Participants

From the normative data of Brazilian validation study of the A Developmental Neuropsychological Assessment (NEPSY-II instrument) (Korkman, Kirk, & Kemp, 2007) translated by the researchers, we selected 203 children from 1,260 individuals with ages between 5 and 6 for school analysis. Hand preference was defined during training of the literacy process in the age group between 5 and 6, reason why we chose this age group (Rigal, 1992).

The NEPSY-II is an instrument for developmental neuropsychological assessment of children between the ages of 3 and 16. The battery is theoretically divided into six sections or Cognitive Domains: Attention/Executive Functioning, Language, Visuospatial Processing, Sensorimotor, Learning and Memory, and Social Perception (Table 1). Of the 32 subtests of the instrument, 26 are applicable to 5- to 6-year-old children and these were used in the present study (see Table 1) (II NEPSY Manual, Korkman et al., 2007). Briefly, given the complexity of the instrument, the NEPSY-II is a flexible battery that is specifically designed for children and it is not an adaptation of a test developed for adults. As an example, in the Design Copying subtest the child has to copy two-dimensional geometric pictures in increasing complexity to observe perceptual motor and visual aspects. In the Visuomotor Precision subtest, the child has to draw a line inside winding paths without touching the sides in the shortest possible time. It evaluates fine motor speed and precision.

Table 1

Note: *Subtests not used for 5- and 6-year olds; **Subtest not used to present score differences for 5- and 6-year olds. NEPSY: Developmental Neuropsychological Assessment.

The individual application of the battery occurred between 2010 and 2012 by a trained staff at the child's school or residence. The parents were asked to sign an informed consent form and the child agreed to participate in the study. Handedness was evaluated by an examiner, in accordance with the American NEPSY-II Clinical and Interpretive Manual, as right-handers, left-handers or undefined.

Descriptive statistics were used to analyze socio-demographic characteristics. In this stage of the study, in addition to the general analysis, the sample was divided into two groups: left-handers and right-handers. To compare the scores between the groups, the normality of sample distribution was tested using Shapiro-Wilk test, histogram normal curve, skewness and kurtosis. Analysis of Variance (one-way Anova) was used to analyze possible differences in performance between right-handers and left-handers for each subtest with normal distribution. For the subtests without normal distribution, data was analyzed using Kruskal-Wallis test. The value of p < 0.05 was used to define the subtests that were statistically different between the two groups. To perform the analysis of the data, the Statistical Package for Social Sciences (SPSS, version 18.0) was used.

The study was approved by the Research Ethics Committee of the Hospital Santa Izabel, report obtained on August 11, 2006 (Sistema Nacional de Informação sobre Ética em Pesquisas Envolvendo Seres Humanos, process nº 98513).

Table 2 describes the socio-demographic characteristics of the sample. There is a predominance of right-handers, totaling more than 90.0% of the sample. When comparing the groups of hand preference, no statistically significant differences were found for any of the socio-demographic data. No significant percentage differences between sex and type of school were found intra-group. In both groups, a high percentage of mothers (47.5%) had 8 to 11 years of education in the left-handed group and 46.8% in the right-handed. In the right-handed sample, we observed that approximately half of the children were in each age group (5 and 6 years old), but in the left-handed group, 63.2% were 5-year-old.

Table 2

Note: 1Twelve families did not disclose this information; 2Ten families did not disclose this information.

Of the 32 battery subtests, four showed statistically significant differences between the two groups (right-handers x left-handers) in the domains Attention/Executive Functioning, Language and Memory. In the Attention/Executive Functioning domain, two subtests showed differences: Auditory Attention Total Comission Errors score; and Design Fluency (DF), two scores - Random Series and Total Score. In Language, the Total Score for the Word Generation (WG) subtest and Learning and Memory, the Total Score for Memory for Faces, but not for Memory for Faces Delayed.

The Auditory Attention Subtest (AARS), in the Attention and Executive Functioning domain, provides scores for the number of right answers, omission errors, commission errors and inhibition errors, as well as qualitative observations (distracted behavior and moving around/getting up from the chair), in task of selective auditory attention and auditory sustained attention. The performance of left-handers was significantly lower in one of the scores (p = 0.01), as they committed more action errors than right-handers (Mean - M = 12.21, M = 6.74, respectively).

Design Fluency, in the Attention and Executive Functioning domain, is a subtest that assesses nonverbal fluency through a paper/pencil task generating original pictures, connecting five points, within a time limit. It consists of two steps: Structured Series and Random Series. In the Random Series, where the points do not follow a simple geometric pattern, left-handers presented a significantly lower performance (M = 1.98, p = 0.008). The Total Score for DF consists of the sum of scores of the Structured and Random Series. In this subtest right-handers produced more drawings than left-handers (M = 12.5, M = 9.63, p = 0.016).

In the Language domain, the WG subtest, which assesses verbal fluency within a time limit, consists of two categories (semantic and phonological). The 5- and 6-year-olds only performed the semantic category. A lower performance was found for left-handers than right-handers (M = 13.47, M = 16.91, respectively, p = 0.002).

In the Memory and Learning domain, the Memory for Faces subtest assesses encoding, discrimination and recognition of facial features, in which a number of faces must be recognized by the child. In addition, the later stage assesses long-term memory for faces. The right-handers recalled more faces in the shorter period than the left-handers (M = 8.6, M = 7.11, p = 0.045) (Table 3).

Table 3

Note: *When the score refers to an error, the higher the mean, the poorer was performance. Kruskal-Wallis p < 0.05. AA: Auditory Attention; DF: Design Fluency; M; Mean; MF: Memory for Faces; SD: Standard Deviation; WG: Word Generation.

The aim of the present study was to identify performance differences between right-handers and left-handers with typical development using NEPSY-II, a developmental neuropsychological assessment instrument. With regard to the socio-demographic variables, the frequency of left-handers (9.4%) was equivalent to the percentage described in the literature (Johnston, Nicholls, Shah, & Shields, 2010).

As for hand preference associated with gender, there was no significant difference, which corroborates the studies of Arteaga and Poblano (2008), Beratis et al. (2009) and Preti et al. (2014). Other authors (Johnston et al., 2007; Milenkovi et al., 2013) found a prevalence of left-handed men. The association between sex and hand preference is still controversial. Because it is related to multifactorial inheritance, other factors may be important for handedness. Sexuality cannot be predominant in men with normal development; however, it could be so in clinical samples of developmental disorders. These disorders have genetic and/or environmental causes. The study of Pogetti, Souza, Tudella, and Teixeira (2013) exemplifies the strength of environmental factors on handedness. In that study, the authors conducted a unilateral visual occlusion and observed that there was a decrease in the use of the ipsilateral limb during and immediately after occlusion. The authors concluded that visual availability interferes with hand choice. Other factors, such as maternal stress (Reissland et al., 2014), are mentioned. Preslar et al. (2014) also found no relationship between gender and handedness.

Left-handers performed poorly in two subtests of the Attention and Executive Functioning domain, AARS and DF. In AARS, the left-handers committed more errors, showing difficulty in inhibitory control. In DF, they presented fewer drawings, indicating low visuomotor fluency. Why auditory attention, verbal fluency and design fluency is more deficient in left-handers in comparison with right-handers is difficult to explain. However, Beratis et al. (2009) showed that, when dealing with tasks that include executive functions and language, left-handers activate both hemispheres in a more balanced way than right-handers. Thus, the lower hemispheric specialization presented by left-handers could be a disadvantage in these cognitive domains. Would performing both functions cause one of them to be deficient? The deficit-attention disorder/hyperactivity disorder is the prototype of disability in attention/executive function. Are left-handed children with this disorder are more likely to have language disorders? Do left-handed children with specific language disorders have a deficiency in attention and executive function tasks? To answer these questions further studies will be needed.

The hypothesis of Arteaga and Poblano (2008) for hemispheric asymmetry development should also be emphasized. Considering the results of the subtests in the language domain, there were significant differences in verbal fluency subtest, WG. Left-handers presented a poorer performance than right-handers in the total semantic score (sum of animal, food and beverage categories) (p = 0.002). As well as the findings of Arteaga and Poblano (2008), in which left-handers took longer to recognize monosyllabic pseudowords and discriminated fewer trisyllabic words, the hypothesis is that incomplete hemispheric specialization may interfere with skills involving language in left-handers. Siengthai, Kritz-Silverstein, and Barrett-Connor (2008), in cross-sectional study with the elderly, found that left-handed women have lower verbal fluency than right-handers. In a longitudinal study conducted by Johnston et al. (2010), one of the tests used with right-handers and left-handers was the Memory for Digit Span Test (operating verbal and short-term memory), in which left-handers presented a poorer performance. The study suggests that other aspects of language, such as verbal memory, may be underdeveloped in left-handers.

Both Design Fluency as Word Generation are subtests that evolve time and productivity, visuomotor and semantics, respectively. It is important to note that both subtests evaluate functions in the Attention/Executive Functioning domain. In the Learning and Memory domain a difference was found for the subtest that assesses Memory for Faces (MF) (p = 0.045), nonverbal short-term test, in which left-handers presented the poorer result. Still referring to the study of Agostini and Dellatolas (2001), left-handers performed more poorly during the task that assessed prosopagnosia, which also involves the ability to recognize faces, as in the subtest used in the present study.

All subtests of cognitive domains that showed differences in the performance between left-handers and right-handers are related to anterior brain structures: prefrontal, premotor and temporal. This is an important finding as these structures tend to develop later.

Through the obtained results, it may be concluded that when comparing right-handers and left-handers with normal development, it was possible to find differences in performance in the subtests involving Attention and Executive Functioning, Language, and Memory and Learning, with no differences between gender, maternal education, socio-economic status and type of school. The predominance of performance differences in subtests of the Attention/Executive functioning, Language and Memory domains suggests involvement of anterior brain structures and development may be delayed. Questions can be raised from this finding: in brain development, may performance differences in these subtests disappear? If so, at what age will development compensate? What influences may these findings have on learning in and out of school? Further studies are needed to answer these questions.

The limitation of the present study was the small sample of left-handers. Since the data were taken from normative data, the frequency of left-handers was the general population. This limitation does not underestimate the value of data, since they derived from a population of individuals with normal development, participating in an extensive neuropsychological research.