Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Neuropsychology. 2024 Nov;38(8):727–739. doi: 10.1037/neu0000966

Sociodemographic predictors and cross-cultural comparisons in tests performance from the Cambridge Neuropsychological Testing Automated Battery (CANTAB) among children aged 6–8 years from Montevideo, Uruguay

Danelly Rodríguez 1, Elena I Queirolo 2, Katarzyna Kordas 1, Daniel Costa-Ball 2, Gabriel Barg 2
PMCID: PMC11897995  NIHMSID: NIHMS2057713  PMID: 39480351

Abstract

Introduction:

Cross-culturally comparative data on measures of executive function (EF) are essential, but the 6–8-year group remains insufficiently described. This study examined the sociodemographic predictors of EF test performance employing the Cambridge Neuropsychological Testing Automated Battery (CANTAB). It also compared developmental trends in EF among children from Uruguay, the US, and Mexico.

Methods:

EFs were assessed with the Intra-dimensional/Extra-dimensional shift (IED), Spatial Span (SSP), and Stockings of Cambridge (SOC) tests from the CANTAB. The study sample consisted of 6–8-year-old children from the Salud Ambiental Montevideo (SAM) cohort in Uruguay. Differences between cohorts were examined, and we performed generalized linear regressions to assess the association between sociodemographic factors, and each EF domain.

Results:

The final sample consisted of 525 participants (mean age in months 82.5 ± 6.0). Across all ages, SAM children had significantly lower performance in the SSP and SOC tasks compared to US and Mexican children. On the IED task, SAM children had similar scores to US and Mexican children. Mother’s IQ (β=0.01 [95% CI: 0.005,0.02]), child’s IQ (0.02 [0.02,0.03]), the HOME total score (0.02 [0.01,0.03]), as well as HOME subscales of accompaniment (0.13[0.07,0.20]), enrichment (0.11 [0.06,0.16]), and physical environment (0.07 [0.03,0.10]) were positively associated with the span length (SSP task). Child’s IQ (0.02 [0.01,0.03]) was positively associated with the number of problems solved on the SOC test.

Conclusion:

Uruguayan children perform lower in working memory and planning tests than US children but similarly to Mexican children, while cognitive flexibility is consistent across all groups. Further, mother and child IQ, as well as the home environment, are important predictors of EF. These differences should be examined in the context of diverse cultural values and sociodemographic factors affecting CANTAB construct validity in this population.

Keywords: sociodemographic predictors, cognition, executive function, child health, home environment

Public significance statement

This research includes the most extensive sample to date on the Cambridge Neuropsychological Testing Automated Battery (CANTAB) performance on children aged 6–8 years, a critical period in the maturation of executive functions (EF), incorporating the study of the association with sociodemographic/contextual factors (motheŕs characteristics, home stimulation and neighborhood context), and comparing this results with other populations (US and Mexico). Our results confirm the utility of the battery as an assessment tool and provide insights into the developmental trajectory of EF in the 6–8 period, but also underscore the importance of considering contextual and cultural factors when studying these cognitive processes.

1. Introduction

Executive functions (EF) are a group of cognitive control processes, mainly underpinned by the prefrontal cortex, enabling goal-directed behaviors. These processes orchestrate lower-level functions, such as perception and motor responses, which manipulate cognitive representations simultaneously and organize action sequences in steps (Fuster, 2013; Snyder et al., 2015). There exists a broad consensus that three core EF domains play a pivotal role in self-regulating behavior: cognitive flexibility, working memory, and the inhibition of prepotent responses to stimuli (Miyake et al., 2000). These three EF domains provide a foundation upon which more advanced functions, like strategic planning and reasoning, are constructed (Diamond, 2016). Considering they are crucial to regulating complex behavior, EFs have been a central focus of neuropsychological assessments.

Comprehensive normative and cross-culturally comparative data on executive functioning, encompassing various stages of child neurodevelopment, is essential. Luciana and Nelson (2002) have described the developmental trends of working memory, planning, and cognitive flexibility in children from 4 to 12 years old, observing that each EF has a different trajectory. Working memory develops steadily during this period, cognitive flexibility matures by age 7, resulting in a ceiling effect in performance in the following years, and strategic planning only begins developing at 11–12 years of age. From a broader perspective, De Luca et al. (2003) identified three developmental spurts at the ages of 8, 15–19, and 20–29 years old.

Despite the significance of longitudinal studies in understanding EF development, specific developmental stages, including the early school years, remain insufficiently described. The 6–8year developmental period is critical for the rapid development of working memory and cognitive flexibility (P. Anderson, 2002), supported by the sustained growth of white matter and metabolic activity in the prefrontal cortex (V. Anderson, 2001; Sowell et al., 2002). Children’s performance on validated measures of EF in this period needs to be clarified to address the scarcity and limited sample sizes in prior studies encompassing this specific stage. No previous study has included more than 100 participants aged 6–8 years per year, except Green et al. (2019). More research is needed to understand the developmental changes during these critical years.

Another issue related to the study of EF developmental trends is the normative samples employed to standardize the assessment tools. One assessment tool, the CANTAB (Cambridge Neuropsychological Test Automated Battery), consists of cognitive tests, designed to assess various aspects of cognitive function, including memory, attention, EF, and visuospatial abilities. These tests are administered using computerized tasks, providing standardized and objective measures of cognitive performance (Robbins et al., 1994). Although raw scores from the battery tests can be used in clinical settings or intra-subject research designs, appropriate normative data is necessary to compare results between different samples. Sociocultural factors could differentially affect childreńs cognitive development in different countries (Feldman & Eidelman, 2009). For the CANTAB, initial normative data were obtained from children living in the US (Luciana & Nelson, 2002), Europe (Lehto et al., 2003), and Australia (De Luca et al., 2003). Comparing the results of children from different countries and sociocultural contexts with the normative standards derived in the US, Europe, or Australia can serve as a reference. However, it may not be useful to characterize the developmental trajectory of their EF. Different cultures have diverse values regarding the skills and abilities children should acquire and the typical ages at which these abilities should manifest. These cultural differences influence expectations regarding performance, which are further shaped by demographic factors such as access to resources, education, gender, and ethnicity. Consequently, when evaluating the validity of a measurement, it is essential to consider culturally specific patterns (Fernald et al., 2009). These differences could also affect the psychometric properties of the battery, including the factorial structure, reliability, and internal consistency (Syväoja et al., 2015).

The interpretation of EF performance based on local norms can contribute to more comprehensive theoretical models of EFs and clarify the influence of socio-cultural factors on neurodevelopment. Efforts have been made to study children’s performance on CANTAB in different societies. For instance, Teixeira et al. (2011) administered the Spatial Span test (SSP), a measure of visual working memory, to 60 Brazilian participants aged 6 to 18 years. Toornstra et al. (2020) evaluated the short-term recognition memory using the Delayed Matching to Sample (DMS) with 184 participants aged 5 to 14 years from Ukraine. Employing a battery approach, Green et al. (2019) evaluated planning (Stockings of Cambridge- SOC), short-term memory (DMS), sustained attention (Rapid Visual Information Processing), the ability to match visual stimuli (Match to Sample Search), cognitive flexibility (Intra-Extra Dimensional Set Shift-IED) and response inhibition (Stop Signal Task) in 826 Mexican participants aged 5–15 years. These studies found different patterns of influence by age, sex, child IQ, and maternal IQ, highlighting the importance of establishing local databases. Nevertheless, other sociodemographic variables with known effects on neurodevelopment, such as the home context, maternal characteristics (age, employment, depressive symptoms), neighborhood variables, or cognitive stimulation (Everson-Rose et al., 2003; Feldman & Eidelman, 2009) were not included in previous studies of children.

The use of the CANTAB battery has improved the assessment of EF through the standardized administration of tasks, precise performance data recording, and the generation of comparable results across the lifespan. However, more availability of normative data that encompass cultural and geographical differences is needed, particularly at stages where rapid neural and cognitive EF development occurs, such as the 6–8 years period. Therefore, the objectives of the present study were to describe the developmental trends in CANTAB performance of 525 children aged 6–8 years from Montevideo, Uruguay, on working memory (SSP test), cognitive flexibility (IED), and planning (SOC), and to compare their performance with available data on children from the US and Mexico.

This comparative analysis can provide valuable insights into potential cross-cultural variations in EF development, as well as potential influences of several sociodemographic characteristics, including those unique to this study: motheŕs characteristics, home stimulation, and neighborhood context. To our knowledge, this research includes the most extensive sample to date on CANTAB performance in children aged 6–8 years that incorporates these sociodemographic factors.

2. Methods

2.1. Participants and recruitment

The present study derives data from the Salud Ambiental Montevideo (SAM) cohort. The initial purpose of the SAM was to examine metal exposures, including arsenic, cadmium, and manganese, among school-age children in Montevideo, Uruguay. (Barg et al., 2018; Desai et al., 2018; Desai et al., 2020; Frndak et al., 2019a; Frndak et al., 2022; Kordas et al., 2015; Kordas et al., 2018; Rodríguez et al., 2023). We followed the recruitment criteria described by Kordas et al. (2018) and Queirolo et al. (2024). In brief, recruitment consisted of advertising via posters, radio, television, newspaper announcements, and informational meetings for parents in private elementary schools in city neighborhoods where lead exposure was reported or suspected. Altogether, participants were recruited from 50 private and public elementary schools that agreed to participate. All first-grade children who regularly attended the participating schools were eligible for the study. A total of 940 first-grade children aged 6–8 years old and their primary caregivers enrolled in the study.

According to the National Administration of Public Education in Uruguay, educational institutions in Montevideo are categorized according to the socioeconomic status (SES) of the students and families they cater to, with levels ranging from 1 (high SES) to 5 (very low SES) (Calvo, Macadar, & Nathan, 2013). The study aimed to reach schools specifically from levels 3 to 5, although most of the schools were situated within the mid-range of SES. Drawing from prior experience with this demographic, the study anticipated that schools with high SES would be less inclined to participate in a study addressing challenges affecting disadvantaged children. Therefore, the inclusion criterion was attendance in the first grade at the participant schools. The sole exclusion criterion was a previous diagnosis of lead poisoning, defined as blood lead level >45 μg/dL; according to public health guidelines at the time, this level would have necessitated medical intervention. None of the children were excluded based on this criterion.

2.2. Procedures

Children and parents completed self-administered questionnaires, neuropsychological tests, and biological sampling of urine and blood. Qualified psychologists conducted the cognitive assessments in a designated room at the school (years 2009–13) or at the study center (2015–19). The children underwent two neuropsychological testing sessions at the school but only one longer session in the study center to avoid more than one study visit by the family. The cognitive evaluations encompassed measures of EF as well as IQ. Mothers participated in a separate evaluation with trained psychologists to assess their IQ and the presence of depressive symptoms. Finally, a home visit was arranged. Study personnel went to the family home to conduct a direct observation of the home environment. At the end of the study, participating families were provided with food vouchers for use at local grocery stores as compensation for their time. Research ethics boards approved the study protocols at the Pennsylvania State University, the Catholic University of Uruguay, and the University at Buffalo.

2.3. Sociodemographic Measures

The caregivers, most often mothers, provided information on demographics related to themselves, their children, and the household setting. Children’s characteristics encompassed gender, age in months, medical history, household density (the total number of individuals in the home divided by the total number of bedrooms), the type of preschool attended (private, public, or no attendance), and the mother’s age in years.

2.3.1. Home Observation for Measurement of the Environment (HOME) Inventory

A trained social worker assessed the developmental environment of the household during the home visit using the Spanish version of the Home Observation for Measurement of the Environment (HOME) Inventory (Bradley et al., 2003). The HOME Inventory is comprised of 59 items organized into eight subscales: (1) Emotional and Verbal Responsivity of Caregiver; (2) Encouragement of Maturity; (3) Emotional Climate of the Home; (4) Growth-Fostering Materials and Experiences; (5) Provision for Active Stimulation; (6) Family Participation in Developmentally Stimulating Experiences; (7) Parental Involvement with Child; and (8) Aspects of the Physical Environment. The total HOME score is a global measure of all eight scales, with higher scores reflecting greater household enrichment. The internal consistency of the HOME inventory score is 0.90, and subscales range from 0.53–0.83 for 6–10-year-old children (Sugland et al., 1995).

2.3.2. Television and Computer Usage

A questionnaire was administered to assess television watching and computer use. Caregivers were asked to report the usual number of hours their children spent watching television per day, and responses were recorded as: 0–1 hours, 1–2 hours, 2–4 hours, and 4 or more hours. Caregivers also reported on the number of hours their child spent on the computer doing activities that were not games per day, responding as: Never, 0–1 hours, 1–2 hours, 2–4 hours, 4 or more hours. Due to low response rates in some categories, we recoded the above options. The new categories for TV watching were: 0–1 hours, 1–2 hours, and 2 or more hours, while the new categories for computer use were: Never, 0–1 hours, and 1 or more hours.

2.3.3. Neighborhood Disadvantage Factor

As described previously (Frndak et al., 2021), a neighborhood disadvantage (ND) index was created via factor analysis using the demographic attributes of all census sections throughout the city of Montevideo. The ND factor demonstrates strong construct validity, as it correlates with maternal education, maternal age, and the HOME inventory score. Each participant was allocated to an ND level based on the specific census segment where they lived at the time of study enrollment. The total population of the census segment was assigned to each participant according to the location of their residence.

2.4. Maternal Depression

Maternal depression in this study was evaluated using the Argentine version of the Beck Depression Inventory II (BDI II) (Richaud & Sacchi, 2001). This assessment tool is designed to measure depressive symptoms in adults, aligning with the criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders IV (DSM IV). The BDI II is widely recognized and frequently employed for assessing depressive symptoms within various populations. In a factorial study of the Argentine sample, factors were similar to those found by others with English versions, indicating strong factorial validity (Richaud & Sacchi, 2001). Because of strong cultural and linguistic similarities, it is considered acceptable to extend the use of the Argentine adaptation of the BDI to Uruguayan adults. We have used it successfully to assess depressive symptoms in Uruguayan low-to-middle-income women (Ardoino et al., 2015).

The instrument consists of 21 statements assessing components of depression, including: sadness, pessimism, failure, feelings of guilt, self-criticism, and suicidal thoughts. The women were asked to think about their feelings in the previous two weeks and choose one of four phrases that best described them during that time. The answers to all the questions were summed to obtain a total BDI score, which is used to determine whether the women have mild (scores 14–19 points), moderate (20–28), or severe (29–63) depressive symptoms.

2.5. Maternal IQ

The study utilized the Wechsler Adult Intelligence Scale III (WAIS III) to evaluate the intellectual abilities of mothers. The WAIS III is considered the benchmark assessment of intelligence in adults. Its underlying structure has been confirmed across various cultures (Shuttleworth-Edwards et al., 2004) and age groups (Taub et al., 2004). Furthermore, it has been validated for Spanish-speaking populations (Bausela Herreras, 2007; Rodriguez et al., 2008) and is commonly employed for research and clinical assessments in Uruguay. Following the criteria adopted with the BDI, the Argentinian version (Brenlla et al., 2002) of the WAIS-III was administered.

In this research, five subtests from the WAIS III were administered: similarities, arithmetic, vocabulary, block design, and object assembly. These subtest scores were then used to calculate an estimated IQ, following the established methodology published previously. (López et al., 2003) To ensure accuracy, two psychologists independently scored the performance of each mother on the test, thereby minimizing measurement errors. Any discrepancies in scoring were resolved through discussions involving the test administrator and the coordinating psychologist. It is worth noting that this abbreviated version of the test has been used before in Uruguayan women with low-to-middle incomes (Ardoino et al., 2015).

2.6. Child intelligence assessment

During the years 2009–13, child intelligence was assessed trough the Woodcock–Muñoz (WM) test battery, a Spanish adaptation of the Woodcock–Johnson battery (Muñoz-Sandoval et al., 2009; Woodcock et al., 2003), commonly employed in clinical settings to evaluate cognitive impairment (Ledbetter et al., 2017; Scantlebury et al., 2016; Taub & McGrew, 2004). Within the cognitive battery, seven tests were administered, encompassing verbal comprehension, visual–auditory learning, spatial relations, sound blending, concept formation, visual matching, and reversed numbers. Each subtest generates a standardized score known as the W score, which considers age and gender, as described previously. (Frndak et al., 2019b). A composite score was calculated to determine general intellectual ability (GIA), which weighed the results from all seven tests. The median internal consistency reliability across all age groups for these tests is 0.87. (Schrank & McGrew, 2001).

In the years 2015–19, intelligence was assessed through an abbreviated procedure based on selected subtests of the Weschler Intelligence Scale for Children (WISC-IV). An estimation of the total IQ is calculated from the working memory and processing speed indexes, summing the results of the Digit Span, Symbol Search, Letter-Number Sequencing, and Coding subtests (Sattler, 2008). This procedure is recommended for research purposes, and has been standardized and validated for Latino populations (Weschler, 2011). Since Weschler and Woodcock-Johnson scores are highly correlated (Chiodi & S.Weschler, 2009; McGrew, 1983), the WM battery offers an equivalent score to the IQ that can be used to integrate the results of both batteries (Muñoz-Sandoval et al., 2009).

2.7. Cambridge Neuropsychological Test Automated Battery (CANTAB)

The CANTAB is a computer-based neuropsychological assessment designed to measure brain-behavior relations associated with EF supported by the frontal lobe (Robbins et al., 1998). It comprises 22 tests that assess 3 cognitive domains (executive function, attention and psychomotor speed, and memory) and 2 additional tests to measure emotional aspects. The use of CANTAB in neuropsychological research allows a highly standardized administration, automated response recording with millisecond precision, and a uniform scoring procedure (Luciana, 2003). As a result, it facilitates comparability of results across diverse participant samples. It was originally proposed to study cognitive deficits in elderly individuals (Robbins et al., 1994). However, it was rapidly applied in research in neuropsychiatric disorders (Fray et al., 1996) and in other health problems, like neurotoxicology (Fray & Robbins, 1996), ADHD (Fried et al., 2015), autism (Ozonoff et al., 2004), Asperger (Kaufmann et al., 2013), Tourette (Rasmussen et al., 2009), fetal alcohol disorder (Green et al., 2009), exposure to environmental contaminants (Desai et al., 2020; Frndak et al., 2019; Rodríguez et al., 2023) among others, in pediatric and adolescent population.

The CANTAB battery was administered on a touchscreen device. The study researcher provided instructions for the CANTAB tests to the children in Spanish. It is worth noting that because the CANTAB tests used in this study involve nonverbal, visual, and auditory stimuli, there was no need to translate the test into another language. Three CANTAB tests were employed to evaluate executive functions: SOC, IED, and SSP. In adult populations, the IED showed a high level of reliability (ICC = 0.79), while the SOC and SSP tasks exhibited moderate reliability (ICC = 0.60 and 0.68, respectively). (Lowe & Rabbitt, 1998). So far, one study has attempted the test-retest reliability of the CANTAB in school-age children. Specifically, a study among school-aged children in Finland reported moderate to good stability for IED and SSP and low for SOC (Syväoja et al., 2015). In another study involving children and adolescents aged 11–17 with ADHD, intraclass correlations (ICC) for the IED ranged from 0.78 to 1.0 (Gau & Shang, 2010).

The SOC assesses the ability to plan spatially. In this task, children were presented with colored balls placed in three “stockings.” One set of stockings was positioned at the bottom of the screen, and the second at the top. The stocking at the top represented the example arrangement of balls that the child had to recreate at the bottom of the screen by placing balls in their chosen positions, one at a time. Importantly, they could not move the balls at the bottom of their stocking until the ball on top had been moved, often to a temporary position in an adjacent stocking. Four different problems were presented, and as the trials progressed, the number of moves and the complexity of the problems increased. Spatial planning ability was evaluated as the number of problems solved correctly with the minimum number of moves.

The IED evaluates cognitive flexibility, specifically the ability to learn and adapt to changing rules. It involves two dimensions: pink shapes and white lines. Children were instructed to choose the dimension they believed to be correct, and implicit rules governed their choices, which they learned through their responses. The computer provided feedback through sound to indicate whether their choices were correct. This task consisted of nine stages. In stages 1 and 2 of the IED, children encountered pink shapes. After a few trials, the rule for selecting the correct pink shape changed. Once children learned the new rule, they proceeded with additional trials. In stages 3, 4, and 5, white lines overlapped the shapes. Stages 6 and 7 presented new shapes and new patterns of white lines, representing an intra-dimensional shift. In stages 8 and 9, white lines became the correct response selection, marking an extra-dimensional shift.

The outcomes of interest in the IED task included the number of completed stages, the number of trials, the number of errors, and errors before and after the extra-dimensional shift. Scoring children’s performance involved an adjustment method described in the CANTAB Administration Guide. Specifically, 25 errors were counted for each failed stage, as children who advanced to more stages could make more errors. Using this scoring approach, five children failed the IED test before reaching stage 4, and two children received zero errors after failing to complete stage 9. The values were replaced with missing errors before and after the extra-dimensional shift for these seven children.

The SSP assesses short-term visual memory, where 10 white squares are randomly displayed on a computer screen, changing color one by one. Children were tasked with touching the squares while remembering the sequence of color changes. As the test progressed, the number of squares increased from 2 to 9, representing working memory as span length. Other outcomes included errors, the number of attempts, and response time. In this study, 27 children (8%) were unable to replicate the pattern during the 2-square trial after three attempts and were assigned a maximum span length of 1.

2.8. Statistical Analyses

Of the 940 children who were recruited and eligible, 264 were excluded from our analysis due to missing baseline enrollment characteristics or missing cognitive measures. Therefore, 696 children were fully enrolled and had complete EF measures. Of these children, 171 did not have data on sociodemographic variables, which resulted in an analytical sample size of 525.

2.8.1. Comparison of analytical vs. excluded sample

To assess potential selection bias, we compared participant characteristics between the analytical sample (n=525) and the children excluded (n=415) due to either not completing minimum enrollment assessments, missing sociodemographic data, or missing EF measures from the study. Summary statistics for sociodemographic and cognitive measures included means (standard deviations) or median (ranges), depending on the distribution. We employed chi-square, independent t-tests, or Wilcoxon tests (for non-normal variables) to assess whether these noted differences were statistically significant.

2.8.2. Developmental trends of the CANTAB

To analyze the developmental performance of EFs within our sample as well as comparisons with US and Mexican children, a new age variable was created where months were converted to years: 6, 7 and 8. We compared EF performance across 6-, 7-, and 8-year-old children. A one-way analysis of variance (ANOVA) or Kruskal-Wallis (for non-normal variables) was conducted to test EF performance differences across the three age groups.

2.8.3. Comparison of the CANTAB with other population groups

To test whether the scores of the US and Mexican normative sample are equivalent to EF performance of the SAM cohort, Welch’s t-tests were conducted utilizing the CANTAB normative scores published for 6–8-year-old children in the US (Luciana & Nelson, 2002) and 6–8-year-old children in Mexico (Green et al., 2019). Only subtests that were the same across all studies were used: SSP span length, SOC problems solved in minimum moves, and IED stages completed.

2.8.4. Associations between sociodemographic factors and the CANTAB

To address our objective of analyzing sociodemographic predictors of EFs, generalized linear models were performed to predict each EF endpoint and commands were selected to match outcome distributions (i.e., normal or Poisson). In this study, the following variables were considered: age (in months), sex (coded as 0 for male and 1 for female), household density (calculated as the total number of persons in the home divided by the total number of bedrooms), ND factor, type of preschool (dummy coded with private, public, and did not attend as the referent category), mother’s age (in years), maternal depression, maternal IQ, child’s IQ, computer use per day (dummy coded with 0–1 hours, 1+ hours, and never as the referent category), TV watching per day (dummy coded with 1–2 hours, 2+ hours, and 0–1 hours as the referent category), HOME total score, and HOME subscales including receptivity, emotional climate, maturity, learning materials, accompaniment, enrichment, integration, and physical environment.

The three assumptions (normality, constant variance, linearity) of linear regressions were evaluated. Normality was investigated by visual inspections of quantile-quantile (Q-Q) plots. Constant variance was assessed using scatter plots of residuals, and linearity was examined using scatter XY plots. We employed Poisson regressions to model the relationship between count response variables (IED endpoints including stages completed, total trials, total errors, pre dimensional errors, and post dimensional errors) and sociodemographic predictor variables. Overdispersion was evaluated by comparing the deviance to its degrees of freedom. We alleviated overdispersion issues by performing negative binomial regressions for all IED endpoints. The Holm-Bonferroni correction method was used for regression analysis to adjust for familywise error rate inflation. (Holm, 1979). All statistical analyses were performed in SAS version 9.4 (SAS Institute Inc., Cary NC, USA) with a type I error rate of 0.05.

2.8.5. Sensitivity analysis: Multiple imputations to account for missing data

To assess our original fully enrolled sample of 696 children with complete cognitive data, we imputed missing sociodemographic data for 171 children. We conducted multiple imputation (MI) of all HOME total and subscale scores, as well as sociodemographic predictors (except for variables not missing at random: maternal depression). Imputation was implemented in SAS across 50 cycles. Using the imputed data, we reran generalized linear models and negative binomial regressions to test the association between sociodemographic predictors and EF domains.

3. Results

3.1. Comparison of analytical vs. excluded sample

Table 1 shows sample characteristics, including sociodemographic factors and EF performance scores, for children with complete data compared to children excluded from the analysis. The sample with complete data had higher mean scores on maternal IQ (81.2 vs 77.5, p=.0002) and child IQ (90.2 vs. 87.0, p=.0018). No other differences were noted.

Table 1.

Comparison of sociodemographic characteristics and CANTAB endpoints between SAM children ages 6–8 years old with complete relevant data (included) and those with incomplete relevant data (excluded).

Variables N Complete data (n=525) N Excluded (n=415)
Sex
 %Female 261 49.7 182 43.9
 %Male 264 50.3 209 50.4
Age (months) 525 82.5 ± 6.0 379 82.3 ± 6.8
ND factor 525 1.4 ± 1.1 366 1.4 ± 1.2
Home density 525 2.2 ± 1.1 320 2.1 ± 0.9
HOME Inventory score 525 40.1 ± 10.2 206 41.6 ± 10.6
Type of Preschool
 %Private 195 37.2 128 30.8
 %Public 314 59.9 190 45.8
 %Did not attend 15 2.9 8 1.9
Maternal age (years) 525 33.8 ± 6.96 335 33.4 ± 7.0
Maternal IQ* 525 81.2 ± 12.9 225 77.5 ± 11.9
Child IQ* 525 90.2 ± 14.1 359 87 ± 16.0
Mother’s Depression 525 10.3 ± 8.9 228 10.3 ± 10.6
Mother employed
 %Yes 308 58.7 152 36.6
 %No 217 41.3 87 21.0
Hours spent watching TV during the day (hours)
 %0–1 64 12.2 48 11.6
 %1–2 213 40.6 113 27.2
 %2+ 248 47.2 148 35.7
Computer use (hours per day)
 %Never 332 63.2 197 47.5
 %0–1 117 22.3 60 14.5
 %1+ 76 14.5 48 11.6
SSP, span length 525 3.6 ± 1.2 362 3.5 ± 1.2
SSP, total errors 525 10.8 ± 4.4 362 10.6 ± 4.5
SOC, problems solved 525 5.3 ± 1.8 359 5.0 ± 2.1
IED, total trials 525 146 (91, 155) 366 139 (85, 155)
IED, stages completed 525 8 (7,9) 366 8 (7,9)
IED, total errors 525 53 (21,60) 366 47 (18,60)
IED, pre dimensional shift errors 525 7 (5,9) 366 7 (5,10)
IED, post dimensional shift errors 525 20 (5,27) 366 17 (5,26)
Open in a new tab

Categorical variables are presented as N and %, continuous variables presented as Mean ± SD, continuous variables, continuous variables that were not normally distributed presented as Median (Q1, Q3); Abbreviations: HOME: Home Observation Measurement of the Environment, SSP: Spatial Span, SOC: Stockings of Cambridge, IED: Intra-Extra Dimensional Set Shift.

*

p=<0.05

3.2. Developmental trends of the CANTAB

Table 2 presents the distributions of EF performance in children with complete data (n=525) by ages 6, 7, and 8. Analysis of Variance (ANOVA) and Kruskal-Wallis tests were utilized to assess differences by age. We found significant differences by age for SSP total errors. 8-year-old children had fewer SSP total errors and higher post dimensional errors compared to 6- and 7-year-old children. We performed the post-hoc Tukey Honestly Significant Difference (HSD) to indicate which groups differed from others. Children ages 6 and 7 did not differ on SSP total errors (Diff=0.74, 95%CI=−0.39 to 1.87, p=0.27), but there were significant differences between age 6 vs age 8 (Diff=−5.68, 95%CI=−7.31 to −4.04, p=<.0001) as well as between age 7 vs age 8 (Diff=−6.42, 95%CI=−7.83 to −5.00, p=<.0001). No other differences were noted.

Table 2.

Comparison of CANTAB outcomes by ages 6, 7, and 8 in SAM children with complete data (N =525)

Age 6 (n= 109) Age 7 (n=353) Age 8 (n=63)
SSP, span length 3.41 ± 0.95 3.62 ± 1.21 3.44 ± 1.12
SSP, total errors* 10.37 ± 3.81 11.11 ± 4.69 9.95 ± 3.62
SOC, problems solved 5.19 ± 1.75 5.26 ± 1.81 5.30 ± 2.0
IED, total trials 146 (94,159) 145 (90,154) 147 (103,155)
IED, stages completed 7.76 ± 1.48 7.85 ± 1.27 7.57 ± 1.52
IED, total errors 53 (22,61) 51 (19,59) 54 (19,61)
IED, pre dimensional shift errors 7 (5,10) 7 (5,9) 7 (5,8)
IED, post dimensional shift errors 17 (6,28) 20 (5,27) 25 (8,27)
Open in a new tab

Variables presented as Mean ± SD, continuous variables, continuous variables that were not normally distributed presented as Median (Q1, Q3); Abbreviations: HOME: Home Observation Measurement of the Environment, SSP: Spatial Span, SOC: Stockings of Cambridge, IED: Intra-Extra Dimensional Set Shift.

ANOVA utilized for normally distributed continuous measures.

Kruskal-Wallis test utilized for skewed continuous measures

*

p<0.05

3.3. Comparison of the CANTAB with other population groups

Table 3 presents EF performance of SSP, SOC, and IED among U.S, Mexican, and SAM children. Across all ages, SAM children performed lower in the SSP and SOC tasks compared to U.S. and Mexican children except for comparable performances in the SOC task to Mexican children. On the IED task, SAM children had similar scores to U.S. and Mexican children, with small differences detected at ages 7 and 8.

Table 3.

Comparisons of CANTAB mean performance in SAM children vs. Mexican and U.S children ages 6–8.

SAM US MEX US vs. SAM MEX vs. SAM
N M SD N M SD N M SD t t
Age 6
SSP 109 3.41 0.95 61 3.84 0.82 - - - 3.09* -
SOC 109 5.19 1.75 56 5.80 1.53 21 5.19 1.57 2.30* 0.00
IED 109 7.76 1.48 54 7.70 2.38 21 7.43 0.81 0.17 0.33
Age 7
SSP 353 3.62 1.21 59 4.73 1.17 - - - 6.71* -
SOC 353 5.26 1.81 50 6.16 1.74 186 5.92 1.58 3.41* 4.38*
IED 353 7.85 1.27 57 7.84 1.98 185 7.64 0.87 0.04 2.26*
Age 8
SSP 63 3.44 1.12 58 4.97 1.12 - - - 7.50* -
SOC 63 5.30 2.0 49 6.37 1.54 84 6.23 1.83 3.20* 2.89*
IED 63 7.57 1.52 61 7.95 2.02 84 7.68 1.1 1.18* 0.49
Open in a new tab

Note: Mexican children among the “age 6” group were 5–6 years old.

Abbreviations: HOME: Home Observation Measurement of the Environment, SSP: Spatial Span, SOC: Stockings of Cambridge, IED: Intra-Extra Dimensional Set Shift.

*

p=<0.05

3.4. Associations between sociodemographic factors and the CANTAB

Table 4 presents the associations between each sociodemographic factor and EF endpoints, applying the Holm-Bonferroni correction for multiple comparisons. Compared to children who reported no time engaged in non-game computer activities, those dedicating 0–1 hours per day exhibited longer span length (on the other hand, computer use for longer than 1 hour was not associated with span length). Child’s IQ was positively associated with the number of problems solved in minimum moves (SOC test). Mother’s IQ, Child’s IQ, and the HOME total score were positively associated with the span length (SSP task). Furthermore, we examined associations of the HOME subscales (Supplemental Table 2) with each EF endpoint. The subscales, accompaniment (β=0.13; 95% CI=0.07–0.20) enrichment (β=0.11; 95% CI=0.07–0.20), and physical environment (β=0.07; 95% CI=0.03, 0.10) revealed positive associations with span length (SSP task). No statistically significant associations existed between other sociodemographic factors, HOME subscales, and EF endpoints.

Table 4.

Associations for sociodemographic predictors with domains of executive functions measured by the CANTAB (n=525)

SSP, span length1 SOC, problems solved1 IED, stages2 IED, total trials adj2 IED, totals errors adj2 IED, predimensional shift errors2 IED, post dimensional shift errors2
Age (months) 0.01 [−0.005, 0.03] 0.001 [−0.02, 0.02] −0.002 [−0.006, 0.003] 0.0018 [0.0007, 0.0031] 0.0048 [−0.0039, 0.01] −0.002 [−0.01, 0.006] 0.007 [0.005, 0.01]
Sex
Male (ref) N=217 - - - - - - -
Female N=308 −0.10 [−0.30, 0.09] 0.48 [0.17, 0.79] −0.02 [−0.08, 0.03] 0.04 [−0.02, 0.09] 0.08 [−0.02, 0.19] −0.004 [−0.10, 0.09] −0.004 [−0.14, 0.13]
ND factor −0.11 [−0.19, −0.02] −0.02 [−0.16, 0.11] 0.004 [−0.02, 0.03] −0.0032 [−0.03, 0.02] −0.01 [−0.06, 0.04] −0.004 [−0.05, 0.04] −0.03 [−0.09, 0.03]
Preschool
Did not attend (ref) N=16 - - - - - - -
Public N=314 −0.01 [−0.19, −0.02] 0.15 [−0.79, 1.09] 0.009 [−0.17,0.19] −0.06 [−0.24, 0.12] −0.06 [−0.39, 0.26] −0.002 [−0.30, 0.29] −0.17 [−0.59, 0.24]
Private N=195 0.25 [−0.35, 0.86] 0.08 [−0.87, 1.04] 0.007 [−0.18, 0.19] −0.07 [−0.24, 0.11] −0.08 [−0.41, 0.25] −0.10 [−0.41, 0.19] −0.07 [−0.50, 0.34]
Mother’s age (years) 0.002 [−0.01, 0.01] 0.006 [−0.02, 0.02] −0.0016 [−0.007, 0.003] −0.003 [0.007, 0.002] −0.004 [−0.01, 0.004] −0.0061 [−0.01, 0.001] 0.007 [−0.004, 0.01]
Mother’s Depression −0.01 [−0.02, 0.0004] −0.02 [−0.04, −0.005] −0.0015 [−0.005, 0.002] 0.0040 [0.10, 0.13] 0.006 [−0.002, 0.01] 0.007 [0.001, 0.01] 0.0006 [−0.007, 0.009]
Mother’s IQ 0.01* [0.005, 0.02] 0.003 [−0.008, 0.015] 0.0006 [−0.001, 0.003] −0.0016 [−0.004, 0.0007] −0.003 [−0.006, 0.001] −0.004 [−0.007, −0.0003] 0.0013 [−0.004, 0.006]
Child’s IQ 0.02* [0.02, 0.03] 0.02* [0.01, 0.03] 0.0013 [−0.0008, 0.004] −0.0027 [−0.005, −0.0007] −0.0048 [−0.008, −0.001] −0.005 [−0.008, −0.001] −0.0007 [−0.005, 0.004]
Mother employed
No (ref) N= 217 - - - - - - -
Yes N= 308 0.27 [0.08, 0.47] 0.30 [−0.01, 0.61] 0.02 [−0.04, 0.07] −0.04 [−0.10, 0.01] −0.07 [−0.018, 0.03] −0.03 [−0.12, 0.07] −0.09 [−0.23, 0.04]
TV watching (per day)
0–1 hours (ref) N=64 - - - - - - -
1–2 hours N=213 0.16 [−0.15, 0.49] 0.61 [0.10, 1.12] 0.06 [−0.04,0.16] −0.10 [−0.20, −0.01] −0.19 [−0.36,−0.01] −0.22 [−0.38, −0.06] −0.007 [−0.23,0.22]
2+ hours N=248 0.13 [−0.18. 0.46] 0.27 [−0.23, 0.76] 0.04 [−0.05, 0.14] −0.09 [−0.19, 0.003] −0.16 [−0.34, 0.004] −0.12 [−0.27, 0.03] −0.04 [−0.26, 0.18]
Computer use (per day)
Never (ref) N=332 - - - - - - -
0–1 hours N=117 0.40* [0.16, 0.64] 0.007 [−0.37, 0.39] −0.02 [−0.09, 0.05] 0.04 [−0.03, 0.11] 0.06 [−0.06, 0.19] 0.09 [−0.02, 0.22] −0.14 [−0.31, 0.03]
1+ hours N=76 0.14 [−0.14, 0.43] 0.41 [−0.04, 0.87] −0.008 [−0.09, 0.08] −0.002 [−0.09, 0.08] 0.01 [−0.14, 0.17] −0.07 [−0.21, 0.07] 0.06 [−0.14, 0.26]
Home density −0.11 [−0.20, −0.02] −0.08 [−0.23, 0.05] −0.002 [−0.03, 0.02] 0.001 [−0.03, 0.03] 0.003 [−0.05, 0.05] −0.008 [−0.05, 0.04] −0.006 [−0.06, 0.05]
HOME Total 0.02* [0.01, 0.03] −0.001 [−0.02, 0.01] 0.0001 [−0.003, 0.003] 0.000 [−0.003, 0.003] 0.0001 [−0.005, 0.005] −0.004 [−.009, 0005] 0.004 [−0.003, 0.01]
Open in a new tab
1

Linear regression

2

Negative binomial regression

3

Value given as β [95% CI]

*

p<0.05 Holm-Bonferroni correction

Abbreviations: SSP: Spatial Span, SOC: Stockings of Cambridge, IED: Intra-Extra Dimensional Set Shift.

3.5. Sensitivity analysis: Multiple imputations to account for missing data

We repeated the analysis (Supplemental Table 1), investigating the relationship between sociodemographic factors and EF task performance in a sample with imputed sociodemographic data (n=696). The results of these sensitivity analyses were consistent with our main findings (Table 4) in terms of the directionality, but CI intervals were more precise, revealing more statistically significant relationships. First, HOME subscales: receptivity, maturity, emotional climate, learning materials, and integration were positively associated with SSP (span length) performance. Second, the HOME subscales, receptivity, and accompaniment demonstrated a negative association, while integration showed a positive association with pre-dimensional errors on the IED task. Third, Child’s IQ was negatively associated with IED endpoints: total trials, total errors, and pre-dimensional errors. Lastly, compared to males, females performed better on the SOC task.

Discussion

Comparative studies of EF performance that include children outside the U.S. and utilize protocolized and automated measures are scarce. We evaluated developmental trends and a range of sociodemographic predictors of performance on three tasks from a computerized battery among urban children aged 6–8 from Montevideo, Uruguay. We also compared Uruguayan children’s performance with similarly aged children from the U.S. and Mexico.

In summary, we observed notable age-related differences in EF performance, with 8-year-old children performing better on the working memory (SSP) task than younger children in the cohort. SAM children also exhibited weaker performance in working memory (SSP) and planning (SOC) tasks across all age groups, but 6-year-olds performed similarly to their counterparts in the US and Mexico in the shifting (IED) task. Maternal IQ, child IQ, and HOME scores were positively associated with span length in the working memory (SSP) task. Children spending 0–1 hours on non-game computer activities also demonstrated enhanced span length. Child’s IQ was positively associated with problem-solving in the planning (SOC) task. Our sensitivity analysis reaffirmed most associations with higher precision in confidence intervals and more significant findings amongst HOME subscales and sex differences in the planning (SOC) task. Overall, our findings emphasize the influence of sociodemographic factors on EF performance in 6 to 8-year-old children and underscore the need to consider these factors in future research and interventions to support cognitive development during this critical period.

Considering the developmental trends in the studied population, our findings replicate previous results of Luciana and Nelson (2002), who described a steady improvement in working memory capacity in the 6–12 age period. This improvement was reflected in our study in the significant difference observed in the working memory task (SSP variable total errors) between ages 6 and 8. This task is based on the Corsi block-tapping task (Berch et al., 1998) and highly loads in visual working memory. In contrast, planning (SOC) and shifting (IED), which exhibit substantial growth in subsequent years (9 to 12) since they require more prefrontal cortex maturation (Anderson, 2001), did not exhibit improvement with age in our 6 to 8 sample. These findings are also consistent with Diamond’s (2013) model of EF development, wherein the maturation of working memory precedes the development of cognitive flexibility (shifting), subsequently leading to the emergence of complex cognitive processes such as planning and problem-solving.

When conducting cross-cultural comparisons, some cohort differences emerged. Specifically, children from the SAM cohort exhibited poorer performance than their U.S. counterparts across the studied age groups in working memory (SSP) and planning (SOC) tasks. Similar outcomes were observed when contrasting SAM children with their Mexican counterparts in the planning (SOC) task, although the disparities were comparatively more minor. In the case of the shifting (IED) task, performance between the different cohorts showed a greater degree of similarity, with occasional exceptions. These discrepancies could be explained by the different characteristics of the samples, including the influence of the sociodemographic factors. For example, regarding gender, SAM females performed better in the SOC task (in the imputed sample), while Mexican males performed better in the shifting (IED) task; no differences were found in the US cohort.

Additionally, the predictive role of child IQ varied across cohorts. It was a predictor of working memory (SSP) performance in SAM. However, this association was only observed in one of two WISC-III subtests administered in the US cohort (block design) (Luciana & Nelson, 2002). Vocabulary, the other subtest administered in the US cohort, was correlated with shifting (IED) performance (Luciana & Nelson, 2002). The association was also found in SAM but not in the Mexican study (Green et al., 2019). Notably, in the context of the planning (SOC) task, all three studies reported an association with IQ. The characteristics of the planning (SOC) task can explain this relationship, as it requires the application of problem-solving and reasoning abilities (Albert & Steinberg, 2011), which are strongly correlated with fluid intelligence (Stadler et al., 2015). Variations can also be observed in the impact of maternal IQ on EF performance. Our study revealed a significant influence of maternal IQ on working memory (SSP task) performance. However, it is noteworthy that a similar association was not documented in the Mexican cohort. While maternal IQ was not directly assessed in the US cohort, the level of maternal education, which can serve as a proxy measure of IQ (Rosander & Bäckström, 2012), was higher in that sample compared to the other cohorts. Notably, EF performance was also higher in the US cohort.

Other findings also highlight the importance of considering sociodemographic factors to understand EF development, particularly the home environment. We revealed an association between parental engagement in actively stimulating verbal communication, imparting knowledge of the alphabet and numerical concepts, as well as providing attention and support for the acquisition of novel skills (as measured by scores on the HOME Active Stimulation and Involvement sub-scales) and the enhancement of working memory performance (SSP). A safe, spacious play environment, measured via the Physical environment HOME sub-scale, was also associated with this EF. This result aligns with previous research (Rosen et al., 2020) and underscores the multifaceted impact of socio-cultural aspects on cognitive development. Considering such impact, some recent approaches (Obradović & Willoughby, 2019; Raver & Blair, 2020) strongly recommend considering culturally responsive assessment practices when measuring EF development. Moreover, Doebel (2020) argues that EFs are better understood when skills are oriented to specific goals that emerge in socio-cultural and individual contexts instead of domain-general processes.

The observed results have some psychometric implications regarding the CANTAB assessment of EF development. The construction of the CANTAB was based on a medical model aimed at identifying signs of dysfunction, a common goal in clinical neuropsychology (Luciana, 2003). For this reason, since its inception, the CANTAB test has not strictly adhered to the usual psychometric standards for the examination of test reliability or internal structure. This is one of the main weaknesses in assessing its psychometric properties, as shown by Karlsen et al. (2022) in adults and Syväoja et al. (2015) in children.

Although some studies of the CANTAB have reported consistent factorial structures in normative populations (De Luca et al., 2003; Lenehan et al., 2015), others have proposed two-factor models (Haring et al., 2015) or four-factor models (Lenehan et al., 2016; Robbins et al., 1994; Smith et al., 2013). Furthermore, some studies have questioned the psychometric properties of the CANTAB by finding modest associations with measures of traditional neuropsychological tests (Lenehan et al., 2015; Smith et al., 2013). Others have reported that the association between established neuropsychologically measured constructs and components of the CANTAB remains poorly investigated (Lenehan et al., 2016) and that psychometric properties of traditional neuropsychological tests may not hold when translated to a computerized format (Smith et al., 2013; Syväoja et al., 2015). These findings suggest that the factorial structure of the CANTAB may not be universally consistent and, therefore, not measuring the same constructs in different populations. These content validity problems hinder the effective comparison between countries (Casaletto & Heaton, 2017) and could contribute to the observed differences across cohorts. Conducting rigorous studies following current guidelines for instrument development and local adaptation (AERA et al., 2014) are recommended to provide more evidence of validity based on the internal structure of the CANTAB.

Our findings need to be interpreted in light of the study’s limitations. Although the 6–8 age period is well represented in our cohort, the distribution of participants per age was not homogenous across years, with the smallest sample at age 8. Besides, given the distinct developmental milestones associated with each EF during various stages of the academic years, an extended follow-up in subsequent years is necessary to obtain a comprehensive understanding of developmental trends. The non-uniform distribution of participants across age groups may have diminished the statistical power of the analyses. Another limitation relates to the partial administration of the CANTAB battery, thereby precluding a direct comparison with studies employing different sub-tests. This fact also limited our ability to analyze our data’s psychometrics characteristics, including examining the battery’s factorial structure.

Furthermore, our main analysis was impacted by statistical power, which was reflected in some differences in the complete case and sensitivity analysis. For example, computer use for non-game activities was associated with working memory for children who engaged in 0–1 hours per day but not for more than 1 hour. However, in the imputed dataset (Supplemental Table 1), however, this association was eliminated. In addition, there were more significant findings amongst home subscales and sex differences in the SOC task. This limitation highlights the critical role of missing data, which may have reduced statistical power and been influenced by random error. The imputation process likely contributed more statistical power.

Regarding the generalizability of our results, the association between computer use and working memory should be considered in light of the substantial evolution of digital environments over the past decade. According to a public report, in 2022 in Uruguay, nearly every child (90%) used the internet daily, mainly for recreational purposes through social networking applications (Pardo et al., 2022). Such findings markedly deviate from our first data collected in 2009–13 years and, therefore, require new research on this topic. Additionally, our sample could be biased regarding SES, considering that we limited recruitment to average to low quintiles of income, thus excluding children from higher-income households. Nevertheless, income is not homogenously distributed in Uruguay, and the SES of the schools we choose to work with represents more than 80% of Uruguayan children in between the ages of 6 and 11 years old (Instituto Nacional de Evaluación Educativa, 2022). Another issue concerning generalizability refers to the diversity of IQ measurement methods employed in our study. A prorated method based on Weschler scales measured maternal and child IQ and, although empirically validated (Sattler, 2008), these assessments should be considered a screening approach and not a definitive measure of IQ.

Regarding child IQ, two assessments were employed in different periods: WISC-IV and Woodcock-Muñóz. Those batteries correlate highly in their general measures and offer comparable indexes (Chiodi & S.Weschler, 2009; McGrew, 1983, Muñoz-Sandoval et al, 2009). Nevertheless, using different methods could bias the results and explain differences in the predictive role of IQ across cohorts. In this context, the consistency of the influence of IQ in planning (SOC) across the three cohorts is significant, thereby providing convergent validity to this finding. A final point regarding IQ is that our analysis of the sample characteristics revealed that children with complete data had significantly higher mean scores in maternal and child IQ compared to children who were excluded, which may influence the generalizability of our results to children and mothers with lower IQ.

In conclusion, this study provides evidence of the developmental trends of EF in early school years for urban children from a South American country. This investigation represents the largest sample to date assessed with the CANTAB within the understudied 6–8-year-old age group. Our findings confirm the efficacy of the battery in capturing EF changes during this developmental period, as observed trends align with those documented in prior literature. Nevertheless, in cross-cultural comparisons, some differences emerge beyond common patterns. These disparities could be attributed to socio-demographic factors and contextual influences. Therefore, conducting psychometric and normative studies of the CANTAB is important, particularly when administered in cultural settings distinct from its original context. Addressing this necessity demands longitudinal studies with adequately sized samples, allowing the assessment of CANTAB psychometric properties (stability, reliability, and factorial structure) and their association with developmental trajectories and sociocultural factors.

Supplementary Material

supplemental table 1
supplemental table 2

Funding:

This study was supported by the NIGMS Grant T-32-Initiative for Maximizing Student Development (T32 GM 144920), the Art Goshin Global Health Fieldwork Award (D.Rodriguez) and the following NIH Grants: R21ES16523, R21ES019949, R01ES023423 (PI: Kordas)

Footnotes

Ethical approval: Research ethics boards approved the study protocols at the Pennsylvania State University, the Catholic University of Uruguay, and the University at Buffalo

REFERENCES

  1. American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for educational and psychological testing. American Psychological Association. https://www.apa.org/science/programs/testing/standards [Google Scholar]
  2. Anderson P (2002). Assessment and development of executive function (EF) during childhood. Child Neuropsychology, 8(2), 71–82. 10.1076/chin.8.2.71.8724 [DOI] [PubMed] [Google Scholar]
  3. Anderson V (2001). Assessing executive functions in children: Biological, psychological, and developmental considerations. Pediatric Rehabilitation, 4(3), 119–136. 10.1080/13638490110091347 [DOI] [PubMed] [Google Scholar]
  4. Albert D, & Steinberg L (2011). Age differences in strategic planning as indexed by the Tower of London. Child Development, 82(5), 1501–1517. 10.1111/j.1467-8624.2011.01613.x [DOI] [PubMed] [Google Scholar]
  5. Ardoino GI, Queirolo EI, Barg G, Ciccariello DA, & Kordas K (2015). The relationship among depression, parenting stress, and partner support in low-income women from Montevideo, Uruguay. Health care for women international, 36(4), 392–408. [DOI] [PubMed] [Google Scholar]
  6. Barg G, Daleiro M, Queirolo EI, Ravenscroft J, Mañay N, Peregalli F, & Kordas K (2018). Association of low lead levels with behavioral problems and executive function deficits in schoolers from Montevideo, Uruguay. International journal of environmental research and public health, 15(12), 2735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bausela Herreras E (2007). Estudio de algunas propiedades psicométricas del WAIS-III en estudiantes universitarios. Revista de Psiquiatría y Psicología del Niño y del Adolescente, 7(1), 1–19. [Google Scholar]
  8. Berch DB, Krikorian R, & Huha EM (1998). The Corsi block-tapping task: Methodological and theoretical considerations. Brain and Cognition, 38(3), 317–338. 10.1006/brcg.1998.1039 [DOI] [PubMed] [Google Scholar]
  9. Bradley RH, Caldwell BM, & Corwyn RF (2003). The Child Care HOME Inventories: Assessing the quality of family child care homes. Early Childhood Research Quarterly, 18(3), 294–309. [Google Scholar]
  10. Brenlla M (2002): Adaptación de los subtests verbales de la WAIS-III al castellano. In Wechsler D. (2002): WAIS-III. Test de Inteligencia para Adultos. Manual Técnico (pp. XV–XXIII). Buenos Aires: Paidós. [Google Scholar]
  11. CANTAB® [Cognitive assessment software]. (2023). Cambridge Cognition. www.cambridgecognition.com [Google Scholar]
  12. Calvo J, Macadar D, & Nathan M (2013). Resultados generales y distribución territorial y por edades de la población con Necesidades Básicas Insatisfechas. In Calvo J (Ed.), Atlas Sociodemográfico y de la desigualdad del Uruguay (pp. 27–40). Montevideo:Trilce. [Google Scholar]
  13. Casaletto K, & Heaton R (2017). Neuropsychological assessment: Past and future. Journal of the International Neuropsychological Society, 23(9–10 Special Issue), 778–790. 10.1017/S1355617717001060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Chiodi M, & Weschler S (2009). Escala de Inteligência WISC- III e Bateria de Habilidades Cognitivas Woodcock Johnson-III: comparação de instrumentos. Avaliação Psicológica, 8(3), 313–324. [Google Scholar]
  15. De Luca C, Wood S, Anderson V, Buchanan J, Proffitt T, Mahony K, & Pantelis C (2003). Normative data from the Cantab. I: Development of executive function over the lifespan. Journal of Clinical and Experimental Neuropsychology, 25(2), 242–254. 10.1076/jcen.25.2.242.13639 [DOI] [PubMed] [Google Scholar]
  16. Desai G, Barg G, Queirolo EI, Vahter M, Peregalli F, Mañay N, & Kordas K (2018). A cross-sectional study of general cognitive abilities among Uruguayan school children with low-level arsenic exposure, potential effect modification by methylation capacity and dietary folate. Environmental Research, 164, 124–131. 10.1016/j.envres.2018.02.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Desai G, Barg G, Vahter M, Queirolo EI, Peregalli F, Mañay N, Millen A, Yu J, & Kordas K (2020). Executive functions in school children from Montevideo, Uruguay and their associations with concurrent low-level arsenic exposure. Environment International, 142(May), 105883. 10.1016/j.envint.2020.105883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Diamond A (2016). Why assessing and improving executive functions early in life is critical. In Executive Function in Preschool-Age Children (Issue August, pp. 9–45). 10.13140/RG.2.1.2644.6483 [DOI] [Google Scholar]
  19. Diamond A (2013). Executive functions. Annual Review of Psychology, 64, 135–168. 10.1146/annurev-psych-113011-143750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Doebel S (2020). Rethinking Executive Function and Its Development. Perspectives on Psychological Science, 15(4), 942–956. 10.1177/1745691620904771 [DOI] [PubMed] [Google Scholar]
  21. Everson-Rose SA, Mendes De Leon CF, Bienias JL, Wilson RS, & Evans DA (2003). Early Life Conditions and Cognitive Functioning in Later Life. American Journal of Epidemiology, 158(11), 1083–1089. 10.1093/aje/kwg263 [DOI] [PubMed] [Google Scholar]
  22. Feldman R, & Eidelman A (2009). Biological and environmental initial conditions shape the trajectories of cognitive and social-emotional development across the first years of life. Developmental Science, 12(1), 194–200. 10.1111/j.1467-7687.2008.00761.x [DOI] [PubMed] [Google Scholar]
  23. Fernald LC, Kariger P, Engle P, & Raikes A (2009). Examining early child development in low-income countries: a toolkit for the assessment of children in the first five years of life. World Bank, Washington DC. [Google Scholar]
  24. Fray P, & Robbins T (1996). CANTAB battery: Proposed utility in neurotoxicology. Neurotoxicology and Teratology, 18(4), 499–504. 10.1016/0892-0362(96)00027-X [DOI] [PubMed] [Google Scholar]
  25. Fray P, Robbins T, & Sahakian B (1996). Neuropsychiatric Application of CANTAB. International Journal of Geriatric Psychiatry, 11, 329–336. [Google Scholar]
  26. Fried R, Hirshfeld-Becker D, Petty C, Batchelder H, & Biederman J (2015). How Informative Is the CANTAB to Assess Executive Functioning in Children With ADHD? A Controlled Study. Journal of Attention Disorders, 19(6), 468–475. 10.1177/1087054712457038 [DOI] [PubMed] [Google Scholar]
  27. Frndak S, Barg G, Canfield R, Quierolo EI, Mañay N, & Kordas K (2019). Latent subgroups of cognitive performance in lead-and manganese-exposed Uruguayan children: Examining behavioral signatures. Neurotoxicology, 73, 188–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Frndak S, Barg G, Queirolo EI, Mañay N, Colder C, Yu G, Ahmed Z, & Kordas K (2022). Do Neighborhood Factors Modify the Effects of Lead Exposure on Child Behavior? Toxics, 10(9), 517. https://www.mdpi.com/2305-6304/10/9/517 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Frndak S, Gallo Y, Queirolo EI, Barg G, Mañay N, & Kordas K (2021). A mixed methods study examining neighborhood disadvantage and childhood behavior problems in Montevideo, Uruguay. International Journal of Hygiene and Environmental Health, 235, 113753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Fuster J (2013). Cognitive functions of the prefrontal cortex. In Stuss D & Knight R (Eds.), Principles of frontal lobe function (pp. 11–22). Oxford University Press. [Google Scholar]
  31. Gau SS-F, & Shang C-Y (2010). Executive functions as endophenotypes in ADHD: evidence from the Cambridge Neuropsychological Test Battery (CANTAB). Journal of Child Psychology and Psychiatry, 51(7), 838–849. 10.1111/j.1469-7610.2010.02215.x [DOI] [PubMed] [Google Scholar]
  32. Green C, Mihic A, Nikkel S, Stade B, Rasmussen C, Munoz D, & Reynolds J (2009). Executive function deficits in children with fetal alcohol spectrum disorders (FASD) measured using the Cambridge Neuropsychological Tests Automated Battery (CANTAB). Journal of Child Psychology and Psychiatry and Allied Disciplines, 50(6), 688–697. 10.1111/j.1469-7610.2008.01990.x [DOI] [PubMed] [Google Scholar]
  33. Green R, Till C, Al-Hakeem H, Cribbie R, Téllez-Rojo M, Osorio E, Hu H, & Schnaas L (2019). Assessment of neuropsychological performance in Mexico City youth using the Cambridge Neuropsychological Test Automated Battery (CANTAB). Journal of Clinical and Experimental Neuropsychology, 41(3), 246–256. 10.1080/13803395.2018.1529229 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Haring L, Mõttus R, Koch K, Trei M, & Maron E (2015). Factorial validity, measurement equivalence and cognitive performance of the Cambridge Neuropsychological Test Automated Battery (CANTAB) between patients with first-episode psychosis and healthy volunteers. Psychological medicine, 45(9), 1919–1929. 10.1017/S0033291714003018 [DOI] [PubMed] [Google Scholar]
  35. Holm S (1979). A simple sequentially rejective multiple test procedure. Scandinavian journal of statistics, 65–70. [Google Scholar]
  36. Instituto Nacional de Evaluación Educativa (2022). Distribución de la población por quintiles de ingresos de los hogares, entre 6 y 11 años. Años 2006 – 2022. Mirador Educativo. https://mirador.ineed.edu.uy/indicadores/distribucion-de-la-poblacion-de-entre-6-y-11-anos-por-quintiles-de-ingresos-de-los-hogares-19-3.html [Google Scholar]
  37. Karlsen RH, Karr JE, Saksvik SB, Lundervold AJ, Hjemdal O, Olsen A, Iverson GL, & Skandsen T (2022). Examining 3-month test-retest reliability and reliable change using the Cambridge Neuropsychological Test Automated Battery. Applied neuropsychology. Adult, 29(2), 146–154. 10.1080/23279095.2020.1722126 [DOI] [PubMed] [Google Scholar]
  38. Kordas K, Ardoino G, Coffman DL, Queirolo EI, Ciccariello D, Mañay N, & Ettinger AS (2015). Patterns of Exposure to Multiple Metals and Associations with Neurodevelopment of Preschool Children from Montevideo, Uruguay. Journal of Environmental and Public Health, 2015, 1–9. 10.1155/2015/493471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kordas K, Burganowski R, Roy A, Peregalli F, Baccino V, Barcia E, Mangieri S, Ocampo V, Mañay N, Martínez G, Vahter M, & Queirolo EI (2018). Nutritional status and diet as predictors of children’s lead concentrations in blood and urine. Environment International, 111, 43–51. 10.1016/j.envint.2017.11.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kaufmann L, Zotter S, Pixner S, Starke M, Haberlandt E, Steinmayr-Gensluckner M, Egger K, Schocke M, Weiss E, & Marksteiner J (2013). Brief report: CANTAB performance and brain structure in pediatric patients with asperger syndrome. Journal of Autism and Developmental Disorders, 43(6), 1483–1490. 10.1007/s10803-012-1686-6 [DOI] [PubMed] [Google Scholar]
  41. Ledbetter C, Moore AL, & Mitchell T (2017). Cognitive effects of ThinkRx cognitive rehabilitation training for eleven soldiers with brain injury: a retrospective chart review. Frontiers in psychology, 8, 825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Lehto J, Juujärvi P, Kooistra L, & Pulkkinen L (2003). Dimensions of executive functioning: Evidence from children. British Journal of Developmental Psychology, 21(1), 59–80. 10.1348/026151003321164627 [DOI] [Google Scholar]
  43. López MJL, González JMR, Vilariño CS, & Linares ET (2003). Utilidad de las formas cortas de la Escala de Inteligencia de Wechsler para Adultos (WAIS). Anales de Psicología/Annals of Psychology, 19(1), 53–63. [Google Scholar]
  44. Luciana M (2003). Practitioner review: Computerized assessment of neuropsychological function in children: Clinical and research applications of the Cambridge Neuropsychological Testing Automated Battery (CANTAB). Journal of Child Psychology and Psychiatry and Allied Disciplines, 44(5), 649–663. 10.1111/1469-7610.00152 [DOI] [PubMed] [Google Scholar]
  45. Lowe C, & Rabbitt P (1998). Test\re-test reliability of the CANTAB and ISPOCD neuropsychological batteries: theoretical and practical issues. Neuropsychologia, 36(9), 915–923. [DOI] [PubMed] [Google Scholar]
  46. Luciana M, & Nelson C (2002). Assessment of neuropsychological function through use of the Cambridge Neuropsychological Testing Automated Battery: Performance in 4- to 12-year-old children. Developmental Neuropsychology, 22(3), 595–624. 10.1207/S15326942DN2203_3 [DOI] [PubMed] [Google Scholar]
  47. McGrew K (1983). A comparison of the WISC-R and the Woodcock-Johnson tests of cognitive ability. Psychology in the Schools, 21, 271–276. 10.1016/0022-4405(83)90023-7 [DOI] [Google Scholar]
  48. Miyake A, Friedman N, Emerson M, Witzki A, Howerter A, & Wager T (2000). The Unity and Diversity of Executive Functions and Their Contributions to Complex “Frontal Lobe” Tasks: A Latent Variable Analysis. Cognitive Psychology, 41(1), 49–100. 10.1006/cogp.1999.0734 [DOI] [PubMed] [Google Scholar]
  49. Muñoz-Sandoval AF, Woodcock RW, McGrew KS, Mather N, & Ardoino G (2009). Batería III Woodcock-Muñoz. Ciencias psicológicas, 3(2), 245–246. [Google Scholar]
  50. Obradović J, & Willoughby MT (2019). Studying Executive Function Skills in Young Children in Low- and Middle-Income Countries: Progress and Directions. Child Development Perspectives, 13(4), 227–234. 10.1111/cdep.12349 [DOI] [Google Scholar]
  51. Ozonoff S, Cook I, Coon H, Dawson G, Joseph R, Klin A, McMahon W, Minshew N, Munson J, Pennington B, Rogers S, Spence M, Tager-Flusberg H, Volkmar F, & Wrathall D (2004). Performance on Cambridge neuropsychological test automated battery subtests sensitive to frontal lobe function in people with autistic disorder: Evidence from the Collaborative Programs of Excellence in Autism network. Journal of Autism and Developmental Disorders, 34(2), 139–150. 10.1023/B:JADD.0000022605.81989.cc [DOI] [PubMed] [Google Scholar]
  52. Pardo V, Dornel S, Hughes C, García M, Cancela V, Blanc MV, … & Carusso LO (2022). Niños, niñas y adolescentes conectados: Informe Kids Online Uruguay 2022. [Google Scholar]
  53. Queirolo EI, Kordas K, Martínez G, Ahmed Z, Barg G, & Mañay N (2024). Secular trends in blood lead concentrations of school-age children in Montevideo, Uruguay from 2009 to 2019. Environmental Pollution, 343(November 2023). 10.1016/j.envpol.2023.123160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rasmussen C, Soleimani M, Carroll A, & Hodlevskyy O (2009). Neuropsychological functioning in children with Tourette syndrome (TS). Journal of the Canadian Academy of Child and Adolescent Psychiatry, 18(4), 307–315. [PMC free article] [PubMed] [Google Scholar]
  55. Raver CC, & Blair C (2020). Developmental science aimed at reducing inequality: Maximizing the social impact of research on executive function in context. Infant and Child Development, 29(1), 1–10. 10.1002/icd.2175 [DOI] [Google Scholar]
  56. Richaud M., & Sacchi C. (2001). Adaptación del inventario de depresión de Beck a sujetos argentinos normales. Revista Oficial de La Asociación Iberoamericana de Diagnóstico y Evaluación Psicológica (AIDEP), 12(2), 11–17. [Google Scholar]
  57. Robbins T, James M, Owen A, Sahakian B, Lawrence A, Mcinnes L, & Rabbitt P (1998). A study of performance on tests from the CANTAB battery sensitive to frontal lobe dysfunction in a large sample of normal volunteers: Implications for theories of executive functioning and cognitive aging. Journal of the International Neuropsychological Society, 4(5), 474–490. 10.1017/S1355617798455073 [DOI] [PubMed] [Google Scholar]
  58. Robbins T, James M, Owen A, Sahakian B, McInnes L, & Rabbitt P (1994). Cambridge neuropsychological test automated battery (CANTAB): A factor analytic study of a large sample of normal elderly volunteers. Dementia, 5(5), 266–281. 10.1159/000106735 [DOI] [PubMed] [Google Scholar]
  59. Rodríguez D, Barg G, Queirolo EI, Olson J, Mañay N, & Kordas K (2023). Pyrethroid and Chlorpyrifos Pesticide Exposure, General Intellectual Abilities, and Executive Functions of School Children from Montevideo, Uruguay. International Journal of Environmental Research and Public Health, 20(7). 10.3390/ijerph20075288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Rodriguez JM, Herrans LL, Pons J, Matías-Carrelo L, Medina G, & Rodríguez M (2008). Proceso de traducción y adaptación para Puerto Rico de la Wechsler Adult Intelligence Scale-III: Escala de Inteligencia Wechsler para Adultos, Versión III (EIWA-III). Revista Puertorriqueña de Psicología, 19, 58–74. [Google Scholar]
  61. Rosander P, & Bäckström M (2012). The unique contribution of learning approaches to academic performance, after controlling for IQ and personality: Are there gender differences? Learning and Individual Differences, 22(6), 820–826. 10.1016/j.lindif.2012.05.011 [DOI] [Google Scholar]
  62. Rosen M, Hagen M, Lurie L, Miles Z, Margaret A, Meltzoff A, & Mclaughlin K (2020). Cognitive stimulation as a mechanism linking socioeconomic status with executive function : A longitudinal investigation. Child Development, 91(4), e762–e779. 10.1111/cdev.13315 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Roy A, Queirolo E, Peregalli F, Mañay N, Martínez G, & Kordas K (2015). Association of blood lead levels with urinary F2–8α isoprostane and 8-hydroxy-2-deoxy-guanosine concentrations in first-grade Uruguayan children. Environmental Research, 140, 127–135. 10.1016/j.envres.2015.03.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Sattler J (2008). Assessment of Children: Cognitive Applications (5th ed.). La Mesa, CA: Jerome Sattler [Google Scholar]
  65. Scantlebury N, Bouffet E, Laughlin S, Strother D, McConnell D, Hukin J, Fryer C, Laperriere N, Montour-Proulx I, & Keene D (2016). White matter and information processing speed following treatment with cranial-spinal radiation for pediatric brain tumor. Neuropsychology, 30(4), 425. [DOI] [PubMed] [Google Scholar]
  66. Schrank FA, & McGrew KS (2001). Technical abstract (Woodcock-Johnson III assessment service bulletin no. 2). Itasca, IL: Riverside Publishing. [Google Scholar]
  67. Shuttleworth-Edwards AB, Kemp RD, Rust AL, Muirhead JGL, Hartman NP, & Radloff SE (2004). Cross-cultural Effects on IQ Test Performance: A Review and Preliminary Normative Indications on WAIS-III Test Performance. Journal of clinical and experimental neuropsychology, 26(7), 903–920. 10.1080/13803390490510824 [DOI] [PubMed] [Google Scholar]
  68. Smith PJ, Need AC, Cirulli ET, Chiba-Falek O, & Attix DK (2013). A comparison of the Cambridge Automated Neuropsychological Test Battery (CANTAB) with “traditional” neuropsychological testing instruments. Journal of clinical and experimental neuropsychology, 35(3), 319–328. 10.1080/13803395.2013.771618 [DOI] [PubMed] [Google Scholar]
  69. Snyder H, Miyake A, & Hankin B (2015). Advancing understanding of executive function impairments and psychopathology: Bridging the gap between clinical and cognitive approaches. Frontiers in Psychology, 6(MAR). 10.3389/fpsyg.2015.00328 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Sowell E, Trauner D, Gamst A, & Jernigan T (2002). Development of cortical and subcortical brain structures in childhood and adolescence: A structural MRI study. Developmental Medicine and Child Neurology, 44(1), 4–16. 10.1111/j.1469-8749.2002.tb00253.x [DOI] [PubMed] [Google Scholar]
  71. Stadler M, Becker N, Gödker M, Leutner D, & Greiff S (2015). Complex problem solving and intelligence: A meta-analysis. Intelligence, 53, 92–101. 10.1016/j.intell.2015.09.005 [DOI] [Google Scholar]
  72. Sugland BW, Zaslow M, Smith JR, Brooks-Gunn J, Coates D, Blumenthal C, Moore KA, Griffin T, & Bradley R (1995). The early childhood HOME Inventory and HOME-short form in differing racial/ethnic groups: Are there differences in underlying structure, internal consistency of subscales, and patterns of prediction? Journal of Family Issues, 16(5), 632–663. [Google Scholar]
  73. Syväoja HJ, Tammelin TH, Ahonen T, Räsänen P, Tolvanen A, Kankaanpää A, & Kantomaa MT (2015). Internal consistency and stability of the CANTAB neuropsychological test battery in children. Psychological assessment, 27(2), 698–709. 10.1037/a0038485 [DOI] [PubMed] [Google Scholar]
  74. Taub GE, & McGrew KS (2004). A Confirmatory Factor Analysis of Cattell-Horn-Carroll Theory and Cross-Age Invariance of the Woodcock-Johnson Tests of Cognitive Abilities III. School Psychology Quarterly, 19(1), 72. [Google Scholar]
  75. Taub GE, McGrew KS, & Witta EL (2004). A confirmatory analysis of the factor structure and cross-age invariance of the Wechsler Adult Intelligence Scale-Third Edition. Psychol Assess, 16(1), 85–89. 10.1037/1040-3590.16.1.85 [DOI] [PubMed] [Google Scholar]
  76. Teixeira R, Zachi E, Roque D, Taub A, & Ventura D (2011). Memory span measured by the spatial span tests of the Cambridge Neuropsychological Test Automated Battery in a group of Brazilian children and adolescents. Dementia & Neuropsychologia, 5(2), 129–134. 10.1590/s1980-57642011dn05020012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Toornstra A, Hurks P, Van der Elst W, Kok G, & Curfs L (2020). Measuring visual matching and short-term recognition memory with the CANTAB® Delayed Matching to Sample task in schoolchildren: Effects of demographic influences, multiple outcome measures and regression-based normative data. Child Neuropsychology, 26(2), 189–218. 10.1080/09297049.2019.1642316 [DOI] [PubMed] [Google Scholar]
  78. Weschler D (2011). WISC-IV : escala de inteligencia para niños-IV de Wechsler. Buenos Aires: Paidós. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplemental table 1
NIHMS2057713-supplement-supplemental_table_1.docx (27.7KB, docx)
supplemental table 2
NIHMS2057713-supplement-supplemental_table_2.docx (22.5KB, docx)

RESOURCES