Analogical reasoning is a foundational skill necessary for enabling learners to draw inferences about new experiences, to transfer learning across contexts, and to make abstractions based on relevant information from daily experiences. Linguistic and executive function (EF) skills may support analogical reasoning ability, as both these skill sets have previously been shown to influence other higher-order cognitive abilities, such as perspective taking. Outside influences such as socio-economic status (SES) backgrounds may also influence analogical reasoning, as they have been shown to affect other cognitive processes. At present, current research offers little information about developmental relations among SES, language, EF and analogical learning. The purpose of this dissertation research was to explore the extent to which the provision of relational language facilitates children’s analogical reasoning, and to investigate the influence of SES, executive function and language skills in regard to such facilitation. Results indicate that the use of relational language indeed aids analogical reasoning. SES significantly predicted analogical reasoning, but interestingly, this was so only when relational language was absent. These findings support that relational language plays a key role in scaffolding analogical reasoning, and this support is particularly beneficial to children whose cognitive skills may be influenced by SES.

Research on analogical problem solving has delineated several factors that impact one's ability to spontaneously generate a correct solution strategy to a target problem. These factors include, but are not limited to, type of analogy provided to subjects (i.e., partial versus complete), the level of isomorphism between analogies and target problems, and the solver's level of analogical problem-solving expertise Recently, researchers have begun to focus on providing solvers with direct instruction on analogical problem-solving processes and strategies in an effort to augment analogical problem-solving ability. The most common type of instruction (i.e., teacher-generated) involves providing direct instruction on problem-solving processes and strategies without input from the solver. A second type of instruction (i.e., learner-generated) that has gained some attention in the literature but has not yet been tested in the realm of analogical problem solving involves learners actively participating in developing analogical problem-solving strategies while being guided by the instructor. Using an experimental design. the present study examined the differential effects of type of analogue (i.e., partial versus complete), level of expertise (i.e., novice versus expert). and type of training (teacher-generated, learner-generated, or no training) on spontaneous generation of correct solution strategies to two target problems. Findings indicate that solvers. regardless of training group or ability level. were better able to solve the target problem to the complete analogies than the target problem to the partial analogies. x2 (1, N = 116) = 18, p < .001; d = .85. Moreover, there was no advantage for expert solvers to participate in problem-solving training. However. when examining novice solvers. findings indicate that direct instruction on problem-solving processes and strategies resulted in better performance when solving the partial analogy than did no instruction (d = .61). Also, active participation in the learning process resulted in better performance when solving the partial analogy than did no instruction (d = .80). Limitations of the study, implications for educators, and recommendations for future studies are provided.

Chemistry textbooks and teachers frequently use a variety of metaphors, analogies and models to describe atomic and molecular structures and processes. While it is widely believed that multiple analogical models encourage students to construct appropriate mental models of chemical phenomena, uncritical use of multiple analogical models may actually be responsible for a number of alternative conceptions in chemistry. Students hear and read about electron clouds and shells, atoms that are like miniature solar systems and balls, and molecules that are simultaneously represented by balls-and-sticks, joined spheres, electron-dot and structural diagrams. A strong case has been made that students try to integrate these diverse analogical models resulting in the generation of unscientific synthetic models. Conceptual change research programs also propose that carefully designed teaching and learning activities can stimulate students to exchange their intuitive and synthetic conceptions for more scientific conceptions.This thesis investigates the occurrence of students' intuitive and synthetic mental models of atoms and molecules at both a general and specific level. The investigations consisted in the first phase of semi-structured interviews with 48 Year 8-10 science students. While the data were predominantly qualitative the interviews also generated simple quantitative data. The second phase was wholly qualitative and involved the researcher as teacher' in the Year 11 class. Portfolios were compiled for each student in the class and six portfolios were interpreted to produce a set of case studies describing the students' learning about atoms, molecules and bonds. These data were derived from transcripts of class discussions and individual interviews; pre-tests, formative tests and post-tests; student essays and worksheets and analogical teaching events. The data were ++ / interpreted from a constructivist viewpoint with attention given to credibility, viability and transferability, and dependability. The desire to collect every piece of useful data was constrained by the ethical need to minimise the disruptive effect of the research on the students' normal learning.The first or general phase of this study investigated the question: With what models of atoms and molecules are lower secondary science students familiar? The interviews about atomic and molecular conceptions held by the Year 8-10 students found, for example, that some students confused atoms with cells because both have a nucleus, while others believed that electron shells enclose and protect the atom. All but two students visualised atoms with large nuclei and close static electrons. A majority of this student sample were confused by ball-and- stick molecular models and had a strong preference for space-filling molecular models because they were more 'real'.The second or specific phase of this study consisted of an in-depth study of the development of mental models of atoms, molecules and bonds by six Year 11 chemistry students over 40 weeks of instruction. This study investigated the question: Do systematically presented multiple analogical models help students change their conceptions of atoms, molecules and bonds in favour of the scientific view? The students' prior mental models of an atom were dominated by a solar system model with the electrons in simple shells. A variety of metaphors, analogical models and explanations emphasising the diffuse spaciousness of atoms helped three students restructure their conceptions in favour of the scientific concept. Students also were encouraged to identify the shared and unshared attributes of familiar molecular models and, in time, three students became competent multiple modellers. It is claimed that these three students ++ / changed their conceptions of atoms and molecules in the sense that they realised that models are thinking and communicative tools, not reality itself. The significant change in these students' thinking was their recognition that atomic and molecular analogical models are context-dependent.The phase two study's pre-occupation with conceptual change or knowledge restructuring raised an important methodological question: Is a multi-dimensional approach a better way to interpret conceptual change learning? or, are the various theoretical perspectives on conceptual change complementary? The study's theoretical framework found that conceptual change learning can be interpreted from epistemological, ontological, motivational, holistic explanatory and developmental perspectives. The collection and analysis of the data showed that student modelling ability and Perry's model of intellectual development were powerful interpretive tools when data needed to be examined from multiple perspectives. The six case studies support the assertion that multi-dimensional interpretive frameworks have superior credibility and viability compared to uni-dimensional studies.Finally, the research raised several questions requiring further investigation. No direct support was found for the claim that dissatisfaction is central to conceptual change. This issue needs much more study due to the popularity of discrepant event teaching. While a multi-dimensional conceptual change model has been synthesised, this model needs further refinement as does the issue of how to monitor the status of students' conceptions. A most promising line of pedagogical research is the value of teaching scientific modelling through the use of multiple systematic analogical models.

Over the past few decades, the dominant view by comparative psychologists of analogical reasoning in nonhuman primates was one of dichotomy between apes, including humans, and monkeys: the distinction between the analogical ape and paleological monkey (Thompson & Oden, 2000). Whereas evidence for analogy proper by representation reinterpretation in monkeys is sparse and debated, the gap between that which is analogic and paleologic has been narrowed by the studies presented here. Representation of relational concepts important for analogy proves difficult for rhesus and capuchin monkeys without the ability to rely on a greater amount of perceptual variability, implicating a perceptually-bound predisposition in problem-solving (Chapters 2-3). A shift in attention from perceptual features to abstract concepts for employment in relational matching is again difficult, but not impossible given cognitive incentive in the form of differential outcomes to refocus attention on conceptual properties (Chapter 4). Finally, chimpanzees unlike monkeys appear more apt to reason by analogy, perhaps due to a more default conceptual focus (Chapter 5). Taken together, these studies provide an account for the emergence of analogical reasoning skills throughout the primate lineage in contrast to views regarding analogy a hallmark of human intelligence.

No description available.

No description available.

The acquisition of relational concepts plays an integral role and is assumed to be a prerequisite for analogical reasoning. Language and token-trained apes (e.g. Premack, 1976; Thompson, Oden, and Boysen, 1997) are the only nonhuman animals to succeed in solving and completing analogies, thus implicating language as the mechanism enabling the phenomenon. In the present study, I examine the role of meaning in the analogical reasoning abilities of three different primate species. Humans, chimpanzees, and rhesus monkeys completed relational match-to-sample (RMTS) tasks with either meaningful or nonmeaningful stimuli. For human participants, meaningfulness facilitated the acquisition of analogical rules. Individual differences were evident amongst the chimpanzees suggesting that meaning can either enable or hinder their ability to complete analogies. Rhesus monkeys did not succeed in either condition, suggesting that their ability to reason analogically, if present at all, may be dependent upon a dimension other than the representational value of stimuli.

Thesis (M.S.)--Worcester Polytechnic Institute. / Keywords: Analogy, Design, Functional Modeling, Functional Reasoning, Knowledge Representation, Repertory Grid, SME, Structure Mapping Engine, AI in design. Includes bibliographical references (p.106).

Le raisonnement par analogie est reconnu comme une des principales caractéristiques de l'intelligence humaine. En tant que tel, il a pendant longtemps été étudié par les philosophes et les psychologues, mais de récents travaux s'intéressent aussi à sa modélisation d'un point de vue formel à l'aide de proportions analogiques, permettant l'implémentation de programmes informatiques. Nous nous intéressons ici à l'utilisation des proportions analogiques à des fins prédictives, dans un contexte d'apprentissage artificiel. Dans de récents travaux, les classifieurs analogiques ont montré qu'ils sont capables d'obtenir d'excellentes performances sur certains problèmes artificiels, là où d'autres techniques traditionnelles d'apprentissage se montrent beaucoup moins efficaces. Partant de cette observation empirique, cette thèse s'intéresse à deux axes principaux de recherche. Le premier sera de confronter le raisonnement par proportion analogique à des applications pratiques, afin d'étudier la viabilité de l'approche analogique sur des problèmes concrets. Le second axe de recherche sera d'étudier les classifieurs analogiques d'un point de vue théorique, car jusqu'à présent ceux-ci n'étaient connus que grâce à leurs définitions algorithmiques. Les propriétés théoriques qui découleront nous permettront de comprendre plus précisément leurs forces, ainsi que leurs faiblesses. Comme domaine d'application, nous avons choisi celui des systèmes de recommandation. On reproche souvent à ces derniers de manquer de nouveauté ou de surprise dans les recommandations qui sont adressées aux utilisateurs. Le raisonnement par analogie, capable de mettre en relation des objets en apparence différents, nous est apparu comme un outil potentiel pour répondre à ce problème. Nos expériences montreront que les systèmes analogiques ont tendance à produire des recommandations d'une qualité comparable à celle des méthodes existantes, mais que leur complexité algorithmique cubique les pénalise trop fortement pour prétendre à des applications pratiques où le temps de calcul est une des contraintes principales. Du côté théorique, une contribution majeure de cette thèse est de proposer une définition fonctionnelle des classifieurs analogiques, qui a la particularité d'unifier les approches préexistantes. Cette définition fonctionnelle nous permettra de clairement identifier les liens sous-jacents entre l'approche analogique et l'approche par k plus-proches-voisins, tant au plan algorithmique de haut niveau qu'au plan des propriétés théoriques (taux d'erreur notamment). De plus, nous avons pu identifier un critère qui rend l'application de notre principe d'inférence analogique parfaitement certaine (c'est-à-dire sans erreur), exhibant ainsi les propriétés linéaires du raisonnement par analogie. / Analogical reasoning is recognized as a core component of human intelligence. It has been extensively studied from philosophical and psychological viewpoints, but recent works also address the modeling of analogical reasoning for computational purposes, particularly focused on analogical proportions. We are interested here in the use of analogical proportions for making predictions, in a machine learning context. In recent works, analogy-based classifiers have achieved noteworthy performances, in particular by performing well on some artificial problems where other traditional methods tend to fail. Starting from this empirical observation, the goal of this thesis is twofold. The first topic of research is to assess the relevance of analogical learners on real-world, practical application problems. The second topic is to exhibit meaningful theoretical properties of analogical classifiers, which were yet only empirically studied. The field of application that was chosen for assessing the suitability of analogical classifiers in real-world setting is the topic of recommender systems. A common reproach addressed towards recommender systems is that they often lack of novelty and diversity in their recommendations. As a way of establishing links between seemingly unrelated objects, analogy was thought as a way to overcome this issue. Experiments here show that while offering sometimes similar accuracy performances to those of basic classical approaches, analogical classifiers still suffer from their algorithmic complexity. On the theoretical side, a key contribution of this thesis is to provide a functional definition of analogical classifiers, that unifies the various pre-existing approaches. So far, only algorithmic definitions were known, making it difficult to lead a thorough theoretical study. From this functional definition, we clearly identified the links between our approach and that of the nearest neighbors classifiers, in terms of process and in terms of accuracy. We were also able to identify a criterion that ensures a safe application of our analogical inference principle, which allows us to characterize analogical reasoning as some sort of linear process.

The language-based analogical reasoning abilities of Deaf children are a controversial topic. Researchers lack agreement about whether Deaf children possess the ability to reason using language-based analogies, or whether this ability is limited by a lack of access to vocabulary, both written and signed. This dissertation examines factors that scaffold the development of language-based analogical reasoning through signed language. First it examines how background factors, such as age, race/ethnicity, or additional disabilities can affect the development of language-based analogical reasoning. Second, it looks at how different kinds of American Sign Language (ASL) vocabulary support the development of language-based analogical reasoning. Five-hundred and fifty-six Deaf children were given five tasks from the ASL Assessment Instrument; one analogies task and four vocabulary tasks: an antonyms task, a synonyms task, a definitions task, and a contextual-based vocabulary task. The data showed that background traits can and do affect how well Deaf children reason using language-based analogies. The most important predictor of performance on the analogies task was ASL vocabulary knowledge, although other factors such as age, race/ethnicity, and additional disabilities can impact task performance. The data also showed that ASL vocabulary knowledge that promotes metalinguistic thinking is the best predictor of language-based analogical reasoning abilities. Potential applications to the classroom and to teacher training are also discussed.