Submitted by Aelson Maciera (aelsoncm@terra.com.br) on 2017-09-06T18:51:39Z No. of bitstreams: 1 DissMAF.pdf: 890318 bytes, checksum: db8d31a4a8b322924c03f6f0c0b7c9d7 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-09-25T14:08:02Z (GMT) No. of bitstreams: 1 DissMAF.pdf: 890318 bytes, checksum: db8d31a4a8b322924c03f6f0c0b7c9d7 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-09-25T14:08:11Z (GMT) No. of bitstreams: 1 DissMAF.pdf: 890318 bytes, checksum: db8d31a4a8b322924c03f6f0c0b7c9d7 (MD5) / Made available in DSpace on 2017-09-25T14:11:54Z (GMT). No. of bitstreams: 1 DissMAF.pdf: 890318 bytes, checksum: db8d31a4a8b322924c03f6f0c0b7c9d7 (MD5) Previous issue date: 2017-03-15 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Self-regulation refers to thoughts, feelings and actions that are planned and cyclically adapted so as to make the individual reach their goals. Within the self-regulation is the metacognitive monitoring, the ability with which the individual follows and evaluates their own cognition. The present study was divided into two studies, both aimed to investigate the metacognitive monitoring of illiterate adults using the metacognitive judgement, a measure of the metacognitive monitoring, in cognitive tasks that evaluate the general factor of intelligence, processing speed and short-term memory. The relation between the real performance and the estimated performance was also investigated, including in which cognitive task the real performance related more to the estimated performance. In addition, we investigated which type of judgment model – the only difference between the two studies – was the most accurate. The participants of Study (1) included 34 functional or fully illiterate adults, from both genders, and age between 40 and 60 years old; and participants of Study (2) included 15 functional or fully illiterate adults, from both genders, and age between 20 and 74 years old. The instruments utilized were: Initial Interview, R-1 Test and subtests Digit Symbol-Coding, Symbol Search and Digit Span from the Wechsler Adult Intelligence Scale III (WAIS III), Judgment Register from 0 to 10, Judgement Register from 0 to 100 and Judgement Register with the use of a metric object. Results indicated that the real performance of the subjects, in both studies, were lower when compared to subjects with a greater level of education, and, according to the Pearson product-moment correlation coefficient, there was a significant correlation between the real performance and the estimated performance in the subtest Digit Symbol-Coding in Study 1, and significant correlation between real and estimated performance in the subtests Digit Symbol-Coding, Digit Span and Symbol Search in Study 2. In addition, results indicated that the judgment made with a metric object presented he most accurate measures. / A autorregulação refere-se aos pensamentos, sentimentos e ações que são planejados e ciclicamente adaptados para o alcance dos objetivos traçados pelo indivíduo. Dentro dessa totalidade, encontra-se o monitoramento metacognitivo, que é a habilidade pela qual o sujeito “acompanha” e avalia sua própria cognição. Foram realizados dois estudos, ambos com o objetivo investigar o monitoramento de adultos analfabetos, por meio do julgamento, entendido como uma medida do monitoramento metacognitivo, sobre o desempenho em tarefas cognitivas que avaliam o fator geral de inteligência, a velocidade de processamento e a memória de curto prazo. Também foi investigada a existência de diferenças nas relações entre desempenhos reais e estimados e em quais tarefas cognitivas o desempenho real mais se relaciona ao desempenho estimado. Além disso, investigou-se qual o modelo de registro de julgamento foi o mais acurado, sendo o tipo de modelo utilizado a única diferença entre os estudos. Participaram do Estudo (1) 34 analfabetos funcionais ou absolutos, de ambos os gêneros e com idade entre 40 e 60 anos, e, no Estudo (2), 15 analfabetos funcionais ou absolutos, de ambos os gêneros e com idade entre 20 e 74 anos. Os materiais utilizados para a coleta de dados foram: Entrevista Inicial (EI), Teste R-1 e os subtestes Código, Procurar Símbolos e Dígitos da Escala Wechsler de Inteligência para Adultos III (WAIS III), Registro de Julgamentos de 0 a 10, Registro de Julgamentos de 0 a 100 e Registro de Julgamentos a partir de um objeto de medida. Os resultados indicaram que, em ambos os estudos, os escores reais dos participantes foram mais baixos quando comparados com sujeitos de maior escolaridade das amostras normativas dos instrumentos. A partir das correlações de Pearson, observou-se no Estudo 1 uma correlação significativa entre o desempenho real e o julgamento do desempenho no subteste Códigos; e, no Estudo 2, correlações significativas entre o desempenho real e o julgamento de desempenho nos subtestes Códigos, Dígitos e Procurar Símbolos. Ainda, os resultados indicaram que o julgamento de desempenho a partir de um objeto de medida foi o que apresentou as medidas mais acuradas.

Improving the efficiency of spectrum access and utilization under the umbrella of cognitive radio (CR) is one of the most crucial research areas for nearly two decades. The results have been algorithms called cognitive radio engines which use machine learning (ML) to learn and adapt the communication's link based on the operating scenarios. While a number of algorithms for cognitive engine design have been identified, it is widely understood that significant room remains to grow the capabilities of the cognitive engines, and substantially better spectrum utilization and higher throughput can be achieved if cognitive engines are improved. This requires working through some difficult challenges and takes an innovative look at the problem. A tenet of the existing cognitive engine designs is that they are usually designed around one primary ML algorithm or framework. In this dissertation, we discover that it is entirely possible for an algorithm to perform better in one operating scenario (combination of channel conditions, available energy, and operational objectives such as max throughput, and max energy efficiency) while performing less effectively in other operating scenarios. This arises due to the unique behavior of an individual ML algorithm regardless of its operating conditions. Therefore, there is no individual algorithm or parameter sets that have superiority in performance over all other algorithms or parameter sets in all operating scenarios. Using the same algorithm at all times may present a performance that is acceptable, yet may not be the best possible performance under all operating scenarios we are faced with over time. Ideally, the system should be able to adapt its behavior by switching between various ML algorithms or adjusting the operating ML algorithm for the prevailing operating conditions and goal. In this dissertation, we introduce a novel architecture for cognitive radio engines, with the goal of better cognitive engines for improved link adaptation in order to enhance spectrum utilization. This architecture is capable of meta-reasoning and metacognition and the algorithms developed based on this architecture are called metacognitive engines (meta-CE). Meta-reasoning and metacognitive abilities provide for self-assessment, self-awareness, and inherent use and adaptation of multiple methods for link adaptation and utilization. In this work, we provide four different implementation instances of the proposed meta-CE architecture. First, a meta-CE which is equipped with a classification algorithm to find the most appropriate individual cognitive engine algorithm for each operating scenario. The meta-CE switches between the individual cognitive engine algorithms to decrease the training period of the learning algorithms and not only find the most optimal communication configuration in the fastest possible time but also provide the acceptable performance during its training period. Second, we provide different knowledge indicators for estimating the experience level of cognitive engine algorithms. We introduce a meta-CE equipped with these knowledge indicators extracted from metacognitive knowledge component. This meta-CE adjusts the exploration factors of learning algorithms to gain higher performance and decrease training time. The third implementation of meta-CE is based on the robust training algorithm (RoTA) which switches and adjusts the individual cognitive engine algorithms to guarantee a minimum performance level during the training phase. This meta-CE is also equipped with forgetting factor to deal with non-stationary channel scenarios. The last implementation of meta-CE enables the individual cognitive engine algorithms to handle delayed feedback scenarios. We analyze the impact of delayed feedback on cognitive radio engines' performances in two cases of constant and varying delay. Then we propose two meta-CEs to address the delayed feedback problem in cognitive engine algorithms. Our experimental results show that the meta-CE approach, when utilized for a CRS engine performed about 20 percent better (total throughput) than the second best performing algorithm, because of its ability to learn about its own learning and adaptation. In effect, the meta-CE is able to deliver about 70% more data than the CE with the fixed exploration rate in the 1000 decision steps. Moreover, the knowledge indicator (KI) autocorrelation plots show that the proposed KIs can predict the performance of the CEs as early as 100 time steps in advance. In non-stationary environments, the proposed RoTA based meta-CE guarantees the minimum required performance of a CRS while it’s searching for the optimal communication configurations. The RoTA based meta-CE delivers at least about 45% more data than the other algorithms in non-stationary scenarios when the channel conditions are often fluctuating. Furthermore, in delayed feedback scenarios, our results show that the proposed meta-CE algorithms are able to mitigate the adverse impact of delay in low latency scenarios and relieve the effects in high latency situations. The proposed algorithms show a minimum of 15% improvement in their performance compared to the other available delayed feedback strategies in literature. We also empirically tested the algorithms introduced in this dissertation and verified the results therein by designing an over the air (OTA) radio setup. For our experiments, we used GNU Radio and LiquidDSP as free software development toolkits that provide signal processing blocks to implement software-defined radios and signal-processing systems such as modulation, pulse-shaping, frame detection, equalization, and others. We also used two USRP N200 with WBX daughterboards, one as a transmitter and the other as a receiver. In these experiments, we monitored the packet success rate (PSR), throughput, and total data transferred as our key performance indicators (KPI). Then, we tested different proposed meta-CE algorithms in this dissertation to verify the productivity of the proposed algorithms in an OTA real-time radio setup. We showed that the experiments’ outputs support our simulations results as well.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2017-07-24T13:11:29Z No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) dissertação_GERDNA_2017.pdf: 3085732 bytes, checksum: 8699e86d5c3fb4cfbf12ddd6e572a599 (MD5) / Made available in DSpace on 2017-07-24T13:11:29Z (GMT). No. of bitstreams: 2 license_rdf: 811 bytes, checksum: e39d27027a6cc9cb039ad269a5db8e34 (MD5) dissertação_GERDNA_2017.pdf: 3085732 bytes, checksum: 8699e86d5c3fb4cfbf12ddd6e572a599 (MD5) Previous issue date: 2016-12-15 / CAPES / Esta pesquisa busca aplicar uma abordagem do ensino de leitura no intuito de contribuir na formação do leitor literário, através de estratégias metacognitivas de leitura que auxiliem os alunos a se tornarem leitores ativos e reconhecerem no conto o discurso universal inerente ao texto literário. Nosso embasamento teórico associa Estética da Recepção com Jauss (1979), Teoria do Efeito Estético, representada por Iser (1979), ensino de Literatura com Bordini &Aguiar (1993), Zilberman (2014) e Cosson (2014) à interação com o uso das estratégias de leitura, respaldado em Flavell (1976), Baker & Brown (1980), Solé (1998), Duke & Pearson (2002) e Kleiman (2007). Sendo assim, os teóricos da teoria da literatura e da psicolinguística dialogam para justificar a nossa proposta de entrelaçamento dessas áreas, visando um ensino de Literatura que respeite os aspectos literários e utilize os conhecimentos linguísticos para construção de significados do texto. A metodologia utilizada é qualitativa. Participaram da pesquisa 30 voluntários, estudantes do 9º ano de uma escola pública. As fases de aplicação das atividades para a coleta de dados foram divididas em diagnóstica, ensino de leitura e teste/avaliação. As atividades foram produzidas com base nos contos (O Crime do Professor da Matemática, de Clarice Lispector; O Conto se Apresenta, de Moacyr Scliar; O Gato Preto, de Edgar Allan Poe) lidos em cada fase, sendo elas o Questionário Avaliativo de uso de Estratégias Metacognitivas de Leitura e o Questionário de Compreensão do Conto, em todas as fases e o Guia de Leitura, nas duas últimas fases. O corpus foi constituído por 234 fichas de atividades (questionários e guias de leitura). Os resultados indicam que é possível a percepção do discurso universal, quando o estudante é capaz de transubstanciar a ficção inerente ao texto literário. / The present study aims to apply a teaching Reading approach in order to contribute to the literary reader formation through metacognitive reading strategies that help students become active readers and recognize in the short story the universal discourse inherent in the literary text. Our theoretical basis associated Aesthetics of Reception with Jauss (1979), Theory of Aesthetic Response, represented by Iser (1979), literature teaching with Bordini & Aguiar (1993), Zilberman (2014) and Cosson (2014) to the interaction with the use of reading strategies, represented by Flavell (1976), Baker and Brown (1980), Solé (1998), Duke and Pearson (2002) and Kleiman (2007). Thus, literary theory and psycholinguistics dialogue to justify our proposal of interweaving these areas, seeking a literature teaching that respects the literary aspects and use the language skills to build text meanings. The methodology is qualitative. The participants were 30 volunteers, students from 9th grade in a State School. The implementation stages for the data collection were divided into diagnostic, teaching of reading and test/evaluation. The activities were produced based on the short stories (O crime do Professor de Matemática by Clarice Lispector; O Conto se Apresenta by Moacyr Scliar; O Gato Preto by Edgar Allan Poe) read at each stage, they were the Assessment Questionnaire on the use of Metacognitive Reading Strategies, Short Story Comprehension Questionnaire, at all stages and the Reading Guide in the last two stages. The corpus consisted of 234 activities sheets (questionnaires and reading guides). The results indicate that the universal discourse perception is possible, when the student is able to transubstantiate fiction inherent in the literary text.

Multiple-choice (MCQs) and short-answer questions (SAQs) are the most common test formats for assessing English reading proficiency. While the former provides test-takers with prescribed options, the latter requires short written responses. Test developers favor MCQs over SAQs for the following reasons: less time required for rating, high rater agreement, and wide content coverage. This mixed methods dissertation investigated the impacts of test format on reading performance, metacognitive awareness, test-completion processes, and task perceptions. Participants were eighty English as a second language (ESL) learners from a Midwestern community college. They were first divided into two groups of approximately equivalent reading proficiencies and then completed MCQ and SAQ English reading tests in different orders. After completing each format, participants filled out a survey about demographic information, strategy use, and perceptions of test formats. They also completed a 5-point Likert-scale survey to assess their degree of metacognitive awareness. At the end, sixteen participants were randomly chosen to engage in retrospective interviews focusing on their strategy use and task perceptions. This study employed a mixed methods approach in which quantitative and qualitative strands converged to draw an overall meta-inference. For the quantitative strand, descriptive statistics, paired sample t-tests, item analyses, two-way ANOVAs, and correlation analyses were conducted to investigate 1) the differences between MCQ and SAQ test performance and 2) the relationship between test performance and metacognitive awareness. For the qualitative strand, test-takers’ MCQ and SAQ test completion processes and task perceptions were explored using coded interview and survey responses related to strategy use and perceptions of test formats. Results showed that participants performed differently on MCQ and SAQ reading tests, even though both tests were highly correlated. The paired sample t-tests revealed that participants’ English reading and writing proficiencies might account for the MCQ and SAQ performance disparity. Moreover, there was no positive relationship between reading test performance and the degree of metacognitive awareness generated by the frequency of strategy use. Correlation analyses suggested whether a higher or lower English reading proficiency of the participants was more important than strategy use. Although the frequency of strategy use did not benefit test performance, strategies implemented for MCQ and SAQ tests were found to generate interactive processes allowing participants to gain deeper understanding of the source texts. Furthermore, participants’ perceptions toward MCQs, SAQs, and a combination of both revealed positive and negative influences among test format, reading comprehension, and language learning. Therefore, participants’ preferences of test format should be considered when measuring their English reading proficiency. This study has pedagogical implications on the use of various test formats in L2 reading classrooms.

Students with specific learning disabilities (SLD) have difficulty with most reading skills, including reading comprehension (Hulme & Snowling, 2011). Improving reading comprehension skills requires efficient interventions that consider both meaning- and code-based skills simultaneously. Using a single-subject multiple-baseline design across participants, with alternating treatment design, this study compared two reading interventions (repeated reading vs. tablet text-to-speech) combined with a meta-cognitive strategy (question generation). Three fourth-grade and third-grade students who had been diagnosed by their school as having reading difficulties (reading one to two grades behind their expected reading levels) participated in the study. Using the index of narrative complexity (Labov, 1973; Petersen, Gillam, & Gillam, 2008) as a major dependent variable, two participants showed improvement in reading comprehension skills as measured by visual analysis and the effect size between means. However, there were slight differences for the RAAC intervention over the tablet intervention for one participant. The time required to administer the tablet intervention was shorter than the time required to administer the RAAC intervention (an average of 12.73 minutes for the RAAC vs. 5.45 minutes for the tablet), which is an important consideration when deciding to use an intervention.

Cognitive adaptability has been conceptualised as the ability to effectively and appropriately change decision policies (i.e. to learn) given feedback (inputs) from the environmental context in which cognitive processing is embedded. Based on a large sample of 2650 established entrepreneurs in South Africa, this study attempts to determine how entrepreneurs cognitively adapt to unpredictable entrepreneurial environments. Multidimensional constructs representing cognitive adaptability and the Big Five personality traits were operationalised and empirically investigated. It was hypothesised that the Big Five personality trait dimensions of openness to experience, conscientiousness, extraversion and agreeableness are positively related to the cognitive adaptability dimensions of goal orientation, metacognitive knowledge, metacognitive experience, and metacognitive choice and monitoring. Neuroticism was hypthesised to be negatively related to the cognitive adaptability dimensions of goal orientation, metacognitive knowledge, metacognitive experience, metacognitive choice and monitoring. Hypotheses were tested using structured equation modelling and correlational and regression analysis. Results provide support for subcomponents of the Big Five personality traits. Intellectual interest (openness to experience), goal striving (conscientiousness), activity (extraversion), prosocial orientation (agreeableness) were found to be positively related to cognitive adaptability. They were found to be negatively related to prior metacognitive knowledge. Self-reproach (neuroticism) was found to be negatively related to cognitive adaptability. It was found to be positively related to prior metacognitive knowledge. This research builds on and extends existing literature on cognitive adaptability in an entrepreneurial context by bringing together two streams of literature from psychology metacognition and personality traits. The implications of the process for dynamic, adaptable thinking are important in an emerging context such as that found in South Africa. The results of this study will inform the practice of policy makers who are trying to encourage start-up entrepreneurs to think about thinking in unpredictable entrepreneurial environments. In terms of methodology, the use of a sample of established entrepreneurs is desirable for this type of research since metacognition is better studied in entrepreneurs who are involved in a series of activities. / Thesis (PhD)--University of Pretoria, 2016. / Business Management / PhD / Unrestricted

This study investigates the potential benefits of reading strategy instruction in the English classroom in an L2 learner context. We examined the effects of reading strategy instruction on reading comprehension for L2 learners and investigated howreading strategies should be taught in the context of Swedish upper secondary school. To this end, six empirical studies were evaluated. We found that reading strategy instruction had a positive impact on reading comprehension in general, but that it was especially effective for certain student groups. Namely, intermediate learners benefitted more than other groups, while the weakest students improved the least in response to instruction. Furthermore, the research we examined suggests that it may be advantageous to focus on teaching a single strategy at a time. However, it is important to eventually expand students’ range ofstrategies long-term, since many pupils tend to over-rely on problem-solving strategies to the detriment of their progress in reading. Thus, emphasizing the less frequently used support strategies during instruction may help students read moreefficiently. Additionally, for reading strategies to best benefit learners, they should be taught in a clear, step-by-step manner. Finally, we argue that using aspects of Vygotsky’s sociocultural theory to inform lesson plans would greatly benefit students, especially those who are less proficient. Due to the lack of research in the increasingly heterogeneous Western world, future research should investigate how reading strategy instruction behaves in a multicultural context in the West.

The purpose of this research was to investigate high school students’ computer self-efficacy, cognitive actions, and metacognitive strategies in a self-regulated learning (SRL) framework while utilizing an interactive learning module. The researcher hypothesized that computer self-efficacy is correlated positively with cognitive actions and metacognitive strategies while the students are engaged with interactive learning modules. This research used a mixed-methods approach to answer the research questions. Two research questions guided this research: (1) How is students’ computer self-efficacy related to cognitive actions and metacognitive strategies while using interactive learning modules?; and (2) How do students plan monitor their cognitive actions, and regulate their monitoring strategies during learning with interactive learning modules?This study utilized self-regulated learning framework that covered self-efficacy, cognitive, and metacognitive components. While self-efficacy was represented by computer self-efficacy, the metacognitive component was represented by planning, monitoring, and regulating strategies. Cognitive actions represent contextual activities while using interactive learning modules. One hundred and thirteen students from two high schools in Northern Utah, USA(i.e., InTech Collegiate High School and Logan High School) participated in this study. Each student worked on three modules: Boolean Logic, Minimum Spanning Tree, and Modeling Using Graphs. Due to the differences in class schedules between both schools, students at InTech Collegiate High School and Logan High School completed the activities within 2 and 4 days, respectively. Three different forms of data were gathered for analysis. These data included questionnaires, screen captured videos, and audio recordings of the interviews. The students completed three questionnaires: demographic, computer self-efficacy, and self-regulated computer-based learning questionnaires.The findings of the study revealed that while computer self-efficacy was not positively correlated with cognitive actions, it was positively correlated with metacognitive strategies. Specifically, the findings revealed a significant positive correlation between computer self-efficacy and planning strategies. Screen-captured video analyses showed that there were different profiles of cognitive actions and metacognitive strategies between high and low computer self-efficacy groups. The findings were confirmed by issues from interview analyses between the groups.

The primary task of learning to program in introductory computer science courses (CS1) cognitively overloads novices and must be better supported. Several recent studies have attempted to address this problem by understanding the role of metacognitive awareness in novices learning programming. These studies have focused on teaching metacognitive awareness to students by helping them understand the six stages of learning so students can know where they are in the problem-solving process, but these approaches are not scalable. One way to address scalability is to implement features in an automated assessment tool (AAT) that build metacognitive awareness in novice programmers. Currently, AATs that provide feedback messages to students can be said to implement the fifth and sixth learning stages integral to metacognitive awareness: implement solution (compilation) and evaluate implemented solution (test cases). The computer science education (CSed) community is actively engaged in research on the efficacy of compile error messages (CEMs) and how best to enhance them to maximize student learning and it is currently heavily disputed whether or not enhanced compile error messages (ECEMs) in AATs actually improve student learning. The discussion on the effectiveness of ECEMs in AATs remains focused on only one learning stage critical to metacognitive awareness in novices: implement solution. This research carries out an ethnomethodologically-informed study of CS1 students via think-aloud studies and interviews in order to propose a framework for designing an AAT that builds metacognitive awareness by supporting novices through all six stages of learning. The results of this study provide two important contributions. The first is the confirmation that ECEMs that are designed from a human-factors approach are more helpful for students than standard compiler error messages. The second important contribution is that the results from the observations and post-assessment interviews revealed the difficulties novice programmers often face to developing metacognitive awareness when using an AAT. Understanding these barriers revealed concrete ways to help novice programmers through all six stages of the problem-solving process. This was presented above as a framework of features, which when implemented properly, provides a scalable way to implicitly produce metacognitive awareness in novice programmers.

This study investigates reading comprehension strategies amongst 298 Swedish ESL students in upper-secondary school, possible relationships between high motivation to learn and high strategy use, and whether more experienced readers tend to use reading strategies more frequently than lesser experienced readers. Data was collected through online surveys, and the findings were analysed using quantitative analysis. The results of the study showed that experienced readers tend to use reading strategies more often than less experienced readers, but that the most employed strategies amongst students seem to be naturally occurring rather than metacognitive and self-regulated ones, indicating that the participating Swedish upper-secondary students lack sufficient knowledge of other reading strategies. The data regarding the possible relationships between high levels of motivation and a more varied use of reading strategies do not indicate a strong correlation. One possible conclusion of the study is that even if students seem to perceive themselves as strategy users, the strategies deemed most important by researchers still end up amongst the least used strategies, indicating that students still need to receive more explicit training in the use and functions of such strategies.