Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Solving Conceptual and Perceptual Analogies with Virtual Reality Among Kindergarten Children of Immigrant Families


by David Passig & Timor Schwartz - 2014

Background: The ability to think analogically is central to the process of learning and understanding reality and there is a broad consensus among researchers that we can improve this ability. Immigrants who have emigrated from developing to developed countries tend to experience tremendous challenges in their early years as immigrants. Their children often find themselves in a situation where it is clear that their low achievements are the result of cultural mediation, which expresses itself not only in a language gap, but also in cultural and basic technological disorientation.

Purpose: The goal of this study is to help find efficient ways of nurturing analogical thinking in children who have emigrated from developing to developed countries and express difficulties in analogical thinking, and to point out the advantages inherent in the use of immersive 3D Virtual Reality technology for this goal.

Population: The participants in this study included 56 children, aged 4 to 7 years, whose parents immigrated to Israel from Ethiopia during the last ten years. The experimental group (n=28) practiced solving analogies that were presented in 3D VR, while the control group (n=28) practiced solving the same analogies with a pictorial version of the items that was presented with cards.

Research Design: The research instrument employed for evaluating Analogical thinking was the CCPAM measure, which includes 10 questions on Conceptual Analogies and 10 questions on Perceptual analogies. We designed the intervention program according to the CCPAM test. The CCPAM test was administered in three intervals: a. Prior to the beginning of the intervention. b. Immediately after the intervention, which included two meetings of 15 minutes each, during which the children were given exercises in solving analogies. c. Three weeks after the end of the intervention, in order to test the ability to preserve the solution strategy (the follow-up test).

Conclusions: The results indicate that both programs of intervention—VR and picture cards— significantly improved the ability to solve both kinds of analogies—perceptual and conceptual. However, the children in the experimental group, who practiced analogies within an immersive VR environment, improved their ability to a statistically significant degree more than did the children who practiced solutions with picture cards. The children in the experimental group preserved the solution strategy three weeks after the intervention significantly better than the control group, and the improvement in solving conceptual analogies was greater than the improvement in solving perceptual analogies.


INTRODUCTION


Immigrants who emigrated from developing to developed countries tend to experience tremendous challenges in their early years in their new country (Shany, Geva, & Melech-Feder, 2010). This research focused on the Ethiopian Jews’ abrupt transition from village life in Ethiopia to Israel, which occurred en masse throughout the past three decades. This challenge has been accompanied by adjustment crises that have in turn immensely affected their learning and integration into Israeli society. Numerous reports show the low educational performance of this population and a high number of dropouts. Ethiopian students are disproportionately represented in special education streams (Berhanu, 2006). On the other hand, studies demonstrate that extracurricular activities with students of Ethiopian origin in order to improve academic abilities, both at the kindergarten stage and at any other stage in school, are highly successful and show that the majority of participants were capable of closing the gap between Ethiopian and Israeli students within a short period (Avinor, 1995; Berhanu, 2006; Kozulin, 2005).


One of the gaps reported in the literature refers to analogical thinking. Similar to the mathematical equation of relations, analogical thinking expresses the ability to identify relationships, e.g., A:B::C:D, which means that A relates to B in the same way that C relates to D (Sternberg, 1977). The ability to think analogically is central to the process of learning and understanding reality. Studies show that analogical thinking abilities at the kindergarten age are predictors of reading and math readiness (Tzuriel & Flor-Maduel, 2010).


According to Piaget (1969), the ability to identify the similarity between the relations of the analogy happens only between 11 and 12 years old. At younger ages children are inclined to think about analogical relationships in successive terms (this after that). They choose one relationship in order to make a connection between terms A and B of the analogy, and another relationship to make a connection between terms C and D. A conclusion of this sort is termed as a lower-order or first-order relational processing.


According to the initial Sternberg’s componential theory of human intelligence, the process of analogical reasoning includes a continuity of several stages from encoding stimuli at the beginning, to response at the end. The more experience the child has in solving analogies, and as age increases, the higher the level of efficiency those stages are employed. Piaget and Sternberg agree that analogical thinking develops as age increases, but according to Sternberg, some level of analogical reasoning already occurs at younger ages (Gentner, Holyoak, & Kokinov, 2001; Sternberg, 1977; Tzuriel, 2006).


In this study we designed an intervention program to improve the analogical thinking ability of kindergarten children of Ethiopian origin. Researchers that examined intervention programs to improve analogical thinking usually used pictorial modes of representation (Tzuriel, 2006).


Bruner described three modes of representation that appear in children in the continuum of their development (Bruner, 1973). The first mode that appears in early childhood is enactive mode, in which objects are represented by means of immediate sensory perception. The second mode that appears at kindergarten age is iconic mode, which requires the use of mental representations and is based on visual images such as pictures, models, and drawings. Iconic representation allows the child to recognize an object that has changed slightly—a mountain with or without snow at its summit, for example. This mode is important for achieving distancing from the enactive representation, and coming closer to the symbolic representation. The third mode is symbolic mode. It expresses conceptual understanding, which is seen also in spoken language. These representations are a sort of tool that people use when they are processing the information needed for transmitting their intentions, and in streamlining their ways of thinking.


Bruner’s model is related to the way the child develops representation modes. According to this model, the child can interpret a picture without difficulty if he has gone through the normative development stages without interruption (Bruner, 1973). In our study we claim that the children of immigrant Ethiopian families did not exercise enough interpretation of pictures in their early childhood stage. Therefore, the virtual reality (VR) exercises of 3D objects close the gap of the enactive stage they are lacking. Our assumption is that this concrete experience assists them in interpreting the pictures.


The unique nature of the VR technology, which we used in this study, lies in the way in which the concrete and three-dimensional presentation of the analogical terms is expressed. This can be attributed to its high level of interactivity, which creates an active learning experience, and in the unique interface of the Head Mounted Display (HMD), which empowers the experience of immersion and assists in the attentiveness of learning (Passig, 2010).


Our hypothesis is that training with VR technology will help to improve analogical thinking of children of Ethiopian origin significantly better then training with picture cards. It is based on earlier studies, which pointed out that training with immersive VR technology brought about improvement in thinking skills significantly better than other modes of representation (Passig & Eden, 2010).


THE ETHIOPIAN JEWS


The story of how the Ethiopian Jews—known as Falasha, meaning “gone to exile,”—survived for so many centuries in exile clinging to their Jewish tradition and how, finally, they came to Israel, is fascinating. As to their history and origin, there are contradictory statements and theories, and it has been an intensely debated issue, especially in Israel among different religious Jewish authorities, pertaining to, for instance, their rights to Israeli citizenship and the authenticity of their Jewishness. It was not until 1973 that they were officially accepted as having the right to “return” to Israel and become Israeli citizens. It is important to recognize that while in Ethiopia, the Ethiopian Jews have lived most of their lives in isolation, both in time and space. They have a singular, defined traditional way of raising and educating children (Berhanu, 2006).


The daily occupations of Ethiopian Jews generally did not require reading and writing, and therefore not many children were sent to school. Learning was executed through oral stories and songs—mainly through mimicking what the elders or teachers were saying or singing, not through exercising general thinking principles such as classification and analogical thinking (Kozulin, 2005). In Israel, immigrants were referred to Hebrew language classes. For many of them it was their first encounter with the written word in any language; that is to say they are illiterate in their mother tongue too. Even though the curriculum is tailored to the needs of this population, a large number of immigrants still lack the ability to read and write effectively. Moreover, most of the parents in these families are living under the poverty line, which adds to their hardship (Shany et al., 2010).


Similarly, the results of various recent studies indicate that kindergarten-age children from the Ethiopian community present statistically significant deficiencies in cognitive literacy, environmental, and language skills. Thus, some researchers maintain that there is a need for children of this community to be exposed frequently at an early age to children’s books, with the mediation of adults that covers conversation about the stories, the vocabulary, and additional contextual knowledge (Shany et al., 2010). As mentioned above, in order to help to find efficient ways to close those gaps, we chose one of the gaps reported in the literature—analogical thinking—and designed this study to examine a way to improve this ability in children from Ethiopian origin, using VR technology.


ANALOGICAL THINKING


A child’s ability to use analogical thinking is one of the measures of his or her cognitive abilities and level of intelligence. This ability is expressed in all aspects of life—from reaching conclusions and expressing them, to creating new hypotheses and theories (Inhelder & Piaget, 1958; Sternberg, 1977).


According to Piaget, analogical reasoning is characterized by thinking from the “second order” that emerges at the stage of formal operations only between the ages of 11 and 12 (Piaget, 1969). In the past few years, this theory has come under criticism (Goswami & Brown, 1989). Nowadays, numerous researchers are in agreement that children already exhibit analogical reasoning abilities at a younger age. This is contingent upon the manner by which the problem is presented to the children and the degree to which they are familiar with the terms of the analogy (Gentner, Holyoak, & Kokinov, 2001). In addition, studies show that with mediation and training, young children improve their ability to solve analogies (Tunteler & Resing, 2007).


The research literature distinguishes between two different kinds of analogies: conceptual and perceptual analogies. Conceptual analogies refer to semantic relationships between objects. A prior knowledge is necessary in order to solve them. In contrast, perceptual analogies express the relationship between visual perceptions and objects. In perceptual analogies all the relevant information is given in the framework of the task at hand, so that there is no need to access information from long-term memory. Various studies have indicated that there are differences between the ability to solve conceptual analogies as opposed to perceptual analogies (Tzuriel, 2006).


There is no consensus, however, over which kinds of problems are solved more successfully by children, or what factors allow for success in one kind or another. Different researchers attribute young children’s success solving different kinds of analogies to the method of the mode of representation, to age, to the subjects’ familiarity with the problem, and to the way in which the intervention was mediated (Goswami & Pauen, 2005). It would appear that there is a distinction between the manners of working through the different processes that are required for solving each of the different kinds. Moreover, many researchers maintain that each kind represents a different aspect of intelligence. According to this, it would seem that conceptual analogies express semantic relationships between objects, so that in order to solve them, prior knowledge is required, while perceptual analogies express visual, perceptual relationships. In that case, all the relevant knowledge must be given in a task, and there is no need to access knowledge from long-term memory (Goswami, 1992; Tzuriel, 2006).


Sternberg (1977) defined the process of analogical reasoning as including a continuity of several stages from encoding stimuli at the beginning, to response at the end of this process of thinking. First the child identifies the significant characteristics of each of the components of the analogy, and stores them in the working memory. Then he detects the relationship between the terms A and B of the analogy, and seeks the higher-order relationship between the first half of the analogy (A and B) and the second half of the analogy (A and C). He seeks the correct answer from among the presented elements. During his process of thinking, he maps out the relations between the term C and each of the elements. He uses the relations between terms A and B of the analogy and the decision made about them to help him make current decisions. He seeks to verify the better or best of the available options. At the end of this process, the child communicates a solution to the analogy. The more experience the child has in solving analogies, and as age increases, the higher the level of efficiency those stages are employed.


However, according to Piaget (1969), as opposed to Sternberg, reaching an analogical conclusion is a much more complex task, and the ability to perform it is developed only in the stage of formal operations: at the age of 11 to 12. According to Piaget, the ability to reach conclusions regarding relations begins to develop around age 7. At that age, children begin to solve problems that require sorting items into groups. The ability to build groups testifies to the ability to understand the connection between objects within a group. When the ability to understand the relations between objects exists, the ability to build new relations between these relations has begun to develop. In other words, the child begins to generate conclusions of a higher-order or second-order thinking skills, and to understand the principle of identity in relationships: identifying the similarity of the relations between terms A and B of the analogy and terms C and D of the analogy. According to Piaget that happens only between 11 and 12 years old. At younger ages children are inclined to think about analogical relationships in successive terms (this after that). They choose one relation in order to make a connection between terms A and B of the analogy, and another relation to make a connection between terms C and D. A conclusion of this sort is termed a “lower-order” or “first-order” relational processing. In Piaget’s opinion, higher-order relational processing demands a higher level of conclusion than does lower-order relational processing.


Over the years, a debate developed between proponents of Piaget’s approach and those of Sternberg’s approach. Researchers who criticize Piaget’s theory of development, for example, claim that it is possible that the youngsters who were examined had difficulty solving the analogies because it was hard for them to understand the relationships with which they were presented, and not because of any difficulty with the process of conclusion making. For the critics, the presentation of abstract relationships (such as black:white is similar to hard:soft, when the distracters are white-solid-blue) or the presentation of connections that are alien to the children’s world (e.g., auto:fuel is analogous to sailboat:wind, when the distracters are cruise-sails-rudder) is the reason for the children’s failure to solve the analogies. More than that, the utilization of pictorial cards is not sufficient to summon up the children’s analogical abilities, as they lacked relevant experiences that would allow the children to explain them (Goswami, 1992).


Two central approaches took form over the years among those who criticized Piaget’s research of analogical thinking. One approach is known as “relational primacy,” which sees the ability to reach conclusions as an innate ability that helps the child acquire knowledge in the first stages of life (Goswami, 1992). According to this approach, while the ability to think analogically develops as the child grows, with the aggregation of knowledge and new experiences, the change in the manner of thinking expresses itself in the areas of knowledge in which the child succeeds in making transitions, and not in the transition between stages of mapping.


The second approach, known as the “relational shift” approach (Rattermann & Gentner, 1998), maintains that children are able to think analogically at an age younger than that posited by Piaget. However, similar to Piaget’s approach, they say that the ability develops with the aggregation of knowledge, experiences, and the enrichment of language. That knowledge includes prepositions and morphological as well as compositional knowledge that the children need for formulating correct answers (Gentner & Rattermann, 1991). According to the relational shift approach, the stage in which the child identifies similarities between objects is the stage, like Piaget said, in which the ability to identify lower-order relationships appears (Inhelder & Piaget, 1958).


Studies that researched the development of analogical thinking emphasize the importance of exposing children to stimuli in an active learning experience for advancing those children’s analogical thinking (Goswami, 1992; Tzuriel, 2000). These studies also indicate the advantage inherent in visual presentation and in concretization by means of concrete concepts chosen from the children’s world for their achievements in understanding analogies. The dynamic assessment approach emphasizes the importance of meaningful experiences in mediated learning for the improvement of cognitive and learning skills. When the experience in mediated learning is inadequate for the child’s individual needs, gaps in learning achievement and in cognitive development occur (Tzuriel, 2000).


For the kindergarten children of Ethiopian origin, the utilization of pictorial cards was not sufficient to summon up the children’s analogical abilities, as they lacked relevant experiences to allow the children to explain them. The conclusions of these studies are that in order to test analogical thinking among young children, particularly among children of Ethiopian origin, there is a need to have them perform tasks adapted to their age and background, and to accompany them with the proper mediation that is relevant to the tasks (Gentner, Holyoak, & Kokinov, 2001; Goswami, 1992).


The basic claim underlying our study is that the contemporary diagnostic tools are built in such a way that the questions that are presented are viewed in a pictorial mode. However, the pictorial mode is best suited to Western cultures (Bruner, 1973), and it does not suit the needs of immigrant children without any mediation. We suspect that the children of immigrants from Ethiopia are not exposed to similar pictures as the children of Western cultures. Thus, when they are asked about something that initially requires some kind of decoding, their achievement in understanding the analogies are lower compared to children that do not need to decode any information before extrapolating the analogies embedded in the pictures. We believe, therefore that these achievements do not reflect the genuine abilities to understand analogies, and they are more linked to the representation mode than to their cognitive abilities.


In many Western homes children are playing with cards to exercise their abilities to memorize or sort things. Such cards were not in the possession of rural children in Ethiopia that used to play with other means at their disposal.


The need for cultural mediation we are pointing at refers to these aspects: the representation mode, and the familiarity with the type of questions. Therefore, we suggest that VR exercises of 3D objects close the gap of the concrete stage they are lacking. Our assumption is that this concrete experience assists children in interpreting the pictures, and helps them to become familiar with the type of questions.


ENHANCEMENT OF LEARNING POTENTIAL


The theory known as the enhancement of learning potential stresses the ability to change via mediating and training. According to this theory, young children could improve their abilities to solve analogies (Feuerstein, Feuerstein, Falik, & Rand, 2002). Thus, intelligence (considered to be an inborn trait) by itself does not promise efficiency in perception, thinking, and learning to problem-solve. For this theory, there are a limited number of basic thinking processes that, together with emotional, motivational, attitudinal factors, and certain work habits, make up the “basic, cognitive processes” that are necessary for perception, thinking, and methodical, efficient problem solving. Therefore, one must acquire the basic cognitive processes through learning in order to further develop his abilities to solve analogies.


From this point of view, it seems that there are two kinds of cognitive learning – direct exposure to things that happen, and mediated learning experiences – and every child needs to experience mediated learning. The quantity, the quality, the degree of intensity, the frequency, and amount of time needed for cognitive development to take place differ according to the individual differences of each child (for example, level of inherited genetic intelligence, soundness of the child’s sensory systems, emotional stability, and environmental support). When there is insufficient mediated learning for all the child's individual needs, the result is delay on cognitive development, or to the point of this study involving children from Ethiopian background, the development of the “socially impoverishment” syndrome, and relatively low social and educational achievements. When mediated learning experience answers the children's individual needs, the negative influences of various situations (poverty, mental retardation, emotional disturbance, and a low degree of parental education) on educational and social achievement are diminished to a significant extent. For the enhancement of learning potential theory, older siblings, grandparents, and parents have a vital role to play in providing mediated learning experiences in the course of cultural transfer across generations. Cross-generational transfer exists in all cultures, and in all cultures, there are elements that are crucial to the cognitive development with which children are provided. The failure to transfer ways of thinking typical of the culture is characteristic of the cultural impoverishment syndrome (Feuerstein et al., 2002).


When certain facets of cognitive development fail to receive sufficient stimulation through mediated learning experiences, it is possible to make up for the deficiencies with carefully planned, constructive teaching. In other words, according to this theory, teachers can exploit opportunities to provide mediated learning experiences, which parents have missed (Haywood & Lidz, 2007).


Studies demonstrate that extracurricular activities with students of Ethiopian origin, in order to improve academic abilities, both at the kindergarten stage and at any other stage in school, are highly successful and show that the majority of participants were capable of closing the gap that existed between Ethiopian and Israeli students within short period (Avinor, 1995; Berhanu, 2006; Kozulin, 2005).


Some researchers have been claiming that it is possible to describe and reconstruct mediated learning processes. For them, these processes make up an important style of teaching that is identified as a mediated teaching style. Mediating the children’s learning experiences includes functions such as choosing stimuli, helping children to minimize the number of stimuli available to them, focusing on the relevant aspects of the available stimulus, multiple exposures to important stimuli, perception of shared and different denominators, cause and effect, shared elements in different experiences, and operations such as comparison, classification, the continuation from the past to the present, and to the future, and transfer (Haywood & Lidz, 2007). From this theory the dynamic assessment model developed (Tzuriel, 2000).


Similar ideas are found in the approach of Zone of Proximal Development (ZPD) (Vygotsky, 1978), and in the theory of scaffolding (Wood, Bruner, & Ross, 1976). It seems that the instruments for testing the abilities for solving different kinds of analogies were developed on the basis of these different theories, according to the traditional approach of static testing (such as the Raven test, Raven, 1965), and, according to the theory of  mediated learning by dynamic testing (Haywood & Lidz, 2007; Tzuriel, 2000).


Researchers who emphasized the importance of the way in which the analogies were presented to the subjects referred to the advantage inherent in concrete, three-dimensional, full color representations that are game-like, and can be manipulated when the analogical problems are presented. They found that when a child is able to perform motor manipulations with parts of the test and to change them until one reaches a satisfactory solution, the achievements tend to increase (Tzuriel, 2000). According to various researchers, the process of change that the child makes in reaching an answer reflects the process of thinking and indicates the cognitive factors that contribute to the solving of a problem or interfere with finding the solution (Tunteler & Resing, 2007; Tzuriel, 2000).


For this reason, we had chosen, in this study, to examine how VR technology can advance achievements in conceptual and perceptual analogies. The VR intervention was based on the model in Passig and Eden’s study (2010). The analogical terms were expressed by concrete and three-dimensional presentation in high level of interactivity that created an active learning experience. The unique interface of the Head Mounted Display (HMD) empowered the experience of immersion, and assisted the attentiveness of learning (Passig, 2010).


The main hypothesis was that there would be a significant improvement in both groups of participants’ achievements after the intervention. We hypothesized that the level of improvement would be higher in the experimental group, who would practice analogies using 3D IVR technology, than in the control group, with children who would practice analogies using pictorial version of the items. We hypothesized this would be true both immediately after the intervention and three weeks following the end of the intervention.


This hypothesis based on the dynamic assessment model suggests that after a teaching and mediating phase we would find an improvement in the participants' achievements (Tzuriel, 2000), but as mentioned above, with the aid of IVR technology, we assumed the improvement in the achievements will be significantly higher (Passig, 2010).


The second hypothesis was that significant interaction would be found between the participants’ scores in the conceptual analogies test and the participants’ scores on the perceptual analogies test, in the different methods of training, before and after the intervention.


This hypothesis is based, as mentioned above, on various studies that indicate that there are differences between the ability to solve conceptual analogies as opposed to perceptual analogies (Tzuriel, 2006). Different researchers attribute success in solving different kinds of analogies in young children to the method of the mode of representation, to age, to the subjects’ familiarity with the problem, and to the way in which the intervention was mediated (Gentner & Rattermann, 1991; Goswami & Pauen, 2005).


Finally, the third hypothesis was that we would find a correlation between the subjects’ ages and their success in solving analogies. The assumption was that the older children would achieve higher scores than those achieved by the younger children, in both kinds of analogies.


This hypothesis is based on the relational shift approach. It suggests that children are able to think analogically at an age younger than that posited by Piaget, but similar to Piaget’s approach, it says that the ability develops with the aggregation of knowledge, experiences, and the enrichment of language, and increases with age (Gentner & Rattermann, 1991).


METHODS


THE PARTICIPANTS


The participants in this study included 56 kindergarten children whose parents immigrated to Israel from Ethiopia during the last 10 years. We communicated with the children’s parents with the help of a liaison assistant – a woman of the Ethiopian community who spoke both Amharic and the native language, who is usually employed in this capacity throughout the school year. In order to obtain the parents’ agreement for their children’s participation in the research, a demonstration meeting was prepared, in which the teachers and the parents were able to experiment putting on the 3D gear and to play with the virtual worlds. After this meeting, the parents’ signatures were collected, expressing their consent to their children’s participation in the research. Only children whose parents agreed to their participation took part in the research.


The children were divided into research and control groups only after the first set of tests, as described below in the research procedure section. Each group consisted of 28 preschoolers. Since the children were from immigrant families, some of them repeated the preschool year as part of the adjustment process that the teachers recommended for them. Thus, some of them were 6 years old and above. The age range of the participants in the research group was 58 to 80 months (4 years and 10 months to 6 years and 6 months). The average was 70.75 months (5 years and 10 months), with a standard deviation of 6.02 months. The age range of the participants in the control group was 57 to 80 months (4 years and 9 months to 6 years and 8 months). The average was 70.78 months (5 years and 10 months), with a standard deviation of 6.43 months


THE RESEARCH INSTRUMENTS


The research instrument we employed for evaluating analogical thinking was the CCPAM measure (Tzuriel and Galinka, 2000), which includes 20 questions of a pre-learning test and 20 questions of a post-learning test.


We also used the classic Raven’s colored matrices test (Raven, 1965) to evaluate the child’s level of intelligence.


RAVEN'S COLORED MATRICES TEST


Raven's test has a high reliability and validity as a measure for intelligence. This test is based on abstract shapes. The child has to choose the most appropriate form to complete the matrix. Using that, the test evaluates abstract thinking abilities such as analogical thinking. There are three series in the test, A, AB, and B, where each has 12 items. We used only series B, which focuses more on analogical thinking. Each correct answer is worth 1 point, so the range of scores was from 0 to 12.


THE CCPAM TEST


The conceptual and perceptual analogies subtest from the CCPAM test that we employed was developed by Tzuriel and Galinka (2000). There are two versions of this test: closed analogies and construction analogies. Both versions are based on Piaget’s developmental theory (1952), Vygotsky's concept of the ZPD (1978), and Feuerstein et al.’s theory of mediated learning experience (2002).


We used the closed version, which includes classic analogies in a pictorial mode. We graded the answers based on correct or incorrect solutions. Each section of the test had 20 items, where 10 items were administered in the pre-learning phase, and 10 administered in the post-learning phase. Each problem was presented in a 2 x 2 matrix (A:B::C:D) in the colored pictorial modality. In each item, the problem was presented at the top of the page, while four possible answers were presented underneath (Figure 1). The child was asked to think about the relationship between the first pair of pictures presented in the problem, and match the picture to a second pair out of the four, which were presented, according to the relation between the pictures of the first pair. For example, the conceptual analogy of bird:nest::dog:doghouse is parallel to the analogy of bee:beehive::parrot:cage, in the pre-learning phase that present identical relationships. In each subtest, the children were given two examples prior to the beginning of the test in order to give them an example of what was expected of them.


Figure 1. A sample of items from the CCPAM (Tzuriel & Galinka, 2000)

[39_17339.htm_g/00002.jpg]                     [39_17339.htm_g/00004.jpg]

Item from the perceptual test                Item from the conceptual test


While Raven’s test is based on abstract shapes, the CCPAM, however, is based on daily objects that are familiar in Western cultures and less so in Ethiopian culture, such as a knife to cut bread. This point further emphasizes our research question of whether the use of pictures without proper mediation for diagnosing the level of analogical thinking is proper for such immigrant children.


CONCEPTUAL ANALOGIES TEST


In the original test of conceptual analogies (Tzuriel & Galinka, 2000), there are three types of connections: categorical relations, part-whole relations, and functional relations. In this study, both the amount of time devoted to administering the test and the practice session were limited to 15 to 20 minutes, in order to not push the children beyond their kindergarten-age attention span (Tzuriel & Galinka, 2000). For that reason, only 10 of the original items on the test were selected from Series A, and 10 from Series B. Eight of the 10 in each series had functional relationships, and 2 had part-whole relationships, but they were selected in such a way as to be understood on the functional level as well. For example, a lid covers the pot, and the roof covers the house. We decided not to use the items that were based on categorical knowledge and focus instead solely on functional relationships. This was because the goal of mediating children’s conceptual analogical thinking in this study was to mediate the syntax and morphological tools they needed for solving analogies (Gentner & Rattermann, 1991). Each test item was based on a short sentence whose subject was shared with the analogy parallel to it. The sentences were presented orally by the mediator during the intervention. The focus of the mediation was the child’s formulation of a parallel question, and finding a matching answer, i.e., identifying the shared subject (for example, “A eats B. What eats C?”). During the test, the child formulated a sentence with the solution on his own, and pointed to the correct answer from among four given alternatives. The distracters were objects that have a relationship with C that was different from the relationship between A and B. If the relationship between A and B was functional in one way, then the distracters relate to C through a different functional relationship, or were relating through categorical relationships or part-whole relationships. During the intervention, after the child formulated the relationship between A and B, the researchers mediated to the child the way to reject the answers that didn’t have the same relationship as the one between A and B, and to choose the object that has a relation that is parallel to the relationship between A and B. During the test the child was asked to point at the right answer and explain the reason for his choice.


PERCEPTUAL ANALOGIES TEST


The perceptual analogies test was based on Goswami’s Geometric Forms Test (Goswami, 1992), but instead of using geometric forms, the forms in our test were based on objects that can be named, for example, a table, a ball, flowers, et cetera. In every analogy, the relationship between the analogies’ terms were based on three types: (a) differences (change in color, place, number, or type of object); (b) presence (whether the object is present or missing); and (c) position (for example, the object is on the chair or under the chair).


It must be noted that despite the conceptual analogies being presented in a visual mode, they are considered conceptual because understanding the relationship between them is not based solely on looking at them. The answer requires understanding of the abstract principles that connect them. In contrast, perceptual analogies are based primarily on relationships that are easily distinguished visually, although a certain degree of conceptual understanding is required (for example, understanding that a ball is on or under a table requires the grasp of the difference between on and under, even though making that distinction is essentially conceptual).


Each correct solution in the perceptual analogies test and in the conceptual analogies test received a score of 1 point. The maximum score was 10 points in conceptual analogies, and 10 in perceptual analogies.


Reliability of Cronbach’s α, which was tested in a number of different kindergarten samples was .64 - .74 on the pre-learning conceptual analogies measure, and .70 - .85 for the post-learning measure. For the perceptual analogies, the pre-learning results were .83 - .87, and for the post-learning .90 - .91 (Tzuriel & Galinka, 2000; Tzuriel, 2000, 2006).


The instrument's external validity can be found in a number of different studies (Tzuriel, 2006). One of them tested the predictability of reading and writing skills of kindergarteners using this instrument. The authors have found that the scores in both subtests of the analogical thinking tests (CCPAM) predicted 19% of the participants’ achievements in the reading and writing tests beyond the other variables such as age and gender (Tzuriel & Flor-Maduel, 2010).


These findings underscore the importance of doing research on the subject of analogical thinking among kindergarten-age children, as well as devoting resources to develop innovative ways of nurturing these thinking skills.


RESEARCH PROCEDURE


We first administered the pre-intervention tests. After realizing that there are no significant differences in the Raven scores among the control and research groups we engaged in the intervention program. They were 28 children in each group that were drawn from 4 kindergartens. In 2 kindergartens the children experienced VR worlds and in the other 2 kindergartens the intervention employed pictures.


These 4 kindergartens were chosen through the opportunity sampling method. We were looking for clusters of children from Ethiopian families and we found a convenient cluster of kindergartens in the Tel-Aviv metropolitan area in which many of the children were of such a background.


We did not include in the sample children who were reported by their teachers to be ADHD and as such received reinforcement by an Amharic language speaker.


The tests and the intervention program were administered in a quiet corner of the kindergarten library under the supervision of the kindergarten staff. The researchers set the hardware needed for the study in the corner. This included a computer and an HMD (Head Mounted Display), with which the children experienced the 3D immersive worlds. The same set-up was provided in the kindergartens in which the program was administered to the control group, but without the HMD. One child in the research group reported difficulties putting on the HMD, and therefore was removed from the study.


The number of meetings and their length was as follows: First we met once to administer Series A of the two tests—conceptual and perceptual. In the first meeting we administered Raven’s test too. After a week we met for two consecutive meetings during which we practiced the analogies. The duration of each meeting was about 15 minutes. In the third week we met to administer Series B of the tests. Three weeks later we administered Series A again.


Table 1 shows the timeline of the meetings with the children and the kind of test or intervention that was administrated in each meeting.


Table 1. Research Plan

Sixth

week

Fifth

week

Fourth
week

Third week

Second
week

First
week

Type of intervention

CCPAM -  Series A

-

-

CCPAM -  Series B

2  intervention meetings of 15 minutes each

Raven &

CCPAM -  Series A

VR

CCPAM -  Series A

-

-

CCPAM -  Series B

2  intervention meetings of 15 minutes each

Raven &

CCPAM -  Series A

Pictures


The order of the tests was similar in both groups. The children were initially tested in the conceptual and then in the perceptual test. The order of the questions was also similar. The study did not aim to compare the two tests, therefore we decided to simplify the procedure and employed a similar order, and the results were analyzed as cumulative scores, so the order of questions was the same for all participants.


The children were given feedback for the mistakes or correct answers only during the intervention and not during the tests. The mediation during the intervention included the reference to the above relationships.


THE INTERVENTION


The learning phase for both groups, the experimental group (N = 28) which practiced the solution of analogies that were presented in 3D VR, and the control group (N = 28) which practiced the solution of the same analogies with a pictorial version of the items, in cards, consisted of the use of 6 items from among the analogical tests from the pre-learning phase, and included the following strategies:


Looking for the required dimension for solving the analogy (for example, in perceptual analogies, distinguishing between the dimensions of color, size, quantity, or kind of object)

Mapping the transformational rules (for example, in perceptual analogies, some of the elements change, but some remain the same in A and B)

Verbal expression of the solution before making a choice

Justifying the correct answer and the incorrect one

Methodical searching for the correct answer (looking at all the alternative answers)


The questions and syntax in the mediated learning were equated between the experimental and control groups.


At the first meeting, Series A of the CCPAM of analogies test was administered by means of pictures. During the next two meetings, which took place one week later, the participants practiced solving analogies for 15 minutes. The experimental group practiced the solution with the use of VR technology while the control group practiced with the use of pictures. One week later an additional meeting was held, in which Series B of the CCPAM test was administered. Three weeks later, the participants did the A series once again (see Table 1).


We chose two meetings of 15 minutes each for practice reflecting the conclusions of those who developed the instrument (Tzuriel & Glinka, 2000). According to their findings, short but intensive training using the pictures produced a significant improvement in the children’s ability to solve analogies. They concluded that examination should also be made following an interval of time after the intervention. An acclimatization period was taken into account, bearing in mind the children’s unfamiliarity with VR technology. In order to make a comparison of the research conditions, we decided that the number of practice sessions with VR would be the same as the number of practice sessions for the control group, who were working with picture cards. Our final decision was that the intervention phase, for the control group and the experimental group, would include two practice meetings of 15 minutes each, and that the ability to preserve solution strategy would be tested by administering Series A a second time, three weeks after the end of the intervention (the follow-up test).


THE INTERVENTION IN THE EXPERIMENTAL GROUP


The VR intervention was based on the model in Passig and Eden’s study (2010). The first intervention meeting with the children in the experimental group was devoted to becoming acquainted with the VR interfaces.


In this initial intervention meeting, first we examined the children’s ability to control the mouse, by pressing the buttons and dragging the cursor in a drawing program. Then, to introduce the child to the HMD (Figure 2) we fit the instrument to his head with an elastic strap tied on to the back. This balanced the HMD on his head and enabled him to be immersed in the virtual worlds without external distractions.


For each exercise we have built a different VR world with its corresponding objects that were presented to the child in 3D through the HMD and were monitored through the laptop screen too, in order to let the researchers follow the child’s actions and to guide him. The experience in the VR worlds involved having the eyes of the children being covered by the HMD, thus they could not see the researchers. In order to communicate with them we used a special application called ActiveInspire, with which we could draw the attention of the children to a special place in the VR world without having them to take of the HMD. For example, when we thought that it is appropriate to draw the attention of a child to the similarity of the colors, we were able to draw a circle around the objects with ActiveInspire’s drawing tools.


Figure 2. The Head Mounted Display (HMD)

[39_17339.htm_g/00005.jpg]


THE VIRTUAL WORLDS OF THE CONCEPTUAL ANALOGIES


In each of the worlds we built for the experimental group, the first presentation was of two objects, A:B, between which a relationship existed; one another object C; and, next to it, an empty space to fill. The relation between A and B was concretized by an action in the virtual world. In order to match a solution D to C the child was asked to describe verbally his actions in the virtual world. The sentences were formulated orally by the researcher and the child. The intervention included having the child identify the similarity between two sentences and activities. For example, chalk is drawing on a blackboard; paint is drawing on a piece of paper. The similar word is “drawing.”


While the activity was going on, we broadened the conversation to touch on the possibilities of using additional actions such as writing and coloring. The researchers helped the child write his or her name on the board in the 3D virtual world using the white chalk. Thereafter, the child was able to color using the colors on the painting page on the screen in the virtual world. The activity’s goal was to sharpen analogical thinking by enriching the child’s world of concepts, and provide him with experience formulating appropriate syntax while minimizing the overload of information. For that, we emphasized sentences in which the action word was the same, as in the example mentioned above: the chalk draws on the blackboard the same as the markers draw on the paper.


After this phase, four different objects were presented on the screen of the virtual world. They included the solution and three distracters (incorrect answers). Each of the distracters has a different relationship to C such as categorical, functional, or part-whole. The child was asked to choose the object that matched the analogical solution with the same relation to C as B to A. The goal of this phase was to practice the ability to transfer and reject incorrect answers. He was then asked to formulate the matching sentence independently, to choose the object, and to drag it to the proper place (Figure 3).


Figure 3. Practicing the chalk and the blackboard analogy in 3D VR world. The chalk draws on the blackboard, just as the colors draw on the paper.

[39_17339.htm_g/00007.jpg]  [39_17339.htm_g/00008.jpg]


In the second exercise (see Figure 4), a plank of wood appeared, and next to it a saw, then a loaf of bread appeared, and next to it a blank space. After naming the items, the child was asked to imagine what was missing from the empty space. Afterwards, four objects were presented: a pair of scissors, a knife, a fork, and the bread. By using the mouse, the child was able to manipulate and play with the objects, while making mistakes and formulating the appropriate sentence with the researchers. The following actions were presented in this way: the saw saws the wood, the knife cuts the bread, the scissors cut the paper, and the fork spears a vegetable. In the next phase, the child was asked to choose the object that should be dragged to the empty space next to the bread.


In the course of the conversation with the child during the VR exercises, the common action was emphasized – the act of cutting – and the common subject, the word "cut,” was said out loud. Stress was placed on the matching pairs: knife-bread, saw-plank, scissors-paper. The correct answer in this exercise was the knife. In addition, the child was given the opportunity to build a different analogy: if, instead of bread, there were paper, what would be the right answer? Scissors, of course. Special attention was given to the understanding of the question: Are we looking for something unusual, or a parallel action? In other words, it is an analogy. The goal of this phase was to practice employing the strategy of rejecting incorrect answers.


Figure 4. Practicing the saw and bread analogy in 3D VR world. The saw cuts the plank just as the knife cuts the bread.

[39_17339.htm_g/00010.jpg] [39_17339.htm_g/00012.jpg] [39_17339.htm_g/00013.jpg]


THE INTERVENTION IN THE CONTROL GROUP ON CONCEPTUAL ANALOGIES


The control group practiced solving analogies by using cards on which items identical to those practiced by the experimental group were depicted. The exercises were performed individually as a game with the researchers for 15 minutes with each child. In the course of the game, each child was asked to name the items in the picture, and to formulate sentences that were questions and answers in accordance with the instructions. Emphasis was placed on identifying similar questions and answers, while focusing on shared subject words – the same as in the experimental group described above.


The participants in the experiment group and in the control group first practiced the conceptual analogies as described above, and then practiced the perceptual analogies.


THE INTERVENTION IN THE EXPERIMENTAL GROUP ON PERCEPTUAL ANALOGIES


In the beginning, the participants were presented with a paved open space in 3D VR world, which looked like a large and open plaza (Figure 5). Two tables were placed in the center of the space, parallel to one another. A red telephone was located on the table on the left, and an identical telephone was located under the table on the right. Another table was placed in front of the table on the left, upon which a purse was set. The child was asked to determine what was supposed to be in the space to the right of the table with the purse on it. After looking at the items, naming them, describing their color and location, on the right side of the open space appeared the following items: an oblong table which was identical to the three tables which had already been placed in the open space, a round table, a phone, and a purse. The child was then asked to drag the appropriate objects to the empty place. The right answer was: the oblong table, with the purse underneath it.


Figure 5. The telephone exercise: perceptual analogies in virtual reality


[39_17339.htm_g/00014.jpg]


In this phase, the child actively built the analogy and was able to test additional possibilities. It was also possible to move around in the open plaza, to approach objects and to move away from them, and even to change the position from which the observer views them. All the actions were performed with the accompaniment of verbal explanations and the naming of concepts of color, shape, and relation.


We employed in this research an additional helping tool in order to better focus the child’s attention on the task—a drawing tool from the ActiveInspire software. Because the children were immersed in the VR worlds through the HMD, they were not able to see the researchers who conversed with them. Therefore, the visual communication in the 3D VR world between them and the researchers was mediated using this instrument. When the researchers wanted to signal a child, and to emphasize the similar and different objects, they signaled him on the screen by means of the drawing tool.


To progress to the next phase of the exercise, the child had to click on one of the keys on the keyboard with the help of the researchers, and then he was presented with four possible answers. They were, essentially, four different combinations of objects on the screen. The child had to identify the correct combination and drag it to the empty place on the screen.


In every exercise that the experiment, the group identified two similar and two different characteristics. Besides the color trait that appeared in all the virtual worlds, in other exercises the child had to identify the condition of presence versus absence of objects (the ball and telephone exercise: Figure 5); quantity of objects (the cake exercise: Figure 6); and the difference between two kinds of the same type of object (the fence and flowers exercise: Figure 7).


Figure 6: The cake exercise: perceptual analogies in a 3D VR world

[39_17339.htm_g/00015.jpg]


Figure 7. The fence and flowers exercise: perceptual analogies in a 3D VR world

[39_17339.htm_g/00016.jpg]


The research assumption in developing an intervention in perceptual analogies in virtual reality was based on a theoretical model that points to the advantage inherent in the use of means which are concrete, 3D, and colorful, and which have an air of manipulating control and games about them. The child can carry out motor manipulations in parts of the test, and can change them until he arrives at a satisfactory and correct decision. The process of changing the answer reflects the thinking process, and indicates the kind of cognitive factors, which contribute to or hinder finding a correct answer (Tzuriel, 2000). In the instructional and mediating phases, the researcher can identify those factors that could, with proper instruction, impart thinking skills to the child. The goal of conceptualizing actions while performing them is to give meaning to the concepts of relations, to the names of colors, and to the counting of the items.


THE INTERVENTION WITH THE CONTROL GROUP ON PERCEPTUAL ANALOGIES


As with the conceptual analogies, the control group practiced solving analogies by using cards (Figure 8), which showed pictures of items identical to those on the test. The exercises that were presented were identical to those presented to the experimental group. The exercises were conducted with each child individually, following practicing with conceptual analogies for the same time period. They included a game with one of the researchers. In the course of the game, the child was required to name the items in the picture and to describe their characteristics, while using prepositions, counting, and the names of colors. The emphasis was on the identification of the similarities and the differences between the pictures while using the same syntax as in the experimental group.


Figure 8. A sample of game cards

[39_17339.htm_g/00018.jpg]                                    [39_17339.htm_g/00020.jpg]

Conceptual analogies:                                              Perceptual analogies:

The chalk and board exercise                                  The ball and telephone exercise


RESULTS


In order to examine the reliability of the tests we did a first analysis, in which we found that the coefficient of Cronbach’s α in the conceptual analogies before the intervention was α = .48; directly after the intervention: α = .83; and three weeks after the end of the intervention (the follow-up test): α = .83.


Similarly, the Cronbach’s α on the perceptual analogies test before the intervention was α = .61; immediately after the intervention: α = .76; and three weeks after the intervention: α = .86. The tests therefore became more reliable with training, suggesting the children’s level of understanding increased with random guessing.


These findings support the dynamic assessment model (Tzuriel, 2000) on which the instrument is based.


In order to test the research hypotheses regarding the impact of the intervention program on the experimental group in comparison with the control group, two separate ANCOVA analyses of difference were performed, with repeated measurement, one for the conceptual analogies, and one for the perceptual analogies, while taking level of intelligence into account by means of the Raven test (Raven, 1965). A 2 x 3 matrix of measurements was carried out, where 2 expressed the kind of intervention: intervention using pictures in comparison with intervention using VR; and 3 expressed the repeated measurements over time: before intervention, immediately after intervention, and three weeks after the end of intervention (the follow-up test). Averages, standard deviations, and sample size are presented in Table 2.


Table 2. Averages and Standard Deviations Distributed According to Kind of Intervention and Type of Test (n=28)

Pictures

Virtual Reality

 

SD

Mean

SD

Mean

Time

1.47

4.42

1.57

4.39

Conceptual

pre-intervention

1,83

4.96

2.02

8.35

Conceptual post-intervention

1.88

6.35

1.54

8.39

Conceptual

follow-up test


Pictures

Virtual Reality

 

SD

Mean

SD

Mean

Time

1.75

2.78

2.39

2.96

Perceptual pre-intervention

2.15

4.57

2.01

6.00

Perceptual post-intervention

3.22

4.32

2.76

5.39

Perceptual follow-up test



In the analyses of differences we found that the interaction between the time factor and the kind of training was significant, F(1,53) = 6.25; p < .01; eta² = .10. In other words, the improvement in the experimental group was significantly higher than that in the control group, in both the conceptual and perceptual tests.


The interaction between the kind of test and the kind of training was also found to be significant, F(1,53) = 3.96; p < .05; eta² = .07. In other words, the subjects responded differently to the intervention in the conceptual and perceptual tests. The level of achievements in the conceptual test was higher than the level of achievement in the perceptual test as a result of the intervention.


Similarly, the three-sided interaction among the time factor, the kind of test, and the kind of training was found to be almost significant: F(1,53) = 3.33; p < .07; eta² = .05, meaning that there were differences as a factor of time, the kind of training, and the type of test.


Post Hoc range test analysis was performed for the 2-way interactions, as it refers to all the means in the analysis to get a meaningful explanation for the results. The analyses pointed to five sources of the differences in the interactions:


1. Immediately after the intervention, significant differences were found between the experimental group and the control group in the conceptual test (M = 8.35 in the experimental group, compared with M = 4.96 in the control group), and in the perceptual test (M = 6.00 in the experimental group, compared with M = 4.57 in the control group).


2. Three weeks after the end of the intervention (at the follow-up test), significant differences were found between the experimental and control groups only in the conceptual test (M = 8.39 for the experimental group, compared with M = 6.35 in the control group). In the perceptual test no significant differences were found three weeks after the termination of the intervention.


3. Similarly, it was found that in the control group significant differences were found on the conceptual test between the pre-intervention (M = 4.42) test and the test administered three weeks after the intervention (M = 6.35), but not between the pre-intervention and post-intervention results (M = 4.96).


4. In the control group, on the perceptual tests, differences were found in the achievements before the intervention and the achievements three weeks after the termination of the intervention (M = 4.32).


5. In the experimental group significant differences were found among the results of the pre-intervention conceptual test (M = 4.39), the results achieved immediately after the intervention (M = 8.35), and the results obtained three weeks after the intervention’s end (M = 8.39). Similarly, significant differences were found among the results of the perceptual test pre-intervention (M = 2.96), the results immediately after the intervention (M = 6.00), and at the follow-up test (M = 5.39). No significant differences were found for the achievements immediately after the intervention and the achievements obtained three weeks after the intervention.


Regression analyses were calculated to examine additional factors, such as age and gender, as predicted variables of the level of achievements, after the intervention. (Tables 3 and 4).


Table 3. Summary of the Regression Analyses of the Predicted Variables: The Score in the Perceptual Follow-up Test (percentages)

Predictor variables

In the experimental group

In the control group

Perceptual test pre-intervention

-

14.6**

Perceptual test post intervention

30**

54.3***

Raven

-

-

Age

-

5.7*

Gender

-

-

*p < .05    **p < .01    ***p < .001


Table 4. Summary of the Regression Analyses of the Predicted Variables: The Score in the Conceptual Follow-up Test (percentages)

Predictor variables

In the experimental group

In the control group

Conceptual test pre-intervention

-

-

Conceptual test post-intervention

38***

-

Raven

9.1*

31.1**

Age

-

-

Gender

13*

-

*p < .05    **p < .01    ***p < .001


DISCUSSION


The main hypothesis was that there would be a significant improvement in the participants’ achievements after the intervention in the experimental group and in the control group, but that the level of improvement in the participants’ analogical thinking tests would be higher in the experimental group, with children who would practice solving analogies using 3D VR technology than in the control group, with children who would practice solving analogies using pictorial version of the items, both immediately after the intervention, and after a period of three weeks following the end of the intervention.


The findings supported this assumption (Table 2). There was a significant improvement in the experimental group and in the control group achievements after the intervention, but the improvement in the experimental group was significantly higher than that in the control group, in both the conceptual and perceptual tests.


As described above, the CCPAM tool with which we assessed the analogical thinking contains 10 questions of conceptual analogies and 10 questions of perceptual analogies. The conceptual questions are based on functional relations. For each question the children were presented with 4 optional answers. Each of them represented a different relationship to the object. The correct answer was the one that pointed to its functional relationship. The distracting answers expressed categorical relationships, part-whole, or identity relationships.


The data collected from the answers before the intervention indicate, as reflected by the low figure of Cronbach’s α analyses, that the children guessed on their answers more than they understood what they were answering. We did not find a pattern for their mistakes. However, in the conceptual test after the intervention, we could see significant improvements in the experimental group that also were significantly higher than in the control group, both immediately after the intervention and after a period of three weeks following the end of the intervention (Table 2). We could also see that the mistakes were mainly in their understanding of relationships, and there was a tendency to find functional relationships among objects in both groups. In the perceptual test we could also see that the level of improvement was significantly higher in the experimental group than in the control group (Table 2).


The similarities in the relations of the perceptual analogies were based on two visual characteristics of the objects, for example, location and color. The distractions were: a picture similar to B, a picture similar to C, and a picture that does not express any relations. The mistakes that the children made were more about pointing to pictures similar to B or C. This pattern of mistakes was much less significant in both groups after the intervention.


These results place our study together with earlier studies, which pointed out that training brings about improvement in thinking skills (Turiel, 2000), but with immersive VR technology the improvement can be significantly higher (Passig, 2010; Passig & Eden, 2010; Passig, Eden, & Bezer, 2009).


The theoretic assumption at the base of these studies was that the advantage of VR technology is its ability to make abstract concepts concrete. This is the nature of presentation in immersive 3D VR, which is unique to this technology.


Also the ability to explore/manipulate was the key to the intervention in our study with VR. The language mediation was carried out through oral description of the actions during their execution. The oral description was carried out while the pictures were presented to both groups, but in the control group the children were asked to imagine the actions by interpreting the pictures, and that’s what made the difference. As we underlined above, the basic claim of our study is that the pictorial mode is best suited to Western cultures and does not suit the needs of immigrant children without any mediation. We suspect that the children of immigrants from Ethiopia are not exposed to similar pictures as the children of Western cultures. Thus, when they are asked about something that requires initially some kind of decoding, their achievement in understanding the analogies are lower compared to children that do not need to decode any information before extrapolating the analogies embedded in the pictures.


In addition, our study joins other studies in the field of the development of analogical thinking in early childhood, which demonstrated that when analogical problems were presented to children in the form of images which have concrete meanings to them, the children succeed in solving the analogies (Goswami, 1992; Goswami & Pauen, 2005; Tzuriel, 2000).


As we remarked above we also assumed that participants who practiced solving conceptual and perceptual analogies with VR would preserve solution strategies better, and would present a higher improvement in solving analogies even a number of weeks after the termination of the intervention in comparison with participants who would practice solving analogies solely with the aid of picture cards. Here, also, the findings supported the research assumption (Table 2).


We did not find many cases reported in the research literature that tested the ability to preserve solving strategies over time after the end of an intervention with mediating for analogical thinking. Most of the research touching on this topic tested the influence of short-term intervention, while conducted with the dynamic diagnosis approach. One such a study conducted recently, tested the ability to preserve the strategies for solutions with analogical thinking in children aged 5 to 7 (Tunteler & Resing, 2007). That research used a microgenetic research method to study the process of change in the domain of analogical reasoning with geometric forms tasks. They note that the variability in children’s strategies indicates that the ability to reason by analogy on this type of task develops over a protracted age range. They underlined the importance of a microgenetic research method in order to gain more insight into the nature of young children’s analogical reasoning ability.


Our research focused on the method of intervention. We suggested that the method of intervention influenced significantly the ability to preserve the strategy of thinking that solving conceptual and perceptual analogies needs, even over time following the intervention. Our study showed that mediation via the means of the VR technology, which was built according to the recommendations of earlier research, and which took into consideration the needs of kindergarten-age children of the immigrant community from Ethiopia, brought about improvement in the participants’ achievements, even after the elapse of a period of time from the date of the intervention.


The second research assumption was that significant interaction would be found between the participants’ scores on the conceptual analogies test and the participants’ scores on the perceptual analogies test, in the different methods of training, before and after the intervention. The research findings indicate that this assumption was corroborated by indicating an advantage for the experimental group on the conceptual test, as opposed to the perceptual test. The subjects reacted differently to the intervention in the conceptual and perceptual tests. The level of achievements on the conceptual test was higher than the level of achievements on the perceptual test following the intervention.


It should be noted that the beginning level of scores on both of the tests were not identical, and indicated a relative advantage on the part of the conceptual. The average scores in the total sample of the conceptual test before the intervention were M = 4.23; SD = 1.84; and in the perceptual test before the intervention M = 2.89; SD = 2.08.


Similarly, differences were found as a result of the time factor, according to the kind of training and type of test.


Significant differences were found in the conceptual test in the achievements of participants who practiced with VR, between the test that was administered before the intervention, and the tests administered immediately after the intervention and three weeks later. Since they preserved the achievement level, no statistical differences were found between the test that was administered immediately after the intervention and the test administered three weeks later. In contrast, in the control group the statistical differences were recorded only between the test that was administered before the intervention and the test, which was conducted three weeks after the intervention. In any event, the achievement level was significantly lower than that of the experimental group, three weeks later as well.


To the best of our understanding, the improvement we have seen in the conceptual scores of the control group after three weeks from the intervention indicates that learning analogies with pictures was also effective. However, it took some time to surface. One possible reason that the control group improved only on the conceptual test at follow up can be the fact that the test administered after three weeks was similar to the test administered initially. Thus, it is possible that this has contributed to the improvement of the scores in both groups. However, it could be also due to the fact that after three weeks the control group absorbed the logic behind conceptual analogies and thus improved their scores when they were retested.


On the perceptual test, both groups improved their achievements significantly in the test conducted immediately after the intervention. It is important to note that the achievements of the experimental group were significantly higher than those of the control group. On the test that was administered three weeks after the intervention, the control group preserved its achievements, so that there were no significant differences between the test conducted immediately after the intervention and the one conducted three weeks after the end of the intervention (the follow-up test). In contrast, in the experimental group, there was a lower level of achievement in the follow-up test. Nonetheless, the achievements of the experimental group were higher in comparison with those of the control group, but the difference in achievement level was not statistically significant (Table 2).


The author of the CCPAM (Tzuriel, 2000) has indicated that he did not find a great difference when sizes of the samples with which he developed and established the reliability of the tools were different. We believe that the size of the sample was the reason the results were not statistically significant.


The differences between the conceptual and perceptual tests could stem from a number of factors. The first factor is the process of absorbing the different kinds of information needed for the two tests, and from a different kind of intelligence that takes place in these processes.


When solving a perceptual analogy, there is a different process of working with information. The mapping process that takes place puts the information into a classification matrix that demands an ability to place things into a hierarchical order. According to Piaget (1969), this ability appears only at about the age of 8 or 9, and still relates to objects and concrete situations. Abstract understanding of this sort appears later, between the ages of 11 or 12. According to Gentner and Rattermann (1991), age is an important factor in the appearance of those abilities. In their opinion, however, the presentation of objects and concrete situations to children can lead even younger children to success in solving analogies of this sort. Problems of matrix classification in abstract forms are typical of Series B in Raven’s colored matrices, which we used as a control variable. Problems of matrix classification of concrete objects characterize the perceptual analogies test of the CCPAM (Tzuriel & Galinka, 2000). It should be noted that the Raven test was not found to be a general predictor in the perceptual test. We suggest putting the mode of representation in focus again. Thus, while the Raven test is based on abstract shapes, the perceptual analogies at the CCPAM test are based on objects that can be named; for example, a table, a ball, flowers, et cetera. and on relationships which are easily distinguished visually, although a certain degree of conceptual understanding is required (for example, understanding that a ball is on or under a table).


Obviously these findings contribute to the discussion about the level of understanding that kindergarten children have with analogies as perceived with different modes of representation and different type of analogies.


Compared with perceptual analogies, a mapping of the knowledge that the child uses for solving conceptual analogies is based on prior knowledge and on familiarity with the terms of the analogy. For example, in the question of the chalk and the board, the child has to understand from the pictures that chalk and a blackboard are before him, and that there is a connection between them. The connection is that with the chalk one may draw on the blackboard. Afterward, when the child is presented with a picture in which colored markers appear, and when he is asked: Which object is missing next to the markers? he must understand that the missing object is a piece of drawing paper, just as one uses chalk to draw on a board. The child must be able to bring the pictures to life, and to tell a story about them. Research findings indicate that active experiencing with VR has a significant advantage to interventions based on verbal mediation concretized by the use of pictures. This is similar to the findings of studies that examined the influence of interventions that employed VR, as opposed to interventions that used pictures, on the ability to arrange events in the correct, chronological order (Passig, Eden, & Bezer, 2009).


In our opinion, the second factor that may lead to differences in the level of improvement, which was achieved by the intervention with VR in both tests, was the degree to which we matched the intervention we built to specific information overload processes, as noted above. It appears that we succeeded in focusing the children’s thinking on the information required for solving conceptual analogies. Emphasis was placed on finding the similarity in the structure of the sentence. Formulation of the similar sentences was performed during the activities in virtual worlds. This method helped the children to solve successfully the tests following the intervention.


At the same time, it’s possible that in the VR worlds in which the children were trained for solving perceptual problems, there was an overburdening of data and information. It may be that the large amount of training sessions—four for the perceptual test, with only two for the conceptual test—caused confusion and became an obstacle to focused thinking.


The third factor that may lay behind the differences in the tests maybe located in differences in the computer skills needed for the training in VR worlds, and in the various kinds of interactions. At the time of the intervention in VR the children donned their HMDs, through which they saw the screen display. The activities in the VR worlds were performed by moving the mouse. This situation required the participants to be able to control a mouse without being able to see it. The intervention in VR for the conceptual test included two exercises. In the chalk and board exercise, the children had to operate the program’s drawing tools, to draw on the blackboard, and to color the drawing paper. The children were quite familiar with these operations, and most of them demonstrated relative mastery of the situation. In the second exercise, we used animation in order to concretize the activity by pressing on a key on the keyboard. This activity, as well, was easy to carry out with the help of the mediator that led the child to press on the keyboard. On the other hand, the practice in VR worlds for the perceptual analogies test consisted of four exercises. The computer skills expected from the children were pressing on the left side of the mouse, selecting and dragging objects with the mouse, turning the objects around by simultaneously pressing a key on the keyboard, and dragging objects with the mouse, approaching and distancing themselves by scrolling backward and forward with the moving wheel of the mouse. We note here that most of the children achieved partial control of the mouse, and needed help in controlling it. It could be that the encounter with new technology, overly focusing on using the mouse, and on mastering the computer skills that the exercises required interfered with the thinking processes, and thereby caused a lowering of the level of improvement more than was the case with the conceptual test.


In summary, there is a consensus among researchers that there are differences between the ability to solve conceptual analogies, as opposed to perceptual analogies, in kindergarten-age children. There is no consensus, however, regarding which of the kinds of problems are solved more successfully by the children, or what are the factors which allow success in one kind or another (Gentner & Rattermann, 1991; Goswami, 1992; Goswami & Pauen, 2005; Tzuriel, 2006).


According to the results of the present research, it is evident that worlds which were built in VR were an effective tool for conceptual thinking, and also for perceptual abilities, although clearly less so.


We feel that it is correct to point out these differences for the sake of continued study in the future.


Finally, the third research assumption was that we would find a correlation between the subjects’ ages and their success in solving analogies. The assumption was that the older children would achieve higher scores than those achieved by the younger children, with both kinds of analogies. The age range of the participants was 4 years and 9 months old to 6 years and 8 months old.


This assumption was based on many studies which indicated that the ability to solve analogies improves with increased age, but opinions are still divided as to whether the ability of analogical thinking develops with age, or if it is a innate ability (Goswami & Pauen, 2005). However, there is a consensus among researchers that the more a child accumulates knowledge and experience, the more his abilities to conclude and to solve analogies will improve (Tunteler & Resing, 2007).


In order to test this assumption, we performed a regression analysis and examined the variables that predicted success in the tests. We decided to examine the participants’ achievements at the follow-up test, as the achievements would express the degree of effectiveness of the intervention.


The results show that age was found to be a predictor only in the perceptual test only in the control group, and with a level of prediction of only 5.7% (Tables 3 and 4).


That shows that the means of intervention have a crucial influence, even more than age, and that focused mediation directed at the individual child’s needs, both with VR technology and with picture cards, prepare children for successful solving of analogical thinking tasks.


SUMMARY


This study addressed the abilities of kindergarten-age children to solve classic analogies – conceptual and perceptual. A review of the research literature raised a question whose answer has yet to achieve consensus: At what age does this ability appear? According to Piaget’s 1969 theory of child development, analogical thinking is abstract thinking, which develops relatively late in childhood, between the ages of 11 and 12 years. Piaget maintains that children of kindergarten age are in the stage of concrete operations, and have difficulty in understanding relationships. Many researchers who came after him take issue with his theory. Today, many agree that a certain level of analogical thinking appears as early as kindergarten age. Beyond that, the findings of studies performed recently show that analogical thinking abilities at kindergarten age are predictors of reading and math readiness (Tzuriel & Flor-Maduel, 2010), and that places the importance of developing analogical thinking from early childhood ages.


In the same direction, the results of the present research supported today’s prevailing opinions that analogical thinking appears as early as kindergarten age. Moreover, it is possible to improve significantly the abilities of kindergarten children of an immigrant community to solve conceptual and perceptual analogies in a short time, with the help of focused mediation and VR technology.


In this respect, the present study joins the ranks of studies that examined the contribution of intervention by means of VR for developing cognitive abilities, and supported its degree of effectiveness. It is clear that there is a need for continued examination of this question in relation to additional cognitive abilities, and for finding ways of making this technology available for educational needs.


References


Avinor, E. (1995). Academic difficulties and early literacy deprivation: The case of Ethiopians in Israel. Language, Culture and Curriculum, 8, 201-210.


Berhanu, G. (2006). Intercultural mediation, the institution of learning and the process of educational ‘integration’ and assimilation: The case of Ethiopian Jews in Israel. Educational Research and Review, 1(3), 62-79.


Bruner, J. S. (1973). The growth of representation processes in childhood. In J. Alin (Ed.) Beyond the information given: Studies in the psychology of knowing (pp. 313-324). New York: Norton.


Feuerstein, R., Feuerstein, R. S., Falik, L. H., & Rand, Y. (2002). The dynamic assessment of cognitive modifiability: The Learning Propensity Assessment Device, theory, instruments, and techniques. Jerusalem: ICELP.


Gentner, D., Holyoak, K. J., & Kokinov, B. N. (2001). The Analogical mind:  Perspectives from cognitive science. Cambridge, MA: MIT Press.


Gentner, D., & Rattermann, M. J. (1991). Language and the career of similarity. In S. A. Gelman, & J. P. Byrnes (Eds.), Perspectives on thought and language: Interrelations in development (pp. 225-277). London: Cambridge University Press.


Goswami, U. (1992). Analogical reasoning in children. Hillsdale, NJ: Lawrence Erlbaum.


Goswami, U., & Brown, A. L. (1989). Melting chocolate and melting snowmen: Analogical reasoning and causal relation. Cognition, 35, 69-95.


Goswami, U. & Pauen, S. (2005). The effects of a ‘family’ analogy on class inclusion reasoning by young children. Swiss Journal of Psychology, 64(2), 115-124.


Haywood, H. C., & Lidz, C. S. (2007). Dynamic assessment in practice: Clinical and educational applications. New York: Cambridge University Press.


Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from childhood to adolescence. New York: Basic Books.


Kozulin, A. (2005). Immigrant parents’ educational aspirations for their children and the required family support system: A lack of confluence. Journal of Cognitive Education and Psychology, 4, 352-361.


Passig, D. (2010). The future of VR in education: A future oriented meta analysis of the literature. Themes in science and technology education, special issue on virtual reality in education, 2(1-2) 269-293.


Passig, D., & Eden, S. (2010) Enhancing time-connectives with 3D Immersive Virtual Reality (IVR). Journal of Educational Computing Research, 42(3), 307-325.


Passig, D., Eden, S. & Bezer, M. (2009) Improving the sequential time perception of teenagers with mild to moderate mental retardation  with 3D Immersive Virtual Reality (IVR). Journal of Educational Computing Research, 40(3), 263-280.


Piaget, J. (1952). The origins of intelligence in children. New York: International University Press.


Piaget, J. (1969). The psychology of the child. London: Routledge & Kagan.


Rattermann, M. J. & Gentner, D. (1998). More evidence for a relational shift in the development of analogy: Children’s performance on causal-mapping task. Cognitive Development, 13, 453-478.


Raven, J. C. (1965). The colored progressive matrices. London: Lewish & Co.


Shany, M., Geva, E. & Melech-Feder, L. (2010). Emergent literacy in children of immigrants coming from a primarily oral literacy culture. Written Language & Literacy, 13(1), 24-60.


Sternberg, R. J. (1977). Component processes in analogical reasoning. Psychological Review, 84, 353-378.


Tunteler, E. & Resing, W. C. (2007). Effects of prior assistance in using analogies on young children’s unprompted analogical problem solving over time: A microgenetic study. Educational Psychology, 77(1), 43-68.


Tzuriel, D. (2000). Dynamic assessment of young children: Educational and intervention perspectives. Educational Psychology Review, 12, 385-435.


Tzuriel, D. (2006). Transfer effects of teaching conceptual versus perceptual analogies. Journal of Cognitive Education and Psychology, 6(2), 194-217.


Tzuriel, D., & Flor-Maduel, H. (2010). Prediction of early literacy by analogical thinking modifiability among kindergarten children. Journal of Cognitive Education and Psychology, 9(3), 207-226.


Tzuriel, D., & Galinka, E. (2000). The Conceptual and Perceptual Analogical Modifiability (CCPAM) Test: Closed analogies—instruction manual. Ramat-Gan, Israel: School of Education, Bar-Ilan University.


Vygotsky, L. S. (1978). Mind and society. Cambridge, MA: Harvard University Press.

Wood, D., Bruner, J., & Ross, G. (1976). The role of tutoring in problem solving. Journal of Child Psychology and Psychiatry, 17, 89-100.




Cite This Article as: Teachers College Record Volume 116 Number 2, 2014, p. -
https://www.tcrecord.org ID Number: 17339, Date Accessed: 1/23/2022 4:10:54 PM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • David Passig
    Bar-Ilan University
    E-mail Author
    DAVID PASSIG is an Associate Professor at the school of Education at Bar-Ilan University in Israel. He is the Director of the Graduate Program in Information and Communication Technology and Education. He teaches graduate courses and conducts research on Educational Futures, Future Technologies, Social Systems Theories, Futures’ Methodologies, and Virtual Reality. He also heads the Virtual Reality Lab aimed at researching and teaching Virtual Reality in Education. He is engaged in studying the cognitive effects of VR in educational settings. Here is a sample of recent papers dealing with his recent work: Passig, D. (2011). The impact of Virtual Reality on educators’ awareness of the cognitive experiences of a dyslectic student. Teachers College Record, 113(1), 181-204. Passig, D. and Eden, S. (2010). Enhancing time-connectives with 3D Immersive Virtual Reality (IVR). Journal of Educational Computing Research, 42(3), 307-325.
  • Timor Schwartz
    Bar-Ilan University
    E-mail Author
    TIMOR SCHWARTZ is a researcher at the Virtual Reality Lab at the school of Education, Bar-Ilan University, Israel.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS