The mechanisms of human cognition are universal although they manifest themselves in a particular way in each culture (^{Villar, 2016}). The situated cognition paradigm, linked to the sociocultural approach of Vygotsky, assumes that knowledge is relational, active, and practical, which is generated in the interaction with the environment, constituting cognitive adaptations according to ecocultural and ecological patterns (^{Restrepo, 2018}; ^{Veloso et al., 2016}).

Vygotsky's contributions have given rise to foundational theoretical and empirical developments in the framework of cultural psychology and anthropology of childhood, among which the LOPI model is highlighted (Learning by Observing and Pitching In to Family and Community Endeavors,^{Rogoff & Mejía-Arauz, 2022}), as well as to the theoretical and empirical contributions of ^{Gaskins (2020}), ^{Lancy (2016}), ^{Lave (2019}), ^{Lave and Wenger (1991)}, Rogoff (¹⁹⁹⁸, ²⁰⁰³, ²⁰¹⁴), Rogoff and Chavajay (²⁰⁰⁴), ^{Veresov (2020}).

From this perspective, an important criticism has been made regarding the teaching-learning process in school environment. For example, some of the educational materials used are decontextualized, and educational practices are based on a monocultural and hegemonic view of learning and child development (^{Quilaqueo & Torres, 2013}). This has implied, among other things, that the cultural ways of learning and the characteristics of cognitive processes in the indigenous context are considered a deficit or an obstacle to learning (^{Villar, 2016}).

The relevance of studying cognitive processes in context, such as working memory and spatial representation, is related to the importance of these processes in the curricular bases and subjects of the formal educational system. Memory plays an essential role as it is involved in reading, text comprehension, arithmetic, and calculus (^{Vernucci et al., 2017}).

^{Ardila and Ostrosky (2012}) define working memory as the ability to maintain information mentally while working with it or updating it, this ability allows remembering plans and instructions, using alternatives, relating ideas and data from the present with the future and the past. Conversely, spatial memory is defined as the ability to store and manipulate information simultaneously using three subsystems coordinated by a central executive, which functions as an attentional control system. The first subsystem is the visuospatial agenda, the second one is the phonological loop, and the third one is the episodic buffer (^{Baddeley, 2010}).

For the purposes of this article, spatial representation will be understood as the ability to organize and use information to orient oneself in space in a global and organized way; it considers spatial landmarks and place representation (^{Ávila & García, 2018}). This capacity is related to the navigational ability used to orient oneself in a given environment (^{Newcombe, 2018}). Navigation is the most complex domain, since it involves managing information to orient oneself in a given space, e.g., locations, knowledge of landmarks, routes, coordination of perspectives and movement. This area has received less attention due to its methodological complexity and the low availability of instruments for its evaluation (^{Newcombe, 2019}; ^{Nys et al., 2015}).

The classic studies by ^{Piaget and Inhelder (1956}) demonstrated the importance of studying how children develop spatial knowledge and perception. Furthermore, it has been shown that the interaction and familiarity of the individual with his or her environment, urban or rural origin, age, and gender are relevant aspects for the development of spatial knowledge (^{Dasen & Mishra, 2010}).

In Brazil, a study conducted by ^{Andrade and Amodeo (2007}) with 12 indigenous Guarani and 12 non-indigenous people, between 20 and 40 years of age, from São Paulo, indicates that non-indigenous people show a tendency to use verbal memory, while indigenous people are more proficient in visual memory elements and visual-constructive skills. In the same line, a study by ^{Fierro et al. (2018}) shows a decrease in verbal aspect scores in indigenous students. In addition, ^{Rogoff et al. (2015}) establish that the cognitive development of indigenous populations focuses on the predominance of holistic thinking and learning strategies based on observation, intense attention, and the preponderance of non-verbal linguistic patterns.

^{De León (1994}, ²⁰⁰¹) and ^{Brown and Levinson (2000}), in research with Tzotzil-speaking children in Mexico, showed that 5-year-old children had a linguistic mastery of the geocentric frame of reference before the egocentric one, in comparison with what is shown in research with Western children who, at the same age, are still unable to recognize the left-right dimension, a situation that, in the case of Mexican children, would be facilitated by the use of a geocentric frame of reference.

^{Enesco (1983}) conducted a cross-cultural study in an area of Peru in which she investigated the development of spatial concepts in 74 children who were between 6 and 15 years old and belonged to the Shipibo-Conibo ethnolinguistic group. Results showed some differences with respect to the performance observed in Western children. In fact, the hypothesis that topological relations prevailed over Euclidean relations in non-Western societies was refuted. On the other hand, in the coordination of perspectives, a state of spatial egocentrism was not present in this group of children.

Regarding rural and urban schools, studies such as the one conducted by ^{Anckermann (2014}), in which there were 40 participants who were between 4 and 5 years old and belonged to an urban private school and a rural public school in Guatemala, reveal that rural schools have low performance in the execution of verbal memory tasks compared to children from urban schools.

In Chile, the results of the study conducted by ^{Denegri (1991}), show a difference between Mapuche and non-Mapuche rural children, between 6 and 7 years old, in terms of the representation of spatial notions of conservation and length. Mapuche children perform better than their non-Mapuche peers, which, according to the author, can be explained by the ecocultural context, considering the socialization process and the knowledge acquired from the surrounding environment, since people conceptualize space according to the predominant frame of reference in their own culture and language.

The study by ^{Alonqueo and Silva (2012}) regarding spatial notions shows that non-Mapuche rural schoolchildren from Bío Bío and La Araucanía Regions use relative and absolute frames of reference, that is to say, spaces are defined according to the physical position and point of view of the individual and according to his or her own body map; at the same time, they use a large-scale geocentric system, such as the cardinal points. On the other hand, Mapuche children who were between 10 and 12 years old, compared to non-Mapuche children, mostly use the absolute frame of reference. This is consistent with the differences identified in both groups in the study by ^{Muñoz and Alonqueo (2017}), since Mapuche schoolchildren, when describing a known path, used more linguistic expressions that encoded an absolute frame of reference.

A recent study conducted in Chile compared the performance between rural and urban schoolchildren, by using the Wechsler Intelligence Scale for Children, fifth edition (WISC V), with a sample of 690 urban students and 47 rural students, demonstrating that the performance of visuospatial reasoning in urban and rural children of medium or low socioeconomic level did not show significant differences, presenting a similar development in both groups (^{Rodríguez-Cancino et al., 2019}).

There is evidence that Western schooling favors the development of the skills required for performance in cognitive tasks that do not take into account the daily context of children. Schoolchildren with low performance in experimental tasks show a good performance in their daily context, unlike what happens in assessment situations (^{Hein et al., 2015}).

The study of spatial representations and working memory has not been studied in rural Mapuche and urban non-Mapuche children, which is why it is a topic of particular relevance. Furthermore, the influence of the types of formats used in the cognitive tasks which assess these abilities has not been considered either. The use of a nonverbal format allows for the possibility that tests can reduce the influence of language, and nonverbal measures have also been shown to produce similar results in people from diverse backgrounds, thereby reducing cultural biases (^{Veloso et al., 2016}).

The results of empirical studies demonstrate the importance of continuing to explore cognitive processes in children, since in Latin America a significant gap in access to educational opportunities among schoolchildren belonging to indigenous peoples has been described in comparison to their non-indigenous peers (^{Treviño et al., 2017}). In the case of Chile, the indigenous child population corresponds to 13 % of the national total (^{Unicef, 2020}), and is mainly distributed in the regions of La Araucanía, Arica and Parinacota, and Aysén (^{Ministerio de Desarrollo Social, 2017}). The aforementioned inequities have also been described for the case of indigenous children living in the national territory, in particular for the Mapuche child population (^{Treviño et al., 2017}).

This idea is reinforced by the low performances obtained in the SIMCE tests and in the Weschler Scales to measure Intellectual Coefficient (IQ), determining the overrepresentation of indigenous children in the School Integration Program (SIP) (^{Fernández et al., 2017}). Mapuche students obtained lower performance compared to non-Mapuche students of the same age and educational level, being below the national and regional average, due to the use of standardized instruments with normative parameters (^{Agencia de Calidad de la Educación, 2016}).

The use of standardized instruments in the indigenous population presents limitations, due to the requirement of a high use of verbal language, influenced by the concepts of time and space (^{Dingwall et al., 2013}). The use of contextually organized materials results in fewer cultural differences in cognitive development (^{Garcés, 2018}); the principle of cultural familiarity operates in the format in which the material is presented, whether it be photographs, drawings or objects to be successfully addressed when they are part of the cultural environment and the structures of the activities are similar to those of everyday life (^{Villar, 2016}).

It is estimated that most of the standardized instruments lack ecological validity, because they do not consider the context in which the assessed behaviors are manifested (^{Sternberg, 2012}). Ecological validity refers to the degree of representativeness of a construct and its relation to the activities the individual performs in his or her natural environment. This type of validity allows the results to be generalized in a context different from their original environment (^{Tirapu-Ustárroz et al., 2011}). In accordance with the aforementioned, the American Psychological Association (^{APA, 2003}) establishes that the instruments must be culturally sensitive, which implies a linguistic adaptation of the instrument that guarantees the equivalence of the measurement in both cultures.

The available theoretical and empirical evidence shows that although cognitive processes such as working memory and spatial representation, in their respective verbal and non-verbal formats, are universal in human experience, the content and form of expression vary according to culture. Therefore, considering the formats is fundamental to overcome the inequities present in standardized assessments and in the content of the school curriculum.

Therefore, it is relevant to explore the following: are there any differences in performance in verbal and nonverbal working memory and spatial representation tasks between Mapuche and non-Mapuche schoolchildren? The general objective of this study was to determine the existence of differences in performance in verbal and nonverbal working memory and spatial representation tasks in Mapuche and non-Mapuche children. The specific objectives were: a) to measure performance in working memory tasks in verbal format in Mapuche and non-Mapuche children, b) to measure the performance in working memory tasks in non-verbal format in Mapuche and non-Mapuche children, c) to measure the performance in spatial representation tasks in verbal format in Mapuche and non-Mapuche children, and d) to measure the performance in spatial representation tasks in non-verbal format in Mapuche and non-Mapuche children.

Results

The results obtained by the participants in both groups for each of the dependent variables are presented in Table 1, displaying the means and standard deviations.

In order to test whether the differences in the scores of the two groups are statistically significant, a MANCOVA was performed controlling the effect of age and sex. The result of the analysis showed a main effect of the cultural group in the model (Λ = .820, F(4,89) = 4.89, p= .001).

Table 1: Means and standard deviations of performance on working memory and spatial representation tasks by cultural group 

In accordance with the specific objectives, in the first place, it was established that performance in verbal memory tasks differed significantly between Mapuche and non-Mapuche children. It was observed that Mapuche schoolchildren scored lower than non-Mapuche schoolchildren in verbal working memory tasks (F(1,92) = 13.52, p< .001, η²= .128).

In relation to the objective of measuring the performance in working memory tasks in non-verbal format in Mapuche and non-Mapuche children, it was found that the differences between both groups of schoolchildren are marginally significant (F(1,92) = 3.81, p= .054, η²= .040).

Regarding the performance in the spatial representation tasks, it was found that in the verbal format there were no significant differences in both groups of children (F(1,92)= 0.325, p = .570, η²= .004). However, in the spatial representation tasks in non-verbal format, Mapuche children scored higher than their non-Mapuche peers (F(1,92) = 6.62, p= .012, η²= .067).F1

Figure 1: Mean scores by cultural group