Available online at www.sciencedirect.

com Procedia
Social and
Behavioral
Procedia
Procedia - Social
- Social and Behavioral
and Behavioral Sciences
Sciences 00 (2011)
33 (2012) 627 –000–000
631 Sciences
www.elsevier.com/locate/procedia

PSIWORLD 2011

The structure of executive functions in preschoolers: An

investigation using the NEPSY battery
Laura Visu-Petraa*, Lavinia Cheiea, Oana Bengaa, Mircea Micleab
a
Developmental Psychology Lab, Department of Psychology, Babeú-Bolyai University, Cluj-Napoca, 400015, Romania
b
Applied Cognitive Psychology Center, Department of Psychology, Babeú-Bolyai University, Cluj-Napoca, 400015, Romania

Abstract

Additional evidence is needed in order to clarify the latent structure of executive functions in young children. The
current study included seven standardized measures of executive functioning from the NEPSY battery (Korkman,
Kirk, & Kemp, 1998), administered to preschoolers (N = 186), mean age = 3 y 6 mo. Results indicated that in the 3-5
years interval, a unitary factor solution best accounted for the data, compared to a 3-factor solution (the NEPSY
subdomains), or a 2-factor solution (working memory and inhibition). Our results support previous research,
revealing that executive functioning is a relatively undifferentiated construct in young children.

©
© 2012 PublishedbybyElsevier
2011 Published Elsevier B.V.
Ltd. Selection
Selection and/or peer-review
and peer-review under responsibility
under responsibility of PSIWORLD2011
of PSIWORLD 2011

Keywords: executive functions; development; structure; preschoolers; NEPSY

1. Introduction

Executive functioning (EF) represents an umbrella-type concept for the complex set of cognitive
processes that underlie flexible, goal-directed responses to novel or difficult situations (Hughes &
Graham, 2002). The few previous investigations using confirmatory factor analyses (CFA; Hughes,
Ensor, Wilson, & Graham, 2010; Wiebe, Espy & Charak, 2008; Wiebe et al., 2011) supported a unitary
model of executive functioning in preschoolers, or a two-component model, with working memory and
inhibition as latent factors (Miller, Giesbrecht, Müller, McInerney, & Kerns, in press). The present study
analyzed the performance of very young children (3-5 years) evaluated with the widely-used NEPSY
battery, searching for the relations between EF-demanding tasks in this battery. Although there have been

*
Corresponding author. Tel/Fax +40 264590967
E-mail address: laurapetra@psychology.ro

1877-0428 © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of PSIWORLD2011
doi:10.1016/j.sbspro.2012.01.197
628 Laura
L.V. Visu-Petra
Petra et al. / Procedia
et al. / Procedia - Social-and
Social and Behavioral
Behavioral SciencesSciences 33 000–000
00 (2011) (2012) 627 – 631

previous exploratory attempts to analyze the factorial structure of the whole NEPSY battery (Jarratt,
2005; Klenberg, Korkman, & Lahti-Nuuttila, 2001; Stinnett, Oehler-Stinnett, Fuqua, & Palmer, 2002),
they focused exclusively on the 5-12 years interval, and found mixed evidence for a unitary model
(Jarratt, 2005; Stinnett et al., 2002), or for a plurifactorial model (Klenberg et al., 2001).

1.1. A developmental neuropsychological approach to EF using the NEPSY battery

The NEPSY test battery (Korkman, Kirk, & Kemp, 1998) was developed based on the flexible model
and diagnostic principles of Luria (Luria, 1973; Korkman, 1999). There are five domains within which
the tests have been construed: Attention/Executive Functions, Language, Visual-Spatial Processing,
Sensoriomotor Functioning and Memory and Learning. Three distinct models were tested with a
confirmatory analysis technique to see which provides the best fit to all the NEPSY tasks with strong EF
demands: a 1-factor (perhaps Language) domain (which has already received some empirical validation),
the NEPSY core domain subdivisions, or an alternative model, based on the three EF factors (WM,
inhibition, and shifting), which received both developmental and adult validation (although not tested for
the NEPSY battery). These three models were tested with a newly-formed EF-battery from the NEPSY
subtests for the 3-5 years interval. Our previous exploratory investigation with a large sample of school-
age children 5-12 years old (Visu-Petra, Miclea, Benga, 2007) supported a two-factor solution, with a
clear inhibition factor, and a mixed fluency and planning factor.

2. Method

2.1. Participants

The analysis has been performed on a subsample from the data collected in the research project of
adapting NEPSY: A Developmental Neuropsychological Assessment for the Romanian population. A
subsample of 186 children who did not lack more than one NEPSY task (age range 3 to 4 years and 11
months, mean age 3 years 6 months; 108 girls) was included in the CFA for preschool age. Missing data
from both samples were estimated using the mean of nearby points algorithm in SPSS 13.00.

2.2. Measures and procedure

For the memory and learning domain, the Sentence Repetition and Narrative Memory tests were used.
Sentence Repetition is a measure of short-term memory, evaluating a sentence span of the child. The
gradual increase in sentence size also tests for the child’s ability to organize the information and to
employ memory strategies for better recall. Narrative Memory is a complex test, requiring attention,
organizing and sequencing, understanding the essence of the story along with recalling specific details,
semantic and syntactic linguistic abilities, as well as the ability to encode, maintain and recall both names
and content.
For the core EF domain, we used the Statue and the Visual Attention tests. In the Statue test, the child
is required to stand still in a position as a “statue” holding a flag over a 75-sec interval. At pre-set
intervals distractors are introduced (e.g. the examiner dropping a pen). For each 5-sec interval, the child is
awarded 2 points for lack of inappropriate responses, and 1 point for one inappropriate response. The
maximum score is 30. In the Visual Attention test, the child is instructed to select only the items that
match the target stimuli on the page containing both targets and distracters. The number of targets, both
on the Cats subtest (max. 20) and on the Faces subtest (max. 20) is scored.
 LauraL.V.
Visu-Petra
Petra et et
al.al. / Procedia- Social
/ Procedia - Socialand
andBehavioral
BehavioralSciences
Sciences0033(2011)
(2012)000–000
627 – 631 629

Tests of Comprehension of Instructions, Phonological Processing, and Verbal Fluency were selected
for the language domain. Comprehension of Instructions involves processing, encoding, and action
control according to gradually increasing linguistic commands. The child was instructed to point some
shapes on a paper, according to increasingly complex rules. Phonological Processing requires in part A to
establish a correspondence between sounds and images, and to select based on a partial sound cue the
corresponding image. Part B requires the child to encode a verbal stimulus and to manipulate its phonetic
structure by either removing a sound, or by replacing a sound with another. During Verbal Fluency, for a
1-minute interval, children are asked to generate words belonging to either two semantic categories
(animals and food/drinks) or two phonemic categories (words beginning with an “S”/“F”). The sum of
correct words across the four categories constitutes the test score.
After obtaining parental informed consent and child verbal assent, the tests were administered by
licensed psychologists who had already undergone a NEPSY training course. Most of the evaluations
were conducted individually in educational centers (kindergartens and schools), in a separate room. The
process of administering the whole NEPSY battery lasted about one hour.

3. Results

Three models were contrasted using CFA for 3-5-year-olds (see Fig. 1 for a comparison of the three
models). The first model tested for the possibility that a 1-factor, model would provide the best account
for the data. A second model grouped the 7 EF tasks according to the NEPSY domains (Attention/EF,
Memory and Learning, and Language). Finally, the 3-factor model of Miyake et al. (2000) could not be
tested, because there was no test that assessed shifting. Instead, a two-factor (Working Memory and
Inhibition), plus Complex EF was assessed. The unitary EF model had a substantial goodness of fit index
(.96), although it was not the best, the 3-factor NEPSY model having a .97 GFI. However, the RMSEA
was similar to the NEPSY model (.06), and it had better parsimony indexes (PNFI and PCFI). All the
subtests loaded significantly (over .30) on this general factor, with lower loadings from the inhibitory
tasks, Statue and Visual Attention (.43 and .47, respectively), perhaps suggesting an early trend towards
differentiation of this EF subcomponent, that is also maturing rapidly, as the analysis of the
developmental pathways in the older sample has revealed (Visu-Petra, Benga, & Miclea, 2007). The other
tasks had good loadings on this general factor, ranging from .52 in the case of Phonological Processing, to
.74 in the case of Narrative Memory. The 3-factor NEPSY model had the best goodness of fit index (.97)
and a reasonable RMSEA (.06); it also had the best NFI and CFI indexes. However, the model has the
poorest parsimony indexes. Indeed, by looking at the model, it is obvious that the three NEPSY factors
are highly interrelated (correlations from .81 to .92), suggesting that their differentiation is artificial with
regard to the true structure of the data. Finally, the 2-factor EF model that groups the working memory
and the inhibition tasks, and relates them to a complex EF-task, although comparable to the other models
in goodness of fit and in NFI and CFI indexes, has a larger RMSEA (.07), and again lower parsimony
indexes than the unitary model. Again, correlations between the three dimensions are high (from .67 to
.87), suggesting that the fractionation of the EF in this age interval is overestimated.
By directly comparing the BIC index, and favouring the model with the lowest value (a 10-points
difference being considered significant), the unitary EF model provides a better fit to the data than the 3-
factor NEPSY model, and than the 2-factor EF model, although the difference between the plurifactorial
models is not significant. In conclusion, based on the goodness of fit and on the parsimony statistical
indexes, for the 3-5 years interval, a unitary framework provides the best account for the NEPSY data.
The matrix for this preferred model is presented in Table 2.
630 Laura Visu-Petra et al. / Procedia - Social and Behavioral Sciences 33 (2012) 627 – 631
L.V. Petra et al. / Procedia - Social and Behavioral Sciences 00 (2011) 000–000

Model 1 Model 2 Model 3

Fig. 1. The three proposed models: Model 1. The 1-factor Language model; Model 2. The 3-factors NEPSY model; Model 3. The
two-factors + complex EF tasks; 3-4 years, N = 186.

Table 1. Goodness of Fit Statistics for Confirmatory Factor Analyses with the EF tasks from NEPSY (3-5 years)

Model Ȥ2 df GFI AGFI RMSEA NFI CFI PNFI PCFI BIC

1-factor Language (/General EF) 25.10* 14 .96 .92 .06 .92 .96 .61 .64 125.51
model
3-factor NEPSY model 19.3 11 .97 .93 .06 .94 .97 .49 .50 141.21
2-factor EF model 23.21* 12 .96 .92 .07 .92 .96 .53 .55 137.96

Table 2. Matrix of the Unitary EF Factorial Model for the 3-5 Years Age Group

Variable 1-factor (Language / General EF)

Narrative Memory .74
Sentence Repetition .61
Statue .43
Visual Attention .47
Comprehension of Instructions .63
Phonological Processing .52
Verbal Fluency .69

4. Conclusions

In the 3-5 years interval, a 1-factor solution best accounted for the data, suggesting that a unitary
approach to EF is justified at this early age. But what is the underlying nature of this unifying factor? An
initial possibility advocated by previous CFA analysis in preschoolers (Wiebe, Espy, & Charak, 2008) is
that EF is indeed a relatively undifferentiated construct in young children, but that it becomes more
modular with age. This conclusion is also supported by theoretical conjectures from data on early
prefrontal cortex functioning: the shift from the more diffuse to more focal activation of this area and the
“frontalization” process are age-dependent processes (Zelazo, Carlson, & Kesek, 2008). A second
possibility is that the underlying variable that integrates all the NEPSY tasks in preschoolers is actually a
linguistic factor. Children with better linguistic abilities better understand the instructions and maintain
easier the rules in a linguistic format throughout the task solving process. This perspective would be
sustained by studies revealing a unitary, (supposedly) linguistic structure of the whole NEPSY battery
 LauraL.V.
Visu-Petra
Petra et et
al.al. / Procedia- Social
/ Procedia - Socialand
andBehavioral
BehavioralSciences
Sciences0033(2011)
(2012)000–000
627 – 631 631

(Jarratt, 2005; Stinnett et al., 2002). Indeed, the non-verbal tasks (Statue and Visual Attention) had the
lowest loadings onto this general factor; this could be equally explained though by their simpler structure
and strong reliance on the earlier maturing function of motor inhibition. However, only a longitudinal
perspective could be truly informative for a potential “developmental shift” from an undifferentiated to a
more modular account of EF.

Acknowledgements

The present research has been supported by a PN-II-RU-PD-2010/427 grant awarded to the first author
by the National Council for Scientific Research.

References

Hughes, C., & Graham, A. (2002). Measuring executive functions in childhood: Problems and solutions? Child and Adolescent
Mental Health, 7, 131–142.
Hughes, C., Ensor, R., Wilson, A., & Graham, A. (2010). Tracking executive function across the transition to school: A latent
variable approach. Developmental Neuropsychology, 35, 20–36.
Jarratt, K. P. (2005). The CAS and NEPSY as measures of cognitive processes: Examining the underlying constructs. Unpublished
doctoral dissertation, Texas A&M University.
Klenberg, L., Korkman, M., & Lahti-Nuuttila, P. (2001). Differential development of attention and executive functions in 3- to 12-
year old Finnish children. Developmental Neuropsychology, 20, 407–428.
Korkman, M. (1999). Applying Luria’s diagnostic principles in the neuropsychological assessment of children. Neuropsychology
Review, 9, 89–105.
Korkman, M., Kirk, U., & Kemp, S. (1998). NEPSY: A Developmental Neuropsychological Assessment. New York: The
Psychological Corporation.
Luria, A. R. (1973). The working brain. New York: Basic Books.
Miller, M. R., Giesbrecht, G. F., Müller, U., McInerney, R. J. , & Kerns, K. A. (in press). A latent variable approach to determining
the structure of executive function in preschool children. Journal of Cognition and Development.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D. (2000). The unity and diversity of
executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology,
41, 49–100.
Stinnett, T. A., Oehler-Stinnett, J., Fuqua, D. R., & Palmer, L. S. (2002). Examination of the underlying structure of the NEPSY: A
developmental neuropsychological assessment. Journal of Psychoeducational Assessment, 20, 66–82.
Visu-Petra, L., Benga, O., & Miclea, M. (2007). Dimensions of attention and executive functioning in 5- to 12-year-old children:
Neuropsychological assessment with the NEPSY battery. Cognition, Brain, Behavior. An interdisciplinary journal, 11, 585–
609.
Wiebe, S. A., Espy, K. A., & Charak, D. (2008). Using confirmatory factor analysis to understand executive control in preschool
children: I. Latent structure. Developmental Psychology, 44, 575–587.
Wiebe, S. A., Sheffield, T., Nelson, J. M., Clark, C. A. C., Chevalier, N., & Espy, K. A. (2011). The structure of executive function
in 3-year-olds. Journal of Experimental Child Psychology, 108, 436–452.
Zelazo, P. D., Carlson, S. M., & Kesek, A. (2008). Development of executive function in childhood. In C. A. Nelson, & M. Luciana
(Eds.), Handbook of developmental cognitive neuroscience, 2nd ed. (pp. 553–574). Cambridge, MA: MIT Press.