; therefore, effect sizes were only computed following Westfall’s (2014) approach 36 . Power Analysis The sample size was determined based on the aims of the larger study that included, among others, the two task-switching paradigms described here. We recruited as many participants as possible based on the availability of resources and access to local schools. Nonetheless, we performed a sensitivity power analyses using the method introduced by Westfall and colleagues 36 for a fully-crossed linear mixed-effects model, conservatively assuming participant and stimulus intercepts and residual variance partitioning coefficients of .2, .1, and .7, respectively 36 . The variance partitioning coefficients for the participant and stimulus slopes and the participant-by-stimulus intercept were set to 0 because those effects were not included in the models we tested. This power analysis revealed that a sample size of 618 participants and 16 stimuli was large enough to detect a very small effect size (Cohen’s d = .036) with a power of .80. It should be noted, however, that this approach is not adequate for complex mixed effect models like the one used in this work 36 , 44 , but it nonetheless provides a useful estimation of the so-called minimal statistically detectable effect for our study (i.e., the lower bound of the range of effect sizes that can be detected 45 ). Indeed, to the best of our knowledge, to date there are no accepted analytical approaches to accurately compute statistical power for such models. Results Aim 1 - Executive vs. Semantic Flexibility The first analysis compared the behavioral effects of executive and semantic flexibility and explored their interplay across development. The conditional R 2 of the trimmed LMM model (M_EvsS_iRT) provided a good fit to the data (Conditional R ² = .33), with 0.65% of observations removed as extreme outliers (> 3 absolute standardized residuals). All confounding predictors modulated participants’ performance (iRT) significantly (see Table 2 for full model details). Table 2 Results of the LMM model assessing the domain-dependent switch effects on iRTs (M_EvsS_iRT) Estimate SE df t p d S d W (Intercept) 0.5910 0.0058 98 102.40 < .0001 10.33 2.50 iRT preceding trial 0.0509 0.0011 40510 46.21 < .0001 0.23 0.22 Trial 0.0118 0.0041 10020 2.90 .0038 0.03 0.05 Age 0.0327 0.0051 852 6.42 < .0001 0.22 0.14 Post-error -0.0415 0.0028 40820 -14.62 < .0001 -0.07 -0.18 Domain 0.0127 0.0070 14 1.82 .0898 0.48 0.05 Condition -0.0660 0.0023 25440 -28.90 < .0001 -0.18 -0.28 Baseline Performance 0.0411 0.0031 12030 13.36 < .0001 0.12 0.17 Trial:Age -0.0118 0.0037 40460 -3.21 .0014 -0.02 -0.05 Age:Post-error -0.0406 0.0029 41040 -14.14 < .0001 -0.07 -0.17 Domain:Condition 0.0350 0.0045 25100 7.79 < .0001 0.05 0.15 Age:Domain 0.0113 0.0032 41070 3.57 .0004 0.02 0.05 Age:Condition -0.0129 0.0029 40470 -4.37 < .0001 -0.02 -0.05 Domain:Baseline Performance -0.0394 0.0032 41020 -12.46 < .0001 -0.06 -0.17 Condition:Baseline Performance -0.0119 0.0028 40440 -4.25 < .0001 -0.02 -0.05 Age:Domain:Condition -0.0089 0.0058 40440 -1.53 .1265 -0.01 -0.04 Domain:Condition:Baseline Performance 0.0113 0.0056 40440 2.03 .0429 0.01 0.05 Notes SE, standard error; df, degrees of freedom computed with the Satterthwaite’s approximation; d S , effect size estimates calculated with Satterthwaite's approach; d W , effect size estimates calculated with Westfall’s approach. The analyses revealed various significant effects. Specifically, performance (iRT) improved significantly with Age and with higher baseline performance. Crucially, there was a main effect of Condition, confirming the presence of a significant switch cost, with better performance on Repeat trials compared to Switch trials. These main effects were qualified by several significant interactions. First, we observed two interactions involving age. The Age*Domain interaction was significant, with an age-dependent greater performance enhancement, especially in the semantic domain (Fig. 2 A). The Age*Condition interaction was also significant, indicating that performance improved with age more steeply for Repeat trials than for Switch trials (Fig. 2 B). Moreover, we found a significant interaction between Domain and the baseline performance (PerfBase), showing that the greater the baseline performance the greater the performance in task-switching, especially in the executive domain. The Condition*PerfBase interaction was also significant, showing that the performance benefit of having higher baseline abilities was more pronounced for Repeat trials than for Switch trials. Also relevant to our research question was the Domain*Condition interaction, which was significant, revealing that the magnitude of the switch cost was significantly larger in the executive domain compared to the semantic domain (Fig. 3 ). Finally, the three-way Age*Domain*Condition interaction was not significant, whereas the p -values for the Domain*Condition*PerfBase interaction was just below .05 (and it is not significant in the analysis on accuracy, see below), thus it will not be discussed further. To further probe the relationship between the two forms of flexibility, we correlated the individual switch cost effects estimated by the model. Despite the difference in magnitude, the executive and semantic switch costs were strongly and positively correlated (Kendall's τ = .817. p < .001; see Fig. 4 ). The analysis on the children’s accuracy substantially confirmed the results reported above, except for the Dom*Age and Dom*PerfBase interactions, whose statistical significance was not confirmed (see Table 3 ). Table 3 Results of the LMM model assessing the domain-dependent switch effects on accuracy (M_EvsS_Acc) Estimate SE z p d W (Intercept) 1.4475 0.0362 40.01 < .0001 3.34 iRT preceding trial -0.0319 0.0114 -2.81 .0050 -0.07 Trial 0.0095 0.0435 0.22 .8281 0.02 Age 0.6157 0.0302 20.39 < .0001 1.42 Post-error -0.4922 0.0259 -19.02 < .0001 -1.14 Domain -0.2955 0.0468 -6.31 < .0001 -0.68 Condition -0.4960 0.0240 -20.65 < .0001 -1.15 Baseline Performance 0.0949 0.0170 5.59 < .0001 0.22 Trial:Age -0.0506 0.0362 -1.40 .1624 -0.12 Age:Post-error -0.1857 0.0234 -7.94 < .0001 -0.43 Domain:Condition 0.3339 0.0475 7.03 < .0001 0.77 Age:Domain 0.0208 0.0256 0.81 .4162 0.05 Age:Condition -0.1634 0.0234 -6.99 < .0001 -0.38 Domain:Baseline Performance 0.0385 0.0295 1.31 .1918 0.09 Condition:Baseline Performance -0.0416 0.0209 -1.98 .0472 -0.10 Age:Domain:Condition 0.0238 0.0465 0.51 .6091 0.06 Domain:Condition:Baseline Performance 0.0091 0.0419 0.22 .8285 0.02 Notes : SE, standard error; d W , effect size estimates calculated with Westfall’s approach. Aim 2 - Semantic Flexibility Our second analysis aimed to characterize the development of semantic flexibility. As the log-likelihood ratio test confirmed that the model including the three-way interaction (M_SemFull_iRT) provided a significantly better fit to the data, we report the results from this more complex model (see Table 4 ). First, the analysis confirmed a significant effect of Condition, indicating a robust switch cost within the semantic domain. We also found a significant effect of Semantic Distance, revealing that performance worsened as the semantic distance between consecutive stimuli increased. Crucially, we found a significant three-way Age*Condition*Semantic Distance interaction. This interaction revealed that the performance worsening associated with greater semantic distance was more pronounced on Repeat trials, and that this effect became stronger with age (Fig. 5 ). In older children especially, this resulted in a notable cost for repeating a task when they had to switch towards more semantically distant stimuli. Table 4 Results of the LMM model assessing semantic switch effect on iRTs (M_SemFull_iRT) Estimate SE df t p d S d W (Intercept) 0.6192 0.0081 39 76.92 < .0001 12.26 2.71 iRT preceding trial 0.0325 0.0017 18570 19.63 < .0001 0.14 0.14 Trial 0.0402 0.0060 5441 6.75 < .0001 0.09 0.18 Age -0.0019 0.0079 924 -0.25 .8051 -0.01 -0.01 Post-error -0.0344 0.0040 18700 -8.57 < .0001 -0.06 -0.15 Condition -0.0602 0.0081 12360 -7.45 < .0001 -0.07 -0.26 Baseline Performance 0.0856 0.0066 510 13.00 < .0001 0.58 0.37 Semantic Distance -0.0784 0.0173 12090 -4.54 < .0001 -0.04 -0.34 Trial:Age -0.0082 0.0055 18330 -1.50 .1350 -0.01 -0.04 Age:Post error -0.0342 0.0042 18810 -8.22 < .0001 -0.06 -0.15 Age:Condition -0.0459 0.0086 18330 -5.36 < .0001 -0.04 -0.20 Condition:Baseline Performance -0.0056 0.0039 18310 -1.42 .1544 -0.01 -0.02 Condition:Semantic Distance 0.0824 0.0339 13200 2.43 .0151 0.02 0.36 Age:Semantic Distance -0.0174 0.0173 18320 -1.01 .3149 -0.01 -0.08 Age:Condition:Semantic Distance 0.1346 0.0344 18330 3.91 .0001 0.03 0.59 Notes : SE, standard error; df, degrees of freedom computed with the Satterthwaite’s approximation; d S , effect size estimates calculated with Satterthwaite's approach; d W , effect size estimates calculated with Westfall’s approach. The analysis on the children’s accuracy confirmed the results reported above, except for the main effect of SemDist (see Table 5 ). Table 5 Results of the LMM model assessing semantic switch effect on accuracy (M_SemFull_Acc) Estimate SE z p d W (Intercept) 1.3769 0.0598 23.02 < .0001 1.92 iRT preceding trial 0.0013 0.0172 0.08 .9391 < 0.01 Trial 0.0669 0.0618 1.08 .2793 0.09 Age 0.5874 0.0547 10.75 < .0001 0.82 Post-error -0.3967 0.0368 -10.79 < .0001 -0.55 Condition -0.4775 0.0831 -5.75 < .0001 -0.67 Baseline Performance 0.3134 0.0370 8.48 < .0001 0.44 Semantic Distance -0.2366 0.1765 -1.34 .1800 -0.33 Trial:Age -0.0804 0.0539 -1.49 .1359 -0.11 Age:Post-error -0.1785 0.0344 -5.18 < .0001 -0.25 Age:Condition -0.4177 0.0794 -5.26 < .0001 -0.58 Condition:Baseline Performance -0.0413 0.0301 -1.37 .1698 -0.06 Condition:Semantic Distance 0.7035 0.3477 2.02 .0431 0.98 Age:Semantic Distance -0.1509 0.1666 -0.91 .3650 -0.21 Age:Condition:Semantic Distance 1.2505 0.3309 3.78 .0002 1.74 Notes : SE, standard error; d W , effect size estimates calculated with Westfall’s approach. Discussion This study investigated the developmental trajectories of executive and semantic flexibility in children using parallel versions of a cued task-switching paradigm to address a fundamental question raised in the adult cognitive control literature: whether these two forms of flexibility rely on distinct or shared mechanisms. By charting these abilities throughout childhood, we sought to determine if they follow distinct or shared developmental trajectories. Our findings reveal a complex picture, characterized by evidence consistent with a strong shared component that is nonetheless modulated by domain-specific factors. These results provide a crucial developmental perspective on the neurocognitive models of executive and semantic control; they also highlight the intricate nature of cognitive development itself. Our first key finding demonstrates a strong correlation between switch costs in the executive and semantic domains, consistent with the idea of shared control mechanisms. This result is consistent with findings in healthy adults 16 , which show a similar moderate-to-strong correlation. Finding this strong association in children suggests the link between these abilities is a fundamental aspect of development, not just a mature state. This aligns with theoretical models which argue that a core set of executive functions drive flexible behavior 46 , 47 . The mental act of switching tasks also taxes working memory to maintain task sets and apply the relevant one, and inhibitory control to stop applying the old one. In essence, both our paradigms relied on this same underlying cognitive machinery for switching. Therefore, a child's developmental level in operating this machinery would naturally lead to similar performance across both domains, explaining the strong correlation. While the correlation suggests a shared component, our findings are consistent with the asynchronous and multi-faceted nature of cognitive control development highlighted in our introduction. The significant Age by Condition interaction revealed that performance on Repeat trials improved more steeply with age than on Switch trials. This pattern points to a growing functional dissociation between more automatized task execution and effortful cognitive control. As children age, they become increasingly efficient at maintaining and executing an active task-set, leading to dramatic performance gains on Repeat trials. However, this emerging task automatization introduces a greater degree of task inertia. A more consolidated cognitive system, while more efficient, naturally offers more resistance to flexibility. Consequently, the process of switching to a newly relevant task-set necessarily requires a costly control intervention to override this emerging automaticity and implement a new goal-directed action. While this flexibility ability also matures, its developmental gains observed on Switch trials are intrinsically less pronounced than those afforded by the powerful process of task automatization on Repeat trials. Our results, therefore, suggest that a key developmental achievement in this age range is the ability to form and execute efficient cognitive routines, which in turn makes the relative cost of volitionally breaking those routines – the switch cost – more apparent. Our results of a significant Domain by Condition interaction also challenge a purely domain-general account by revealing domain-specific modulations. The larger switch cost in the executive domain than in the semantic domain is consistent both with findings with adults 16 and developmental research showing that switching between highly interdependent perceptual dimensions (like color and shape) is particularly difficult for young children 22 , 48 . As discussed in our introduction, prior studies have shown children successfully switching between perceptual rules (shape) and semantic ones (function) 22 . Our finding extends this, suggesting that a switch between two competing perceptual rules may be uniquely demanding. Furthermore, the steeper age-related performance improvement in the semantic task, evidenced by the Domain by Age interaction, strongly supports the idea that development is shaped by the richness of the knowledge base upon which control operates. Crucially, this developmental effect was driven entirely by processing speed, not accuracy. This speed-accuracy dissociation is highly informative: it suggests that the conceptual knowledge required for the semantic task –whether stimuli were 'living' or 'moving'– was already firmly in place even for our youngest participants. Therefore, the observed developmental gains are not attributable to new knowledge acquisition but rather reflect the maturing efficiency with which this established knowledge is accessed, organized, and managed. This aligns perfectly with the notion that as children’s conceptual knowledge becomes more organized, the task is "scaffolded", reducing the cognitive load on control processes and boosting their operational speed 28 , as well as with a developmental shift toward more automatic semantic processing, which would free up control resources with age 6 . Therefore, our findings reveal a key developmental distinction: while the perceptual processing remains a significant bottleneck for efficiency throughout childhood, the conceptual processing benefits greatly from the ongoing organization and automatization of the underlying semantic knowledge base. Our most novel contribution is the finding of a three-way interaction between Age, Condition, and Semantic Distance, that speaks directly to the CSC framework 8 and our second aim of characterizing semantic flexibility itself. The finding that older children showed a performance cost for larger semantic distance on Repeat trials, but not on Switch trials, is consistent with the interplay between the semantic "hub" (the knowledge store where conceptual distance is represented) and the "control" system 8 . This suggests that as children grow older, their semantic representations become more structured and, thus, increasingly similar to the adults’ one 16 . As such, even when repeating the same task, traversing this space requires a form of flexibility to navigate through the stimuli in the semantic space (see stimulus-level flexibility) 16 . The masking of this effect on switch trials is what the CSC model would predict: the resource-intensive process of reconfiguring the entire control network leaves fewer resources for finer-grained, representation-level modulations. This echoes the broader theme that developmental trajectories are often not linear but are marked by increasing complexity and dynamics. Of course, this interpretation must be tempered by the possibility of a statistical floor effect. Performance on switch trials is already under such high load that our behavioral measures may lack the sensitivity to detect a more subtle, additional cost from semantic distance. Yet, this finding is consistent with the idea raised by research on early conceptual development 22 : understanding the development of flexibility requires understanding the development of the knowledge it controls 22 . The potential confounds discussed above represent the most significant limitations of our study. We must explicitly state that while our model controlled for baseline performance, the larger executive switch cost could stem from differences in switch-specific task difficulty or rule type. Future work must aim to equate the interference demands between paradigms to isolate the specific effects of information content. Furthermore, our cross-sectional design only allows for inferences about group-level trends, not the individual trajectories that require longitudinal study. Finally, while our behavioral data are informative, their greatest value may be in generating specific, testable hypotheses for future neuroimaging research. The strong behavioral correlation between switch costs is consistent with the hypothesis of a shared neural basis, which should be mirrored by correlated activation in the domain-general multiple-demand network that supports a wide range of effortful cognitive tasks 12 . Correspondingly, the unique variance in semantic task performance, particularly the modulation by semantic distance, is consistent with the possibility of a distinct neural basis uniquely explained by activity in semantic-specific control regions, such as the left inferior frontal gyrus and posterior middle temporal gyrus, consistent with the CSC framework 6 , 8 . However, while our data point towards a shared foundation, we must be cautious. This correlation could also be driven by the maturation of a more general ability to consciously represent and apply paired-rule structure, regardless of their content 48 . A child who has mastered this fundamental logical operation will succeed at both tasks, while a child who has not will struggle, creating a strong correlation driven by general rule-use ability rather than a specific switch mechanism. Therefore, while our data point towards a shared foundation, its precise nature, a specific flexibility module, a set of core executive functions, or a general rule-use capacity, remains ambiguous. Therefore, the precise nature of this shared resource remains a key question for future research. In conclusion, by employing the cued, trial-by-trial methodology, this study offers the first direct comparison of the developmental trajectories of executive and semantic flexibility. Our findings are consistent with a model where a shared cognitive foundation underpins both abilities, as observed in adult findings 16 , but where performance may be shaped by distinct developmental timelines for its subcomponents and modulated by domain-specific factors. These findings highlight that a complete understanding of adaptive behavior requires focusing on the dynamic interplay between maturing general control processes and specialized knowledge systems. Declarations Competing interests: The authors declare no competing interests. Funding: This study was supported by the European Union (HORIZON-MSCA-2023-PF-01-01, CTRL-ALT-DEV, Grant 101150190 to MM and ERC-2021-STG, IN-MIND, Grant 101043216 to SB-V). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. Acknowledgements: We thank Marina Mancuso and Chiara Maria Migliorin for their assistance with data collection. We are also deeply grateful to the children, parents, and teachers who participated in this study for their invaluable collaboration. Data Availability Statement: The datasets generated and analyzed during the current study are available in the OSF repository, https://osf.io/zr4dn/overview . AUTHOR CONTRIBUTIONS STATEMENT Giada Viviani: Data Curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing - Original Draft, Writing - Review & Editing. Ettore Ambrosini: Data Curation, Formal analysis, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing - Original Draft, Writing - Review & Editing. Annamaria Porru: Resources, Writing - Review & Editing. Silvia Benavides-Varela : Funding acquisition, Writing - Review & Editing. Erin M. Buchanan: Writing - Review & Editing. Irene Di Pietro: Investigation, Writing - Review & Editing. Daniela Lucangeli: Writing - Review & Editing. Maria Montefinese: Conceptualization, Data Curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing - Original Draft, Writing - Review & Editing. References Chiew KS, Braver TS (2017) Context Processing and Cognitive Control: From Gating Models to Dual Mechanisms. in The Wiley Handbook of Cognitive Control (ed. Egner, T.) 143–166Wiley. 10.1002/9781118920497.ch9 Monsell S (2003) Task-set reconfiguration processes do not imply a control homunuculus: Reply to Altmann. Trends Cogn Sci 7:341–342 Viviani G, Visalli A, Finos L, Vallesi A, Ambrosini E (2023) A comparison between different variants of the spatial Stroop task: The influence of analytic flexibility on Stroop effect estimates and reliability. Behav Res 56:934–951 Egner T (2023) Principles of cognitive control over task focus and task switching. Nat Rev Psychol 2:702–714 Sulpizio S, Job R, Leoni P, Scaltritti M (2022) Prepotent task-irrelevant semantic information is dampened by domain-specific control mechanisms during visual word recognition. Q J Experimental Psychol 75:390–405 Jefferies E (2013) The neural basis of semantic cognition: Converging evidence from neuropsychology, neuroimaging and TMS. Cortex 49:611–625 Montefinese M, Hallam G, Thompson HE, Jefferies E (2020) The interplay between control processes and feature relevance: Evidence from dual-task methodology. Q J Experimental Psychol 73:384–395 Lambon Ralph MA, Jefferies E, Patterson K, Rogers TT (2017) The neural and computational bases of semantic cognition. Nat Rev Neurosci 18:42–55 Reilly J et al (2024) What we mean when we say semantic: Toward a multidisciplinary semantic glossary. Psychon Bull Rev. https://doi.org/10.3758/s13423-024-02556-7 Badre D, Poldrack RA, Paré-Blagoev EJ, Insler RZ, Wagner AD (2005) Dissociable Controlled Retrieval and Generalized Selection Mechanisms in Ventrolateral Prefrontal Cortex. Neuron 47:907–918 Davey J et al (2016) Exploring the role of the posterior middle temporal gyrus in semantic cognition: Integration of anterior temporal lobe with executive processes. NeuroImage 137:165–177 Duncan J (2010) The multiple-demand (MD) system of the primate brain: mental programs for intelligent behaviour. Trends Cogn Sci 14:172–179 Thompson HE et al (2018) The contribution of executive control to semantic cognition: Convergent evidence from semantic aphasia and executive dysfunction. J Neuropsychol 12:312–340 Almaghyuli A, Thompson H, Lambon Ralph MA, Jefferies E (2012) Deficits of semantic control produce absent or reverse frequency effects in comprehension: Evidence from neuropsychology and dual task methodology. Neuropsychologia 50:1968–1979 Montefinese M, Hallam G, Stampacchia S, Thompson HE, Jefferies E (2021) Deficits of semantic control disproportionately affect low-relevance conceptual features: evidence from semantic aphasia. Aphasiology 35:1448–1462 Viviani G et al (2025) Exploring semantic and executive flexibility interplay in task switching. Sci Rep 15:31609 Ruffini C, Traverso L, Marzocchi GM, Pecini C (2025) Enhancing preschoolers’ cognitive flexibility and pre-literacy skills through dialogic reading: evidence from an Italian teacher-led intervention. Early Child Dev Care 195:847–868 Buttelmann F, Karbach J (2017) Development and Plasticity of Cognitive Flexibility in Early and Middle Childhood. Front Psychol 8:1040 Crone EA, Donohue SE, Honomichl R, Wendelken C, Bunge SA (2006) Brain Regions Mediating Flexible Rule Use during Development. J Neurosci 26:11239–11247 Uddin LQ (2021) Cognitive and behavioural flexibility: neural mechanisms and clinical considerations. Nat Rev Neurosci 22:167–179 Schettini E, Hiersche KJ, Saygin ZM (2023) Individual Variability in Performance Reflects Selectivity of the Multiple Demand Network among Children and Adults. J Neurosci 43:1940–1951 Deak GO (2000) The Growth of Flexible Problem Solving: Preschool Children Use Changing Verbal Cues to Infer Multiple Word Meanings. J Cognition Dev 1:157–191 Deák GO, Ray SD, Pick AD (2004) Effects of age, reminders, and task difficulty on young children’s rule-switching flexibility. Cogn Dev 19:385–400 Chami S, Charalambous C, Knijnik SR, Docking K (2022) Language and executive function skills as predictors of semantic fluency performance in pre-school children. Int J Speech Lang Pathol 24:626–635 Ruffini C, Osmani F, Bigozzi L, Pecini C (2024) Semantic fluency in 3–6 years old preschoolers: which executive functions? Child Neuropsychol 30:563–581 Memisevic H, Biscevic I, Pasalic A (2017) Developmental trends in semantic fluency in preschool children. Cogent Psychol 4:1403064 Kavé G, Kigel S, Kochva R (2008) Switching and clustering in verbal fluency tasks throughout childhood. J Clin Exp Neuropsychol 30:349–359 Betz N, Coley JD (2020) Development of Conceptual Flexibility in Intuitive Biology: Effects of Environment and Experience. Front Psychol 11:537672 Finger H, Goeke C, Diekamp D, Standvoß K, König P (2017) LabVanced: A unified JavaScript framework for online studies. in 1–3 (University of Osnabrück Handbook of Human Multitasking . (Springer International Publishing, Cham, (2022) 10.1007/978-3-031-04760-2 Jackson RL (2021) The neural correlates of semantic control revisited. NeuroImage 224:117444 R Core Team (2021) A language and environment for statistical computing. R Foundation for Statistical Computing Baayen RH, Davidson DJ, Bates DM (2008) Mixed-effects modeling with crossed random effects for subjects and items. J Mem Lang 59:390–412 Baayen RH, Milin P (2010) Analyzing reaction times. Int j psychol res 3:12–28 Rabbitt PM (1966) Errors and error correction in choice-response tasks. J Exp Psychol 71:264–272 Westfall J, Kenny DA, Judd CM (2014) Statistical power and optimal design in experiments in which samples of participants respond to samples of stimuli. J Exp Psychol Gen 143:2020–2045 Bartoń (2025) k. MuMIn: Multi-Model Inference Kim S (2015) ppcor: An R Package for a Fast Calculation to Semi-partial Correlation Coefficients. CSAM 22, 665–674 Buchanan EM et al (2025) Measuring the semantic priming effect across many languages. Nat Hum Behav. https://doi.org/10.1038/s41562-025-02254-x Montefinese M, Zannino GD, Ambrosini E (2015) Semantic similarity between old and new items produces false alarms in recognition memory. Psychol Res 79:785–794 Joulin A et al (2016) FastText.zip: Compressing text classification models. Preprint at. https://doi.org/10.48550/ARXIV.1612.03651 Bojanowski P, Grave E, Joulin A, Mikolov T (2016) Enriching Word Vectors with Subword Information. Preprint at https://doi.org/10.48550/ARXIV.1607.04606 Grave E, Bojanowski P, Gupta P, Joulin A, Mikolov T (2018) Learning Word Vectors for 157 Languages. Preprint at https://doi.org/10.48550/ARXIV.1802.06893 Brysbaert M, Stevens M (2018) Power Analysis and Effect Size in Mixed Effects Models: A Tutorial. J Cognition 1:9 Lakens D, Equivalence Tests (2017) A Practical Primer for t Tests, Correlations, and Meta-Analyses. Social Psychol Personality Sci 8:355–362 Diamond A (2013) Executive Functions. Annu Rev Psychol 64:135–168 Miyake A et al (2000) The Unity and Diversity of Executive Functions and Their Contributions to Complex Frontal Lobe Tasks: A Latent Variable Analysis. Cogn Psychol 41:49–100 Zelazo PD (2015) Executive function: Reflection, iterative reprocessing, complexity, and the developing brain. Dev Rev 38:55–68 Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7965304","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":536013299,"identity":"e073cca0-bdf9-4c5e-95fc-58e3485799dc","order_by":0,"name":"Giada Viviani","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Giada","middleName":"","lastName":"Viviani","suffix":""},{"id":536013300,"identity":"101e254d-a39c-4a0f-836f-666bed60e42d","order_by":1,"name":"Ettore Ambrosini","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ettore","middleName":"","lastName":"Ambrosini","suffix":""},{"id":536013301,"identity":"ffebe8e4-984a-491b-b8e1-7be2572eea26","order_by":2,"name":"Annamaria Porru","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Annamaria","middleName":"","lastName":"Porru","suffix":""},{"id":536013302,"identity":"1929faf6-0c44-448b-b819-2006d9546a05","order_by":3,"name":"Silvia Benavides-Varela","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Silvia","middleName":"","lastName":"Benavides-Varela","suffix":""},{"id":536013303,"identity":"5a5b233c-24c3-41dd-8e52-6c6617ab5bbf","order_by":4,"name":"Erin Buchanan","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Erin","middleName":"","lastName":"Buchanan","suffix":""},{"id":536013304,"identity":"2616206f-9073-4bf9-8d14-05404c32509f","order_by":5,"name":"Daniela Lucangeli","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Daniela","middleName":"","lastName":"Lucangeli","suffix":""},{"id":536013305,"identity":"95c3aa6f-02ab-44be-89a7-4d7fbf493605","order_by":6,"name":"Maria Montefinese","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIie3PsUrDQBjA8S8EzuUwawqSPIFwUkiQinmVHMJ16SyOJ4V08QEqvkSnkvHCQacTV+WWZunUwW4WBD2PtkNJmtXh/sPxEfLjuwNwuf5pYj985n/nGQfIwQfUSRB4U0uw2JFT5kDsEFroQ9uay8lrLb5gHqXxuOJ1eZMFvXWyrMsBnMe8kSRqSKon0P3rAlFOFaPPL6OUUDVsvVgiGAgMms4WOF3SQuZEj5LQDO3kbQXVtyXBhtPiJ8s+VAd5ZyB3WzxDhDcLcRdZgbwguk8W7MqQOzpV7D40b8EI5S0XY/5m/aAjImX9uC1us2Ai571tOYjisWheYyPHHzwO+MT/TRnicrlcrn2/kixgUTv86/MAAAAASUVORK5CYII=","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Maria","middleName":"","lastName":"Montefinese","suffix":""}],"badges":[],"createdAt":"2025-10-28 12:40:04","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7965304/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7965304/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94639233,"identity":"ab6d578b-f5dc-455e-b108-5b25b93887b6","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2172014,"visible":true,"origin":"","legend":"","description":"","filename":"VivianiKidsTSms1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/fc29628f069ebdc461a826cb.docx"},{"id":94639232,"identity":"4fc0b8c9-60d6-4bb1-a09a-40b8fb3201d5","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7965304.json","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/68ce4544af1a2819efd949c1.json"},{"id":94671977,"identity":"0937b43d-8a10-4e5f-983f-0aec76c4f43f","added_by":"auto","created_at":"2025-10-29 13:31:55","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175061,"visible":true,"origin":"","legend":"","description":"","filename":"rs79653040enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/ba42c6fa7de28c7d835bd593.xml"},{"id":94639244,"identity":"70735422-4fe1-402e-8f56-225ee27b405d","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1730292,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/1ee4f45e41f4538f8b28e471.png"},{"id":94727599,"identity":"7d3b9865-5228-4f6c-8a39-fd6f1eb47b1e","added_by":"auto","created_at":"2025-10-30 06:50:16","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157787,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/62438b5314fc46c16b7a1661.png"},{"id":94671987,"identity":"687ad8ec-089e-4b1f-8f88-9efe64157762","added_by":"auto","created_at":"2025-10-29 13:31:57","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83495,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/c8de3de15853feb3d72dc455.png"},{"id":94639241,"identity":"cd9a565d-aa1a-4613-9c25-1934e416c2e9","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19837,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/26e5782281313232dd66106c.png"},{"id":94671989,"identity":"515e3c57-bc7b-4d5c-b264-67454630c059","added_by":"auto","created_at":"2025-10-29 13:32:01","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":85100,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/be4da57966cd31b592d62f7e.png"},{"id":94639250,"identity":"c5598697-409e-4438-bd67-4a546d2956c2","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":158014,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/68b20d0d6fd4b6a64148d439.png"},{"id":94727634,"identity":"c1258102-c762-4a63-82b2-1a24a9743532","added_by":"auto","created_at":"2025-10-30 06:50:34","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29218,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/d0ebdd727005c336e5029144.png"},{"id":94639246,"identity":"d228e214-d0f0-4aa2-832f-213cdef6b356","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15360,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/d7f70a1d42604f893ee13e95.png"},{"id":94727582,"identity":"314b6b47-2a09-4fcf-87ba-fd717dfa45b8","added_by":"auto","created_at":"2025-10-30 06:50:15","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8324,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/a7f45d05080edf337c00b005.png"},{"id":94671984,"identity":"f3934114-3693-489d-bd27-8f7a85c872c4","added_by":"auto","created_at":"2025-10-29 13:31:56","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16775,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/bbcf88f0cca0440aa065c102.png"},{"id":94639251,"identity":"fd3f5275-6d01-41d0-afd7-d0be14f1e59a","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":171976,"visible":true,"origin":"","legend":"","description":"","filename":"rs79653040structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/8088d000eadf702aca27782f.xml"},{"id":94639248,"identity":"d37c6bc8-c44f-4702-931a-b53b79f3b63f","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182866,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/7301e6c87df729ebcd2d4524.html"},{"id":94671985,"identity":"5c418a82-8649-4f8d-967a-a950a84991be","added_by":"auto","created_at":"2025-10-29 13:31:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":461007,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTask-switching experimental paradigms.\u003c/strong\u003e Top panel: Executive task-switching. Children judged whether a cartoon alien was red/blue (color task) or had a squared/rounded head (shape task), based on a visual cue (paintbrushes for the color task; shapes for the shape task). Responses were made by tapping on one of two extraterrestrial houses, which had signs displaying the feature-response mapping (left house: red and squared; right house: blue and rounded). Bottom panel: Semantic task-switching. Children judged whether an image referred to a living/non-living or moving/non-moving concept, based on a visual cue (a child/robot for the living/non-living task; a stationary/spinning star for the moving/non-moving task). Responses were made by tapping on one of two areas displaying the concept-response mapping (left area: living and moving; right area: non-living and non-moving). For both paradigms, three example trials are shown. The second trial is a Switch trial, where children were required to switch from the task performed in the previous trial to the other task (e.g., from discriminating alien’s color to discriminating alien’s head shape, or from classifying the concept as living/non-living to classifying it as moving/non-moving). The third trial is a Repeat trial, where children kept performing the same task as in the previous trial (e.g., discriminating the alien’s head shape or classifying the concept as moving/non-moving).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/097c5b0bd9016f9345c9af34.png"},{"id":94639239,"identity":"5558f8ee-d899-420f-becc-4184b0322d9c","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":171360,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLMM results (M_EvsS_iRT).\u003c/strong\u003e \u003cstrong\u003eA)\u003c/strong\u003e Age by Domain interaction. The plot shows the behavioral performance (iRTs) corresponding to the marginal effects estimated by the linear mixed model as a function of Age (x-axis) and Domain (Executive, blue vs Semantic, red lines). \u003cstrong\u003eB)\u003c/strong\u003e Age by Condition interaction. The plot shows the behavioral performance (iRTs) corresponding to the marginal effects estimated by the linear mixed model as a function of Age (x-axis) and Condition (Repeat, green vs Switch, fuchsia lines). Shaded regions indicate the 95% confidence interval.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/9141b890b9fb229733ff464f.png"},{"id":94671983,"identity":"f8627ae6-4529-48be-83ad-a85cba929938","added_by":"auto","created_at":"2025-10-29 13:31:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":89640,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLMM results (M_EvsS_iRT), Domain by Condition interaction. \u003c/strong\u003eThe plot shows the behavioral performance (iRTs) corresponding to the marginal effects estimated by the linear mixed model as a function of Condition (x-axis) and Domain (Executive, blue vs Semantic, red lines). Shaded regions indicate the 95% confidence interval.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/e5fcbb7e34bce6df1ad95242.png"},{"id":94639234,"identity":"4f7c4e63-7267-4b1c-95de-ac25f1632732","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":73324,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelation between the switch costs in the two Domains.\u003c/strong\u003e The figure shows the correlation between participants’ switch costs in the Semantic (y-axis) and Executive (x-axis) Domains.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/44107ed5eae9ea04faac37ec.png"},{"id":94639237,"identity":"c59642d3-80f4-40c4-a64e-9dd65f963286","added_by":"auto","created_at":"2025-10-29 07:36:45","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":115870,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLMM results (M_SemFull_iRT), three-way interaction between Age, Condition and Semantic Distance.\u003c/strong\u003e The plot shows the behavioral performance (iRTs) corresponding to the marginal effects estimated by the linear mixed model as a function of Semantic Distance (x-axis), Condition (Repeat, green vs Switch, fuchsia lines), and Age (z-scored, left and right panels, corresponding to 4 years, 5 months and 10 years, 10 months, respectively). Shaded regions indicate the 95% confidence interval.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/688d48b50957638dd9982113.png"},{"id":94822683,"identity":"aa1b11ba-4add-42f9-85b2-805362c115e9","added_by":"auto","created_at":"2025-10-31 06:32:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2029111,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7965304/v1/2ca34933-a9f8-4f11-87de-979dba02788f.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDevelopmental Trajectories of Executive and Semantic Flexibility Using Task-Switching\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTo effectively navigate a constantly changing environment, we rely on cognitive control \u0026ndash; a family of top-down mechanisms that organize thoughts and actions to align with our goals\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. A fundamental component of cognitive control is stability, which contributes to maintaining focus on a relevant \"task set\" (the rules for a task)\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e while shielding it from distraction, as typically assessed by conflict paradigms such as the Stroop task\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. While crucial, stability alone is insufficient. Our dynamic world often requires cognitive flexibility, a complementary component to rapidly switch between task sets as demands change\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Thus, the crucial role of cognitive control is to balance stability and flexibility to effectively achieve our goals. For instance, while driving, we must maintain the stable goal of following a route, yet be able to flexibly react to an unexpected road closure by quickly planning a new one. Given its central role in adaptive cognition, flexibility is extensively investigated using task-switching paradigms\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, whereby participants alternate between two (or more) tasks. The performance difference between \"Switch\" trials (where the task changes) and \"Repeat\" trials (where the task remains the same) is known as the \"switch cost\". This cost serves as a key behavioral measure of flexibility, with smaller switch costs indicating greater efficiency in shifting between tasks.\u003c/p\u003e\u003cp\u003eWhile the task-switching paradigm provides a robust method for measuring cognitive flexibility, many versions typically rely on stimuli that lack real-world meaning, such as simple geometric shapes or digits. This approach reduces the ecological validity of the findings, as it does not fully capture how we flexibly manipulate meaningful information in everyday life, where we frequently need to adapt our use of semantic knowledge based on context. For example, we may be required to shift from the meaning of the word 'jam' referring to a fruit preserve to its alternative meaning referring to a traffic obstruction. Flexibly interpreting such meanings requires a form of cognitive control known as semantic control\u003csup\u003e\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e\u0026ndash; a mechanism that flexibly prioritizes some of a concept\u0026rsquo;s features or meanings while actively suppressing others that are irrelevant to the current goal.\u003c/p\u003e\u003cp\u003eThe concept of semantic control being distinct from domain-general cognitive control (hereafter called executive control) is a central tenet of the Controlled Semantic Cognition (CSC) framework\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. This framework posits that semantic control is \u0026ldquo;a set of executive control processes that regulate the activation and deployment of semantic knowledge. These allow flexible, context- and task-appropriate responses by ensuring that only relevant aspects of semantic representations are used to direct thought and behavior.\u0026rdquo;\u003csup\u003e9\u003c/sup\u003e (p. 259).\u003c/p\u003e\u003cp\u003eA dilemma exists, however, regarding the independence of these two control systems. While semantic control is supported by specific brain regions (such as the left posterior middle temporal gyrus and the inferior frontal gyrus)\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, neuroimaging evidence also shows partial overlap with the multiple-demand network that underpins executive control\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. The distinction primarily originates from research on patients with semantic aphasia, who show deficits in using conceptual knowledge despite intact semantic representations\u003csup\u003e\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. However, these patients also exhibit deficits in executive control, challenging the assumption that the two mechanisms are entirely independent\u003csup\u003e\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. To better untangle this relationship, it is crucial to consider that much like its executive counterpart, semantic control comprises components of both stability (maintaining focus on relevant meanings) and flexibility (shifting between meanings as the context demands). To date, research exploring the link between the two mechanisms has primarily focused on the stability dimension, whereas the interplay between executive flexibility and semantic flexibility remains largely under-explored. An exception is our recent work with healthy adults, which showed a moderate-to-strong correlation between switch costs in executive and semantic tasks, suggesting shared underlying mechanisms\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eStudying the developmental trajectories of executive and semantic control from childhood offers a unique window into the architecture of cognitive control, allowing researchers to determine whether the observed associations in adulthood reflect shared developmental origins or later functional convergence. This developmental perspective moves beyond correlational evidence to uncover the mechanisms by which cognitive systems differentiate and interact over time. Given that the link between the two control systems is particularly under-explored for the flexibility component, in both adults\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e and children, charting the developmental course of both executive and semantic flexibility is a critical step forward.\u003c/p\u003e\u003cp\u003eIn children, executive flexibility is a core component of cognitive control that is crucial for school readiness and later academic success\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Its developmental trajectory is not a simple, linear improvement. Although the behavioral ability to switch between tasks matures relatively early, showing significant gains between ages 7 and 11 and reaching adult-like performance around age 12, the underlying neural systems supporting this skill undergo a far more prolonged and complex maturation into adolescence\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Neuroimaging research has sought to pinpoint the specific brain regions that drive this maturation. One influential line of evidence proposes that the sub-processes of task switching are supported by distinct prefrontal regions that mature at different rates\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. This model suggests the (pre-)supplementary motor area, a medial frontal region critical for suppressing the previous task set, shows an adult-like activation pattern by early adolescence. In contrast, the ventrolateral prefrontal cortex, a lateral region essential for retrieving and maintaining the current task rule, continues to show immature activation patterns throughout the teenage years\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. However, this region-specific dissociation is debated. An alternative, network-based perspective suggests that even young children recruit a broad frontoparietal \u0026ldquo;multiple-demand network\u0026rdquo; that is spatially similar to the one used by adults\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. From this viewpoint, development is characterized not by different regions coming online at different times, but by a process of network-wide refinement, where activation becomes more efficient and neurally selective with age and improving ability\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. In summary, while there is a clear consensus that the neural substrates of executive flexibility undergo a prolonged, non-linear development, the literature presents an active debate. It remains unresolved whether this maturation is best described by the staggered development of specific prefrontal regions handling distinct sub-processes, or by the holistic refinement of a pre-existing, domain-general cognitive control network.\u003c/p\u003e\u003cp\u003eResearch\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e has also explored the development of semantic flexibility using stimuli with semantic content. Studies have shown that between the ages of 3 and 6, children improve in switching between verbal rules for sorting cards and in flexibly using changing semantic cues to infer the meanings of new words\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. For example, preschoolers can learn to switch between sorting objects by a perceptual feature like shape and a more semantic feature like function\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Similarly, word-learning tasks have demonstrated that 4- to 6-year-olds can flexibly use different verbal cues to decide whether a new word refers to an object's shape, material, or another part, a skill that is less developed in 3-year-olds\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Verbal fluency tasks, which measure how many words a child can generate from a specific semantic category, have also been used as an indicator of semantic flexibility, with performance improving with age\u003csup\u003e\u003cspan additionalcitationids=\"CR25 CR26\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. These paradigms, while providing valuable insights, often differ from the trial-by-trial cued task-switching methodology typically used to measure executive flexibility. This methodological divergence makes it difficult to determine whether observed improvements are driven by the same underlying control mechanisms isolated by classic switch-cost designs. The development of semantic flexibility is further complicated by its interaction with other cognitive domains. For instance, a child\u0026rsquo;s growing ability to flexibly use concepts likely reflect the combined contributions of both executive control and the acquisition of domain-specific semantic knowledge\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eDespite this existing body of work, a significant gap remains. No experimental study has explicitly charted the developmental trajectory of semantic flexibility in children using a cued, trial-by-trial task-switching paradigm, nor has one directly compared it to domain-general executive flexibility using a parallel design. This methodological gap leaves a fundamental developmental question unanswered: are semantic and executive flexibility distinct processes that mature independently, or do they rely on a shared cognitive process that follows a common developmental timeline? Addressing this question has broader theoretical implications, as it informs our understanding of how domain-general and domain-specific control mechanisms interact during development to produce the mature forms observed in adulthood.\u003c/p\u003e\u003cp\u003eTo address this question, the present study investigates the developmental trajectories of both executive and semantic flexibility. As previously validated with adult participants\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, we employed a task-switching methodology with two parallel paradigms: a classical visuospatial task with minimal semantic content to measure executive flexibility, and a novel task requiring semantic judgments about pictures of meaningful concepts to measure semantic flexibility. Our study has two primary aims. First, we aim to elucidate the relationship between these two core cognitive abilities across development, assessing their developmental interplay. By directly comparing performance on these parallel tasks, we can provide crucial evidence as to whether they are supported by distinct or shared control mechanisms. Second, we aim to gain a deeper understanding of semantic flexibility itself, characterizing its unique developmental features. We will examine how its development is modulated by semantic distance between consecutive concepts, a factor intrinsic to the organization of semantic information.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e In line with the journal\u0026rsquo;s guidelines for transparency and openness, this section details the sample size determination, all criteria for data and participant inclusion/exclusion, and all experimental manipulations and measures. Exclusion criteria were established a priori, before the start of data analysis. While the study was not preregistered, we provide full access to the research materials. The anonymized raw data, materials and codes for the experimental task and analyses are publicly available in our project repository on the Open Science Framework (OSF) platform at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/zr4dn/overview\u003c/span\u003e\u003cspan address=\"https://osf.io/zr4dn/overview\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e for reuse by other researchers.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eParticipants\u003c/h2\u003e\u003cp\u003eA total of 655 children participated in the first session of a larger two-session study, which included the experimental paradigms reported here. From these, we then excluded 37 participants (see Data analysis). The final experimental sample for the present study consisted of 618 children (290 females, 328 males, see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) aged between 3 years and 10 months and 10 years and 10 months. Within this sample, 531 children provided valid data for both the executive and semantic tasks, 76 for the executive task only, and 11 for the semantic task only.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e\u003cb\u003eDistribution of the final sample by grade level and gender\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrade\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePreschool\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e116\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1^ grade\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e77\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2^ grade\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e151\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3^ grade\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e127\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4^ grade\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e106\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5^ grade\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e41\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTotal\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e290\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e328\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e618\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNotes\u003c/em\u003e: F, females; M, Males\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eParticipants were recruited from several preschools and primary schools in the Northern part of Italy, within the province of Padova. All children were Italian speakers with typical development, as reported by their parents, no history of neurological, neurodevelopmental, or learning disorders, nor any uncorrected sensory impairments (vision or hearing). Written informed consent was obtained from the parents or legal guardians of all participants, and verbal assent was obtained from each child. The study was conducted in accordance with the ethical standards of the 2013 Declaration of Helsinki for human studies of the World Medical Association and received approval from the Ethical Committee for the Psychological Research of the University of Padova (protocol number: 5272).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eGeneral procedure\u003c/h3\u003e\n\u003cp\u003e Participants performed two versions of a two-choice cued task-switching paradigm: an executive and a semantic version. While targeting different domains, both paradigms were designed with an identical structure. These experimental paradigms were part of a broader battery, designed for a larger study aimed at comparing the developmental trajectories of different executive and semantic control processes. The task-switching paradigms were both administered during the first of two experimental sessions; data not pertinent to the current research question will be presented elsewhere.\u003c/p\u003e\u003cp\u003e To avoid participants by task order interactions, the order of presentation of the two task-switching paradigms was kept fixed, with all participants performing the executive one before the semantic one. Furthermore, the sequence of trials within each paradigm was also fixed, ensuring all children completed the same trial list.\u003c/p\u003e\u003cp\u003eThe experiment was administered using the online version of Labvanced\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, displayed in full-screen mode on tablets with a 1280 x 800 pixel resolution (landscape orientation). Data collection took place in a quiet room within the children\u0026rsquo;s schools. In each session, the battery of tasks was administered collectively to the entire class, with each child working on an individual tablet under the supervision of a team of at least three experimenters. Children responded by tapping on one of two designated response areas on the left or right side of the tablet screen (see details below).\u003c/p\u003e\u003cp\u003eThe procedure for each domain (executive and semantic) followed a structured sequence. Before each task-switching paradigm, participants first completed two single-task blocks (20 trials each). In these blocks, they performed two discrimination tasks in isolation. These blocks were designed to assess baseline visuospatial and semantic abilities of each child. While the results from these single-task blocks are not reported directly, performance metrics derived from them were used as predictors in our main analyses to control for baseline abilities and obtain more precise estimates of the experimental effects of interest. Once the two single-task blocks were completed, the task-switching was administered in a single, continuous block of 48 trials. Specifically, the two simple discrimination tasks were mixed, and participants had to perform the relevant one based on a trial-by-trial visual cue (see below).\u003c/p\u003e\u003cp\u003eTo ensure sustained engagement from our young participants, a gamified approach was adopted throughout the procedure. Each experimental block (executive and semantic single-task and task-switching ones) was preceded by an instructional and training procedure. The lead experimenter (always the same, the author GV) provided standardized instructions to the entire group, adapting the language to the age of the class and using visual aids (slides). To capture the children\u0026rsquo;s attention and enhance compliance and motivation, the instructions were framed within a short, engaging narrative. This was followed by several practical examples where the experimenter collectively guided the children, asking different children to indicate the correct response to ensure the rules were fully understood. After the group instructions, each child completed a brief, individual training phase (4 trials for the single-task blocks, 6 trials for the task-switching blocks) with immediate, visually intuitive on-screen feedback (a happy face for correct answers; a sad face for incorrect/slow responses). During the experimental blocks, motivational images (e.g., a superhero character) were displayed every eight trials to maintain focus and compliance. Finally, both paradigms used child-friendly stimuli, as detailed in the following sections.\u003c/p\u003e\n\u003ch3\u003eExecutive Task-Switching\u003c/h3\u003e\n\u003cp\u003eTo assess executive flexibility, we adapted a classic color/shape task-switching paradigm\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, implementing a gamified version with child-friendly stimuli. The task required children to sort stimuli representing cartoon aliens based on one of two visuo-spatial features: their color (red, RGB: 192, 0, 0; or blue, RGB: 0, 112, 192) or the shape of their head (squared or rounded) (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Across all stimuli, the body and facial elements were held constant. To promote the use of abstract rules and enhance comparability with the semantic paradigm, four unique variations of each stimulus were created. These variations were generated by combining two internal color-fill patterns (zigzag vs. striped) and two head contours (e.g., square vs. rectangle for the squared-head aliens; circular vs. oval for the rounded-head aliens).\u003c/p\u003e\u003cp\u003eThe target stimuli (180 x 255 pixels) were presented at the center of the screen. The experiment was presented on a background depicting a lunar landscape with two extraterrestrial houses, one on the left and one on the right side of the screen, which served as the response areas. Crucially, signs outside each house indicated the correct feature-response mapping. The sign on the left house displayed a red paintbrush and a black square, while the sign on the right house showed a blue paintbrush and a black circle (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The color task therefore required children to sort the alien by color, tapping the left house for red and the right house for blue aliens. The shape task required them to sort it by the shape of its head, tapping the left house for squared-head aliens and the right house for rounded-head aliens. Therefore, the employed stimuli were bivalent, that is, they afforded both a color and a shape judgment, requiring children to flexibly select the relevant stimulus feature based on the currently relevant task, which was signaled by a visual task cue (255 x 106 pixels) presented at the top-center of the screen: an image of two paintbrushes for the color task or two geometric shapes for the shape task (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Moreover, we used only the two stimulus types that were task-incongruent: the red alien with the rounded head, requiring a left response for the color task but a right response for the shape task, and the blue alien with the squared head, requiring a right response for the color task but a left response for the shape task. This was critical to maximize flexibility demands through cross-task interference, while ensuring that correct responses on Switch trials required an actual task-set shift.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe entire procedure was framed within a narrative where children had to help the lost aliens return to the correct house. They were told that to select the correct house first they had to follow a clue (paintbrushes or shapes). Each trial began with the presentation of the lunar landscape for 1000 ms. The task cue then appeared for 750 ms, after which the target alien appeared in the center while the cue remained visible. Both the cue and stimulus stayed on screen until a response was made or for a maximum of 5000 ms. Trials could either be Repeat (same task as the previous trial) or Switch trials (different task) (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe executive task-switching paradigm consisted of a single block of 48 experimental trials, preceded by one buffer trial. Repeat and Switch trials were interspersed and equally probable (50% switch probability). The trial sequence was pseudorandomized to prevent immediate repetitions of the exact same stimulus image and to avoid more than four consecutive repetitions of the same task or response.\u003c/p\u003e\n\u003ch3\u003eSemantic Task-Switching\u003c/h3\u003e\n\u003cp\u003eSemantic flexibility was assessed using a semantic task-switching paradigm adapted for children from a version previously validated with adults\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. While maintaining the same underlying structure as the executive task, this paradigm used stimuli with semantic meaning (instead of visuo-spatial information) to tax flexible control within the semantic domain, requiring participants to flexibly switch between different meaning aspects of a concept\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Children were required to categorize cartoon images based on one of two conceptual dimensions: whether the object depicted was living or non-living, or whether it was capable of moving or non-moving.\u003c/p\u003e\u003cp\u003eThe target stimuli were colorful, child-friendly cartoon images (255 x 255 pixels) presented at the center of the screen (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These images depicted concepts belonging to two categories, chosen to be familiar to children across the entire age range of the sample: Plants (living and non-moving concepts) or Vehicles (non-living and moving concepts). Four exemplars were selected for each category (Plants: tree, sunflower, cactus, potted plant; Vehicles: bicycle, tractor, boat, helicopter. Each image was created ex-novo to be engaging and colorful, and all shared a consistent visual style. As in the executive task-switching paradigm, the selected categories made target stimuli bivalent and task-incongruent (i.e., stimuli that both tasks can be performed on and that required different responses based on the task cue).\u003c/p\u003e\u003cp\u003eThe experiment was presented on a rainbow-themed background with two response areas on the left and right. Each response area displayed images representing the possible conceptual features. The left area showed both a cartoon child (representing living concepts) and a spinning star (representing moving concepts). The right area showed both a cartoon robot (representing non-living concepts) and a stationary star (representing non-moving concepts). Therefore, for the living/non-living task, children had to tap the left area for living images (i.e., Plants) and the right area for not-living images (i.e., Vehicles), while for the moving/non-moving task, they had to tap the left area for moving images (i.e., Vehicles) and the right area for non-moving images (i.e., Plants) (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The task to be performed was signaled by a cue stimulus (300 x 160 pixels) presented at the top-center of the screen. The cue for the living/non-living task was an image showing both the child and the robot side-by-side, while the cue for the moving/non-moving task showed the two stars. The narrative instructed the children to sort the pictures based on the specific cue they saw on each trial.\u003c/p\u003e\u003cp\u003eAll other methodological details, including trial timing and the pseudo-randomization of the 48-trial sequence, were kept identical to the executive task-switching paradigm.\u003c/p\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eData analysis\u003c/h2\u003e\u003cp\u003eStatistical analyses were performed using R (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org/\u003c/span\u003e\u003cspan address=\"https://www.r-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, version 4.4.1\u003csup\u003e32\u003c/sup\u003e) on inverse-transformed RTs (iRTs), computed as 1000/RTs. This transformation was used to mitigate the heavy right skewness common in RT distributions and normalize their residuals. Crucially, iRTs serve as a measure of processing speed or performance, where higher values correspond to better performance (i.e., faster responses, interpreted as responses-per-second).\u003c/p\u003e\u003cp\u003eThirty-seven participants (5.65% of the recruited participants) were excluded from the analyses because they did not complete the two single-task and the two task-switching blocks or showed poor compliance/performance (i.e., failed to provide a response in more than 80% of trials, provided the same response in more than 90% of trials, and/or provided the incorrect response in more than 90% of trials).\u003c/p\u003e\u003cp\u003ePractice trials and the first trial of each block were excluded from analyses. We then excluded experimental trials with incorrect responses (n\u0026thinsp;=\u0026thinsp;10472, 18.98% of trials), missed responses (n\u0026thinsp;=\u0026thinsp;3058, 5.54% of trials) and anticipations (i.e., trials with RTs\u0026thinsp;\u0026lt;\u0026thinsp;150 ms, n\u0026thinsp;=\u0026thinsp;79, 0.14% of trials), which were both considered as errors, as well as trials with extreme iRTs in the preceding trial (i.e., those with an absolute z-scored iRT\u0026thinsp;\u0026gt;\u0026thinsp;3; n\u0026thinsp;=\u0026thinsp;195, 0.035% of trials).\u003c/p\u003e\u003cp\u003eTo address both our aims, we used linear mixed-effects model (LMM) analyses\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eAim 1 - Executive vs Semantic Flexibility\u003c/h2\u003e\u003cp\u003eFirst, to compare the effects of executive and semantic flexibility on behavioral performance and their interplay across development, we performed a within-subject analysis testing an a-priori defined model (M_EvsS_iRT) on the performance data (iRT) from both the executive and semantic paradigms. The model's Wilkinson-notation formula was: \u003cem\u003eiRT\u0026thinsp;~\u0026thinsp;iRTpre\u0026thinsp;+\u0026thinsp;Trial*Age\u0026thinsp;+\u0026thinsp;postERR*Age\u0026thinsp;+\u0026thinsp;Dom*Cond*Age\u0026thinsp;+\u0026thinsp;Dom*Cond*PerfBase + (1|SS) + (1|Stim)\u003c/em\u003e.\u003c/p\u003e\u003cp\u003e\u003cem\u003eControl variables.\u003c/em\u003e The fixed part of the model included several confounding predictors: i) a continuous predictor for the iRT of the preceding trial (iRTpre) to account for temporal dependency in iRTs\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e; ii) a continuous predictor for trial number (Trial) to account for time-on-task effects (e.g., learning and/or fatigue); iii) a dichotomous predictor for an error in the preceding trial (postERR) to account for post-error slowing\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. The latter two predictors were tested in interaction with Age to capture any age-related modulation of these effects.\u003c/p\u003e\u003cp\u003e\u003cem\u003eFixed effects.\u003c/em\u003e To test how executive-semantic flexibility interplay modulated the switch cost across development, we included predictors for Domain (Dom: Executive vs. Semantic), Condition (Cond: Repeat vs. Switch) and the continuous predictor for Age, modeling them in a three-way interaction (Dom*Cond*Age). To control for how the Dom*Cond interaction was modulated by children's baseline abilities, we also included the three-way interaction between Dom, Cond and baseline performance (PerfBase), where PerfBase is a continuous predictor representing each participant's mean iRT in executive and semantic domain when no flexibility was required (i.e., in single-task blocks).\u003c/p\u003e\u003cp\u003e\u003cem\u003eRandom effects.\u003c/em\u003e The random effect structure included a by-participant (SS) random intercept to account for individual differences in overall performance, and a by-stimulus (Stim) random intercept to account for variability attributable to specific stimuli.\u003c/p\u003e\u003cp\u003e\u003cem\u003eVariable coding.\u003c/em\u003e To facilitate model convergence and interpretation, the continuous confounding predictors (Trial and iRTpre) were centered and scaled (from \u0026minus;\u0026thinsp;.5 to .5) and z-scored, respectively, at the participant level. The categorical predictor postErr was factor-coded (0\u0026thinsp;=\u0026thinsp;after correct, 1\u0026thinsp;=\u0026thinsp;after error), with the 0 acting as the reference level. Age was z-scored across the entire sample and PerfBase by-domain. The predictors of interest, Dom and Cond, were treated as continuous numerical variables and centered around zero (-0.5 for Executive/Repeat, 0.5 for Semantic/Switch), so that the estimated main effects and interactions refer to the average across conditions.\u003c/p\u003e\u003cp\u003e\u003cem\u003eOutliers.\u003c/em\u003e Finally, after fitting the initial model, we examined the residuals to check for the presence of extreme outliers and re-fitted a trimmed version, excluding data points with absolute standardized residuals greater than 3.\u003c/p\u003e\u003cp\u003eFor the trimmed model, we reported the estimated coefficients (\u003cem\u003eb\u003c/em\u003e), standard error (\u003cem\u003eSE\u003c/em\u003e), \u003cem\u003et\u003c/em\u003e and \u003cem\u003ep\u003c/em\u003e values for each fixed effect. Satterthwaite's approximation of degrees of freedom was used to calculate \u003cem\u003ep\u003c/em\u003e-values and to derive effect size estimates (\u003cem\u003ed\u003c/em\u003e\u003csub\u003eS\u003c/sub\u003e) for the experimental effects of interest, which were also calculated using the Westfall\u0026rsquo;s approach (\u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e)\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. We additionally calculated the conditional \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e values using the MuMIn package\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. An alpha level of .05 was set as the cut-off for statistical significance.\u003c/p\u003e\u003cp\u003e\u003cem\u003eSwitch cost correlation.\u003c/em\u003e Then, to directly compare executive and semantic flexibility effects on performance and further assess their commonalities, we extracted the individual effects of interest (executive and semantic switch costs) as estimated by the LMM model. Covariates other than Age were set to their mean value or reference level, while Age was kept at the child\u0026rsquo;s observed value. We then computed the Kendall\u0026rsquo;s \u003cem\u003eτ\u003c/em\u003e correlation coefficient between the by-participant executive and semantic switch costs using the package pcor.test\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eAim 2 - Semantic Flexibility\u003c/h3\u003e\n\u003cp\u003eOur second aim was to characterize the development of semantic flexibility by examining how it is modulated by the organization of semantic information. To this end, we conducted a focused analysis on iRT during the semantic paradigm alone. We tested an a priori defined model (M_Sem_iRT) based on our theoretical assumptions and previous findings\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. The model's Wilkinson-notation formula was: \u003cem\u003eiRT\u0026thinsp;~\u0026thinsp;iRTpre\u0026thinsp;+\u0026thinsp;Trial*Age\u0026thinsp;+\u0026thinsp;postERR*Age\u0026thinsp;+\u0026thinsp;Cond*Age\u0026thinsp;+\u0026thinsp;Cond*PerfBase\u0026thinsp;+\u0026thinsp;Age*SemDist + (1|SS) + (1|Stim)\u003c/em\u003e.\u003c/p\u003e\u003cp\u003e\u003cem\u003eControl variables.\u003c/em\u003e As in the previous model (M_EvsS_iRT), we included confounding predictors (iRTpre, Trial*Age, postERR*Age) and two key interactions: Cond*Age, to test how age modulates the semantic switch cost, and Cond*PerfBase, to assess the role of baseline abilities.\u003c/p\u003e\u003cp\u003e\u003cem\u003eFixed Effects.\u003c/em\u003e Crucially, building on our previous evidence showing that the representational distance between concepts influences performance\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, we also included a continuous predictor for semantic distance (SemDist). SemDist reflects the cosine distance between the concepts of the current and previous stimulus (trial n vs. n\u0026ndash;1), derived from fastText, a distributional semantic model trained on Italian corpora\u003csup\u003e\u003cspan additionalcitationids=\"CR42\" citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Our previous results revealed that when stimuli change on every trial (like in our paradigm), participants are required to flexibly navigate through the semantic space. When stimuli are semantically more distant from each other, they require greater flexibility \u0026ndash; a mechanism we termed stimulus-level flexibility because it operates regardless of the task condition. Therefore, we tested SemDist in interaction with Age to explore whether this stimulus-level flexibility changes across development.\u003c/p\u003e\u003cp\u003eHowever, in addition to this stimulus-level effect, semantic distance might also modulate the degree of the type of flexibility required to switch between tasks, reflected in the switch cost. Therefore, to test whether SemDist modulates the semantic switch cost and whether this modulation changes with age, we specified a second, more complex model (M_SemFull_iRT), which added a three-way interaction (Cond*Age*SemDist). The formula was: \u003cem\u003eiRT\u0026thinsp;~\u0026thinsp;iRTpre\u0026thinsp;+\u0026thinsp;Trial*Age\u0026thinsp;+\u0026thinsp;postERR*Age\u0026thinsp;+\u0026thinsp;Cond*Age\u0026thinsp;+\u0026thinsp;Cond*PerfBase\u0026thinsp;+\u0026thinsp;Cond*Age*SemDist + (1|SS) + (1|Stim)\u003c/em\u003e. We then formally compared the fit of M_Sem_iRT and M_SemFull_iRT using a log-likelihood ratio test to determine if the inclusion of this critical interaction was statistically justified. For this analysis, the SemDist predictor was centered, and the rest of the model fitting procedure was identical to that described above.\u003c/p\u003e\n\u003ch3\u003eSensitivity Analysis on Accuracy\u003c/h3\u003e\n\u003cp\u003eTo control for possible speed-accuracy trade-offs, the analyses detailed above were also performed on accuracy as the dependent variable, but in this case using a generalized LMM (with the binomial family) and considering the mean accuracy in the two single-task blocks for the PerfBase predictor. Therefore, the M_EvsS_Acc model compared the effects of executive and semantic flexibility on children\u0026rsquo;s accuracy and their interplay across development, while the M_SemFull_Acc model assessed SemDist modulations of the semantic flexibility. Note that in these analyses, Satterthwaite's approach could not be used to compute the effect sizes because this method is not applicable to generalized mixed models; therefore, effect sizes were only computed following Westfall\u0026rsquo;s (2014) approach\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003ePower Analysis\u003c/h2\u003e\u003cp\u003eThe sample size was determined based on the aims of the larger study that included, among others, the two task-switching paradigms described here. We recruited as many participants as possible based on the availability of resources and access to local schools. Nonetheless, we performed a sensitivity power analyses using the method introduced by Westfall and colleagues\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e for a fully-crossed linear mixed-effects model, conservatively assuming participant and stimulus intercepts and residual variance partitioning coefficients of .2, .1, and .7, respectively\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. The variance partitioning coefficients for the participant and stimulus slopes and the participant-by-stimulus intercept were set to 0 because those effects were not included in the models we tested. This power analysis revealed that a sample size of 618 participants and 16 stimuli was large enough to detect a very small effect size (Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.036) with a power of .80. It should be noted, however, that this approach is not adequate for complex mixed effect models like the one used in this work\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e, but it nonetheless provides a useful estimation of the so-called minimal statistically detectable effect for our study (i.e., the lower bound of the range of effect sizes that can be detected\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e). Indeed, to the best of our knowledge, to date there are no accepted analytical approaches to accurately compute statistical power for such models.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eAim 1 - Executive vs. Semantic Flexibility\u003c/h2\u003e\u003cp\u003eThe first analysis compared the behavioral effects of executive and semantic flexibility and explored their interplay across development. The conditional \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e of the trimmed LMM model (M_EvsS_iRT) provided a good fit to the data (Conditional \u003cem\u003eR\u003c/em\u003e\u0026sup2; = .33), with 0.65% of observations removed as extreme outliers (\u0026gt;\u0026thinsp;3 absolute standardized residuals).\u003c/p\u003e\u003cp\u003eAll confounding predictors modulated participants\u0026rsquo; performance (iRT) significantly (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e for full model details).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of the LMM model assessing the domain-dependent switch effects on iRTs (M_EvsS_iRT)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003et\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cem\u003ed\u003c/em\u003e\u003csub\u003eS\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Intercept)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.5910\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0058\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e102.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e10.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e2.50\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eiRT preceding trial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0509\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40510\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e46.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0118\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10020\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0327\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e852\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e6.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePost-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0415\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-14.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0127\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0070\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0898\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0660\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0023\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e25440\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-28.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.28\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBaseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0411\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0031\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12030\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e13.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial:Age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0118\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0037\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40460\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-3.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Post-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0406\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e41040\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-14.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0350\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e25100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e7.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Domain\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e41070\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0004\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0129\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0029\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40470\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-4.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0394\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e41020\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-12.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0028\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40440\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-4.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Domain:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0089\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0058\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40440\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-1.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.1265\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.04\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Condition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0056\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40440\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0429\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eNotes\u003c/strong\u003e\u003cp\u003eSE, standard error; df, degrees of freedom computed with the Satterthwaite\u0026rsquo;s approximation; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eS\u003c/sub\u003e, effect size estimates calculated with Satterthwaite's approach; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e, effect size estimates calculated with Westfall\u0026rsquo;s approach.\u003c/p\u003e\u003c/p\u003e\u003cp\u003eThe analyses revealed various significant effects. Specifically, performance (iRT) improved significantly with Age and with higher baseline performance. Crucially, there was a main effect of Condition, confirming the presence of a significant switch cost, with better performance on Repeat trials compared to Switch trials.\u003c/p\u003e\u003cp\u003eThese main effects were qualified by several significant interactions. First, we observed two interactions involving age. The Age*Domain interaction was significant, with an age-dependent greater performance enhancement, especially in the semantic domain (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The Age*Condition interaction was also significant, indicating that performance improved with age more steeply for Repeat trials than for Switch trials (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eMoreover, we found a significant interaction between Domain and the baseline performance (PerfBase), showing that the greater the baseline performance the greater the performance in task-switching, especially in the executive domain. The Condition*PerfBase interaction was also significant, showing that the performance benefit of having higher baseline abilities was more pronounced for Repeat trials than for Switch trials.\u003c/p\u003e\u003cp\u003eAlso relevant to our research question was the Domain*Condition interaction, which was significant, revealing that the magnitude of the switch cost was significantly larger in the executive domain compared to the semantic domain (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFinally, the three-way Age*Domain*Condition interaction was not significant, whereas the \u003cem\u003ep\u003c/em\u003e-values for the Domain*Condition*PerfBase interaction was just below .05 (and it is not significant in the analysis on accuracy, see below), thus it will not be discussed further.\u003c/p\u003e\u003cp\u003eTo further probe the relationship between the two forms of flexibility, we correlated the individual switch cost effects estimated by the model. Despite the difference in magnitude, the executive and semantic switch costs were strongly and positively correlated (Kendall's \u003cem\u003eτ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.817. \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001; see Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe analysis on the children\u0026rsquo;s accuracy substantially confirmed the results reported above, except for the Dom*Age and Dom*PerfBase interactions, whose statistical significance was not confirmed (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of the LMM model assessing the domain-dependent switch effects on accuracy (M_EvsS_Acc)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ez\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ed\u003csub\u003eW\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Intercept)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.4475\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0362\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e40.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e3.34\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eiRT preceding trial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0319\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0114\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-2.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.0050\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.07\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0095\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0435\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.8281\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.6157\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0302\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e20.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.42\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePost-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.4922\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0259\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-19.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-1.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.2955\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0468\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-6.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.68\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.4960\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0240\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-20.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-1.15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBaseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0949\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0170\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial:Age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0506\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0362\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.1624\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Post-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.1857\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0234\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-7.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.43\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.3339\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0475\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.77\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Domain\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0208\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0256\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.4162\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.1634\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0234\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-6.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.38\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0385\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0295\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.1918\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.09\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0416\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0209\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.0472\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.10\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Domain:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0238\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0465\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.6091\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.06\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDomain:Condition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0091\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0419\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.8285\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003eNotes\u003c/em\u003e: SE, standard error; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e, effect size estimates calculated with Westfall\u0026rsquo;s approach.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eAim 2 - Semantic Flexibility\u003c/h2\u003e\u003cp\u003eOur second analysis aimed to characterize the development of semantic flexibility. As the log-likelihood ratio test confirmed that the model including the three-way interaction (M_SemFull_iRT) provided a significantly better fit to the data, we report the results from this more complex model (see Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFirst, the analysis confirmed a significant effect of Condition, indicating a robust switch cost within the semantic domain. We also found a significant effect of Semantic Distance, revealing that performance worsened as the semantic distance between consecutive stimuli increased.\u003c/p\u003e\u003cp\u003eCrucially, we found a significant three-way Age*Condition*Semantic Distance interaction. This interaction revealed that the performance worsening associated with greater semantic distance was more pronounced on Repeat trials, and that this effect became stronger with age (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). In older children especially, this resulted in a notable cost for repeating a task when they had to switch towards more semantically distant stimuli.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of the LMM model assessing semantic switch effect on iRTs (M_SemFull_iRT)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003edf\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ep\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003ed\u003csub\u003eS\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003ed\u003csub\u003eW\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Intercept)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.6192\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0081\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e76.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e12.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e2.71\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eiRT preceding trial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0325\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0017\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18570\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e19.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0402\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0060\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5441\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e6.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0079\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e924\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-0.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.8051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePost-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0344\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0040\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18700\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-8.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0602\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0081\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12360\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-7.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.26\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBaseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0856\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0066\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e510\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e13.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.37\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSemantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0784\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0173\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12090\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-4.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.34\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial:Age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0082\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0055\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18330\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-1.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.1350\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.04\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Post error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0342\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18810\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-8.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0459\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0086\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18330\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-5.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.20\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0056\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18310\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-1.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.1544\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0824\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0339\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e13200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0151\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.36\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0174\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0173\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18320\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e-1.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.3149\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.08\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.1346\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0344\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e18330\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.59\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003cem\u003eNotes\u003c/em\u003e: SE, standard error; df, degrees of freedom computed with the Satterthwaite\u0026rsquo;s approximation; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eS\u003c/sub\u003e, effect size estimates calculated with Satterthwaite's approach; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e, effect size estimates calculated with Westfall\u0026rsquo;s approach.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe analysis on the children\u0026rsquo;s accuracy confirmed the results reported above, except for the main effect of SemDist (see Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of the LMM model assessing semantic switch effect on accuracy (M_SemFull_Acc)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ez\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ed\u003csub\u003eW\u003c/sub\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Intercept)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.3769\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0598\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e23.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.92\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eiRT preceding trial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0013\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0172\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.9391\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.0669\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0618\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.2793\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.09\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.5874\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0547\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePost-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.3967\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0368\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-10.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.55\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.4775\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0831\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-5.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.67\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBaseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.3134\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0370\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e8.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.44\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSemantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.2366\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.1765\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.1800\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.33\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTrial:Age\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0804\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0539\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.1359\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.11\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Post-error\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.1785\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0344\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-5.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.25\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.4177\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0794\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-5.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.0001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.58\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Baseline Performance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.0413\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.0301\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.1698\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.06\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCondition:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.7035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.3477\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.0431\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.98\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.1509\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.1666\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.3650\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e-0.21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge:Condition:Semantic Distance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.2505\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.3309\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e.0002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.74\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e\u003cem\u003eNotes\u003c/em\u003e: SE, standard error; \u003cem\u003ed\u003c/em\u003e\u003csub\u003eW\u003c/sub\u003e, effect size estimates calculated with Westfall\u0026rsquo;s approach.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study investigated the developmental trajectories of executive and semantic flexibility in children using parallel versions of a cued task-switching paradigm to address a fundamental question raised in the adult cognitive control literature: whether these two forms of flexibility rely on distinct or shared mechanisms. By charting these abilities throughout childhood, we sought to determine if they follow distinct or shared developmental trajectories. Our findings reveal a complex picture, characterized by evidence consistent with a strong shared component that is nonetheless modulated by domain-specific factors. These results provide a crucial developmental perspective on the neurocognitive models of executive and semantic control; they also highlight the intricate nature of cognitive development itself.\u003c/p\u003e\u003cp\u003eOur first key finding demonstrates a strong correlation between switch costs in the executive and semantic domains, consistent with the idea of shared control mechanisms. This result is consistent with findings in healthy adults\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, which show a similar moderate-to-strong correlation. Finding this strong association in children suggests the link between these abilities is a fundamental aspect of development, not just a mature state. This aligns with theoretical models which argue that a core set of executive functions drive flexible behavior\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. The mental act of switching tasks also taxes working memory to maintain task sets and apply the relevant one, and inhibitory control to stop applying the old one. In essence, both our paradigms relied on this same underlying cognitive machinery for switching. Therefore, a child's developmental level in operating this machinery would naturally lead to similar performance across both domains, explaining the strong correlation.\u003c/p\u003e\u003cp\u003eWhile the correlation suggests a shared component, our findings are consistent with the asynchronous and multi-faceted nature of cognitive control development highlighted in our introduction. The significant Age by Condition interaction revealed that performance on Repeat trials improved more steeply with age than on Switch trials. This pattern points to a growing functional dissociation between more automatized task execution and effortful cognitive control. As children age, they become increasingly efficient at maintaining and executing an active task-set, leading to dramatic performance gains on Repeat trials. However, this emerging task automatization introduces a greater degree of task inertia. A more consolidated cognitive system, while more efficient, naturally offers more resistance to flexibility. Consequently, the process of switching to a newly relevant task-set necessarily requires a costly control intervention to override this emerging automaticity and implement a new goal-directed action. While this flexibility ability also matures, its developmental gains observed on Switch trials are intrinsically less pronounced than those afforded by the powerful process of task automatization on Repeat trials. Our results, therefore, suggest that a key developmental achievement in this age range is the ability to form and execute efficient cognitive routines, which in turn makes the relative cost of volitionally breaking those routines \u0026ndash; the switch cost \u0026ndash; more apparent.\u003c/p\u003e\u003cp\u003eOur results of a significant Domain by Condition interaction also challenge a purely domain-general account by revealing domain-specific modulations. The larger switch cost in the executive domain than in the semantic domain is consistent both with findings with adults\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e and developmental research showing that switching between highly interdependent perceptual dimensions (like color and shape) is particularly difficult for young children\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. As discussed in our introduction, prior studies have shown children successfully switching between perceptual rules (shape) and semantic ones (function)\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Our finding extends this, suggesting that a switch between two competing perceptual rules may be uniquely demanding.\u003c/p\u003e\u003cp\u003eFurthermore, the steeper age-related performance improvement in the semantic task, evidenced by the Domain by Age interaction, strongly supports the idea that development is shaped by the richness of the knowledge base upon which control operates. Crucially, this developmental effect was driven entirely by processing speed, not accuracy. This speed-accuracy dissociation is highly informative: it suggests that the conceptual knowledge required for the semantic task \u0026ndash;whether stimuli were 'living' or 'moving'\u0026ndash; was already firmly in place even for our youngest participants. Therefore, the observed developmental gains are not attributable to new knowledge acquisition but rather reflect the maturing efficiency with which this established knowledge is accessed, organized, and managed. This aligns perfectly with the notion that as children\u0026rsquo;s conceptual knowledge becomes more organized, the task is \"scaffolded\", reducing the cognitive load on control processes and boosting their operational speed\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e, as well as with a developmental shift toward more automatic semantic processing, which would free up control resources with age\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Therefore, our findings reveal a key developmental distinction: while the perceptual processing remains a significant bottleneck for efficiency throughout childhood, the conceptual processing benefits greatly from the ongoing organization and automatization of the underlying semantic knowledge base.\u003c/p\u003e\u003cp\u003eOur most novel contribution is the finding of a three-way interaction between Age, Condition, and Semantic Distance, that speaks directly to the CSC framework\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e and our second aim of characterizing semantic flexibility itself. The finding that older children showed a performance cost for larger semantic distance on Repeat trials, but not on Switch trials, is consistent with the interplay between the semantic \"hub\" (the knowledge store where conceptual distance is represented) and the \"control\" system\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. This suggests that as children grow older, their semantic representations become more structured and, thus, increasingly similar to the adults\u0026rsquo; one\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. As such, even when repeating the same task, traversing this space requires a form of flexibility to navigate through the stimuli in the semantic space (see stimulus-level flexibility)\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. The masking of this effect on switch trials is what the CSC model would predict: the resource-intensive process of reconfiguring the entire control network leaves fewer resources for finer-grained, representation-level modulations. This echoes the broader theme that developmental trajectories are often not linear but are marked by increasing complexity and dynamics. Of course, this interpretation must be tempered by the possibility of a statistical floor effect. Performance on switch trials is already under such high load that our behavioral measures may lack the sensitivity to detect a more subtle, additional cost from semantic distance. Yet, this finding is consistent with the idea raised by research on early conceptual development\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e: understanding the development of flexibility requires understanding the development of the knowledge it controls\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe potential confounds discussed above represent the most significant limitations of our study. We must explicitly state that while our model controlled for baseline performance, the larger executive switch cost could stem from differences in switch-specific task difficulty or rule type. Future work must aim to equate the interference demands between paradigms to isolate the specific effects of information content. Furthermore, our cross-sectional design only allows for inferences about group-level trends, not the individual trajectories that require longitudinal study.\u003c/p\u003e\u003cp\u003eFinally, while our behavioral data are informative, their greatest value may be in generating specific, testable hypotheses for future neuroimaging research. The strong behavioral correlation between switch costs is consistent with the hypothesis of a shared neural basis, which should be mirrored by correlated activation in the domain-general multiple-demand network that supports a wide range of effortful cognitive tasks\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Correspondingly, the unique variance in semantic task performance, particularly the modulation by semantic distance, is consistent with the possibility of a distinct neural basis uniquely explained by activity in semantic-specific control regions, such as the left inferior frontal gyrus and posterior middle temporal gyrus, consistent with the CSC framework\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. However, while our data point towards a shared foundation, we must be cautious. This correlation could also be driven by the maturation of a more general ability to consciously represent and apply paired-rule structure, regardless of their content\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. A child who has mastered this fundamental logical operation will succeed at both tasks, while a child who has not will struggle, creating a strong correlation driven by general rule-use ability rather than a specific switch mechanism. Therefore, while our data point towards a shared foundation, its precise nature, a specific flexibility module, a set of core executive functions, or a general rule-use capacity, remains ambiguous. Therefore, the precise nature of this shared resource remains a key question for future research.\u003c/p\u003e\u003cp\u003eIn conclusion, by employing the cued, trial-by-trial methodology, this study offers the first direct comparison of the developmental trajectories of executive and semantic flexibility. Our findings are consistent with a model where a shared cognitive foundation underpins both abilities, as observed in adult findings\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, but where performance may be shaped by distinct developmental timelines for its subcomponents and modulated by domain-specific factors. These findings highlight that a complete understanding of adaptive behavior requires focusing on the dynamic interplay between maturing general control processes and specialized knowledge systems.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests:\u003c/h2\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e\u003cp\u003eThis study was supported by the European Union (HORIZON-MSCA-2023-PF-01-01, CTRL-ALT-DEV, Grant 101150190 to MM and ERC-2021-STG, IN-MIND, Grant 101043216 to SB-V). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them.\u003c/p\u003e\u003ch2\u003eAcknowledgements:\u003c/h2\u003e\u003cp\u003eWe thank Marina Mancuso and Chiara Maria Migliorin for their assistance with data collection. We are also deeply grateful to the children, parents, and teachers who participated in this study for their invaluable collaboration.\u003c/p\u003e\u003ch2\u003eData Availability Statement:\u003c/h2\u003e\u003cp\u003eThe datasets generated and analyzed during the current study are available in the OSF repository, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://osf.io/zr4dn/overview\u003c/span\u003e\u003cspan address=\"https://osf.io/zr4dn/overview\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\n\u003ch2\u003eAUTHOR CONTRIBUTIONS STATEMENT\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eGiada Viviani:\u0026nbsp;\u003c/strong\u003eData Curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing - Original Draft, Writing - Review \u0026amp; Editing.\u003cstrong\u003e\u0026nbsp;Ettore Ambrosini:\u0026nbsp;\u003c/strong\u003eData Curation, Formal analysis, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing - Original Draft, Writing - Review \u0026amp; Editing. \u003cstrong\u003eAnnamaria Porru:\u0026nbsp;\u003c/strong\u003eResources, Writing - Review \u0026amp; Editing.\u003cstrong\u003e\u0026nbsp;Silvia Benavides-Varela\u003c/strong\u003e:\u0026nbsp;Funding acquisition, Writing - Review \u0026amp; Editing.\u003cstrong\u003e\u0026nbsp;Erin M. Buchanan:\u0026nbsp;\u003c/strong\u003eWriting - Review \u0026amp; Editing.\u003cstrong\u003e\u0026nbsp;Irene Di Pietro:\u0026nbsp;\u003c/strong\u003eInvestigation, Writing - Review \u0026amp; Editing. \u003cstrong\u003eDaniela Lucangeli:\u0026nbsp;\u003c/strong\u003eWriting - Review \u0026amp; Editing. \u003cstrong\u003eMaria Montefinese:\u0026nbsp;\u003c/strong\u003eConceptualization, Data Curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing - Original Draft, Writing - Review \u0026amp; Editing.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eChiew KS, Braver TS (2017) Context Processing and Cognitive Control: From Gating Models to Dual Mechanisms. in \u003cem\u003eThe Wiley Handbook of Cognitive Control\u003c/em\u003e (ed. Egner, T.) 143\u0026ndash;166Wiley. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/9781118920497.ch9\u003c/span\u003e\u003cspan address=\"10.1002/9781118920497.ch9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMonsell S (2003) Task-set reconfiguration processes do not imply a control homunuculus: Reply to Altmann. Trends Cogn Sci 7:341\u0026ndash;342\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eViviani G, Visalli A, Finos L, Vallesi A, Ambrosini E (2023) A comparison between different variants of the spatial Stroop task: The influence of analytic flexibility on Stroop effect estimates and reliability. Behav Res 56:934\u0026ndash;951\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEgner T (2023) Principles of cognitive control over task focus and task switching. Nat Rev Psychol 2:702\u0026ndash;714\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSulpizio S, Job R, Leoni P, Scaltritti M (2022) Prepotent task-irrelevant semantic information is dampened by domain-specific control mechanisms during visual word recognition. Q J Experimental Psychol 75:390\u0026ndash;405\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJefferies E (2013) The neural basis of semantic cognition: Converging evidence from neuropsychology, neuroimaging and TMS. Cortex 49:611\u0026ndash;625\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMontefinese M, Hallam G, Thompson HE, Jefferies E (2020) The interplay between control processes and feature relevance: Evidence from dual-task methodology. Q J Experimental Psychol 73:384\u0026ndash;395\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLambon Ralph MA, Jefferies E, Patterson K, Rogers TT (2017) The neural and computational bases of semantic cognition. Nat Rev Neurosci 18:42\u0026ndash;55\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eReilly J et al (2024) What we mean when we say semantic: Toward a multidisciplinary semantic glossary. Psychon Bull Rev. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13423-024-02556-7\u003c/span\u003e\u003cspan address=\"10.3758/s13423-024-02556-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBadre D, Poldrack RA, Par\u0026eacute;-Blagoev EJ, Insler RZ, Wagner AD (2005) Dissociable Controlled Retrieval and Generalized Selection Mechanisms in Ventrolateral Prefrontal Cortex. Neuron 47:907\u0026ndash;918\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDavey J et al (2016) Exploring the role of the posterior middle temporal gyrus in semantic cognition: Integration of anterior temporal lobe with executive processes. NeuroImage 137:165\u0026ndash;177\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDuncan J (2010) The multiple-demand (MD) system of the primate brain: mental programs for intelligent behaviour. Trends Cogn Sci 14:172\u0026ndash;179\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eThompson HE et al (2018) The contribution of executive control to semantic cognition: Convergent evidence from semantic aphasia and executive dysfunction. J Neuropsychol 12:312\u0026ndash;340\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlmaghyuli A, Thompson H, Lambon Ralph MA, Jefferies E (2012) Deficits of semantic control produce absent or reverse frequency effects in comprehension: Evidence from neuropsychology and dual task methodology. Neuropsychologia 50:1968\u0026ndash;1979\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMontefinese M, Hallam G, Stampacchia S, Thompson HE, Jefferies E (2021) Deficits of semantic control disproportionately affect low-relevance conceptual features: evidence from semantic aphasia. Aphasiology 35:1448\u0026ndash;1462\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eViviani G et al (2025) Exploring semantic and executive flexibility interplay in task switching. Sci Rep 15:31609\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRuffini C, Traverso L, Marzocchi GM, Pecini C (2025) Enhancing preschoolers\u0026rsquo; cognitive flexibility and pre-literacy skills through dialogic reading: evidence from an Italian teacher-led intervention. Early Child Dev Care 195:847\u0026ndash;868\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eButtelmann F, Karbach J (2017) Development and Plasticity of Cognitive Flexibility in Early and Middle Childhood. Front Psychol 8:1040\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCrone EA, Donohue SE, Honomichl R, Wendelken C, Bunge SA (2006) Brain Regions Mediating Flexible Rule Use during Development. J Neurosci 26:11239\u0026ndash;11247\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUddin LQ (2021) Cognitive and behavioural flexibility: neural mechanisms and clinical considerations. Nat Rev Neurosci 22:167\u0026ndash;179\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchettini E, Hiersche KJ, Saygin ZM (2023) Individual Variability in Performance Reflects Selectivity of the Multiple Demand Network among Children and Adults. J Neurosci 43:1940\u0026ndash;1951\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeak GO (2000) The Growth of Flexible Problem Solving: Preschool Children Use Changing Verbal Cues to Infer Multiple Word Meanings. J Cognition Dev 1:157\u0026ndash;191\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDe\u0026aacute;k GO, Ray SD, Pick AD (2004) Effects of age, reminders, and task difficulty on young children\u0026rsquo;s rule-switching flexibility. Cogn Dev 19:385\u0026ndash;400\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChami S, Charalambous C, Knijnik SR, Docking K (2022) Language and executive function skills as predictors of semantic fluency performance in pre-school children. Int J Speech Lang Pathol 24:626\u0026ndash;635\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRuffini C, Osmani F, Bigozzi L, Pecini C (2024) Semantic fluency in 3\u0026ndash;6 years old preschoolers: which executive functions? Child Neuropsychol 30:563\u0026ndash;581\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMemisevic H, Biscevic I, Pasalic A (2017) Developmental trends in semantic fluency in preschool children. Cogent Psychol 4:1403064\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKav\u0026eacute; G, Kigel S, Kochva R (2008) Switching and clustering in verbal fluency tasks throughout childhood. J Clin Exp Neuropsychol 30:349\u0026ndash;359\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBetz N, Coley JD (2020) Development of Conceptual Flexibility in Intuitive Biology: Effects of Environment and Experience. Front Psychol 11:537672\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFinger H, Goeke C, Diekamp D, Standvo\u0026szlig; K, K\u0026ouml;nig P (2017) LabVanced: A unified JavaScript framework for online studies. in 1\u0026ndash;3 (University of Osnabr\u0026uuml;ck\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e\u003cem\u003eHandbook of Human Multitasking\u003c/em\u003e. (Springer International Publishing, Cham, (2022) \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-3-031-04760-2\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-04760-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJackson RL (2021) The neural correlates of semantic control revisited. NeuroImage 224:117444\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eR Core Team (2021) A language and environment for statistical computing. R Foundation for Statistical Computing\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaayen RH, Davidson DJ, Bates DM (2008) Mixed-effects modeling with crossed random effects for subjects and items. J Mem Lang 59:390\u0026ndash;412\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaayen RH, Milin P (2010) Analyzing reaction times. Int j psychol res 3:12\u0026ndash;28\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRabbitt PM (1966) Errors and error correction in choice-response tasks. J Exp Psychol 71:264\u0026ndash;272\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWestfall J, Kenny DA, Judd CM (2014) Statistical power and optimal design in experiments in which samples of participants respond to samples of stimuli. J Exp Psychol Gen 143:2020\u0026ndash;2045\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBartoń (2025) k. MuMIn: Multi-Model Inference\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim S (2015) ppcor: An R Package for a Fast Calculation to Semi-partial Correlation Coefficients. \u003cem\u003eCSAM\u003c/em\u003e 22, 665\u0026ndash;674\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBuchanan EM et al (2025) Measuring the semantic priming effect across many languages. Nat Hum Behav. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41562-025-02254-x\u003c/span\u003e\u003cspan address=\"10.1038/s41562-025-02254-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMontefinese M, Zannino GD, Ambrosini E (2015) Semantic similarity between old and new items produces false alarms in recognition memory. Psychol Res 79:785\u0026ndash;794\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJoulin A et al (2016) FastText.zip: Compressing text classification models. Preprint at. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/ARXIV.1612.03651\u003c/span\u003e\u003cspan address=\"10.48550/ARXIV.1612.03651\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBojanowski P, Grave E, Joulin A, Mikolov T (2016) Enriching Word Vectors with Subword Information. Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/ARXIV.1607.04606\u003c/span\u003e\u003cspan address=\"10.48550/ARXIV.1607.04606\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrave E, Bojanowski P, Gupta P, Joulin A, Mikolov T (2018) Learning Word Vectors for 157 Languages. Preprint at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/ARXIV.1802.06893\u003c/span\u003e\u003cspan address=\"10.48550/ARXIV.1802.06893\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrysbaert M, Stevens M (2018) Power Analysis and Effect Size in Mixed Effects Models: A Tutorial. J Cognition 1:9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLakens D, Equivalence Tests (2017) A Practical Primer for \u003cem\u003et\u003c/em\u003e Tests, Correlations, and Meta-Analyses. Social Psychol Personality Sci 8:355\u0026ndash;362\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiamond A (2013) Executive Functions. Annu Rev Psychol 64:135\u0026ndash;168\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiyake A et al (2000) The Unity and Diversity of Executive Functions and Their Contributions to Complex Frontal Lobe Tasks: A Latent Variable Analysis. Cogn Psychol 41:49\u0026ndash;100\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZelazo PD (2015) Executive function: Reflection, iterative reprocessing, complexity, and the developing brain. Dev Rev 38:55\u0026ndash;68\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Padua","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Cognitive Control, Semantic Flexibility, Task-Switching Paradigms, Executive Flexibility, Controlled Semantic Cognition","lastPublishedDoi":"10.21203/rs.3.rs-7965304/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7965304/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCognitive control supports adaptive behavior through stability and flexibility, with executive flexibility typically assessed through task-switching paradigms. However, while executive flexibility is a well-studied construct, it is unclear whether it relies on the same mechanisms as semantic flexibility \u0026ndash; the ability to switch between meanings based on context. An ideal approach to arbitrate this debate is to compare their developmental trajectories, a method hampered by the fact that semantic flexibility\u0026rsquo;s development remains largely uncharted. Here, 4- to 10-year-old children performed parallel task-switching paradigms: a classic visuospatial paradigm assessing executive flexibility and a novel semantic task assessing semantic flexibility by requiring them to alternate between semantic judgments of meaningful concepts. Results revealed a strong correlation between executive and semantic switch costs, suggesting shared control mechanisms, alongside domain-specific differences and age-related modulations influenced by semantic distance, revealing a growing interplay between semantic knowledge and control as children's conceptual systems mature. These findings provide novel insights into the maturation of cognitive control components in childhood, highlighting the interplay between domain-general executive processes and semantic control mechanisms in flexible cognition.\u003c/p\u003e","manuscriptTitle":"Developmental Trajectories of Executive and Semantic Flexibility Using Task-Switching","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-29 07:36:40","doi":"10.21203/rs.3.rs-7965304/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dc47391c-ae19-4c6f-86b4-8baab207aa4f","owner":[],"postedDate":"October 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-29T07:36:40+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-29 07:36:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7965304","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7965304","identity":"rs-7965304","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}