Children's Perceptions of the Learning ProcessbyGILLIAN VALERIE BICKERTONB.Ed., The University of British Columbia, 1977A THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSinTHE FACULTY OF GRADUATE STUDIES(Educational Psychology and Special Education)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIASeptember 1993© Gillian Valerie Bickerton, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of ^5 P 5 EThe University of British ColumbiaVancouver, CanadaDate^0 Gt • g,^1 q q3DE-6 (2/88)ABSTRACTThe present study set out to investigate children's perceptions of themselves as learners.An understanding of how learning occurs in children is of central importance for planningcurriculum and designing instruction. Although recent research is providing a more completepicture of the learning process, what is apparent is the relative absence of any in-depth inquiry intowhat "learning" itself means to children.The objective of the study was to determine if there are stages of development in children'sunderstanding of the learning process, such that children of the same age think in a similar fashionabout learning and there is increasing complexity in understanding with age. A neo-Piagetianmodel of intellectual development (Case, 1985,1992) was used as a theoretical framework.A random selection of 83 children from a suburban elementary school participated in thestudy. The sample included children at the ages of 6 years, 8 years, 10 years and 12 years. Theseages were specifically chosen because they represent the mid-points of the substages ofdevelopment in middle childhood and beginning adolescence proposed by Case (1985,1992).Children's explanations for the meaning of learning were elicited in a semi-structuredinterview, modelled after Piaget's technique. Specific questions were designed to extrapolatechildren's conceptions of two related aspects on learning in terms of a definition for the meaning oflearning and a definition for the "source" of learning, that is, whether learning comes from aninternal or external source or both.The results of the study are based on qualitative and quantitative analyses of children'sresponses transcribed for this purpose. Scoring criteria reflect the levels of structural complexityhypothesized by Case (1985, 1992) in which children progress from simple conceptions oflearning defined in terms of intentionality, in other words, learning as a behavioural act combinedwith an internal feeling or judgment state to more complex notions of intentional behaviour as theyrelate to the process of learning. In early adolescence, learning is defined as an internal state ofmind, denoting a shift from concrete to abstract thinking. An interpretive understanding of thelearning process characterized responses at this age level which is consistent with the postulates ofthe theoretical model. A statistical analysis of the responses showed significant differencesbetween each age group.IIThe complexity of the "source" responses also increased with age. In the early stages,there was a clear distinction between external and internal sources. For older children, there wasan awareness of learning taking place in a sequential manner from external to internal.Interrelatedness of the two sources was generally recognized by 12 years of age.The pattern of understanding was age-related and hierarchical, consistent with theoreticalpredictions ( Case,1985,1992). By revealing common, age-typical patterns of understanding ofthe learning process, this study suggests that educational methods and materials should beconsistent with children's levels of conceptual development. Such procedures would help solve"the problem of the match" (Donaldson,1979) between a learner's developmental stage andinstructional methods.The study presents an educational perspective on the metacognitive task of reflecting aboutone's own learning. If the goal in education today is to assist children in becoming independentlearners, educators need to first of all understand children's conceptions of learning and theneducate from the child's point of view.TABLE OF CONTENTSPageAbstract ^Table of Contents ^ivList of Tables ^viList of Figures ^viiCHAPTER ONE: INTRODUCTION ^1Objective of the Study ^3Significance of the Study 3Research Questions ^4CHAPTER 2: REVIEW OF THE LITERATURE ^5Perspectives on Learning and the Learner ^5Theories of Mind 9Metacognitive Awareness ^10Externalization Hypothesis 11Children's Understanding of Mental Verbs ^11Research in the Piagetian Tradition 16Piagetian Theory ^16Methodology in the Piagetian Tradition 18Summary ^21Research in the neo-Piagetian Tradition 24Intentional Knowledge in Middle Childhood ^26Interpretive Knowledge in Adolescence 31Bruner's Social Perspective on Learning ^32A Constructivist Perspective on Learning 34Conclusion ^35Hypotheses 35ivTABLE OF CONTENTS (continued)CHAPTER 3: METHODOLOGY ^37Pilot Study ^37Procedures ^38Subjects ^38Interview ^39Scoring ^41Scoring Criteria ^42Scoring Levels ^47CHAPTER 4: RESULTS ^55Analysis of the "Definition" Task ^55Analysis of the "Source" Task ^59CHAPTER 5: DISCUSSION ^69"Definition" Task ^71"Source" Task ^78Summary of the Results ^80Summary ^82Conclusion ^87APPENDIX A: Stages and Substages of Case's neo-Piagetian Model ^88of Cognitive DevelopmentAPPENDIX B: Letter to Parents Requesting Permission ^89APPENDIX C: Student Interview Protocol ^90APPENDIX D: Setting the Stage 91REFERENCE LIST ^92VLIST OF TABLESPageTable 3.1.^Descriptive Statistics for Age Group^ 39Table 4.1^Mean (SD) Level Scores by Age Group 56(Predicted and Obtained) on "Definition" TaskTable 4.2^Analysis of Variance Summary Table on^ 56"Definition" TaskTable 4.3^Percentages of Responses Falling At, Above and^ 58Below the Prototypic Level for Each Age GroupTable 4.4^Mean (SD) Scores on the "Source" Task by Age Group^60Table 4.5^Analysis of Variance of Summary Table on^ 61"Source" TaskTable 4.6^Pattern of Performance of Each Individual Subject's^ 68Response to the "Source" Task in Relation to Scoreobtained on "Definition" TaskviLIST OF FIGURESPageFigure 1.^Means Scores Achieved by Four Age Groups^57on "Definition" Task M Relation to thePredicted Linear TrendFigure 2.^Mean Scores Achieved by Four Age Groups^62on "Source" Task in Linear FormFigure 3.^Distribution of "Source" Responses (Age 6)^64Figure 4.^Distribution of "Source" Responses (Age 8)^65Figure 5.^Distribution of "Source" Responses (Age 10)^66Figure 6.^Distribution of "Source" Responses (Age 12)^67vii1CHAPTER 1: INTRODUCTIONThe nature of learning has been the subject of an extensive body of research over the pastforty years. An understanding of how learning occurs in children is obviously of centralimportance for planning curriculum and for designing instruction. A missing factor in curriculumdevelopment has been the apparent lack of consideration of learning from a child's point of view.This has been accompanied by a relative absence of research into children's perceptions of whatlearning is. Consequently, there is a clear need to search for a better understanding of the meaningof learning from a child's perspective.Educators in British Columbia today are facing a major paradigm shift towards a newconceptualization of what learning is and what education should be. The old belief system viewedthe learner as a passive recipient of knowledge, fostering a sense of dependency upon someoneelse. Learning is now postulated from the cognitive constructivist viewpoint. From thisperspective, learning means the active construction of knowledge through interactions with othersin the environment. In other words, the learner is perceived as an active creator of understandingin the sense that learning is more self-directed. One of the factors providing impetus to the newparadigm is current research about how children learn. With the implementation of the BritishColumbia Ministry of Education document, Year 2000: A Framework for Learning (1990),teachers are examining education in a new perspective and are attempting to build a neweducational construct.For example, "learning dimensions" defined as knowledge, skills and attitudes developalong a continuum over time and "increase in complexity, maturity, and level of sophistication"(p.13). If this is the case, then it should follow that children's perceptions and conceptions ofwhat learning is should also increase in complexity, maturity and level of sophistication. Inaddition, the document also states that "generalizations can be made about the learning of a specificage group" (p.8). This raises the question, "Do children of similar ages have a common patternof thought or conceptualization regarding the concept of learning?" For instance, do the responsesof younger children indicate a limited conceptualization of learning relying on experiences that areimmediate? Subsequently, do older children demonstrate a more elaborate and personalizedresponse and has their concept of learning become more refined?Over the past decade, educators have been turning to cognitive psychologists for answers2to the question, "How do children learn?". Unfortunately, they have been overwhelmed with amyriad of contradictory findings; the conclusion being that there is no straightforward answer tothis question. Behaviourism, for example, defines learning as the "passive acquisition of facts,skills and concepts usually through the processes of drill, guided practice, rewards andpunishments" (Marshall, 1992, p.6). Rote memory is considered to be a form of learning in thebehaviourist tradition. The information-processing approach to learning portrays the mindoperating like a computer (Slavin 1986, cited in Marshal1,1992). According to this view, the mindactively operates on information (input) and produces responses (output). Its "mechanistic"interpretation to learning, combined with its emphasis on individual learning, indicates that thistheory has its roots in behaviourism. In contrast, a number of cognitive approaches to learninghave provided alternative views to the behaviourist perspective.What much of today's educational research indicates is that former premises aboutchildren's learning, both how they learn and what they need to learn, are no longer valid in today'sworld and a new vision of what learning is and how it occurs is needed (Hilbom,1990). Previousconceptions of learning assumed that learners were simply "empty vessels to be filled withknowledge" (Marx, cited in Research Forum, 1990).Marx claimed that "knowledge acquisition is not just the accumulation of information; itinvolves construction by the learner" (p.9). This new belief supports the cognitive constructivistview. For Piagetians, learning always involves construction and comprehension. In consequence,rote memory or memorization is not a part of intellectual development from a Piagetian perspective,since memorization does not require comprehension. Very briefly, constructivists believe thatindividuals actively construct knowledge and understanding, rather than passively receiveinformation in response to external forces such as rewards. Learning is constructed in the sensethat children develop an understanding by building on what they already know.Construction of knowledge by the learner is one of the core postulates of a neo-Piagetianmodel of intellectual development (Case, 1985, 1992). According to this theory, learning is bothconstructed and strategic. It is constructed in the sense that children develop understanding bybuilding upon what they already know about the world. "Learning is strategic in the sense that thissearch for understanding is not random" (Marx, 1990, p.9). Learners follow particular approachesto develop a conceptual understanding. This theory of intellectual development accepts the basicpsychological tenets of Piaget's theory that "children are highly active thinkers, who are constantly3engaged in constructing ever-improved models of the world around them" (Case & McKeough,1990, P. 835). This theory also supports the notion that "children's spontaneous cognitiveactivity leads to stage-characteristic understandings or models of the world that transcend anyparticular task" (Case & McKeough,1990, p.837). In other words, Case's model preserves auniversal sequence of cognitive-developmental stages and substages which are age-related.However, one way in which the theory differs from Piaget's is in the detail and manner inwhich the cognitive structures are defined at the different stages. Case's neo-Piagetian theory isbased on variations of both cognitive and social constructivism. It takes into consideration thatlearning is constructed not only in the individual's head, but in social interactions amongindividuals or between individuals and materials over time (Bruner,1986; Vygotsky, 1978). Assuch, it seems a valuable framework within which to investigate how children conceptualize thelearning process.OBJECTIVE OF THE STUDYThe objective of this study was to provide an opportunity for children to verbalize theirown thoughts about what learning means to them and the types of learning strategies employedduring the learning process. When students engage in conversations like these, they are likely toview thinking as an integral part of lifelong learning because they are included in the process andrecognize that their perceptions are also valued.Case's (1985, 1992) theory was used as a framework not only to gain more insights intothe possible hierarchical development of children's conceptions of learning from the ages of 6-12years, but also to determine whether there are prototypic behaviours at different age levels.SIGNIFICANCE OF THE STUDYAlthough recent research is providing a more complete picture of the learning process, whatis becoming most apparent is the relative absence of any form of in-depth inquiry into what"learning" itself means to children. In other words, the most natural resource - the child - hasbeen left untapped in our attempt to answer this question. Research has offered sophisticatedtaxonomies of different possible approaches to learning, but these are essentially descriptive andnot explanatory when it comes to articulating the child's understanding of the nature of learning4itself.The commencement of any educational reform, as proposed by the Year 2000 document,must begin by rethinking the nature of learning. There is often a tendency to overlook the centralrole played by the learner as well as the different conceptions of how children learn. Investigationinto learning from the child's perspective should be the starting place for improving education.RESEARCH QUESTIONSTwo general research questions guided the study:1. Are there stages of development in children's understanding of the Learning Process; that is,do children of the same age have a common pattern of thought or conceptualization regardinglearning?2. Are there cross-age differences in children's conceptualization of the meaning of learningsuch that children's responses increase in complexity with age?Literature pertaining to these research questions will be reviewed in the following chapter.5CHAPTER 2: REVIEW OF THE LITERATUREThe major goal of the study is to determine whether or not children's conceptual levels ofunderstanding of the learning process change systematically with age in a progressive mannerconsistent with neo-Piagetian stages of cognitive development as described by Case (1985).Relevant literature is reviewed under four major headings describing perspectives on learning andthe learner, theories of mind, research in the Piagetian tradition, and research in the neo-Piagetiantradition.Perspectives on Learning and the LearnerLearning is a complex and challenging mental process. Since learning is not onephenomenon but many, depending on what one is learning about, the context itself plays anintegral part in the learning process. The difficulty with understanding the process of learning isthat it functions at a variety of levels depending upon the type of situation in which it occurs.Learning is an interactive process, a constant engagement with the environment and with otherpeople. Social forces are at work steering a child's early attempts at learning, for example,parents, teachers, language, customs, and the media.Piagetian and Vygotskian theories raise two different aspects of learning. On the one hand,Vygotsky recognized that a key factor in social learning was the young person's ability to learn byimitation while Piaget built a strong case for children being able to independently construct theirown knowledge through their active interaction with the environment. In other words, Piagetbelieved children are capable of developing self-constructed schemes in their attempts to makesense of the world. In contrast, Vygotsky stressed that much of what we learn we learn fromothers and that individual development and learning are influenced by communication with othersin social settings, a fact that Piaget seemed to downplay. To acquire a more complete picture ofthe learning process there is a need to consider both social and individualist aspects of learning.In the works of Margaret Donaldson (1978) there are two very clear messages. The firstconcerns the need to consider the whole child when trying to discover a child's capabilities orunderstanding. The second is the need to consider the situation from the child's point of view.Traditionally, research into human learning is conducted from a "first-order perspective" (Marton& Svensson, 1979). This means that the emphasis is placed on the researcher's description of6children's learning behaviour. Marton and Svensson define this perspective thus:We the researchers observe the learner and describe him as we seehim and we observe the learner's world and describe it as we see it.We frequently relate our description of the student to our descriptionof his world and generally do this within an explanatory framework.(p. 472)A more recent approach to human learning has shifted from a descriptive explanation tofirst-hand information of the learner's point of view; in Marton's words (1981) a "second-orderperspective." With this type of research, information is extrapolated from the learner and thencategorized by the researcher to reveal the various ways in which people view or experienceimportant aspects of the world around them. In this latter approach, the perspective of the learnerand not the researcher becomes the starting point in any investigation with the emphasis ondescribing and understanding and not on explaining human learning.The nature of inquiry in studies of learning and thinking are changing from "mentalmechanisms" and "information-processing devices" to the study of "conceptions of reality."Marton (1984) clarifies what is meant by this latter phrase:Conceptions and ways of understanding are not seen as individualqualities. Conceptions of reality are considered as categories ofdescription to be used in facilitating the grasp of concrete cases ofhuman functioning. (p. 74)One of the pioneers of research into the area of subjective conceptions of phenomena suchas knowledge and learning is William Perry (1970, cited in Richardson, E ysenck & Piper, 1987).His work was based on participant observation of students trying to adapt their learningprocedures to the world of the university. Perry suggests that "behind the learning difficultiesencountered at university there may not necessarily be insufficiencies in 'processing capacities' or'motivation' but in conceptions of knowledge that are at variance with those held by the faculty"(p.104).Perry claimed that first-year university students tended to equate knowledge with theacquisition of facts, that is, "statements about the world that are accepted as true and correct" (p.104). Consequently, students interpreted their task at university as one of essentially memorizingthe "Answers" to a set of "Questions." In Marton and Saljo's (1984) research, as in that of others(van Rossum & Schenk, 1984; Watkins, 1983), it has become evident that there is a "functional7relationship between the mode in which people subjectively construe learning and the way they goabout dealing with learning tasks" (p.115).A good example of research from a second-order perspective is a study conducted by Saljo(1979). He investigated adult views about learning and in an interview situation asked the subjectsthe following question: "What do you actually mean by learning?".From the responses, Saljo identified five qualitatively different conceptions of learning.Learning was seen as:1. A quantitative increase in knowledge2. Memorizing3. The acquisition of facts, methods, etc., which can be retained and used when necessary4. The abstraction of meaning5. An interpretive process aimed at understanding reality.According to Saljo, in the first three definitions, learning is conceptualized as an "activityof reproduction." In other words, the meaning of learning is to transfer units of information orpieces of knowledge into the head. The last two conceptions defined learning as a constructiveactivity - an interpretive process of abstracting meaning and making sense of the world one livesin.Until recently, teachers' beliefs about the nature of knowledge and the nature of thelearning process seemed to emanate from behaviourist views of learning, that is, an emphasis onfactual and procedural knowledge. Cobb (1988) claimed that "such an emphasis on isolated factsand procedures seems to constrain conceptual understanding" (cited in Marshall, 1992, p.19).This could be one of the reasons for so many university students in Saljo's study conceptualizinglearning as activities of reproduction.Marton and Saljo (1976) discovered two distinctive ways in which first-year universitystudents tackled the studying of a text. Some conceived the task as mainly reproductive andapproached it by learning and memorizing the text itself. Other students took a more constructiveview of the task and tried to understand the intention of the author of the text. The researcherscalled the first approach surface-level processing and the latter deep-level processing.In another study, conducted by Van Rossum and Schenk (1984), the authors concludedthat variations in approaches to learning were linked with variations in conceptions of learning.Their results indicated that deep approaches were associated with Saljo's conceptions four andfive and surface approaches with his conceptions one to three. In addition, a very strong8relationship was found between these two levels of processing and the quality of the learningoutcomes. A learning outcome of relatively high quality was usually associated with both a deep-level approach and a constructive learning conception. Saljo himself has suggested thatconceptions one, two and three are most common among students adopting a surface approach toa learning task. Conceptions four and five, on the other hand, are usually those of studentsadopting deep approaches to learning tasks.According to diSessa (1988), current educational approaches place less emphasis onstudents' factual knowledge and more on their conceptual understanding in the areas of scienceand mathematics at the late high school and early college years. She reports that physicsstudents, in particular, possess "intuitive physics," a fragmented collection of ideas looselyconnected that constitute a series of "independent layers of understandability" (p. 55). FromMarton's (1976) perspective, these students lack any deep-level approach to understanding thephysical world. They are functioning at a surface-level and believe that problem solving simplyrequires finding the solution or right equation.Teachers are now focusing their attention not only on students' factual knowledge in thesciences but also on their procedural knowledge as students attempt to acquire "meta-cognitiveabstractions" generated from internal experience (diSessa, 1988, cited in Forman, 1988).Therefore, diSessa suggests that "knowledge fragments" inherent in many college studentsprovide the basic material to develop scientific understanding; the intention being to "build a newand deeper systematicity" (p. 62) in order to integrate these pieces of knowledge.Continuity becomes the key factor in this process as students progress through stages ofunderstanding over a period of time, like a "flow system" towards acquiring a "deep-level"conceptual understanding passing beyond initial perceptions and conceptions.At the other end of the educational system, Ingrid Pramling (1983) conducted a study forthe purpose of investigating and describing the nature of young Swedish children's conceptions oflearning. Responses were categorized in terms of what they think they have learned and how theythink they have learned it. Pramling's findings indicated three developmental levels of conceptionswere present within the ages of 4 to 8 years. In the very early stages of children's awareness oftheir own learning, their first notion is that they learn something by doing something. The nextstage of development indicates a raised level of awareness to the realm of knowing something andthe final conception is construed as understanding something.9Regarding the question of how children think learning occurs, Pramling reported that in theearly stages children have not yet distinguished between "doing and learning to do something".The next stage in their awareness relates learning to growing older and the highest level ofawareness was captured in the idea that learning takes place through experience.The overall results of this study indicated that the majority of four-year-olds believed that"you learn to do something by gaining experience in the form of your own actions"(Pramling,1983, p.80). According to Pramling, children at this age have not yet reached a level ofawareness of their own inner mental state. Instead knowledge and knowing is connectedexternally to some adult telling them something.This result supports Pillow's (1988) hypothesis that young children initially conceive of themind as a passive recipient to external influence and that it is not until they are older that theydevelop a conception of their own active mental experiences. This hypothesis will be discussed inmore detail in the section titled Theories of Mind.Theories of Mind The development of children's knowledge about the mind is a relatively recent area ofinvestigation in cognitive psychology. Research into the development of children's understandingof a variety of mental phenomena establishes the criteria for the selection of the second category ofstudies to be examined under the following subheadings: metacognitive awareness, externalizationhypothesis, and children's understanding of mental verbs.Developmental psychologists have conducted research into children's understanding of theexistence of internal states, cognitive monitoring, intentions and the conception of emotion.Although these topics usually have been considered independently, they all involve understandingor awareness of psychological events or characteristics. Considerable attention has focused on thedevelopment of what children know about cognition, that is, their metacognition. As Donaldson(1978) pointed out, "If a child is going to control and direct his own thinking he must becomeconscious of it" (p. 94). Metacognitive research has attracted increasing attention in recent years,since it is assumed that children's metacognitive knowledge and skills play an important role intheir ability to learn.1 0Metacognitive Awareness. Children's thoughts about various cognitive activities involved in the process of learningsuch as knowing, remembering, understanding and problem solving are generally regarded asmetacognitive aspects of thinking. The significant role of metacognitive awareness in learning isbecoming more prevalent (Costa, 1985). Encouraging students to reflect on the process oflearning appears to be one possible method of raising a child's awareness of different learningstrategies that can be drawn upon to help them develop new knowledge, concepts or ideas. It hasbeen maintained that helping children to become metacognitive about their learning, and at the sametime teaching them a repertoire of strategies to choose from, will enhance their learning (Biggs,1985).The forerunner of metacognitive research is Flavell (1979). He defines metacognition as"children's knowledge of their own cognitive processes" (p.221), that is, what they know aboutcognition. Most of Flavell's studies are experimental in design and have been concerned withpreschool and early school-aged children. As a developmental psychologist, Flavell reached theconclusion that children progress through general stages of development, similar to those describedby Piaget. Flavell's research made no connection to any particular content or situation.From a different perspective, recent metacognitive research (e.g., Griffin, 1992;Pramling,1988) used the child's point of view as a framework for interpretation. With thisapproach, content and situation are key factors. The nature of inquiry is dependent upon howchildren think about and understand specific phenomena in the world around them. Qualitativelydifferent categories of conceptions are regarded as levels of metacognition. For example,children's awareness of their inner world, in terms of conceptual understandings of differentfeelings, can be identified in age-related stages of development (Griffin,1992). The results ofGriffin's study suggest that children conceptualize internal states quite differently at the ages of 4years, 6 years and 8 years.The metacognitive hypothesis used in Prarnling's (1988) study is that metacognition is aquestion of how children think about their own learning. She set out to discover different ways ofinfluencing preschool children's conceptions of their own learning. Her main intent was todetermine how conceptions of learning can be developed by directing children to use their ownthoughts and reflections on learning. Children's conceptions of learning were identified frominterviews at the beginning of the year and provided the basic starting point on which to conduct11the study. Positive results were obtained from interviews at the end of the school year todetermine the possible effect of a metacognitive component built-in to the preschool programme.There were noticeable shifts or changes in children's conceptions of learning indicating anincreased awareness of themselves as learners.In conclusion, metacognition, considered in terms of conceptual understandings, is alwayspresent and differs qualitatively from one individual to another and within one individualdepending on the content considered.Externalization Hypothesis. Pillow (1988) in his report, The Development of Children's Beliefs about the MentalWorld, claims that children's understanding of certain mental states such as perceptual experience,memory, intentions, and emotions, suggests a general developmental hypothesis:Young children view the mind as passive in relation to the externalworld and regard external events as determining subjectiveexperience, whereas older children know many ways thatpsychological activities influence experience. (p.1)Pillow (1988) claims that this "passive-to-active" hypothesis "captures a general developmentaltrend in children's conception of the mind" (p.2).Piaget provided the most general account of children's understanding of mental events. Hemaintained that young children initially externalize the mental world. Wellman (1985a,1985b) haslabelled this claim as the externalization hypothesis : children under the age of 7 years are unable todiscriminate between psychological and physical events. They perceive psychological events asexternal and physical (Piaget,1929,1930). For example, in Piaget' s study, children under 7 yearsidentified thought with the act of speaking.Children's Understanding of Mental Verbs. Studies investigating children's comprehension of mental verbs indicate that preschoolersoften treat verbs that refer to mental states as if they refer to overt actions (Misciones, Marvin,O'Brien & Greenberg, 1978; Shatz, Wellman & Silber,1983). Misciones et al. (1978) stated that:a) children under the age of 4 years were "unsystematic" in their use of theterms know and guess ;b) between the ages of 4 to 6 years, they increasingly used know to refer tocorrect choices of a hidden object's location and guess to refer to incorrect12choices, regardless of prior knowledge; andc) at the age of 6 years and above, children were able to differentiate these twoverbs on the basis of "epistemic states" or internal knowledge rather thanobservable outcomes. (cited in Pillow,1988, p.3)Misciones et al. concluded that children initially use both know and guess to refer toexternal, perceptible aspects of a situation, and gradually acquire their true meanings as theybecome aware of psychological processes.Wellman and Johnson (1979) reached similar conclusions about young children'sunderstanding of the verbs remember and forget. In their study, 4-year-olds tended to judge that astory character had remembered the location of a hidden object whenever the correct location waschosen, regardless of whether the character had previous knowledge of where it was hidden. Both5- and 7-year-olds used remember to refer to the presence of prior knowledge accompanied bymaking the correct choice of the location. The term forget was solely judged by all ages on thebasis of whether the correct location was chosen.Wellman and Johnson (1979) concluded that remember is understood earlier than forget,and that children progress from understanding these verbs in terms of overt behaviours tounderstanding them in terms of inferred cognitive states. In summary, the results reported in thesetwo studies confirm Pillow's "passive to active" hypothesis in which he claims that there is adevelopmental trend from perceiving the mind as passive to active in relation to its influence on theexternal world.However, Johnson and Wellman (1980) conducted another experimental study whichcarefully varied the factors of behavioural outcome and knowledge state. The results indicated that4-year-olds can go beyond reference to overt performance in their use of remember, know andguess. For instance, in a trick condition, children who had previously seen an item hiddeninsisted that they knew where it had been concealed even though their attempts to find the objectwere consistently unsuccessful. Johnson and Wellman (1980) concluded, "In limited contexts,preschool children are able to distinguish between "mental" and external states and betweendifferent uses of mental terms. Yet only during the early school years do children typically exhibita more definitive understanding of the cognitive implications of the terms" (p.1102).Using a more naturalistic approach in determining children's understanding of mentalstates, Shatz et al. (1983) reached similar conclusions about young children's acquisition of mental13verbs. They used a "multi-pronged" method to investigate the early use of mental terms innaturally occurring speech to assess young children's ability to communicate about a mental state.Language samples of children's spontaneous conversations were examined in detail. A codingsystem was designed to characterize the function of each mental term used during the conversation.The authors claimed that the coding system helped to distinguish clearly conversational uses from apotential mental reference. For example, they found that 3-year-olds made "explicit contrasts"between mental states and external reality, such as, "I thought there wasn't any socks, but when Ilooked I saw them." In addition, language samples were assessed to determine syntacticcompetence regarding each subject's "linguistic preparedness" for producing mental stateexpressions. Shatz et al. (1983) claimed that this method is capable of determining the time atwhich the understanding of mental states emerges.Results suggest that the earliest uses of mental verbs are specifically for conversationalfunctions rather than for mental reference. It is not until the second half of the third year that firstattempts at mental reference become evident in some children's speech. Converging evidence fromexperimental tasks and analyses of natural speech provides a consensus on the abilities of childrenaged 4 and above to differentiate mental states and processes from external behaviours and events.The discrepancy between findings for and against Pillow's externalization hypothesis mayreflect both methodological differences and an incomplete understanding of children's earlyknowledge of the mental world. Evidence for externalization comes from two sources: interviewstudies and studies of semantic development. The use of open-ended questions in interviewstudies may underestimate children's understanding because children's performance may reflectpoor verbal skills or inadequate understanding of what constitutes a satisfactory explanation, ratherthan lack of knowledge (Donaldson 1979; Pillow, 1988).Donaldson (1979) reported that research "tended both to underestimate children'scompetence as thinkers and to overestimate their understanding of language" (p.155). Sheclaimed that too much attention was directed towards the grammatical level of sophistication ofchildren's speech and not enough to its meaning. This problem was solved in a recent studyconducted by Griffin (1992). Children from the ages of 4 to 8 years were individually asked toprovide explanations for the internal states happy, sad, good and bad. To prevent children frombeing penalized for immature language, Griffin developed a scoring criteria permitting children toobtain a score with a "bare-bones" response. Griffin defined a "bare-bones" response as "one1 4which met the criteria for the postulated structure in a minimally articulated form" (1992, p.204).In this way, her study was able to accommodate children with difficulties in expressing theirunderstanding of feeling terms.Wellman (1985b) suggests, because mental states are internal and invisible, they may beless salient than external events. Thus children may have an understanding of mental states, butthis understanding often may be obscured because their judgments are influenced by more salientexternal factors. In contrast, the findings from a number of studies using a variety of methodsindicate that some appreciation of the distinction between the physical and mental worlds isapparent during the preschool years. Nevertheless, the generality and robustness of thisunderstanding remains open to further investigation.Children's understanding of mental verbs appears to provide a more sensitive index of theirability to distinguish mental states from external events, but has not yet portrayed a clear picture ofchildren's overall understanding of mental activities.Pillow (1988) reported, "Not only is the differentiation of physical and mental eventsnecessary for the acquisition of detailed beliefs about the mind, but further learning aboutpsychological processes may help children to understand how particular mental representations andobjective reality can occur" (p.6). Pillow (1988) claimed that there are two patterns ofconceptualizing the mind and that they are developmentally ordered and are not distinct stages ofdevelopment. He believes that there is a transition between conceiving the mind as passive toconceiving the mind as active and that the "transition may be a gradual trend toward increasedknowledge of the involvement of mental activities in the experience of external events" (p.14 ).He states that "young children think of the mind as passive in its relationship with theexternal world. That is, they do not understand that psychological activities can transform,organize or select information received from the environment. Instead they view the mindprimarily as a passive recipient and storage place of information and experiences" (p.12).This idea corresponds to Chandler and Boyes (1982) "copy theory" of knowledge in whichobjects are thought of as "transmitting objective knowledge directly into the mind of any observerwho looks in their direction" (cited in Pillow,1988, p.9).Sodian and Wimmer (1987) concluded that children younger than 6 years seemed tounderstand that "knowledge may be received through perceptual contact with the environment, butthey did not seem to understand that knowledge may be constructed through mental activities such15as inferencing" (cited in Pillow,1988, p.15).In contrast, older children conceive of the mind as active. They begin to develop anunderstanding of different ways the mind acts upon information. The crucial feature of an activeconception of mind, according to Pillow, is "understanding that psychological characteristics caninfluence the experience of external events" (p.14).For instance, older children are aware of short-term memory limitations that influence thenumber of items a person can recall and that psychological characteristics and activities impact onmemory performance. Brown (1978) found that children under the age of 8 or 9 years do notrealize that "active memorization strategies such as rehearsal or categorization are more effectivethan passive strategies such as merely looking at the items to be remembered" (p. 86).In addition, Miller and Weiss (1982) found that 7- and 10-year-olds believe that thepresence of "distracting objects," a "person's goals," and "ongoing thoughts" can haveconsequences for learning. Thus, older children's appreciation of psychological influences onlearning and attending is consistent with the proposal that, as children grow older, they considerthe mind to play a more active role in the reception of information.For the most part the studies discussed in this review were not designed as investigationsof Pillow's passive-active hypothesis; in retrospect many of them provided only indirect evidenceto support it. Nevertheless, some consistency was achieved in findings across the different areasof children's understanding of mental phenomena.For instance, the studies investigating young children's acquisition of mental verbsprovided evidence that children as young as 4 years were able to differentiate mental states. Inother words, mental reference in speech reflects a conceptual acquisition of mental processes.The term "learning" itself is a mental verb, by definition a complex one, that encompasses avariety of mental acts. In a study conducted by Johnson and Wellman (1982) titled Children'sDeveloping Conceptions of the Mind, there was significant evidence that children are learningabout the mind as a whole, distinguishing a category of mental acts. Results from this studyindicate that by 4 years of age children demonstrate awareness that the brain is an internal body partassociated with a variety of mental acts such as think, know, remember and dream. This runscounter to previous characterizations of young children as failing to distinguish mental from overtbehavioural acts. In summary, children are naturally making inferences about the differencebetween mental states and external events.1 6In the Swedish studies (Marton & Saljo,1976; Pramling,1983; van Rossum &Schenk, 1984) previously discussed, learning was described as it related to specific content, thatis, a descriptive analysis of the students' qualitatively different views on specific content, typicalof the phenomenographic tradition. However, in the work of Biggs (1980) there is an attempt todescribe the quality of the learning outcome on a specific task. Bigg's view on learning is lesscontent-directed and more focused on analyzing performances and responses in terms of theircognitive structural complexity. To indicate that the students' performances on a specific task candiffer enormously from each other as a result of the way students approach the task, he developeda taxonomy named the SOLO (Structure of Observed Learning Outcome) in which Piagetiandevelopmental stages are clearly recognizable. Five categories were created ranging from simpleand concrete to complex and abstract. The levels of performance are successively categorized as"pre-structural, uni-structural, multi-structural, relational and extended abstract."Biggs maintained that the quality of the learning outcome is possibly related to what hecalled the student's "Hypothetical Cognitive Structure (HCS)." Marton and Saljo's (1976) studydid not take into account that the quality of the learning outcome could be directly related to thelevel of a student's intellectual development. Although hierarchical levels of conceptualdevelopment were identified, there was no attempt to determine a relationship between age andlevel of structural complexity. Piagetian and neo-Piagetian analyses of cognitive developmentprovide educational researchers with a theoretical model by which age-related differences inchildren's knowledge, skills and concepts can be sequentially defined in a hierarchical order.These analyses will now be considered for their relevance to this study.Research in the Piagetian TraditionQualitatively different categories are established which are consistent with age-level normspostulated by Piagetian or neo-Piagetian theory. The categories characterize different ways ofthinking about various phenomena, and can be critically analyzed to determine what constructions(rules and generalizations) students have regarding content or concept under discussion. Theseare described in terms of cognitive structures.Theory.. According to Piaget, there are three components that constitute intelligence: content,1 7function, and structure. Content refers to observable behaviours, specifically sensori-motor andconceptual that reflect intellectual activity. Function, as the name suggests, refers to the goals orpurposes that direct cognitive development. These cognitive functions remain constant throughoutthe course of development and do not change with age. As a result of this, Piaget refers to them asfunctional invariants that guide cognitive activity during all stages of development. He proposedthat there are two functional invariants: a) organization and b) adaptation that are as relevant tobiological and physical growth as they are to intellectual development. Piaget believed thatintelligent behaviour does not seem to be a random or trial-and-error process but is characterizedby forethought and planning. In Piaget's terms, it is organized into "coherent and discerniblepatterns" or "structures." As a trained biologist Piaget was comfortable with the idea thatimportant functions are carried out by biological structures, for example, the function ofrespiration is carried out by structures known as lungs or gills Therefore, it is not surprising thatPiaget approached the function of thinking and learning in terms of mental and cognitive structuresthat make it possible. Although they are unobservable_ and completely abstract, Piaget regardedthese structures as being quite real. In essence, Piaget believed that the developing child was busyconstructing cognitive structures. Structures are not considered to be "innate"; that is, they do notexist in children's thinking in themselves, nor do they exist in the content to be learned. Marton(1988) suggested that structures are created by the child when he or she thinks about something.In other words, children create their own structures on the basis of experience and development.This notion is extended by Pramling (1990) who claimed that children have the ability to relate onething to another in the sense that they perceive various relationships which in turn forms the basisof understanding. Probably the most important implication of Piaget's theory is that childrenconstruct knowledge from their concrete actions on the environment.Piaget's second functional invariant, adaptation, is instrumental in helping to induceconceptual change. Piaget borrowed the biological notions of assimilation, accommodation andequilibration to explain how cognitive structures develop. Assimilation is the cognitive processthat continually integrates new stimuli into existing structures (schemata). Accommodation is thecreation of new structures or the modification of old ones. Either one of these actions results in achange in or development of cognitive structures (schemata). Cognitive structures accommodatein a special manner called equilibration. The equilibration process is an example of adaptation atwork. In a way the process can be described as a recurring cycle whenever there is a loss of1 8equilibrium in understanding a new concept. First assimilation is attempted leading toaccommodatory change and finally a return to mental equilibrium.The development of these cognitive structures and knowledge is an "evolutionaryprocess" (Wadsworth, 1984) that takes place within every individual. It is manifested in anindividual's schemata, that is, a network of mental structures that are constantly undergoingchange. As a child develops, cognitive structures become more differentiated, less sensory andmore numerous; the network they form becomes increasingly more complex.Flavell summarizes Piaget's theory and interpretation of what is meant by structure as:Interposed between function and content, Piaget postulates theexistence of cognitive structures. Structure, like content and unlikefunction, does indeed change with age, and these developmentalchanges constitute the major object of study for Piaget What are thestructures in Piaget's system? They are the organizational propertiesof intelligence (schemata), organizations created through functionand inferrable from the behavioural content whose nature theydetermine.(Flavell, 1963, cited in Wadsworth, 1984, p. 21).Piaget concerned himself primarily with the structure of intelligence, although he dealt withfunction and content to a lesser extent. He described and analyzed qualitative changes in thedevelopment of these cognitive structures but in very broad and general terms. Piaget suggestedthat cognitive development is a function of the interaction of maturation, physical experience,social interaction and equilibration which Piaget defines as a "self-regulatory process".At the beginning of this chapter learning was described as essentially a social cognitiveprocess by nature. Children learn by trying to make sense of the world around them. Inquiry intochildren's attempts to construct meaning of specific phenomena of the "social world" may enablecognitive psychologists to gain insight into the nature of learning from a child's perspective.Further, such an investigation may contribute a clearer understanding of how children intuitivelyconceptualize learning as a process. Theoretical predictions postulated by Piaget enable researchersto plot children's growth in their conceptual understanding of different concepts as they maturewith age.Methodology in the Piagetian Tradition. A common research strategy that developmental studies tend to adopt when addressing their1 9inquiry into children's conceptual knowledge of the social world is to take some knowndevelopment task, such as experimental tests for conservation, to provide additional support for thestage theory. Performance on these tasks serve as "anchors" or indices of a child's cognitivedevelopmental level by providing an independent measure. If, as developmental psychologistsclaim, children's concepts of specific phenomena are embedded in a matrix of cognitive structuralvariables (physical or social causality or identity concepts), then there should be a correlationbetween these matrices and the qualitative analysis of age-related conceptual differences of aspecific social concept. This method of research ensures reliability that the new concept underinvestigation follows a developmental sequence compatible to Piaget's stages. For the purposes ofanalysis and description, Piaget divided the continuum of development into four stages: sensori-motor (0-2 years), preoperational thought (2-7 years), concrete operational thinking (7-11 years),and formal operational thinking (11-15 years).For example, Sparks and Cantor (1983) not only interviewed children after watching eachvideo segment taken from the television program The Incredible Hulk to determine their emotionalresponse, but also administered a standard liquid conservation task to correlate each subject'sidentity conservation ability with their level of fear response. The ability to perform identity-conservation tasks was considered to be related to the ability to decenter from perceptual cues andto the comprehension of transformations. This corresponded to the transformation stage in thehero's appearance from human to monstrous form. The results of the study confirmed that therewas a high association between the child's performance on the conservation task and thepreoperational and concrete operational stages of Piaget's theory.Similar methodology was applied in a study investigating children's concept of age andaging. Although Piaget provided the basis for researchers to study the child's concept of age andaging, there is is only a small amount of literature addressing this area. One study conducted byGalper, Jantz, Seefeldt, and Serock (1981) utilized the questions posed by Piaget(1969) anddeveloped a Concept of Age Test designed to provide a standardized procedure for placingchildren's concepts of age into a cognitive developmental framework. The researchers claimed thatthe administration of this instrument to 180 children ages 3 to 11 would provide "systematicnormative data for the support of such a theoretical framework" (p.150).In order to provide an "indice of criterion validity," the researchers also administered threeconservation tasks involving the ability to conserve two-dimensional space, substance and20discontinuous quantity. They believed that the child's performance on these tasks would providean independent measure of cognitive level of ability. If children's concepts of age are embeddedin a cognitive-developmental matrix as was hypothesized in Sparks and Cantor's (1983) studyassessing emotional cognition, then performance on both tasks should be correlated.A modified clinical technique was adopted by the interviewers in which children wereasked to reply "Yes" or "No" to a set of 12 statements. Children were then prompted to justifytheir responses so that the interviewers could locate the developmental level of each child. Thefindings of this investigation indicated that the total Concept of Age score highly correlated withconservation level (r = .79), supporting the hypothesis that children's understanding of age andaging proceeds through a Piagetian cognitive-developmental sequence of three stages.The researchers also identified a developmental progression within the stages; responsesusually given by younger children scored in the initial phase of the stage while older children wereconsistently rated toward the upper stage hierarchy. This consistent increase in stage scores withincreasing age supports the notion that children's concepts are both progressive and sequential andincrease in accuracy with age.The weakness of this study was in the test instrument. The types of statements presentedto the children were as follows:You will grow older but your father will stay the same age.If someone was born first, then they are older than you.If someone is five years older than you, they will always be fiveyears older than you.Someone is two years older than you, but you will catch up to themand be the same age someday.How old were you when you were born?Statements such as these have some degree of syntactical and semantic complexity, and may be thecause of cognitive confusions and inaccuracy. Such statements most likely narrow the perspectiveof young children's thinking about the topic in general and force reasoning powers to interpret themeaning of the statements.Language is, in Donaldson's (1978) terms, "embedded" in a context of relationships andpersonal concerns. Children would have been less context-bound if they had been asked torespond to open-ended questions enabling them to focus on a more holistic explanation of age.This would have enabled the children to respond within a more general and broader scope on thetopic of age. The distinction drawn by Donaldson between "embedded" and "disembedded" tasks21would seem to provide an important message regarding conditions under which researchers placechildren to assess their cognitive abilities. She also reminds us that young children may havedifficulty disembedding the language in such situations as these.When children describe their understanding of such abstract concepts as age and agingembedded in the matrix of time, they are not expressing the ideas which they have obtained directlyfrom experience; it is through inference they are constructing their knowledge. It may be possibleto conclude that children are not only taking in verbal information about aging, but also processingthis information according to their cognitive structural level. What was interpreted as children'smisconceptions about age by these researchers, may have been in part a result of the child'sreasoning and processing of information which was too complex for his or her developmentallevel.Summary. In summary, two major approaches were used in research conducted in the classicalPiagetian tradition that not only impact on each other but may determine future directions inresearch into children's social concept development. The first concerns itself with the relianceupon children's verbal responses to open-ended or semi-structured questions on a specific topic.Piaget claimed that "observation of natural phenomena permits children to use their"concrete" (logical) intelligence, whereas in order to form such notions as "family" and "nation"children are forced to employ verbal intelligence which causes them to develop ideas and imagesbased in large part on the words they hear spoken by adults"(Piaget, 1924, cited in Berti, Bombi &Lis, 1982, p. 222). This was particularly relevant to the Berti et al. (1982) study inquiring intochildren's understanding of economic concepts. Background information provided in the parentquestionnaires confirmed that children constructed knowledge about the local factory based onverbal information provided by the parents. This second type of intelligence received very littleattention from Piaget. His principal concern had been with concrete intelligence, concentratingmore on children's ability in "logico-mathematical operations" in their attempt to understand thenatural physical world. Most of Piaget's studies were experimental in design requiring students toperform formal tasks. Piagetian conservation and perception tasks were criticized by Bruner(1986) and Donaldson (1978) as being too contrived. According to Donaldson (1979), suchformal tasks limit thinking skills due to the emphasis they place on disembedded thought and22language. She claims there is a disparity between children's skills as thinkers in everydaysituations and those in formal experimental tasks.In recent years, educational research has been "more sensitive to the experimental aspectsof cognition, and more open to using personal introspective accounts of learning processes"(Richardson,1987, p.5). Miller (1976) argued that Piagetian clinical interviews "rely too heavilyupon verbal skills" (cited in Green,1978, p.1045). However, a truly complete or holistic pictureof the child's thought development must also take into consideration the way in which the childperceives different aspects of reality in the social world. Language becomes the vehicle researchersdepend upon to determine children's level of understanding of different social phenomena.Empirical investigations can yield this type of information about the child's conceptions of thesocial world, revealing both similarities to and differences from their conceptions about thephysical world.This leads directly to the second major approach often used in developmental studies, thosewhich examine children's performance on experimental tasks as well as their verbal responses inan interview. Several studies in this category employed this common research strategy in the areaof cognitive development. Researchers administered pertinent Piagetian-type developmental taskswith the intention of using them both as "anchors" or indices to locate a previously unexplored areaof thought.As previously stated, children's conceptual understanding of a specific phenomena isembedded in a matrix of cognitive-structural variables constrained within the level of anindividual's processing capacity. This being the case, then the results from performance taskswhich assess a subject's ability in "logico-mathematical" operations and the qualitative categorywhich gauges a subject's level of conception of a specific phenomena should be correlated. Such acorrelation would thereby strengthen a study's reliability in its claim that children's ability tounderstand a specific concept corresponds to the stages of development in Piagetian theory.Results from the studies discussed in this section confirm a high association between children'sconceptual and cognitive development.A cross-analysis of these studies also reveals similar conclusions regarding children'sconcept development of different phenomena in social and emotional domains:1. Children's conceptual understanding of a specific topic systematically23increased in complexity with age supporting the hypothesis that the conceptsare ordered in a developmental sequence.2. Concept development progresses through a series of general and universalstages. Further, differences within each stage portrayed a developmentalhierarchical sequence from the initial to the final part of the stage.3. Age-related developmental changes that took place in children's conceptualunderstanding of a specific notion were compatible with the three stages ofPiaget's theory of cognitive development.4. Specific social concepts are embedded in a matrix of cognitive structuralvariables such as physical or social causes, conservation or identityconcepts. Consequently, conceptual understanding is limited by the child'slevel of cognitive growth. Children process and construct new ideasaccording to their cognitive structural level; in other words according to theirlevel of ability in "logico-mathematical operations." Information about agiven topic is not simply taken in ; it is assimilated or transformed to thechild's present cognitive level.5. In Piagetian theory, a new stage grows out of and subsumes the one beforeit in a slowly evolving developmental process.6. Qualitative data, extracted from the children's interviews, illustrated thecontent of the concepts in a particular area and the necessary "components"or "units" of thought required for its developmentThe classical Piagetian analysis is essentially a descriptive model and Piagetian explanationby way of assimilation, accommodation, and equilibration is perceived as intrinsically weak byneo-Piagetian theorists such as Case (1985), Demetriou (1987), and Pascual-Leone (1987). Inparticular, Piaget's somewhat mysterious regulatory process known as equilibration has beencriticized as logically unfounded (Fodor,1976, cited in Wellman, 1990). Evolving from the reviewof studies in this section is a growing awareness for the need of a more "fine-tuned" model thatspecifically describes and explains the developmental process of how cognitive structures increasein complexity within each stage. Case's (1985,1992) neo-Piagetian model attempts to provide amore indepth explanation of the development of increasingly complex cognitive structures.Marton (1981,1988) argued for a phenomenographic approach to learning and instruction24which stemmed from his observation that students hold different conceptions of what learning is.Phenomenography is the study of the quality of the conceptions which people have of variousphenomena. In this theoretical framework, learning is about the "shifts or changes in thoseconceptions", and instruction is about "inducing this conceptual change" (Marton,1989). Thisperspective on the nature of learning is congruent to the fundamental beliefs upon which Case's(1985, 1992) neo-Piagetian theory is founded.As a developmental psychologist, however, Case attempts to analyze the cognitivestructures that bring about these changes in conceptual understanding. Little progress was madetowards creating a theory that explained the cognitive structures indicative of developmentalchanges in children's concepts of specific phenomena until a series of revised and refined neo-Piagetian models was developed in the early 1980's. The neo-Piagetian model presented in thenext section provides a more systematic structural model of cognitive growth within substages.Further, this theory also offers a description of the major structural change from one stage to thenext.Research in the neo-Piagetian Tradition Piaget's theory of intellectual development has played an influential role in psychology andeducation for the last sixty years. However, all theories in psychology are constantly changing;they are being reorganized, refined, tested and validated. Critics of Piaget's theory reached acommon conclusion that its weakness lies within a "domain-general" interpretation of logicalstructures (Wellman, 1990).In the light of new research in developmental psychology, intellectual development appearsnot to evolve in such a broad-based manner as Piaget assumed; it is highly content, domain andtask specific (Case & Mckeough,1990). The neo-Piagetian model developed by Case (1985)enables researchers to conduct a "fine-grained" analysis of children's intellectual developmentacross social and cognitive domains. As its name implies, neo-Piagetian theory was designed withthe intent to revise Piaget's theory after acquiring new information on cognitive development fromtask-related studies in a variety of different domains. Case's intent was to preserve the strengths ofclassical Piagetian theory while eliminating its weakness. He retained a core set of postulates thatexplained the universal features of cognitive development. These include the notion that children'scognitive operations proceed through a universal sequence of stages which postulate characteristic2 5understandings of the world according to age. Piaget claimed that these stage-characteristic orlevels of understandings would transcend any particular task. As previously stated, Piaget'stheoretical model was primarily descriptive and offered only a very general explanation of thedevelopmental cognitive processes that took place within stages and to the next stage. This lack ofexplanation of non-universal aspects of cognitive development was considered a weakness bymany developmental psychologists, including Case.Case (1985, 1992) developed a set of "structural transformation processes" or sub-stagesthat explained the cognitive shifts that occur within each major stage. He accomplished this task byclaiming that knowledge can be conceptualized as being charted along two axes. On the verticalaxis he attempted to explain actual performance variation within each stage as sequentially ordered"discrete steps" or "structural levels". The horizontal axis involved the application of thesestructures across other domains.Case maintained that, in any given content domain, children progress through a series ofcognitive-developmental stages and substages. He believed that children's conceptualunderstanding has a distinctive organization or structure at each of the following age levels:Infancy, early childhood, middle childhood, and adolescence. Consequently, four major stagesare still hypothesized, following the classical Piagetian theory: Sensorimotor (0-18 months),Relational (1.5 - 5 years), Dimensional (5 - 11 years) and Vectorial (11 -19 years).Progression from one stage of development to the next occurs when two qualitativelydifferent conceptual structures or "mental units" are coordinated and consolidated at the end of theprevious stage. Transition from one substage to the next takes place by the differentiation andsubsequent integration of two structures that have been brought into a higher-order form ofrelationship (Case, 1991).For example, in the Dimensional Stage, behavioural events in the physical world aredifferentiated from but related to mental states in the form of feelings, judgments or processes.The cognitive structure describing social cognition in the Dimensional Stage is termed the"Intentional" structure. It represents a causal relationship between the external world of action andthe internal world of mental states and processes.There are three substages defined within the Dimensional Stage. From 5 to 6 years of age,children are able to differentiate and coordinate a behavioural event with one mental state and thiscognitive level is described as the "unidimensional" stage. By the ages of 7 to 8 years, children are2 6able to differentiate and coordinate two mental states to a behavioural event and this level is defmedas the "bidimensional" stage. The last of the substages (9-11 years) is described as an elaboratedcoordination of the "bidimensional" stage. The intentional structure is characterized by a higher-order relationship in its attempt to integrate internal and external states in a more complex fashion.An abstract model of this structural development within each substage of the Dimensional Stage isshown in Appendix A.These three substages constitute the "structural levels" that lead to the consolidation andcoordination of mental units. They attempt to explain developmental increases in structuralcomplexity consistent with the theory of cognitive development. Within each of the four stages,Case (1985) hypothesized the occurrence of this same dialectical cycle of identical substages. At amore general level, this assumption existed within the classical Piagetian tradition, but in thecontext of Case's neo-Piagetian theory it exists in a more stringent form. A cyclic recursionsupports the notion that "in the construction of any understanding at any level of development,there is a progression through exactly the same sequence of structural steps as at previous stages"(Case,1987, p.779).The social cognitive structure specifically related to the Vectorial Stage is defined as the"Interpretive" structure and corresponds to formal operational thinking in the classical Piagetiantradition. In other words, this stage marks the initial development of abstract control structures inadolescence. Thought at this level assumes a psychological dimension in the social domain and inCase's terms is defined as "abstract dimensional" (Case,1985).For example, in the verbal domain, children are able to make analogies. "To perceive ananalogy, one must perceive a higher-order vector along which two lower-order dimensions may becompared" (Case,1985, p.109). In other words, children are able to compare one dimensionaloperation with another and this process in turn may be described as a "second-order" or "abstractdimensional" operation. The next section will discuss the developmental progression of children'ssocial cognition as it relates to neo-Piagetian theory.Intentional Knowledge in Middle Childhood. Studies reported in this section derive from neo-Piagetian theory. As previously stated,this theory places greater emphasis on domain-specific experience. Although it utilizes constructsfrom contemporary cognitive science to characterize high-level structures which children develop,27the model is defined in such a fashion that it is not restricted to any particular domain or type ofmental functioning. Unlike the classical tradition, less emphasis is placed upon logico-mathematical structures and more attention has been directed to capacities upon which thesestructures are believed to be dependent, such as working memory.The following study provides an example of children's development of narrativeknowledge from the ages of 4 years to 10 years and how they relate to the theoretical predictionsproposed by Case's (1985) model. Further, comparisons will be made with this study to thework of Goldberg-Reitman (1992) who investigated the conceptual development in children'sunderstandings of a mother's role. The purpose of this comparison is to reveal similar age-typicalpatterns of structural development in children's understanding of two different concepts.McKeough (1984) conducted a study that attempted to relate the way in which childrenstructure stories at different age levels to the general stage characteristics proposed by Case.Further, she set out to determine whether a relationship existed between the subjects' performanceon two measures of short-term memory and the narrative compositions task.Four-year-olds, 6-year-olds, 8-year-olds and 10-year-olds participated in the study. Theseages are believed to typify the general cognitive strategies applied at the different substages in theDimensional Stage proposed by Case. Different story structures were discovered at the ages of 4years, 6 years and 8 years with some indication of further development taking place at the age of10 years.A prototypic 4-year-old story grammar appeared in the form of separate integrated script-like event sequences in which action and characters' feelings are fused or blended within whatBruner (1986) calls the "Landscape of Action". McKeough concluded that four-year-olds possessan "action event schema" (1993) which Case classifies as a pre-intentional stage of cognitivedevelopment. There is also evidence that 4-year-olds are able to apply this cognitive procedure toother tasks, as is indicated in their explanation of a mother's actions when her daughter was insome kind of discomfort or danger. Goldberg-Reitman (1992) showed a series of cartoon stripsthat portrayed a child in different problem situations and asked each subject what the girl's motherwould do and why. Goldberg-Reitman's findings showed that 4-year-olds explained a mother'sactions in terms of one-step actions and when asked to explain the reasons for the mother'sbehaviour, responses still remained in Bruner's "Landscape of Action".For example, one cartoon showed a little girl beginning to slide off a roof and consequently28she calls out for help. A typical 4-year-old's prediction is that "the mother will catch her" and thecommon explanation for why is "because she's falling down." The majority of responses made noreference to either the mother's or daughter's internal state.By the age of 6 years, children's stories became more coherent by introducing a problem,goal, or desire that links the event sequences with a common theme or point to their story. Actionsequences are now coordinated with the character's mental states in the form of feelings, thoughtsor desires. For instance, in the first episode of a prototypic 6-year-old response, a baby lamb islost and lonely. Then a horse comes along and rescues the lamb and they go away together. Thefirst sentence typifies unidimensional thought consistent with Case's theory. Six-year-old childrenare able to differentiate and coordinate an event (a lost lamb) with one mental state which in thiscase is the lamb's feeling of loneliness. McKeough (1993) described 6-year-old narrative structureas an "intentional" schema "because the intentions of the characters motivate the action" (p. 5).This episode also demonstrates a simple version of Bruner's (1986) "dual-landscape" narrative inwhich the Landscape of Action is related to an intentional state (the lamb's feeling of loneliness)which Bruner describes as the Landscape of Consciousness.This integrated intentional structure also reflects 6-year-old responses in Goldberg-Reitman's (1992) study in which children were asked to predict a mother's reaction and feelings toa problem situation in which her daughter was involved. In the cartoon task previously discussed,a common response of 6-year-olds was "a mother would catch her little girl because she didn'twant her to get hurt."In contrast to a 4-year-old response, a typical 6-year-old response provided a rationale forthe mother's actions by making additional reference to the mother's intentional state in terms of animmediate plan of action. In the context of neo-Piagetian theory, this coordination of externalevents or actions with the characters' internal world of feelings,ideas or plans is characteristic ofchildren at this age.Narrative at the 8-year-old level had a tendency to "flesh out" the problem, goal or desireinto an event sequence but still relate it to the major plot. In addition, a second focus is introducedinto the story in the form of a sub-plot that complicates a straightforward resolution to the story.For example, a lost little lamb is found by a girl. Her goal is to care for it (major plot). Howeverher father prevents her from achieving this goal by refusing her request to keep it (sub-plot). Theaddition of this sub-plot introduces a second intentional state and thereby demonstrates29bidimensional thought. Children are able to differentiate and coordinate two mental states to abehavioural event at the bidimensional stage. In this case, two problems now surface that need tobe resolved: a lamb in need of care and a father to be "gotten around" (McKeough,1984, p.7).Several unsuccessful attempts are usually made before both problems are finally resolved.Problem 1 ^An attempt is made to solve the problem(A house is built in the woods)Problem 2^A successful resolution(Father is happy because little girl agrees to return the lambto its rightful home!)At the ten-year-old level, children elaborate upon the 8-year-old structure by composingadditional episodes that develop the resolution more extensively. As a result, stories contain acharacteristically well-developed plot that is associated with a more elaborate network ofcharacters' intentional states woven into the action of the story.This integration is also evident in the Goldberg-Reitman (1992) cartoon task. Ten-year-old responses provided alternate behaviours in which the mother might respond to a particularproblem situation. These alternate behaviours were integrated under the umbrella of a moreabstract "maternal disposition," that of the mother's sense of "caring" for their daughter. Furtherreference was made to a more elaborated set of intentional states that directed the mother's selectionof a particular action sequence. Goldberg-Reitman claims that such responses are consistent withCase's theory in general, in the sense that 10-year-old children are beginning to construct anabstract notion of the mother's role. In fact, Goldberg-Reitman claimed that by the ages of 12 or13 years, they were conceptualizing the role in explicit abstract terms.By way of summary, the results from these two studies can be compared to those studiesconducted within the classical Piagetian tradition. As was identified in children's conceptualunderstandings in other domain-specific tasks, elementary school-aged children's narrativecompositions and understandings of a mother's role proceeded through a series of increasinglycomplex stages of development. Further, a relationship was established between performance onthe narrative task and on two working-memory tasks. In support of this result, Goldberg-Reitman(1992) suggests that from her findings "a sequence of central conceptual structures may exist forthe domain of social cognition" (p.151).As in other areas of cognitive growth, development is easier to describe than to explain.30Case reports that one possible mediator that could be partly responsible for the transition from onestage to the next is an increase with age in information-processing capacity. As children mature,their short-term memory capacity increases making it easier for them to hold more than onerepresentation in mind at the same time. This might be the controlling factor in determining age-related levels of structural complexity in the amount of elaborative detail not only in children'snarrative composition but across other social domains.From a slightly different perspective, Flavell (1988) and his colleagues who studytheories of mind claim that children begin their discovery of the mental world by learning thatthey and other people have internal experiences that are cognitively connected to external objectsor events. As children mature, they gradually realize that these "cognitive connections" entailinner, "mental representations." Griffin (1992) conducted an inquiry into children'sunderstanding of mental states. Her study confirms the developmental trend of young children'stheory of mind from cognitive connections to external events to mental representations. Childrenaged 4 years, 6 years, and 8 years were specifically asked to explain the meaning of four feelingterms: happy, sad, good and bad. The purpose of this task was to determine whether there wouldbe age-related differences in levels of conceptual understanding of these affective states. Griffin(1992) predicted that the structural complexity of these three levels would be consistent with thedevelopmental progression postulated by neo-Piagetian theory.Between the ages of 4 and 6 years, children's conceptions moved from defining a feeling as anexternal event to conceptualizing a feeling as an internal state arising from an event or a sequence ofevents. For example, one 4-year-old response defined happy as playing, and sad as mommyleaving. On the other hand, a typical 6-year-old response combines Bruner's Landscape of Actionwith the Landscape of Consciousness, as this example shows: "Happiness means, well, like I'mall alone and then my feelings turn from sad to happy when I get a friend over". Eight-year-oldresponses still remained in the "dual landscape view" in their explanations, but tended to coordinatean action event with two distinct internal responses. Griffin (1992) claims that "responses such asthese lend support to a suggestion that is implicit in the model (Case's neo-Piagetian): that thepostulated structures are actually inside children's heads, and are used by them to make sense oftheir internal and external worlds" (p. 204).31Interpretive Knowledge in Adolescence. This review of literature was able to locate only two studies that investigated adolescents'conceptions of different phenomena in the social domain from a neo-Piagetian perspective. Aspreviously stated, Case predicts a shift in cognitive development from the "intentional" modewhere actions are differentiated from and coordinated to internal states (e.g. feelings, desires orbeliefs) to an "interpretive" mode. This "interpretive" cognitive structure exists within the VectorialStage (12 years -18 years) in Case's theoretical model.In studies conducted by McKeough (1987,1993), adolescent narrative knowledge takes ona psychological theme by placing more emphasis on the inner world of the character's traits andpsychological make-up by describing the reason behind the character's intentions. In other words,the story-line is developed from the character's perspective. The inclusion of literary techniquessuch as flashbacks or foreshadowing enabled the reader to "interpret the intentional states of storycharacters" (McKeough,1993, p.7). Consequently, adolescent narrative marks a shift to a higherplateau of thought defined as the Interpretive structure.A more detailed description of the Interpretive structure is found in a recent studyconducted by Salter (1993). She analyzed adolescents' interpretation of "family stories"(significant accounts of a past event involving a particular family member). Students aged 10 to 18years recounted in writing a particular family story relating to themselves or some other member ofthe family. Each subject was then interviewed and responded to a series of questions that probedinto each student's personal interpretation concerning the message they gleened from the story.Results supported the hypothesized developmental progression proposed by Case in the VectorialStage. In addition, there was significant evidence of a qualitative shift from an "intentional" levelof understanding, characteristic of 10-year-olds, to an "interpretive level" of understandingreflected in 12-18 year-olds. This qualitative shift in conceptual understanding marks the transitionfrom the Dimensional to the Vectorial Stage of Case's model.Prototypical 10-year-old responses described an action and then attributed a social rule inthe form of a judgmental statement to that action (e.g., you shouldn't leave little kids alone). Onthe other hand, 12-year-old interpretations applied a psychological dimension to the social rule inthe sense that the students related it to their own personal life. Further, an understanding for therule was articulated (e.g., I shouldn't play on construction sites because I might get hurt like myuncle did).32It can be concluded from these studies that the following Piagetian postulates have beenpreserved in Case's neo-Piagetian theory:1. Children are highly active thinkers and are in control of their learning;2. Children construct knowledge from their actions on the environment;3. Cognitive operations proceed through a series of developmental stages;4. Stage-characteristic understandings that transcend any particular task can beidentified.The new theory differs from the classical tradition in a number of different ways:1. Greater emphasis is placed on "domain-specific" experience in general. Ananalysis of the content relating to a specific task is considered in more detail.2. A more explicit description of the processes by which children progressfrom one stage to the next is provided.3. Minor substage shifts in structural complexity are also clearly defined withineach major stage.4. The structural model presented in this new theory is flexible enough toincorporate the effects of environmental influences, social experience inparticular, in cognitive development.Bruner's Social Perspective on LearningBruner (1986) postulated two modes of thought or "cognitive functioning", each providing"distinctive ways of ordering experience, of constructing reality" (p.11):1. Paradigmatic or Logico-Scientific. "A formal mathematical system ofdescription" (p.13), similar to Piaget's logic and rational thinking;2. Narrative - "deals with human or human-like intention and action and thevicissitudes and consequences that mark their course" (p.13).The latter mode denotes an area of mind we know very little about in any formal sense and as itstitle infers, Bruner compares it to creating good "stories." A good story, according to Bruner,must construct two "landscapes" simultaneously:1. Landscape of Action - This corresponds to a story grammar describingevents that take place in the physical world;2. Landscape of Consciousness - This corresponds to the mental states ofthose involved in the action: in other words, what the characters know, thinkor feel.33In this Narrative mode, Bruner postulated a "dual landscape view" matching the "innervision and outer reality" (p.21). In other words children are mindful actors with plans andintentions that they attempt to put into action in their physical setting.This analogy has significant reference to cognitive functioning in the intentional structure ofCase's Dimensional Stage. Thought represents a causal relation between the external world ofphysical states and events (Landscape of Action) and the internal world of feelings and mentalstates in the form of ideas or plans (Landscape of Consciousness). A brief reference to Bnmer's"narrative" mode of thought has already been mentioned in relation to 6-year-old cognitiveoperations. The structural model developed by Case (1985,1992) enables researchers to take intoconsideration Bruner's social perspective on the nature of learning and on the process ofknowledge construction.Explanations in the intentional mode impose both cognitive and linguistic demands on thechild. It is cognitively demanding in that the child has to distinguish between the reason and theresult, despite the fact that they are interdependent. This would require a good understanding ofthe conception of intention. On the other hand, children need to acquire an understanding oflinguistic expressions of intention to explain their thinking. Donaldson's studies indicate that evenfive-year-olds have the cognitive and linguistic abilities for giving and understanding explanationsof events and actions, particularly those involving intentions, motives and purposes; in otherwords, situations which make "human sense."Bruner (1986) states, "Human mental activity depends for its full expression upon beinglinked to a cultural tool kit - a set of prosthetic devices"(p. 15). He emphasized the importance oftaking into account the "tools" employed in that activity when studying mental activity. "Societyprovides a tool kit of concepts, ideas, theories that permit one to get to higher ground mentally"(p.73). Piaget didn't deny that social interaction and communication help facilitate cognitivedevelopment. He claimed that they did so "by creating disagreement and cognitive conflict whichinduce self-constructed cognitive changes that advance a child to the next stage of development"(cited in Wadsworth, 1984). Piaget accepted verbal reasoning as a major vehicle upon whichlogical operations operate.34A Constructivist Viewpoint on Learning. An alternative conception of the nature of the learning process is that learning isconstructed, not only in an individual's head, but in interactions among individuals and materialsas they occur over time (Bruner,1986; Vygotsky,1978). Very briefly, constructivists believe thatindividuals actively construct knowledge and understanding (cf. Piaget), rather than passivelyreceive information in response to external stimuli.Constructivist theory emphasizes the search for meaning and understanding in learning. Attentionis focused upon the process of learning and subsequently developing an understanding of one'sown learning strategies through metacognition (McGuinness,1991).Marshall (1992) said: "Learning consists of building on what the learner brings to thesituation and restructuring initial knowledge in widening and intersecting spirals of increasinglycomplex understanding" (p.11). For example, Paley (1986), acting as researcher in her ownclassroom, came to the conclusion that she was unable "to teach the children that which they don'talready know" (p.126). Paley (1986) discovered that children try to connect what they alreadyknow to what they don't know.Although constructivists view learning as an active process, social and cognitiveconstructivists vary in how they regard the nature and influence of the social world in the processof knowledge construction. On the one hand, cognitive constructivists perceive learning as aprocess within an individual's mind, even though they acknowledge that the learner is often underthe guidance of an adult or expert.On the other hand, social constructivists place greater emphasis on the role of socialinteraction through which "contexts, knowledge, and meanings in everyday life are constructedand reconstructed" (Marshall, 1992, p.11). In a social constructivist framework, learning andthinking are situated in social contexts rather than occurring solely in an individual's mind.Bruner, Vygotsky, Donaldson and Wood support the tenet that knowledge is constructedthrough negotiation with others. Wood (1988) also claimed that learning derives from "vicariousexperiences, interaction, media and tutors." In addition, Bruner emphasized that "human learningpresupposes a specific social nature and a process by which children grow into the intellectual lifeof those around them" (1986, p.88). From this perspective, learning is viewed as a social processin which:• children's knowledge is constructed through social interaction of35participants around them;• learning is cooperatively established by the members of community; ashared mutual understanding;• Shared meanings are established through participating in communicativediscourse involving explanation, justification and negotiation of meaning.(adapted from Wood, Cobb, & Yackel, cited in Marshal1,1992, p.181).Wood (1988) also claimed that children tend to draw upon the closest and most relevantexperiences when they attempt to construct new meaning or tackle new and unfamiliar problems.He also concurs with Vygotsky that "knowledge and expertise are often a product of sharedconstructions by teachers and children"(p. 224). From a social constructivist's viewpointtherefore, the importance of communication and instruction through social interaction is of centralimportance in facilitating children's intellectual development. Case concurred with this andconsequently incorporated a social constructivist viewpoint on learning into his model.ConclusionFrom an educational perspective, valuable information is often extrapolated from a criticalanalysis of the data collected from children's responses to a particular topic. It is possible todetermine what constructions or generalizations children have regarding content under discussion.The type of content and the common patterns of thought that are generated by children of similarages provide insight into how children intuitively approach a problem or construct a conceptualunderstanding.Children construct, in diSessa's (1988) terms, "an intuitive knowledge system" which isrooted in their experience with the environment. DiSessa claims that this "intuitive knowledge"provides the basic material with which to develop a deeper understanding through appropriateinstruction.If the goal of teaching is to support children's learning then Lindfors (1984) recommendsthat, "We would do well to try to understand what children's learning is like, what the child istrying to do" (p.605).Three hypotheses were generated as a result of the literature review.1. Children's conceptual levels of understanding at the ages of 6 years, 8years, and 10 years will correspond to the developmental sub-stages of"dimensional thought" in middle childhood as described by Case (1985,1992) such that there will be significant age-related differences in level ofunderstanding of the meaning of learning.2. Children's conceptual levels of understanding at 12 years of age will indicatea transition to the Vectorial Stage. There will be a significant differencebetween 10-year-olds and 12-year-olds, indicating a qualitative shift inunderstanding of learning from intentional to interpretive.3. Children's conceptual levels of understanding of whether learning comesfrom an external source, internal source, or both sources will show adevelopmental progression from the ages of 6 years to 12 years in ahierarchical fashion.36Methodology used to test these hypotheses is presented in the following section.37CHAPTER 3 : METHODOLOGYChildren's explanations for the meaning of learning were elicited using a semi-structuredinterview technique. A brief overview of the pilot study conducted in the Fall of 1992 will first bediscussed followed by a detailed account of the final study procedures. The chapter will concludewith a description of the scoring criteria established to interpret the results of the study.Pilot StudyA pilot study was conducted in September 1992 to develop specific interview guidelinesand to field test both the questions and types of responses obtained from children at the fivedifferent age levels.A total of 28 students from an elementary school in the Greater Vancouver Regional Districtparticipated in an interview with the researcher. Six 4-year-olds who lived in the school'scatchment area were also included in the pilot study. A small sample of children representing eachof the following five age-levels responded to a set of-four questions in an informal interviewsituation: 4 years, 6 years, 8 years,10 years and 12 years.The keystone to the success of the final study is based on subjects' responses to a set ofquestions and the interviewer's skillful use of probes. Therefore, the researcher felt it necessary togain insight into the nature of the responses and to confirm research predictions prior tocommencing the study. By conducting pilot interviews, procedures and questions could be refmedto maximize the type of information required for the study.During the pilot interviews, it became evident that different kinds of probes were requiredto either clarify or extend children's responses. Consequently a set of probing questions andstatements was developed to be used in the final interviews whenever they were deemed necessary.The structured questions used in the pilot study required very little revision, but the order in whichthey were presented was changed for the final interviews.In addition, the data collected from these pilot interviews helped to provide insight into thegeneral parameters of thought units emerging at each of the different age groups, so that scoringcriteria could be developed. Although 4-year-olds were not included in the major study, theirresponses enabled the researcher to gain insight into the nature of the responses at this age in order38for a base to be established on which to build a hierarchy of scoring levels to categorize children'sconceptual understanding of learning.In summary, the pilot study confirmed research predictions that the resulting data could bequantified and analyzed in the manner intended.Procedures Subjects. Eighty-three subjects ranging from 6-years to 12- years old were individually selected fromthe same elementary school in which the pilot study was conducted. A stratified random samplingprocedure using gender and age as the two factors for subject selection was used. The populationincluded 20 six-year-olds (mean age = 6.7), 20 eight-year-olds (mean age = 8.8), 20 ten-year-olds(mean age =10.7) and 23 twelve-year-olds (mean age = 12.7). Each age group was evenlydivided by sex (10 males and 10 females) with the exception of the 12-year-old age group (10males and 13 females). Due to the small population of 12-year-old students enrolled at the school,subjects at this age-level came from one class. Every student in this division received a letter forparental permission to participate in this project and consequently all students who returned theirpermission slip wanted to be interviewed. Subjects were randomly selected from four differentclassrooms for each of the the other age groups due to the preponderance of primary students atthis school. Descriptive statistics are presented in Table 3.1. The mean age is calculated in monthsfor each age group.The interviews took place between January and March of 1993 and were individuallyconducted in a room set aside by the school for the purpose of this study. The school setting forthe study is a suburban, lower-middle class neighbourhood adjacent to an Industrial Park (thesurrounding area was once rural and has been recently developed into many housingsubdivisions). The Elementary school has a population of 485 students and is in its second year ofoperation.Prior to commencing the interviews, a letter requesting parental consent was sent homewith all subjects selected (See Appendix B). Written parental permission was required to intervieweach student. In addition, the interviewer informally asked for verbal consent from each subject.Ecological validity was established by conducting the interviews at the school in which thechildren were enrolled. Not only was the school considered to be the most convenient setting in39which children's classroom routines would be the least disturbed, but also the most naturalisticsetting for children to talk about their understanding of learning.Interview. In order to standardize procedures, all interviews were conducted in essentially the samemanner. These interviews, described as "informal conversations," in Piagetian terms, weremodelled after his approach. For instance, when Piaget was asked by an interviewer how heknew when to end the interview, he responded : "I have only one criterion. I consider aninvestigation finished when we no longer find out anything new, that's all." (Bringuier,1980, citedin Arlin,1990, p. 82).TABLE 3.1Descriptive Statistics for Age GroupsAge Group^N^ Mean Age(M, F) in Months(SD)6^ 20^ 79.1(10,10) (3.8)8 20 103.8(10,10)^(3.3)10^ 20 127.3(10,10) (2.7)12 23^ 151.5(10,13) (3.7)Piaget's clinical method (Piaget, 1965) is a special type of interviewing technique intendedto reveal the nature of children's thoughts. Interviewers using this method are also observers asthey watch how children react to the questions posed to them.Feedback from the pilot interviews confirmed the need to support the structured questionswith open-ended probes to clarify and extend responses when needed (See Appendix C). In otherwords, these unstructured questions facilitated the explanation and understanding of the responsesto the structured questions. A letter was attached to each type of probe to serve as a coding systemwhen transcribing each subject's taped responses.40Based on this description, the interviews conducted in this study used a semi-structured approach.The researcher in the role of an interviewer elicited verbal responses to the following set of fourquestions:1) What does learning mean?2) What is happening when you are learning?3) When you are learning, where is your learning coming from?4) Can it come from anywhere else?The first two questions selected were intentionally designed to elicit each subject'sexplanation of what learning means to them. A definition for the "source" of learning wasassessed by asking children a third question: When you are learning, where is your learningcoming from? The intent of this question was to determine individual and cross-age perceptions ofwhether learning comes from an external or internal source or both.The interviews took place over an eight week period from January 8 1993 to March 51993. At the onset of each interview time was taken to engage in friendly conversation to not onlyestablish rapport but to put subjects at ease. It was crucial to establish from the very beginning thatthe interview was based on a non-evaluative activity. The interviewer communicated to eachsubject that it was not a matter of looking for the right answer, but provided each subject withencouragement to share his or her thoughts about what learning means. A preamble to eachinterview was administered as an introductory statement prior to asking the set of four questions(See Appendix D - Setting the Stage).Each subject was individually interviewed by the researcher for about 10 - 15 minutes in asmall room located near the school library. The order of the structured questions remainedconstant throughout, but the use of probes naturally varied according to each subject and interviewsession.Subjects' responses to the set of four questions were tape-recorded during the interviewsession and subsequently transcribed verbatim on to a protocol sheet (Appendix C). All probingquestions were included where applicable by means of a code. A letter was attached to each typeof probe to serve the purpose of identifying the type of probe used by the interviewer to clarify orextend a subject's response at different times during the session.41ScoringCoding the interview transcripts involved both judgment-making and the development ofcategories prototypic of each age group. Whereas quantitative studies are usually based on presetcategories for both analysis and statistical hypotheses testing, qualitative researchers avoidestablishing categories prior to the data collection so that they will be open to the patterns ofbehaviour that emerge from the data itself. The latter was the procedure followed in this study.The first step undertaken by the researcher was to examine the responses by age todetermine commonalities within each age group. Children's conceptions of learning wereoperationally defined from the pooled responses to Question 1 and 2. Common patterns of thoughtwere then extracted from each age group, which enabled the researcher to develop age-relatedcategories. Theoretical predictions, proposed by Case's (1985, 1992) neo-Piagetian model, helpedto generate these categories for each age group. Using this model as a theoretical framework, age-related differences could be explained in terms of age-level postulates of the general theory. Basedon these characteristics, prototypical response categories were created, which could potentiallyreflect any child's conception of learning at any age. A description of these categories is presentedin the section describing scoring levels.Regarding the "Source" task, in which subjects were asked to determine where theirlearning was "coming from", pooled responses to Questions 3 and 4 were analyzed and rated asone of the following categories:1. External source2. Internal source3. Both external and internal sources mentioned4. An integrated responseA more detailed description of these categories can be found in the section outlining scoring levels.Each response protocol was then scored with respect to its conformity to one of thecategories in its definition of learning. In addition, the "source" response was also classified andnumerically recorded. Although language could be considered an uncontrolled variable in thisstudy, the researcher ensured that each subject's response was not penalized for languageimmaturity. This was accomplished by paying minimal attention to the surface language structuresin the response such as the levels of grammatical, syntactical and vocabulary sophistication, and bydevoting more attention to the overall meaning. The scoring criteria permitted a subject to obtain a42score at any level, with a "barebones" response, as long as the "thought units" met the criteria forthe postulated structure. Very few subjects in this study displayed language as a barrier inconveying their thoughts to the interviewer.Scoring Criteria. A detailed description of the scoring criteria is stated in this section. First, the researcherholistically examined all 83 protocols and from the response productions was able to identify threedifferent types of information across the four age groups:1. Responses defining learning as a sequence of specific behavioural events2. Responses defining learning as a mental act3. Responses defining learning as a combination of both a behavioural eventand mental actWithin each category, more specific "thought units" were defined and a list of descriptorswas developed to help identify the type(s) of information children refer to when asked to explainwhat learning means. These descriptors were created from the "pooled" responses to Questions 1and 2. Three general categories were established from the data and within each category a list ofmore detailed and refined descriptors were generated. They are presented in order of complexity.1. Type "B" Information - behavioural events. (Bruner's Landscape of Action)1. Response relates learning to an external action or an event performed by self as active agent.These acts can be either:-a) specific behavioural acts:For example :- "Going to school""Learning how to read or write""Learning about penguins""Printing new words such as "you""Counting up to 10""Going to the library and taking out books."or b) general behavioural acts :For example :- "Doing something new"43"Trying to do your best at something."2. Response relates learning to an external action performed by another person; subject becomespassive agent- receiver of information.For example:-^"Someone teaching you something""Listening to the teacher""It's where you get taught things by people who have alreadyexperienced those things."3. Response relates learning as a social interaction between self and others, both being activeparticipators :-For example:-^"My Dad showed me how to do^then I tried to do it on my own.""My teacher teaches me new things and then I can tell it to someone else.""I can learn from people who have already learned it."In other words, there is a recognition of the essentially social and interpersonal nature oflearning. However the level of conceptual understanding in this response is that learning takesplace in the form of a sequential event, that is, subject is first a passive recipient of informationand then becomes an active agent of the newly acquired knowledge.4. Response incorporates the interactive component of #3, but acknowledges that learning can bereciprocal.For example:-^"You can learn from each other""You share your ideas with each other during class discussion."There is also reference to cooperative learning.For example:-^"You can practise for a test with a partner. You could have somebody giveyou a little test that has the same kind of questions but not exactly the samequestions and you try to answer them. Then you test your partner."With this response, "Self" is clearly stated as a self-motivated active agent who draws upon avariety of external sources to maximize learning.For example:-^"Other people will have suggestions and answers and you listen to44what they have to say. You might get an idea from what you've heardfrom other people or you just use your own."2. Type "M" Information - a specific mental state or process.(Bruner's Landscape of Consciousness)1. A mental state may be represented as a feeling or judgment (Griffin, 1992) which is related toan action or to learning itself. Usually judgmental statements fall into two categories :a) Personal Judgments - for example, responses that infer personal values such as thefollowing :"Learning is very special to me""I like learning""Learning is fun when you get to play games""Learning is good for you... it helps you...""Learning can be pretty hard or fun stuff..."or b) Social Judgments - for example, learning judged in terms of social values and/orimplications :"Learning is important because you need to get a good job.""You've got to learn because you've got to get a good educationto get a good job so you can get money to support a family."2. A mental process may be defined as an internal act functioning separately or concurrently withan external event. Reference to a mental process may be represented by the following terms:"knowing", "understanding", "remembering", "memorizing", "thinking" and "listening"; orphrases such as: "making things up", "getting good ideas", "Your brain is absorbing theinformation" and "You have all that knowledge and it's going around and around in yourbrain."3. Type "B + M" Information - a combination of a behavioural event with a specific mental state or process. Responses that are classified as this type demonstrate that behavioural events in the physical45world are differentiated from but related to mental states.1. Response demonstrates an awareness of an internal or mental state in terms of a feeling orpersonal judgment related to an action or to a learning experience in general. This responsedemonstrates an attempt to coordinate two qualitatively different structures - a behaviouralevent with a mental state in a unidimensional manner. One structure (behavioural event) isused as a means to draw a conclusion about the other. Generally the event is related to apersonal judgment, (e.g.,"getting smarter", "If I do good work I get happy.").2. Response refers jointly to a behavioural event and two mental states instead of one. Ajudgment statement and a personal component are related to an event. Information is receivedfrom external events or agents and then internalized, (e.g., "I'm thinking about it (work)","Your brain receives all this information."). External acts are differentiated from but related tomental acts, (e.g., "You're thinking of something and trying it out."). In addition, an attemptto coordinate the two states is made by implying a cause and effect relationship, (e.g.,"Youpractise a lot so you get better at it" ; "You're getting smarter because you're learning more.").There is a focus on a second dimension in this response. Learning is becoming morepersonalized. Therefore, this response demonstrates an attempt to coordinate the two structures in a bidimensional manner.The integration that is beginning to take place involves the subordination of one structure(external state) to another (internal state), in a means-end fashion. There is a qualitativedifference in the concept of "self" as a result of a learning experience with the understandingthat personal effort plays a key role in this new "inner" concept of self, (e.g.,"You get betterat it if you try your hardest.";"Most of the time you're trying to see what you can do.").3. Response demonstrates a higher-order relationship in its attempt to integrate internal andexternal states and/or experiences during the learning process. The two intentional states - ajudgment and a personal component are coordinated in an elaborated bidimensional manner.The "personal component" is more differentiated and explicitly defined such that preferredlearning styles (seeing, hearing, doing) or different sources are incorporated into thedescription of learning.46There is more flexibility in the definition of learning. For instance, responses are becomingless specific and more generalized in their descriptions.For example:"To me learning means knowing how to do something without having anyproblems doing it.""It means getting an education so when you grow up you cangraduate... .get a good job make lots of money so you can have agood living."Different modalities of learning are explained and cause and effect relationships becomemore complex, (eg.,"If you memorize something, you remember it."; "Someone explains it,and you remember it.").In addition, responses to question 2: What is happening when you are learning? - include_a number of alternative behaviours (both external and internal) that take place during a learningexperience. There is also an indication of a conceptual understanding that learning takes placein a sequential fashion from external to internal, (e.g.,"The person is telling you what thisthing is, and you're putting it into your head.").4. Response relates learning to a self-motivated intrapersonal act. In other words, learning isdefined as an internal state of mind while at the same time the learning process isconceptualized as reciprocal (external to internal ; internal to external). The followingexamples characterize this type of response:"Learning is developing a smarter mind.""Learning is gaining knowledge ^expanding your horizons^taking in information^ it's when things go into your brain and youjust remember it.""If you learn something, you can see it in a new light, a differentopinion from what you used to have."There is also an understanding of the interactive nature of learning between one's learningexperiences in the physical world and an individual's psychological state. According to neo-47Piagetian theory, social cognition at the Vectorial Stage may be defined as an interpretivestructure. This is evident in the individual's description of their personal learning style.Conceptual understanding of what learning means has moved from the concrete to theabstract at this level. The implication of this response is that learning is an individual process,the subject being the active agent (e.g.,"You take the instruction and make it work."). There isalso a recognition that learning can be transferred to "real life" situations in adulthood. Forexample, "You've got to do math, so you can do your taxes when you're older." Theseresponses demonstrate the beginning of hypothetical thinking.Scoring Levels.The "Definition" Responses. Responses were scored and categorized into five levels of structural complexity consistentwith the age-level characteristics postulated by Case's neo-Piagetian theory. Using thesetheoretical characterizations, and the data obtained from the pilot study, a 4-year-old's definition oflearning is described as a sequence of behavioural events (e.g., playing, going to school, buildingtowers) a the presence of a learning "agent", such as Mom, Dad, or a teddy bear. Even if there isa conceptual awareness of an internal state at this age, (e.g.,"You learn something you neverlearned before."), it is predicted that 4-year-olds do not generally relate an intentional state toexternal events. This level of thinking is defined as the pre-dimensional stage.Six-year-old thought is labelled unidimensional, with only one intentional dimensionconsidered at a time. An intentional dimension can be defined as a causal relation betweenbehavioural acts and internal states and feelings.Eight-year-old thought, in contrast, is labelled bidimensional, suggesting that two intentional dimensions will be simultaneously considered and coordinated with a behavioural actbut only at a rudimentary level. Prototypic responses at this age level revealed two intentionalstates: a judgment and a self-component which are both related to a behavioural act.Ten-year-old responses can be defined as "elaborated" bidimensional. Like the previoussubstage, two coordinated intentional states are present, but are now represented in a higher-orderrelationship by integrating and differentiating the "personal component" in the description oflearning.48Twelve-year-old thought manifests the emergence of formal-operational thinking and atransition occurs to the Vectorial Stage. New structures transcend those of the previous stage.Learning is defined as an internal state of mind and is conceptualized as both an interactive andreciprocal process between the physical world and the psychological world.Level 0 (Age 4 years)A score of 0 was assigned if the subject's response conformed to the theoretical predictionsof 4-year-old conceptions of learning. For instance, if a child referred exclusively to only one typeof information, specifically a learning experience in the form of an observable external action orevent then he or she received a score of 0. In addition, a score of 0 was awarded if there was theexclusive mention of a learning "agent" (Mom, Dad, or other person(s) or even an object such as ateddy bear).Responses at this level referred only to a sequence of external events and there was noreference to an internal state. As a result, this level is labelled the pre-dimensional stage.PROTOTYPIC 4-YEAR-OLD RESPONSES'CONCEPTIONS OF LEARNINGPre-dimensional StageBehavioural Event Going to schoolPlaying and growing upBuild a zoo, towerLearning to readTeddy BearsIntentional StateI don't knowLevel 1 (Age 6 years)A score of 1 was assigned if the response conformed to the predictions for 6-year-oldconceptions of learning. At this level, a child referred jointly to a learning experience in the formof a behavioural event or learning "agent" (Teacher, Mom, Dad) and a personal judgment thatindicated an awareness of an internal state as an outcome to the learning experience.49For example:"Sometimes it (learning) gets me frustrated because I make so manymistakes""If I do good work I get happy""I learn lots of things when I do my math and spelling, so I can pretty soonget to High School".This level of thought parallels the unidimensional stage hypothesized by Case (1985,1992).If the term "fun" was included in the response, it was not considered a personal judgmentas the researcher considered that subjects were not referring to a mental state but a behavioural act.PROTOTYPIC 6-YEAR-OLD RESPONSES'CONCEPTIONS OF LEARNINGUnidimensional StageBehavioural EventLearning how to read/writeColor or drawLearning about penguinsHaving funGetting ready for Grade 2Intentional State (Feeling or judgment)Getting smarterGetting to know stuffYou have to spellSpecial, something goodIt's when you get good ideas.Level 2 (Age 8 years)A score of 2 was assigned if the response conformed to the predictions for 8-year-oldconceptions about learning. These responses referred jointly to a behavioural event and/or alearning "agent" with an attempt to coordinate two intentional states. In a similar manner to 6year-old responses, a judgmental attitude is mentioned but in addition, a personal component isnow incorporated into the response as a second intentional state. This information is described byincorporating "self" in the experience with the understanding that some personal effort is involvedto bring about learning. As a result of this personal effort, a new "inner" concept of self isrecognized.For example:"You get better at it if you try your hardest""Most of the time you're trying to see what you can do."50The implication of a cause and effect relationship is inferred between the physical andmental states, in an attempt to attribute an internal state as a consequence to this externalexperience.For example:"We do better things 'cause we learn more about it.""You learn the question off by heart and then you can get better andbetter at the thing you're learning.""You grow better in learning and get more things done.""If you keep on learning , you'll probably know more things you doright now.""You listen to the teacher and that's when you learn how to do stuff."The following terms are used in this age's responses to describe what is happening duringlearning:figuring out questionsthinking about your workmemorizing itIt takes practise to learn how to^At this level, there is evidence that an internal process is happening during learning. The twointentional dimensions are simultaneously considered and coordinated to the behavioural event in arudimentary manner. This level of thought parallels the bidimensional stage hypothesized by Case(1985, 1992).PROTOTYPIC 8-YEAR OLD RESPONSES'CONCEPTIONS OF LEARNINGBidimensional StageBehavioural EventLearning different thingsWhen you learn how to do somethingSomeone's helping you learnSomeone's helping you learn stuff(Math, Social Studies)You learnWhen you learn somethingYou'll just get better at itIntentional State You're most of the timeTrying to see what you can doMy things are going into my brainIt goes into your memory and itmeans working.. .that helpsYou can't forget itYou're thinking, trying to figure outquestions51Level 3 (Age 10 years)A score of 3 was assigned to responses that conformed to the prediction for 10-year-oldconceptualizations. These responses demonstrated a higher-order relationship by integrating andelaborating both internal and external states during the learning process. For instance, the"personal component" becomes more differentiated such that different modalities (seeing, hearing,being shown/kinesthetic) and/or different sources (books or people) are incorporated into a morecomplex description of learning.The underlying concept in these responses relates learning to a social interactive process.Often self becomes the passive agent (e.g., "Someone explains it and you remember"; "Beingtaught by the teacher").Cause and effect relationships include judgmental statements with social implications,(e.g.,"Learning means getting an education so when you grow up you can graduate and get agood job, make lots of money so you can have a living").The definitions of learning are now becoming more generalized and both external andinternal processes are more varied and flexible. There is also a transition towards a conceptualunderstanding that learning takes place in a unidirectional fashion from external experiences tointernal/mental processes. (See description B + M #3)PROTOTYPIC 10 YEAR-OLD RESPONSES'CONCEPTIONS OF LEARNINGElaborated Bidimensional StageBehavioural EventLearning means getting a goodeducation because when yougrow up you need to learnGetting to know new thingsYou've got to learn 'cause you'vegot to get a good education toget a job to workLearning means people teachingIntentional StateSomeone explains it and youremember itYou listen and write things downEvery time I do it, I get a little better at itYou're knowing moreStuff is flowing through your brainYou have to think in order to learn,learn in order to think52Level 4 (Age 12 years)A score of 4 was assigned to responses that conformed to the predictions of 12-year-oldthoughts about learning. These responses manifested a more "internalized" sense of what learningis. There is an understanding of the "inner world" of learning which marks a shift from an"intentional" level of understanding to an "interpretive" level of understanding consistent withCase's neo-Piagetian model of cognitive development. Learning is described as the increase ofknowledge, as "memorizing things so you can remember it in your brain until you need it later on",as "knowing and understanding things you didn't know before". At this level of conception,learning is perceived as an increase in knowledge together with the concept of understanding.An intrapersonal perspective relating to a subject's personal style is evident in his/herresponse to Question 2: What is happening when you are learning? The interaction of self in thephysical world with reference to a variety of sources and experiences is more differentiated at thislevel. Further, there is a notable transition from "practising" to "memorizing" and a developingawareness of building on prior knowledge.PROTOTYPIC 12-YEAR- OLD RESPONSES'CONCEPTIONS OF LEARNINGUnivectorial StageBehavioural EventExpanding your horizons so whenyou're older you'll know what to doGaining new knowledgeBeing able to experience new thingsLearning is thinkingIt's keeping things organized and storedin your head for when you need them,like in a testRemembering what people tell youInterpretive StateIf it's something I like, it stays in mymind and I remember itYou have to understand it before youcan learn itThink it through before I can understand itTrying to understand what you're thinkingThink about other things that I know willrelate to thatStudying, you memorize it so you'llknow itObserving, listening, trying it53The " Source " Responses. Children's responses to the question, "When you are learning,where is your learningcoming from?", were scored as follows:1. An External Source. If learning was located in either an action or a "learning agent" (e.g.,going to school, an object such as a book, or another person usually an adult (Mom, Dad, orteacher), a score of 1 was assigned.2. An Internal Source. If location was made within the physical or psychological self (eg., head,brain, mind), a score of 2 was assigned.3. External and Internal. If both Sources are recognized independently of each other, a score of 3was assigned.4. An Integrated Response. External sources are distinctly combined with an internal sourcedemonstrating a conceptual understanding of their interrelatedness.For example:-"You read it and then you think it over.""Learning can come from a number of places but practice comes from you,from the brain, from things around you.. .like when you're listening, you justremember it.""It's coming from words. It's like the brain is talking ...talking outside andtalking inside.""From my brain and people and things""Somebody talks to me, my brain will store it and explain it to me.""It's like taking something from the outside into the inside.. .like takingsomething from outside of a box and putting it inside of the box and keeping itthere".A score of 4 was assigned for such a response.The complexity of the "Source" responses varies with age. In the early ages there is a widegap between external and internal sources. They are stated in isolation of each other but as thestudent matures intellectually, there is a significant closing of this gap. There is an increasingawareness of learning taking place in a sequential manner from the external to the internal. Fromthe ages of 6 to 12 years, a funnelling process occurs until the interrelatedness of the two sourcesis clearly identified. This synthesis of an external and internal source is generally recognized by 12years of age. In addition, there is also evidence that 12-year-olds are beginning to make a shifttowards developing a conceptual understanding that learning may also be initiated from the internalto the external.54A second rater, unfamiliar with the study, was brought in to measure the interraterreliability of the scoring criteria. The researcher provided the rater with a copy of the scoringprocedures and gave a brief explanation of the different levels of performances. A practice sessionthen followed until the researcher felt the rater was both comfortable and competent to scoreprotocols independently.A random sample of six interview protocols representing each of the four age-groups wasscored by the second rater. Initially 5 out of 6 scores were confilmed in the 6-year-old group, anddiscrepancy over the one protocol was resolved through discussion, resulting in 100% agreement.There was total interrater agreement on the scores for both the 8-year-old and 10-year-old agegroups. Regarding the scoring 12-year-old protocols, an initial 67% agreement was obtained.Most discrepancies in scoring were resolved through discussion resulting in 83% agreement forthis age group. The overall independent ratings of the children's responses resulted in 96%agreement.Statistical analyses of the children's scores will be presented in the next chapter.55CHAPTER 4: RESULTSThe purpose of this chapter is to determine if there is empirical support for the hypothesisthat children follow a cognitive developmental progression in their conceptual understanding ofwhat learning means. To recap briefly, the main goal of this study was to confirm the predictionthat there would be significant age-related differences in level of understanding of the meaning oflearning.To test this hypothesis, qualitative data generated from the interview protocols were codednumerically to be analyzed quantitatively.The quantitative analyses in this chapter are based on two scores extrapolated from theinterview data. The first section discusses the results of a statistical analysis of children'sresponses to the "Defniition" task which requested a meaning for the mental term 'learning'. Thesecond section provides an analysis of age-level responses to the "Source" questions whichrequired the students to tell where learning "comes from".Analysis of the "Definition" TaskDescriptive statistics. Table 4.1 provides the means and standard deviations of the level scores attained by each ofthe four age groups for the "Definition" task. These mean scores are compared with the predictedscore based on the expected age-level responses consistent with Case's model and operationalizedin the scoring criteria.Analyses of variance. A one-way analysis of variance (ANOVA) for gender was performed to determine theeffect of gender on the mean level scores. The results indicated no gender effect and the samplewas analyzed as a whole.A one-way analysis of variance (ANOVA) was performed on mean level scores todetermine whether there was a significant difference due to age. The results indicated a significanteffect for age (F (3,1) = 115.98, p < .0001). A summary of these results is provided in table 4.2.The Bartlett test was administered to test homogeneity of variance. Results indicated thatthe assumption was not violated (Bartlett-Box F (3,1) = .42, p = ns).56TABLE 4.1Mean (SD) Level Scores by Age Group (Predicted and Obtained) on "Definition" TaskAge PredictedMeanObservedMean(SD)6 1 1.1 [.55]8 2 2.1 [45]10 3 3.05 [.51]12 4 3.7 [.45]TABLE 4.2Analysis of Variance Summary Table for the Mean Level Scores on the "Definition" TaskSource Sum of DF Mean F FSquare Square Ratio ProbabilityBetween 83.62 3 27.87 115.98 .00GroupsLinear 83.25 1 83.25 347.6 .00TennQuadratic .50 1 .50 2.08 .15TennCubic .05 1 .05 .19 .67TermWithin 18.99 79 .25Post-hoc comparisons were then calculated to specifically locate where the age differencesoccurred. The Scheffe Post-hoc Multiple Comparison method with alpha set at .05 was applied forthis purpose and the results showed significant differences between each age group.Test for linear trend. Figure 1 graphically represents in linear form the observed relationship between age andmean level scores attained. A comparison is made between the observed and predicted age-levelsix^eight^ten^twelve• predictedM obtained43.532.521.510.5057means for each of the four groups. The predicted line was based on the expected mean level scorefor each age group according to Case's developmental model.FIGURE 1Mean Scores achieved by Four Age Groups on "Definition" Taskin Relation to the Predicted Linear Trend58A trend analysis was performed to determine the linear relationship between age and levelinteraction to see if the predicted developmental progression was attained. The results indicated asignificant linear trend (E (3,1) = 347.60, g = < .0001) with non-significant deviations from thepredicted linear trend (Table 4.2). Although there is a slight indication at the 12-year-old level of adecrease in mean level score, no curvilinear trend was observed. The quadratic term was notsignificant. The reason for a lower mean level score for this age group could be a result of aceiling effect since subjects' response productions could only be allocated a score of 4 or less. Nocriteria was established to determine a level 5 response that would rate above this predicted agelevel. As a consequence, no 12-year-old response received a score of 5.Within all four age groups there was a certain percentage of response productions thatscored above or below the prototypic level. Table 4.3 summarizes the percentages of subjectsscoring above, below and at the prescribed level. The results indicate that each group performedclose to age-level expectations on the "Definition" task. Although the performance of the 12-year-old group is slightly lower than the theoretical expectation, this deviation may be accounted for bythe presence of a ceiling effect in the scoring criteria. However the results in general suggest thatthe hypothesized prototypical knowledge components unique to each age level were empiricallyvalidated.Table 4.3Percentages of Responses Falling At, Above and Below thePrototypic Level for Each Age Group.Years 6 8 10 12Predicted 70% 80% 75% 74%LevelPercentage 20% 15% 15% *0%AbovePercentage 10% 5% 10% 26%Below* A result of a ceiling effect, since scoring criteria above this predicted age level were notestablished.59In summary, the results of this statistical analysis indicate significant differences inresponse performance on the "Definition" task between all four age groups. This verifies theprediction of age-related differences in conceptual understanding of the meaning of learning acrossall four age groups.The significant linear trend supports the hypothesis that children's levels of understandingof the meaning of learning increase in complexity with age. The predicted developmentalprogression through the four age groups was established. Further the significant linear trendconfirms that the majority of children responded at the predicted level according to age.Analysis of the "Source" TaskIdentical procedures were performed on the responses related to the "Source" task. Torecap briefly, children's responses to the question: "When you are learning where is your learningcoming from?" were numerically categorized as follows:A score of 1 was assigned if the response referred to an external source;A score of 2 was assigned for an internal source;A score of 3 was assigned for the recognition of both an internal and an externalsource; andA score of 4 was assigned to an integrated response that demonstrated a conceptualunderstanding of their interrelatedness.Descriptive statistics. Table 4.4 presents the means and standard deviations of the level attained at each age forthe "source" responses.Analyses of variance.A one-way analysis of variance (ANOVA) for gender was performed to determine theeffect of gender on the mean level scores. The results indicated no gender effect and the samplewas analyzed as a whole.60TABLE 4.4Mean (SD) Scores on the "Source" Task by Age GroupAge N Mean (SD)6 20 1.85 [.88]8 20 2.15 [.99]10 20 2.85 [1.09]12 23 2.9^[1.38]A one-way analysis of variance (ANOVA) for age was performed on the mean level scoresto determine whether there was a significant difference due to age. The results are summarized inTable 4.5. The analysis indicated a significant effect for age (F. (3, 1) = 4.63, p < .01).The Bartlett test was administered to test homogeneity of variance. Results indicated thatthe assumption was not violated (Bartlett-Box F (3,1) = 1.55, p = ns).Multiple comparisons (Scheffe) were conducted to locate pairwise age differences. TheScheffe procedure revealed a significant difference (p < .05) between 6-year-olds and 12-year-oldsonly.Test for Linear Trend.Figure 2 graphically shows the observed relationship between age and mean scores on the"Source" Task. A significant linear trend (F(3,1) = 13.02, p <.001) was obtained with nosignificant deviation from this trend.TABLE 4.5Analysis of Variance Summary Table for the Mean Level Scores on the"Source" TaskSource Sum of DF Mean F FSquare Square Ratio ProbabilityBetween 17.13 3 5.71 4.63 .01GroupsLinear 16.01 1 16.01 13.02 .00TermQuadratic .29 1 .29 .24 .63TermCubic 1.08 1 1.08 .88 .35TermWithin 97.48 79 1.2361Figure 2Mean Scores achieved by Four Age Groups on the "Source" Task in Linear Form6263The following bar graphs (Figures 3, 4, 5, and 6) graphically display the distribution of thefour levels of "Source" responses in percentages at the ages of 6 years, 8 years, 10 years and 12years respectively.An analysis of the responses provided by 6-year-olds indicated that:45% expressed an external source25% expressed an internal source30% expressed a combination of bothNo subject at this age provided an integrated response.The distribution of 8-year-old responses was as follows:35% provided an external source20% provided an internal response40% provided a combination of both5% provided an integrated responseThe distribution of 10-year-old responses was as follows:20% provided an external source -5% provided an internal source45% provided a combination of both30% provided an integrated responseThe distribution of 12-year-old responses was as follows:30% provided an external response4% provided an internal source9% provided a combination of both57% provided an integrated responseTable 4.6 presents the pattern of performance of each subject, across four ages and fourcomplexity levels for the "Source" task. They are plotted in relation to each subject's scoreobtained on the "definition" task.Figure 3Distribution of "Source" ResponsesAge 664Figure 4Distribution of "Source" ResponsesAge 865Figure 5Distribution of "Source" ResponsesAge 1066Figure 6Distribution of "Source" ResponsesAge 126768Table 4.6Pattern of Performance of Each Individual Subject'sResponse to the "Source" TaskLevel Source 6 8 10 12Response4 Integrated *** ************Both **InternalExternal ***3 Integrated * *** *Both ********Internal * *External ** *** ****2 IntegratedBoth * ******* *Internal ** ****External * ***** *1 IntegratedBoth ***** *Internal ***External ******0 IntegratedBothInternalExternal **In summary, the results of a statistical analysis on children's "Source" responses indicate asignificant difference between the 6-year-old group and 12-year-old group only. Although notstatistically different across all age groups, the "source" attributions showed a steady increase incomplexity as children matured with age.An interpretation of the results of this study will now be discussed in the final chapter.69CHAPTER 5: DISCUSSIONThe chapter will begin with a detailed interpretation of the results of this study. It willexamine children's conceptions of the meaning of learning followed by children's conceptions ofthe "sources" of learning, that is,whether learning comes from an internal or external source orboth. Cross references and further research implications will be included when deemedappropriate. This section will be followed by a summary of the overall results of the study,approached from different perspectives. The chapter will conclude with a short discussion onpossible educational implications as a result of this study.The purpose of this study was to chart conceptual understandings of the meaning oflearning at the ages of 6, 8,10 and 12 years and to describe qualitatively different substages inchildren's development of ideas of learning. Four different conceptions of learning wereidentified which became systematically more complex with age. These conceptions wereconsistent with cognitive development as modeled by Case (1985,1992), a neo-Piagetian theorist.The predicted developmental progression in children's conceptualization of the meaning of learningwas supported. The major limitation of the study is that the participants in this project wererandomly selected from only one elementary school. This therefore restricts the possibility ofgeneralizing the results to a broader population of students of similar ages.This study revealed that children's conceptions of learning from the ages of 6 years to 12years take distinct forms. These forms were analysed from both a structuralist and content-focusedperspective using Case's model of cognitive development as a theoretical framework. To recapbriefly, neo-Piagetian theory relies heavily upon content-related information in response to aspecific concept under investigation, in order to analyze the levels of performances in terms of theircognitive structural complexity. From a structural standpoint, prototypical responses of 6-, 8-, and10- year-old subjects reflected the "intentional" structure hypothesized as typical of theDimensional Stage of Case's theory, while prototypical responses of 12-year-old subjects markeda distinct transition to the "interpretive" structure typical of the Vectorial Stage.During the interview sessions, it became very apparent that this study presented childrenwith a challenging task by asking them to explain the meaning of learning. The requirement of anexplanation of such a phenomenon entails the understanding of learning as a whole. Sincelearning is not one phenomenon but many, depending on what learning is about, both the context70and the content had to be taken into consideration when defining structural units of thought.Structures represent a relation between the child's way of thinking and the content on which thethinking is focused. Content itself depends heavily upon a child's experience and development.As a consequence of taking the learner's perspective on the nature of learning, everyday learningsituations have to be in focus as children draw upon their experiences to explain such aphenomenon. In other words, children construct a conceptual understanding of learning byattempting to relate what they already know or have experienced about learning.Surface differences in content were generally ignored in the age-level scoring criteria sincethe postulates of the neo-Piagetian model are couched at the level of deep structure and generaldomain content. The coordination between the external world of action and the internal world ofmental states and processes captured the general pattern of subject responses in their attempts toarticulate their conceptions of learning. Further, by representing the underlying structure thatcomprises children's responses, a basis is provided for developing scoring criteria. For example,if intentional state information in the form of feelings, judgments, or mental processes was relatedto a sequence of behavioural events in a child's response, an appropriate score based on theoreticalpredictions was assigned.Initially, many subjects, particularly at the younger age levels, found this task quite difficultbut with the help of probes and time to think, all subjects were able to respond to the set of fourquestions. The majority of subjects admitted that the opportunity to explore their own thinkingabout the meaning of learning was a new experience to them in terms of articulating theirunderstanding of and clarifying their beliefs about what learning means. As one studentperceptively mentioned during the conversation, "Sometimes you don't even know you'relearning. ..Like probably right now I'm learning."Children's conceptions of learning were dealt with in terms of two related aspects: what islearning, dealing with children's understanding of learning as a whole and the "where" aspect,investigating children's perceptions of the "source" of learning, that is, whether they think learningis coming from external or internal sources or both. The first part of the chapter will discuss theresults of the study as they relate to these two aspects.71The"Definition"TaskChildren's responses to the "definition" questions supported the age-level predictions ofCase's (1992) model. To ensure that the results did not simply reflect language competence aschildren matured rather than their level of understanding of learning, "bare-bones" responses wereaccepted at any level as long as they complied with the scoring criteria for the postulated structure.Verbal fluency, therefore, did not become a contributing factor to the increase in complexity ofolder children's responses. This concept of a "bare-bones" response was taken from the Griffin(1992) study briefly described in Chapter 2. General characterizations of children's developmentin their conceptual understanding of learning will now be summarized.Children's concept of learning at the age of 6 years is generally associated with schoolactivities such as learning to read, write, or do math. This could indicate that learning is perceivedas a phenomenon that mainly takes place within the school context. The definition of learning ischaracterized by a sequence of behavioural events related to school. Learning as a consequence ofdoing an activity or a skill represented the common pattern of thought of this age group. In otherwords, learning was defined as a sequence of external actions that resulted in a changed internalstate such as "getting smarter" or "getting to know how to do something." Generally speaking, a6-year-old's conception of learning is equated with a positive mental state as evidenced in theirmental reference "to know." This supports Flavell's (1988) belief that young children have learnedthat they can be "cognitively connected" to things in the external world generally through sight, andin light of this study, through action also, for example: "You need to write a lot. ..to help you knowstuff." The term "know" was generally used in response to question 2, "What is happening whenyou are learning?" In the context of Bruner's (1986) theory, this sort of coordination ischaracteristic of thought in the narrative mode - Landscape of Action in the form of a behaviouralevent and the Landscape of Consciousness in the form of an internal state. Although 6-year-oldswere able to demonstrate a clear distinction between external events and internal states, they wereunable to explain psychological activities involved in the learning process that promote the"knowing" of something. In fact they indicated a very limited repertoire of mental processes asthey relate to learning. From a structural perspective, responses typically reflected the firstsubstage of Case's Dimensional period of cognitive development, which postulates twoqualitatively different structures (action and consciousness) coordinated in a unidimensionalfashion.72Children at this age were unable to define learning in general terms. Their responsestended to rely heavily upon their immediate experiences, for instance, "learning about penguins";"printing new words" and "learning how to share books". The child's world at 6 years of age is aworld of actions in which they are actively involved and this is clearly reflected in their way ofthinking. The framework within which a 6-year-old's understanding can be categorized is thatlearning means to become able through experience in terms of action. Thus learning is perceivedas a change from one state to another. For example, learning in terms of "how to do something" isseen as a distinction between not being able and being able. This indicates a transition from onestate to another and children as young as 6 years are developing an awareness of this gradualchange occurring within them. When this happens, an important step has been taken in a child'smanner of thinking. The acknowledgment of an improvement in ability implies the understandingof a process progressing from being able to becoming more able. Underlying this conception,there is the idea of continuity: learning taking place over time along a continuum, which isadvocated in the ministry document, Year 2000: A Framework for Learning (1990) and is alsoconsistent with research (Bruner,1986; Pramling,1990; Saljo,1976; Vygotsky,1978). This studycan conclude that children as young as 6 years are aware of their own active role in learning andalso of their own thinking, but do not yet relate their own thinking to learning, nor their active roleto thinking.Whereas 6-year-olds had a very limited repertoire of mental processes associated withlearning, 8-year-olds demonstrated an awareness of different cognitive processes which interactwith external information to promote the acquisition of knowledge. Definitions of learning weremore general and less specific, but the concept of learning by doing is still present in an 8-year-old's thought. In response to Question 2, "What is happening when you are learning?" differentmental references were incorporated in appropriate contexts, such as "figuring out thequestions";"thinking about it (the work)" and "memorizing so you can't forget it." This indicatesthat, at the age of 8 years, children are consciously aware of mental activities and are beginning todevelop an understanding of the different ways the mind acts upon information during the learningprocess.The most significant qualitative difference from 6-year-old conceptions is the inclusion ofpersonal effort in the learning process. It may be that it is at this point in a child's development thatthe true concept of learning is born. When a child becomes aware of his/her own active role in the73process of learning, an important step has been taken. This more "mature" conception of learningimplies that 8-year-olds have developed a deeper understanding of the role of experience inlearning as compared to 6-year-old conceptions. Although children at 6 years of age are cognizantof being active learners, there was not a conscious awareness of the significance of personal effortin the learning process. The concept of practice in order to learn something was an active indicatorof 8-year-olds' understanding of their role in the learning process. From this, it can be concludedthat children at this age believe that an activity has to be consciously exercised in order for learningto take place. When children grasp this concept of learning as being able to do something throughpractice, continuity in terms of gradual improvement is understood. The data suggest a significantdifference between 6-year-old and 8-year-old conceptions of learning. With the former, theconcept of knowing is attributed to external events which supports Pillow's (1988) externalizationhypothesis (discussed in Chapter 2), while the latter relate learning to practice and demonstrate anunderstanding of cognitive activities such as memorizing involved in the process. In other words,there is a significant difference in exercising a skill on purpose and learning it due to the fact thatone happens to be involved in its execution.In the context of neo-Piagetian theory, prototypic responses of 8-year-olds are consistentwith the Bidimensional Stage of Case's model. A behavioural event was coordinated with twointentional states, that is, a mental process and a judgment statement associated with a "personaleffort" component.During the early 1980s, inquiry into children's beliefs about effort and interest as theyrelate to learning was conducted. A number of studies indicated that children believe that interestand effort influence learning. For example, Wellman, Collins, and Glieberman (1981) found that5-year-olds understood that effort and the number of items to be recalled contributed to memoryperformance; effort was regarded as the more important variable. The authors concluded thatthese results suggest that effort may be a particularly salient aspect of mental activity for youngchildren. This notion is consistent with the findings of the present study.Similarly, children tend to judge interest as influencing learning and also as affectingperformance (Miller & Zalenski,1982). In Miller and Zalenski's study, interest was conveyed as apositive characteristic by 3- and 4- year-olds, similar to trying hard, and both were correlated withpositive outcomes. The authors predicted that "children may conceive of variations in the intensityof mental activity as affecting performance before they understand specific activities and their74consequences" (p.871). What can be determined from these studies is the fact that children relatepositive attributes to positive outcomes but are yet unable to understand the underlying causalrelationships involved.However, by the ages of 6 and 8 years, the findings of this study suggest that children aredeveloping a repertoire of psychological activities associated with learning. It is also evident fromthe response productions that, during these two years, children make a dramatic increase in theirunderstanding of their own involvement in the process of learning. However, this raises theconcern that the younger children may have conceptions of learning which they do not expressspontaneously. In the interview situation the main idea was to enter into a developing interactionwith each subject, in terms of motivating and stimulating the child to explore his/her thoughts to thegreatest extent possible. It may be quite possible that children's conceptions are dependent on thequestions being asked. For many of these children it was their first reflection about thephenomenon of learning since generally speaking, learning is usually taken for granted. All of asudden, the children were confronted with a task in which they had to think about and verbalizewhat is usually "taken for granted."The study confirms a hierarchical structure of qualitatively different conceptions as childrenmature which are consistent with the age-related postulates of Case's model. In one respect, onecould say that these conceptions are additive. Older children revealed a greater variety ofmodalities (seeing, hearing, being shown) and sources of learning (teacher, books, T.V.); they hadaccess to a wider repertoire of cognitive strategies to promote learning. To have a moredifferentiated repertoire means a greater flexibility with regard to content. Such was the case with10-year-old responses which reflected a knowledge of a variety of learning strategies and preferredlearning styles. The general conception of learning at this age can be characterized as informationabsorbed through listening, watching or doing, which is subsequently internalized and stored inthe brain. In other words, there was a common understanding of learning taking place in asequential fashion from external stimuli to an internal state of knowing with an increasedknowledge of mental processes involved in the learning process.Counter to 6- and 8-year-olds' conceptions of learning, the general pattern of 10-year-oldthought was the conception of self as the passive recipient of knowledge as was evidenced in theirexplanations of the meaning of learning. Comments such as the following reflect this conceptionof self as a passive learner: "listening to the teacher"; "getting taught how to do things (math, or75science etc.); "Learning means people teaching you"; "Learning is when a teacher asks or assigns aquestion and you have to answer it"; "Learning is when you get taught things by people who havealready experienced those things." The results of a comparison between the general pattern ofresponses to question 1 and 2 suggested a possible contradiction of understanding of whichsubjects may have been unaware. For instance, prototypic responses to question 2: "What ishappening when you are learning?" expressed an awareness of the mind as an active receiver ofinformation, for example, "stuff (information) is flowing thorough your brain all the time whileyou're learning."; "Your brain is working harder...it's thinking."; "When you have a mathquestion to answer, you put your brain to work to sort the question out to find the answer." ; "Youhave to think in order to learn, and you have to learn in order to think." Such comments indicatethat 10-year-olds are developing a conception of their own active mental experiences even thoughthey view themselves as mainly passive learners. This contradiction of thought suggests that themore salient or visible aspects of learning continue to dominate 10-year-old conceptions. Forexample, often when asked, "What are you doing when you are learning?" responses tended tolean towards the visible or more overt external behaviours such as "paying attention in class";"taking notes if the teacher is talking"; "looking in books." Conceptions of learning at this age-level could be considered in a transitory phase of bridging the gap in conceptualizing learning interms of concrete thought to conceptualizing it abstractly or symbolically.The concept of learning at the 10-year-old level corresponds to the fmal substage in Case'sDimensional Stage of cognitive development. Responses represent an elaborated bidimensionallevel of understanding of learning by integrating internal and external states in a more complexfashion. This is reflected by references to preferred or alternative cognitive processes childrenexperience during the learning process.Another common factor that can be extracted from 10-year-old response data is theemphasis on the importance of learning from a social judgmental perspective, for example,"Learning means getting a good education, so when you grow up you can graduate, get a goodjob...make lots of money so you can have a living." Children's conceptions of the meaning oflearning are a consequence of their reflections on experienced reality and the impact of social andenvironmental influences on their beliefs about learning. Children's learning experiences stemfrom two different milieu, their home situation and elementary school. Within these two socialsettings, children are exposed to a variety of learning tasks, experiences, and influences and, as a76consequence, they reflect upon different realities. In spite of individual varied learningexperiences, it was possible to chart common patterns of thought at each of the four age-levels.Learning defined from a social judgmental viewpoint emerged at the 10-year-old age-level andextended over into the next age group. It seems reasonable to conclude that children's conceptionsat these age-levels may be strongly affected by social and parental influences. Further investigationin this area would be of value.Learning as a social interactive process is also characterized in 10-year-old responses. Thisconcept is very clearly summarized by one subject's comment: "Other people will have suggestionsand answers.... you listen to what they say and you might get an idea yourself." The data suggestthat children are beginning to acknowledge and appreciate others' viewpoints in order to makesense of their own learning about something.Although it is relatively straightforward to describe the minor transitions that occur withinone stage of neo-Piagetian theory, the analysis of a stage transition is often difficult to interpret.According to Case (1985), " The appearance of new structures at each stage appears to have acertain mysterious quality, as though they arose out of the depths of the child's intrapsychic life,by some little-understood transformation of the cognitive system as a whole" (p.72).The results of this study indicate the development of new structures as an outgrowth of theprevious stage. The two new structural units of thought that characterized the responses of 12-year-olds were:a) Psychological Self and its relationship to learning; andb) Social Interactive Process of Learning.As Case (1985) articulated: "The emergence of formal-operational groups has to await theconsolidation of more concrete operational groupings." (p.72). Children at 10 years of ageperceive learning taking place from external experiences to internal processes which would accountfor the combination of concrete definitions of learning in terms of salient observable learningbehaviours (e.g. "Learning to me is going to school, sitting down in a desk and learning from theteacher what she thinks.") and the emergence of abstract explanations of mental processes,(e.g.,"If you memorize something you remember it.") The link between the outer world and theinner world is expressed in terms of acquiring information through listening, watching, or doing.Twelve-year-olds conceptualized learning in abstract terms such as, "Learning is77developing a smarter mind." Learning was generally perceived as an internal state of mind thatwas associated with the concept of understanding. A higher-order conception emerges with thelink between gaining knowledge and understanding. The mental term "understanding" was presentin the majority of protocols at this age and was applied in the context that, without understanding,learning (in school) cannot be transferred to "real life" situations in the big world, for example,"You've got to learn math so you can do your taxes when you're older." Similar to 10-year-oldprotocols, the "personal component" perspective continues to remain a prominent feature in 12-year-old responses to the set of questions. However, there is one significant difference observedbetween the two age-groups with regard to the context in which it is described. A transition hasbeen made from the concept of self as a passive to an active agent during the learning process.Subjects at 12 years of age presented themselves as learners intrinsically motivated to gainknowledge through experiences in the physical world. As a consequence, conceptualization oflearning transcended from the unilateral direction (external to internal) to an understanding oflearning as a reciprocal interactive process (external to internal or internal to external) possibly as areflection of the changed image of self as the active agent in the learning process.-Whereas 10-year-old responses represented the predicted "intentional" structure (i.e.learning activities were qualified in terms of mental states that motivate them, such as personal orsocial judgments), 12-year-old responses indicated the presence of the predicted "interpretive"structure by providing a meta-perspective of one's interplay of psychological activities andphysical experiences to promote learning. In this sense, learning assumes a psychologicaldimension which reflects formal-operational thinking.At this point in the discussion, it is important to mention that a possible ceiling effect maybe the contributing factor towards a relatively low mean score of 3.7 at the 12-year-old level ofconception. Scoring criteria for a Level 5 (a predicted score for a14-year-old conception oflearning) was not generated and consequently no 12-year-old subject was allocated a level 5.However three protocols were rated as 4+ due to the presence of additional information thatsurpassed the criteria expected at Level 4. For example, included in these responses werereferences to the following:a) the concept of building on prior knowledge, for example,"I'm trying to understand and think about other things that I already know thatwill relate to that."78b) the concept of restructuring knowledge, for example,"I rehearsed it (new information) in my brain and put it how I want it....forexample if somebody teaches me something like math, I find an easier way.... Iwould do my own way rather than theirs."c) the concept of learning as the abstraction of meaning, for example,"If you learn something, you can see it in a new light.. ..have a different opinionfrom what you used to have." or "My brain takes in the information andexplains it to me, if I didn't understand it, I guess the brain would work itthrough more and help you know."These conceptions of learning assume a psychological dimension in their understanding ofhow cognitive processes are able to "interpret" or restructure information received from theenvironment. In other words, learning is conceived as an interpretive process with the intention ofmaking sense of the world in which they live. Responses incorporating information of this natureindicate the presence of a second dimension to the definition of learning. An attempt has beenmade to coordinate two interpretive structures to the Psychological Self (conceptualized as an activeagent in restructuring knowledge) in a bivectorial manner which raises the conception of learning tothe next substage in Case's Vectorial Stage.Psychological Self (PS) --- Social Interactive Process (SI)Psychological Self (PS) --- Interpretive Process (I)(Active Agent in Restructuring knowledge)The Source TaskChildren's responses to the "source" questions provided information on a secondperspective of learning. In this task children were asked to define their "sources" of learning, thatis, whether learning comes from an external source (located in an action or "learning agent"), aninternal source (located within the physical or psychological self), or both sources. Thecomplexity of the "source" responses increased with age but a statistical analysis indicated morevariance within the groups than the "definition" task, which suggests the presence of an individuallearning style variable as a possible contributing factor to a variance among responses of similarages. However, age-related patterns of thought were identified in the data.A brief discussion of the results in operational terms will now be presented. At the 6-year-79old level, sources of learning were generally attributed to personal "learning agents" such asteachers, moms and dads. One subject's response succinctly summarized the general pattern ofconception: "It's (Learning) coming from people you're learning from." Consequently, childrenidentify learning to know as a question of external influence (communication of other people'sexperience), while they regard learning to do things as a question of personal experience, forexample, "If I do good work, I get happy." From an educational perspective, it is important totake note of 6-year-olds' perception of the adult playing a significant role in the learning process.When an internal source was identified, children initially referred to their "head" which wassubsequently clarified as the "brain" and sometimes the "mind". Probing questions confirmed thatthe terms "mind" and "brain" were interchangeable in meaning at the ages of 6 and 8 years. Forwhenever an initial reference was made to the "mind," it was usually qualified as the "brain."Further probing into what children at this age knew about the brain, resulted in the identification of"thinking" as a function of the brain. When asked how thinking was associated with learning, thegeneral response alluded to the idea that "it helps you learn get ideas." It is worth mentioningthat a few 6- and 8-year-old responses expressed the "voice" or "mouth" as a source of learning,but it usually was an extension of their reference to the "brain," for example, "It's (Learning) goingaround and around in your brain and then comes out of your mouth." Empirical data such as thisrun counter to Piaget's belief that children fail to distinguish mental acts such as "thinking" fromovert behaviours such as "speaking". The data suggest that children recognize a sequence ofevents involving these two processes: thinking transformed into words and then spoken.However, further inquiry is necessary to confirm this hypothesis. The majority of 6-year-oldswere aware that their learning is transmitted to the brain, but their responses indicate that theyconceive of the mind as passive in relation to the external world even though they recognize aninternal change as a consequence of a learning experience.Similar results were identified with 8-year-old responses when reference was made toexternal sources. However, responses were more differentiated; there was a greater range of"learning agents" with a more frequent reference to friends, brothers and sisters. There was also agrowing awareness of learning taking place at home and at school with a small number of childrenmaking references to specific resources such as books or computers.An internal source was generally equated with the "brain" and the concept of "thinkinghard" which reflects the inclusion of "personal effort" in their definition of learning. In addition80there were several references made to the "body" and body parts such as "fingers" to count on,which reflect the concept of practice "in order to get it right." The general concept of the brain atthis age is a "place for storing things" as one subject expressed: "It's (brain) searching for moredata...just like a computer does in a game".Sources of learning are generally perceived as the media (books, T.V.) and theenvironment in general by the age of 10 years. An underlying concept of learning through bothperceptual and physical experiences surfaces at this age which suggests a growing understandingof learning as an interactive process between self and the outside world. Similar to 8-year-olds,the brain continues to be conceived as a storage place for incoming information through the senses.Some children made analogies in reference to the brain. For example, one subject related the brainto lots of bottles of knowledge: "It's just like people storing things in little bottles. It's thinking ofsomething and then keeps it and whenever you come back to it, you just have to open up that bottleremember it and then write it down."With 12-year-old responses, there was reference to a greater variety of sources andmodalities by which knowledge could be acquired as one interacts internally and externally withnew information in order to understand it. Perceptual experiences were more clearly defined aschildren were able to distinguish between firsthand and secondhand knowledge. The lattergenerally referred to verbal information communicated to them, for example, "Older people tellingyou about their experiences." Internal sources were frequently defined as "thinking aboutsomething in order to understand it" in contrast to references to the "brain" in younger children'sresponses. Only a few protocols referred to the process of building on prior knowledge as anextension to the meaning of "thinking".Summary Of The Results. In the early stages of development there is a clear distinction between external and internalsources. As children mature there is a developing awareness of the reciprocal process that takesplace between external and internal experiences as they are subject to and manipulated by a varietyof psychological activities during the learning process (e.g., "You read it and then think it over.";"Someone explains it and you remember it."; "Observing, listening or watching and trying it outyourself.") By 10 years of age, children clearly identified learning taking place in a sequential81manner from external to internal. By 12 years of age, the interrelatedness of the two sources wasgenerally recognized and, since children at this age have developed a conception of their own activemental experiences that control learning, there is an additional understanding of learning takingplace from internal to external. The developmental progression of the "source" responses from theages of 6 years to 12 years reflects a funnelling process; the two sources initially are conceptualizedas two isolated sources and as children mature, external and internal sources become siphoned intoone combined source. This synthesis is generally recognized by 12 years of age. In summary,these conceptions of the "source" of learning, although not statistically different across all age-groups, showed a steady increase in complexity which was hierarchical in nature.On close examination of the content of children's protocols, it can be concluded that 6- and8-year-olds' scope of general knowledge is limited whereas 10- and 12-year olds showed a moresophisticated understanding of this concept. Young children related knowledge to facts learned atschool which may be a consequence of their tendency to draw upon their closest and immediateexperiences to attach meaning to something (Wood,1988). Older students extended theirknowledge to the world in general with particular reference to nature and animals and alsodemonstrated an understanding of acquiring knowledge through a variety of resources such aspeople, books, and T.V. By 12 years of age, children had begun to realise that they could beginto self-direct their own learning (e.g.,"If I want to find out about Australia, I go to an encyclopediaand look it up").However, children's responses to the interview questions included a coordination of alltypes of knowledge. From the findings of this study, it can be hypothesized that the levels ofawareness will increase with age. For example, the relationship between declarative (facts,information) and procedural (strategies, rules) knowledge to induce learning was identified andapplied in many ways to define the meaning of learning. Conceptual knowledge was also presentboth in terms of beliefs about learning and where learning "comes from" and in their process ofconstructing meaning (schemes) to make sense of the learning phenomenon. Knowledge "inpieces" is meaningless and redundant in helping children develop a conceptual understanding ofspecific phenomena (diSessa,1988). It is feasible to predict from the results of this study thatyounger children may not be conscio usly aware that they are able to integrate and articulatecognitive activities in their explanations about their perceptions of learning. Conceptualunderstanding of specific phenomena demands an integration of all types of knowledge.82Unfortunately cognitive research attempts to isolate rather than integrate mental processes whichoften proves unproductive in terms of educational practice. If research programs placed greateremphasis on integrating rather than isolating cognitive acts in their inquiries into the nature oflearning, educators would be provided with information that would help in designing learningopportunities that enable such integration.SummaryIn summarizing the results of this study, it can be concluded that 6- and 8-year-olds' wayof thinking is dominated by the notion of action. In these early years a child approaches learning interms of concrete experiences which impact on his or her conception of the meaning of learning.Learning is most commonly expressed by children as becoming able to know something bydoing. The most significant difference between these two age groups is the differentiationbetween learning activities. Six-year-olds perceive learning as a sequence of events while 8-year-olds equate learning with practice involving personal effort. With the former conception, externalcircumstances create the learning, usually in the form of a learning agent (e.g.,teacher or parent),while self-regulation is the dominant aspect of learning by practice and recognized by 8 year-olds.By 10 years of age, social implications in terms of expectations and the importance of learningsurface as a key factor impacting on their conception of learning. Conceptions of self in thelearning process move from self as active learner to passive recipient of knowledge reflected intheir definition responses (e.g.,"Listen and watch to learn"; "Someone teaching you"). Althoughchildren at this age have an understanding of knowledge as a variety of mental activities, the actualconception of the mind as active remains in a transitory or dormant stage awaiting theconsolidation of concrete thought towards learning. It is not until 12 years of age that thedefinition of learning is expressed in symbolic or abstract terms. There is a conscious awarenessof the mind's active role in learning to ensure that all incoming knowledge is understood.It can also be concluded from this study that children's responses to the meaning oflearning from the ages of 6 years to 12 years support a cognitive constructivist viewpoint oflearning. Explanations were drawn from experiences in children's attempts to make sense of thephenomenon called learning. As children matured, new conceptions were an outgrowth of oldconceptions and yet previous conceptions did not disappear; they were woven into a more intricatedescription of the meaning of learning. For every new level of conception, the child has a broader83repertoire of previously conceived ideas from which to construct a more sophisticated andadvanced level of conception. This corresponds to the postulates of a cognitive constructivisttheory of development which maintains that new levels of conceptual understanding are built onwhat the children already know.There is a general trend towards more advanced conceptions of learning and yet, in light ofa similar study conducted by Saljo (1979) with first year university students, there have to be somereservations, since these earlier conceptions of children are also to be found among adults. Incomparing Saljo's (1979) results with this present study, there appears to be very little progressionin conceptual understanding from the ages of 12 to 18 years. It seems apparent that future researchinto adolescent conceptions may yield valuable information regarding this issue. The extension ofthis study to subjects aged 14, 16, and 18 years of age would complete the chart of plotting school-aged children's conceptions of learning in B.C.'s educational system.The developmental trend in children's conceptions of learning can be analyzed fromdifferent perspectives. For instance, the results of this study support Pillow's "passive to active"hypothesis in which he claims that there is a gradual developmental transition from conceiving themind as passive to conceiving of it as active. Whereas young children's view of the mind ispassive in relation to external experiences, older children's view of the mind is active whichreflects an increased awareness of their own psychological involvement in the learning process.This is of central importance in cognitive development in general as it empowers children tobecome active and independent learners.In looking at the results from a Piagetian point of view, it can be concluded that, on ageneral cognitive level, children's conceptions become more decentered. As the child matures,learning is seen from a broader perspective involving different perceptions and processes; there isan increase in abstraction in children's conceptions from 6 years to 12 years.The mapping of children's conceptions of learning as portrayed in this study is an exampleof what Marton (1981) described as an approach to research from a second-order perspective:children's conceptions of learning are elucidated from an internal perspective. The child's view oflearning at four different age-levels is described in qualitatively different categories that can helpeducators understand children's conceptions of learning at each age level and the developmentalstages in conceptual understanding of the meaning of learning from 6 years to 12 years of age.Current research in learning supports the notion that learning may be perceived in terms of84changed conceptions or changed meaning of the phenomenon learnt about. Subjects as young as6 years inferred that an internal change does occur from not knowing to knowing. The centralfocus of attention in older children's (10 and 12 years) responses to the question, "What ishappening when you are learning?" was the idea that learning is the reproduction of knowledge(e.g., "memorizing facts so you will remember them"). Rote learning and constructive learning aredistinctly different cognitive processes and, from the perspective of educational practice, are theunderlying principles for two differently structured teaching approaches. In addition, these twoconcepts of learning respectively represent the old and new beliefs of our educational system inB.C. The acceptance of children's learning in terms of memorization and rote learning promotes orsupports the concept of learning as an "activity of reproduction" (Saljo, 1979). Educators areexperiencing today a paradigm shift from conceptualizing learning as reproducing knowledge, a"surface-level" conception (Marton,1984]), to learning as a process of constructing meaning fromexperiences. In other words, "As students acquire new information and filter it through priorknowledge, they transform their individual beliefs into deeper, more coherent understandings"(The Intermediate Program: Foundations,1992, p.58).The research conducted by Marton and Saljo (1976a) and Van Rossum and Schenk (1984)suggested that a person's conception of the learning situation has implications for what one thinksone has to learn and how one will accomplish it. This therefore raises a very important question:Are children's conceptions of learning a reflection of the type of learning experienced in school as aresult of instructional methods and teachers' expectations or do they truly represent children'sattempts to make sense of what learning means to them?From another perspective: to what extent do variables such as parental and socialexpectations and teaching methods affect children's conceptions of learning? A comparative studybetween elementary or secondary school childrens' conceptions of learning and teachers'conceptions of learning (as reflected in their instructional methods) would provide valuable insightsinto the frequent misconceptions of the interactive teaching-learning process. There is often adramatic gap between the teacher's way of thinking and the student's way of thinking(Lindfors,1984). Lindfors analyzes this dilemma as follows:Often teaching-learning connections are not apparent at all, withteacher's efforts to increase children's knowledge and skills, andchildren's efforts to make sense of their world, going on quiteindependently as defined here, (they) are two distinct ventureswhich often get confused with one another. (p.600)85The most valuable skill we can teach a child is how "to think" and not just simply what tothink. To develop the child's thinking demands an awareness of the child's way of constructingknowledge. Conversations and interviews with students provide teachers with avenues to gaininsight into children's level of understanding of different concepts while at the same time providingan opportunity for children to reflect upon their own perspective on learning. A missing factor inmany educational programs today is the tendency to ignore student levels of understanding ofspecific concepts and how students organize and reconstruct new information in their efforts tomake sense of it.Identification of a child's existing level of conceptual understanding of the meaning oflearning will assist educators in designing effective instruction that promotes meaningful learningin conjunction with the child's conceptions of learning. By listening to, and observing children,the teacher should work with aspects of the child's reality which are meaningful and important tohim or her. It is important that teachers take the child's way of thinking seriously as the foundationstone on which to build curriculum. As Paley (1986) states, "The first order of reality in theclassroom is the student's point of view Someone must be there to listen, respond and add adab of glue to the important words that burst forth" (p.127). If we want to develop the child's wayof thinking, then we must begin by learning about how the child actually thinks. One of thecurrent goals in education is to support student learning. By providing opportunities during theday for children to think and reflect about their own learning, teachers are helping to developchildren's awareness towards a positive attitude in learning. There is a need in elementary schools'curriculum for teachers to encourage children to reflect on their own knowledge and learning.Elementary teachers often take for granted that learning is achieved as a consequence of anactivity. If the learning is seen in terms of basic skills this is probably true, but if learning isperceived in terms of a qualitative change in the child's way of thinking, there has to be anotherelement involved in the activity. Research provides evidence of positive outcomes whencurriculum incorporates the component of learning-to- learn interventions into programs.Pramling (1990) set out to determine whether children learn "better" if they are encouragedto become more aware of their own learning. This "descriptive-experimental" study involved fourgroups of preschool children between 5 and 6 years of age: two groups were experimental and theother two served as control groups. The teachers in the two experimental groups used a "didactic"approach with their students by directing the students' thoughts towards reflecting upon their86conceptions of learning in general and their own understanding of the content related to differentthemes. In everyday situations created by the teacher or the children, time was taken to reflectupon the particular activity in a way that promoted the children's metacognitive awareness abouthow they themselves and others think about the specific content.The results of this study showed that the children in the experimental group significantlyoutperformed the children in the control group in relation to their preparedness towards learningnew content. They also indicated an increased level of awareness of how they learn.A similar study examined two programmes designed to improve student learning in twohistory departments at British universities. Martin and Ramsden (cited in Richardson, Eysenck &Piper,1987) obtained undergraduate students' pre-course and post-course conceptions of learningby asking the question: "What do you actually mean by learning?"This information was then used to evaluate two different programs of learning skillsinterventions designed to enhance the process and outcome of student learning while taking ahistory course. Students who registered in the historysourse also participated in either a studyskills program or a learning-to-learn program which ran concurrently with their course. Theresults indicated a definite movement towards higher level conceptions among the learning-to-learnstudents. They progressed from initially conceptualizing learning as an "increase in knowledge ormemorization" to conceiving of it as "the abstraction of meaning or understanding reality"(Saljo,1979, p.158).Further, an examination of the relationship between students' conceptions of learning andtheir first-year performance revealed a positive association between grade and conception. Theresearchers emphasized that if interventions of this kind are to be of any success, they need to becarefully linked to both the content of what is being learned and to the context in which it is to belearned. However, they remained skeptical as to whether any significant development in thestudent's conception of learning or in learning strategies can actually be achieved in this manner.Although their results offer tentative support for further experimental interventions of a similarholistic nature, Martin and Ramsden reached the conclusion that activities designed to improvestudent understanding of learning are really inclusive in the whole process of teaching.This study presents an educational perspective on the metacognitive task of reflecting aboutone's own understanding of the meaning of learning. Children do not reflect spontaneously abouttheir learning and yet if the goal in education today is to assist children in becoming independent87learners, educators need to first of all understand children's conceptions of learning and theneducate from the child's point of view.CONCLUSIONBy revealing common patterns of understanding of the learning process in children ofsimilar ages, this study suggests that educational methods should be consistent with children'slevels of conceptual development (cf. Case & Mckeough, 1990; Case, Sandieson, &Dennis,1986). Galaper (1981) further claims that education should be designed to take intoaccount a cognitive-developmental sequence and the patterns of reasoning of children at differentstages in their development. Such procedures would help solve "the problem of the match"(Donaldson, 1979) between the learner's developmental stage and instructional methods.Children's understandings of what learning means and where it "comes from" should be a guidingprinciple in education.By engaging students in a conversation that enables them to explore their own thinking byarticulating their understanding and clarifying their own beliefs about the meaning of learning,teachers can assist children to become powerful thinkers and independent learners. The followingstatement from the Sullivan report, A Legacy For Learners (1988) provides an appropriateconcluding remark:To be effective, an educational system must respond to children inways that are suitable to their specific levels of intellectual, (social,emotional, and physical) development; this implies that these waysshould change as children mature. (p.103)Substage 3Integiated 84:11mens•ortalTNatight 19-11 Y•ars)• Ne.[SYNTHESIS]■■,. ..mos ••■••••^•••■ ••■•^■•■• twain. .a.m.^••■■••• Ammo ommia•Substage 2:134climerisionalThought (7-9 years)[THESIS & POTENTIALANTITHESISSubstage 1:UnidimensionatThought (5-7 ytarS)[THESIS]3 0.-ensoot06.anta88APPENDIX ADIMENSIONAL STAGEFigure 1 Abstract model of logico-mathematical progression during dimensional stage. Solid horizontallines indicate major transition in thinking. Dotted lines indicate more minor transition, as the fieldof centration expands and implications of a major transition are worked out. (Note: The generalform of notation in this figure is taken from Fischer (1980). The particular representation ofsubstage 3 thought is taken from Pascual-Leone (1969).) [Taken from Case, 1988189APPENDIX B[etter to Parents ies&humiewDear Parent or Guardian,Over this past summer I have been working on my thesis to complete my Master's Degree at UBC.My research is based on children's perceptions of themselves as learners, what they believelearning is. I would appreciate greatly your child's contribution to this project and requesttherefore your permission for him/her to participate in it.The purpose of this study is to provide an opportunity for children to talk about their ownthoughts about learning so that we as teachers may gain more insights into learning from a child'sperspective. Hopefully, such information will prove useful to teachers when planning instructionor designing programs.The study involves your child being informally asked some questions about what learning meansto him/her. I would like to meet with your child on an individual basis. Each interview will takeapproximately fifteen minutes and will occur at Gordon Greenwood School.Responses will be tape-recorded. To ensure confidentiality, no identifying information will berecorded and all of the data will be coded by number. Your child's participation is voluntary andyou may refuse permission if you wish, or withdraw your child from the study at any time.Refusal or withdrawal will not influence your child's class standing.I would very much appreciate your assistance with this study. Please sign this letter in the spaceprovided below indicating whether you do or do not agree to let your child participate, and return itto your child's teacher. Please also sign and retain the second copy for your own records. Shouldyou have any questions , please feel free to call me or Mr. Sasaki at the school at 882-0114. Inaddition, you may also contact the District Administrator for Planning at the School Board Officeor my faculty advisor Dr. Marion Porath who would both be pleased to discuss my study withyou. Dr. Marion Porath can be reached at 822-6045. Thank you very much for your cooperation.Sincerely yours,Gillian Bickerton^ , parent or guardian of^ do^ do not consent toallow my child to participate in this study. I acknowledge that I have received a copy of thisconsent form.Signature : ^ Date :^90APPENDIX CSTUDENT INTERVIEW PROTOCOLRESPONSE SHEETSubject # ^Age:^Date : ^Birthdate:^Question # 1: What does learning mean?Question # 2 : What is happening when you are learning?Question # 3 : When you are learning, where is your learning coming from?Question # 4 : Could it come from anywhere else?Possible probes that could be used during the interviewa) What do you mean by^b) Tell me more about it.c) What are you doing when you are learning?d) Ask for clarification for vague words such as "things", "stuff",Whatkindof^What do you mean by^e) If response to #4 is "your brain", probe: What is your brain doing when you're learning ?0 Extend a response by asking:Then what do you do with ^ (eg., information, ideasetc..).91APPENDIX DSetting the StagePreamble to Each InterviewEach child will be given the following introductory statement prior to being asked the set ofquestions.The interviewer:"Fm interested in how children learn and what they think learning is. It is veryimportant for me to find out what your thoughts are. I am writing a book and willbe using your ideas to help me understand what children your age think aboutlearning. Questions I'll be asking you today, I will also be asking other children ofthe same age. I want you when you answer these questions to be as honest as youcan be. Tell me what you think. There are no wrong answers. All I'm interested inis what you think."92REFERENCE LISTArlin, P. K. (1990). Teaching as conversation. Educational Leadership, October, 82-84.Astington, J.W., Harris, P.L., & Olson, D.J.R. (1989), Developing theories of mind. New York:Cambridge University Press.Berti, A., Bombi, A. & Lis, A. (1982). The child's conceptions about means of production andtheir owners. European Journal of Social Psychology , 12 (3), 221-239.Biggs, J. B. (1980). The relationship between developmental level and the quality of schoollearning. In S. Modgil & C. Modgil (Eds.), Toward a Theory of Psychological Development. Slough, Bucks: NFER.Biggs, J. B. (1985). The role of metacognition in enhancing learning skills. Paper presented atthe Annual Conference of the Australian Association for Research in Education, Hobart.British Columbia Ministry of Education. (1990). Year 2000 : A framework for learning. Victoria,B.C.: Queen's Printer for British Columbia.British Columbia Ministry of Education. (1992). The intermediate program : Foundation. Victoria:Ministry of Education.Brown, A.L. (1978). Knowing when, where and how to remember: A problem of metacognition.In R. Glaser (Ed.), Advances in instructional psychology (pp. 77-165). New York: HalstedPress.Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press.Case, R. (1985). Intellectual development: Birth to adulthood. New York: Academic Press.Case, R. (1987). Neo-Piagetian theory: Retrospect and prospect. International Journal ofPsychology , 22 (5-6) , 773-791.Case, R. (1988). Dialectical cycles in the development of creative intelligence. Journal of Social and Biological Structures, 11(1), 71-76.Case., R. (1991). A developmental approach to the design of remedial instruction. In McKeough,A. & Lupart. J. (Eds.), Toward the practice of theory-based instruction (pp. 117-147)Hillsdale, NJ: Lawrence Erlbaum.93Case, R. (Ed.) (1992). The mind's staircase: Exploring the conceptual underpinnings of children'sthought and knowledge. Hillsdale, NJ: Erlbaum.Case, R. & McKeough, A. (1990). Schooling and the development of conceptual knowledgestructures : An example from the domain of children's narrative. International Journal ofEducational Research, 13 , 835- 855.Case, R. , Sandieson, R. & Dennis, S. (1986). Two cognitive developmental approaches to thedesign of remedial instruction. Cognitive Development, 1 , 293-333.Costa, A. L. (1985). Developing minds: A resource book for teaching thinking. Alexandria, VA:Association for Supervision and Curriculum Development.Demetriou, A. (1987). The neo-Piagetian theories of cognitive development : Toward anintegration. International Journal of Psychology , 22 (5-6) , 679-728.diSessa, A. (1988). Knowledge in pieces. In G. E. Forman & P. Pufall (Eds.),Constructivism in the computer age. Hillsdale, NJ : Lawrence Erlbaum.Donaldson, M. (1978). Children's minds. New York: Norton.Donaldson, M. (1979). The mismatch between school and children's minds. Human Nature,March, 155 -159.Flavell, J. (1979). Metacognition and cognitive monitoring. American Psychologist , 34.Flavell, J.H. (1988). The development of children's knowledge about the mind: From cognitiveconnections to mental representation. In J.W. Astington, P.L. Harris, and D.R. Olson (Eds.),Developing theories of mind. (pp. 244-267). New York: Cambridge University Press.Galper, A., Jantz, R. K., Seefeldt, C. & Serock, K. (1981). The child's concept of age andaging. International Journal of Aging and Human Development, 12 (2),149-157.Goldberg-Reitman, J. (1992). Young girl's conception of their mother's role: A neo-structuralanalysis. In R. Case (Ed.), The mind's staircase: Stages in the development of humanintelligence. (pp.135-152). Hillsdale, NJ: Lawrence Erlbaum Associates.9 4Griffin, S. (1992). Young children's awareness of their inner world; A neo-Structuralist analysisof the development of intrapersonal intelligence. In R. Case (Ed.), The mind's staircase: Stagesin the development of human intelligence. (pp.75-97). Hillsdale, NJ : Lawrence ErlbaumAssociates.Hilborn, K. (1990, Spring) . Looking at learning : Facing a paradigm shift. Research Forum, 6,35-37.Johnson, C. N., & Wellman, H. M. (1980). Children's developing understanding of mentalverbs: "remember", "know" and "guess". Child Development , a , 1095-1102.Johnson, C. N., & Wellman, H. M. (1982). Children's developing conceptions of the mind andbrain. Child Development, 53 , 222-234.Lindfors, J. (1984). How children learn or how teachers teach? A profound confusion. LanguageArts, a (6), 600-606.Marshall, H.H. (1992). Seeing and redefining student learning. In H.H.Marshall (Ed.) , Redefining Student Learning : Roots of Educational Change. Norwood, NJAblex Publishing Corporation.Martin, E. & Ramsden, P. (1987). Learning skills or skill in learning. In John T. E.Richardson, M. W. Eysenck & David W. Piper (Eds.), Student Learning : Research in education and cognitive psychology (pp. 155-169). Milton Keynes : The Society for Researchinto Higher Education & Open University Press.Marton, F. (1981). Phenomenography - describing conceptions of the world around us.Instructional Science ,1Q , 177-200.Marton, F. (1988). Revealing educationally critical differences in our understanding of the worldaround us. Paper presented at the American Educational Research Association. New Orleans.Marton, F. & Saljo, R. (1976). On qualitative differences in learning : II-Outcome as a function ofthe learner's conception of the task. British Journal of Educational Psychology , 46, 115-127.Marton, F. Sz. Saljo, R. (1984). Approaches in learning. In F. Marton, D. Hounsell & N. J.Entwistle (Eds.) , The Experience of Learning. Edinburgh : Scottish Academic Press.Marton, F. & Svensson, L. (1979). Conceptions of research in student learning. HigherEducation, 8 , 471-486.95Marx, R. (1990, Spring). Learning, motivation and teaching. Research Forum , , 8-15.McGuiness, C. & Nisbet, J. (1991). Teaching thinking in Europe. British Journal of EducationalPsychology , 61(2) ,174-186.McKeough, A. (1984). Developmental stages in children's narrative composition. Paperpresented at the American Educational Research Association Annual Meeting, New Orleans.McKeough, A. (1986). Narrative composition from four to ten years of age: A neo-Piagetianperspective. Paper presented at the Canadian Psychological Association Annual Convention,Toronto.McKeough, A. (1987). Stages in storytelling: A neo-Piagetian analysis. Paper presented at theBiennial Meeting of the Society for the Study of Behavioural Development, Tokyo.McKeough, A. (1992). Testing for the presence of a central social structure: Use of the transferparadigm. In R. Case (Ed.), The mind's staircase: Exploring the conceptual underpinnings ofchildren's thought and knowledge. Hillsdale, NJ: Erlbaum.McKeough, A. (1993). Narrative knowledge across the school years. Paper presented at theCanadian Society for the Study of Education, Calgary.Miller, P. H.. & Weiss, M.G. (1982). Children's and adults' knowledge about what variablesaffect selective attention. Child Development, 53, 543- 549Miller, P.H. & Zalenski, R. (1982). Preschooler's knowledge about attention. DevelopmentalPsychology, 18 , 871 -875.Misciones, J. L., Marvin, R. S., O'Brien, R. G. & Greenberg, M. T. (1978). A developmentalstudy of preschool children's understanding of the words 'know' and 'guess'. ChildDevelopment, 48 ,1107-1113.Paley, V. (1986). On listening to what the children say. Harvard Educational Review, 56 (2),122-131.Pascual-Leone, J. (1987). Organismic processes for neo-Piagetian theories : A dialectical causalaccount of cognitive development. International Journal of Psychology , 22 (5 -6) , 531 -570.Piaget, J. (1926). The language and thought of the child. New York : Harcourt, Brace.96Piaget, J. (1929). The child's conception of the world. London: Kegan Paul.Piaget, J. (1930). The child's conception of physical causality. London: Kegan Paul.Piaget, J. (1969). The child's conception of time. New York : Ballantine Books.Pillow, B. H. (1988). The development of children's beliefs about the mental world. Merrill-Palmer Quarterly, M (1), 1- 32.Pramling, I. (1983). The child's conception of learning [Monograph]. Recent Research inPsychology. New York : Springer-Verlag.Pramling, I. (1988). Developing children's thinking of their own learning. British Journal ofEducational Psychology , 5 . , 266-278.Pramling, I. (1990) Learning to learn: A study of Swedish preschool children [Monograph].Recent Research in Psychology New York : Springer-Verlag.Richardson, J. T. E. (1987). Research in education and cognitive psychology. In J. T. E.Richardson, M. W. Eysenck & D. W. Piper (Eds.), Student Learning (pp. 3-12). MiltonKeynes : Open University Press.Saljo, R. (1979). Learning in the learner's perspective, 1 - Some common sense conceptions(Report No. 77). Sweden : Institute of Education, University of Gothenburg.Salter, D.J. (1993). Interpreting family stories: Narrative meaning constructed throughoutadolescence. Paper presented at the Canadian Society for the Study of Education, Calgary.Sandieson, R. & Case, R. (1987). Horizontal and vertical enrichment: A developmental approach of the teaching of central conceptual skills. (Final Report No. 1020-86-87). Ministry ofEducation, Province of Ontario, Canada.Shatz, M., Wellman, H. M., & Silber, S. (1983). The acquisition of mental verbs: A systematicinvestigation of the first reference to mental states. Cognition, 14, 301-321.Sodian, B., & Wimmer, H. (1987). Children's developing understanding of inference as a sourceof knowledge. Child Development,^, 424-433.97Sparks, G., & Cantor, J. (1983, May). Developmental differences in responses to the Incredible Hulk : Using Piaget's theory of cognitive development to predict emotional effects. Paperpresented at the Annual Meeting of the International Communication Association , Dallas,Texas.(ERIC Document Reproduction Service No. 236737).Sullivan, B. (1988). A legacy for learners : The report of the royal commission on education.Victoria : Province of British Columbia.Van Rossum, E. & Schenk, S. (1984). The relationship between learning conception, studystrategy and learning outcome. British Journal of Educational Psychology, 54, 73-83.Vygotsky, L. S. (1978). Mind in society. Cambridge, Mass : Harvard University Press.Wadsworth, B. (1984). Piagees theory of cognitive and affective development (3rd. Ed.) . NewYork : Longman.Watkins, D. (1983). Depth of processing and the quality of learning outcomes. InstructionalScience, 12 , 49-58.Wellman, H. M. (1985a). The child's theory of mind : The development of conceptions ofcognition. In S. R. Yussen (Ed.), The Growth of Reflection in Children (pp. 169-206). NewYork : Academic Press.Wellman, H. M. (1985b). The origins of metacognition. In D.L. Forrest-Pressley, G. E.MacKinnon & T. Waller (Eds.), Metacognition, cognition and human performance : Vol. 1.Theoretical perspectives (pp. 1 -31). New York : Academic Press.Wellman, H. M. (1990). The child's theory of mind. Cambridge, Mass : MIT Press.Wellman, H. M. , Collins, J. & Gliebermann, J. (1981). Understanding the combinations ofmemory variables : Developing conceptions of memory limitations. Child Development,52,1313-1317.Wellman, H. M. & Johnson, C. N. (1979). Understanding of mental processes: A developmentalstudy of "remember" and "forget". Child Development , 50 , 79-88.Wood, D. (1988). How children think and learn. New York : Basil Blackwell.