Les différences interindividuelles en mémoire de travail (MDT) ont un lien stable et largement documenté avec la performance dans les tâches de cognition de haut niveau. Dans la mesure où l'utilisation de stratégies efficaces joue un rôle aussi bien dans les tâches de MDT que de cognition de haut niveau, on peut faire l'hypothèse que les stratégies médiatisent le lien entre ces deux construits. De fait, certaines données suggèrent que le comportement stratégique au sein de tâches de cognition de haut niveau pourrait être en lien avec la capacité de MDT. L'objectif de ce travail de thèse était d'évaluer ce lien de façon plus poussée. Cette approche nous a conduit à deux résultats particulièrement significatifs. Le premier concerne le développement et la formalisation du paradigme de Hard Fall Effect (HFE), défini comme étant la chute de performance plus importante en situation de double tâche pour les participants avec une forte capacité de MDT. L'hypothèse sous-jacente à cet effet est que ces participants utilisent des stratégies afin d'améliorer leurs performances en situation de simple tâche, et que la situation de double tâche perturbe l'utilisation de ces stratégies. Au cours de ce travail, le HFE a notamment été mis en évidence dans une tâche de mémoire visuospatiale, et a pu être attribué à l'utilisation de stratégies d'encodage plus efficaces par les participants avec une forte capacité de MDT au sein de tâches de mémoire complexes. Notre second résultat significatif correspond au test direct de l'hypothèse de médiation du lien entre MDT et cognition de haut niveau par l'utilisation de stratégies efficaces en cognition de haut niveau. Nous avons montré que lorsqu'on contrôle la variance associée à l'utilisation de stratégies efficaces dans la tâche des Matrices Avancées de Raven, la relation entre capacité de MDT et intelligence fluide diminue. Dans l'ensemble, ces deux résultats permettent de renforcer l'idée selon laquelle le comportement stratégique impliqué dans les tâches de cognition de haut niveau intervient dans la relation entre MDT et cognition de haut niveau. / The relationship between individual differences in working memory capacity (WMC) and high-level cognition has been extensively documented. Since efficient strategy use plays an important role in the performance of both working memory and high-level cognitive tasks, it could be the case that strategies mediate the WM-high-level cognition relationship. Importantly, it has been observed that the use of efficient strategies in high-level cognitive tasks is related to individual differences in WMC. The goal of this research work was to investigate this association between WMC and the use of efficient strategies in more detail. Firstly, we formalized a paradigm termed the Hard Fall Effect (HFE), defined as a larger decrease of performance in dual task conditions for high WMC individuals. The underlying interpretation of this effect is that these individuals use efficient strategies under simple task conditions to improve their performance, and these strategies are disrupted by dual tasking. The present work evidenced the HFE in a visuospatial memory task, and attributed the effect to the higher tendency of high WMC individuals to use efficient encoding strategies in complex memory tasks. Secondly, we provided evidence that efficient strategy use in high-level cognitive tasks partly mediates the WMC-high-level cognition relationship. More precisely, when statistically controlling the use of efficient strategies in Raven's Advanced Progressive Matrices, the WMC-fluid intelligence relationship was significantly reduced. Overall, these two results provide evidence that the use of efficient strategies in high-level cognitive tasks is involved in the WMC-high-level cognition relationship.

This thesis deals with the relationship between Working Memory, Working Memory Span tasks and general factor g and Broad cognitive abilities. Measured constructs are introduced in the theoretical part, with their evolution, various methods of their measurement and studies investigating the relation between them. The empirical part of the research has been conducted to verify the relationship between Working Memory and general intelligence factor g. It has been done to reveal the relationship between Working Memory Span tasks and Broad cognitive abilities as well. The question concerning the influence of the use of strategy while performing the automatic version of Working Memory Span tasks has been investigated as well.

Self-paced study involves choosing items for (re)study and determining how much time will be allocated to those items so as to maximize later recall, making it a viable venue for examining whether there are age-related differences in metacognitive control. Two prominent models have been proposed to account for item selection and study time allocation behaviors during self-paced study. The Discrepancy Reduction Model (DRM; Dunlosky & Hertzog, 1998; Nelson & Leonesio, 1988) suggests individuals will always select and allocate the most time to items that have not yet been learned, whereas the Region of Proximal Learning model (RPL; Metcalfe, 2002) predicts individuals will select the easiest unknown items and will only later select and allocate time to the more difficult items if time constraints permit, thus making distinctions among unlearned items graded by difficulty. Two experiments were conducted to examine whether younger and older adults item selection and study time allocation behaviors would be more consistent with DRM or RPL model predictions. Across both experiments younger and older adults initially selected easier items for study, providing the first evidence to date that the RPL model would extend to older adults self-paced study of heterogeneously difficult Spanish-English vocabulary pairs. However, both younger and older adults allocated more time to difficult than easier items. The assignment of point values to items in Experiment 2 affected how likely participants were to pursue each of four experimenter-determined task goals that either stressed the number of words recalled, points earned, or both. Whether point values initially favored recall of easy or difficult items interacted with time constraints to influence the basis (objective versus subjective difficulty) and order of participants item selections (Experiment 2). However, younger adults were better able to effectively allocate their study time to achieve self-determined (Experiment 1) and experimenter-determined goals (Experiment 2), indicating age-related differences in metacognitive control despite younger and older adults having similar memory self-efficacy ratings and encoding strategy use behaviors.

Strategy instruction can improve memory performance, but some training programs are more effective than others. Some scholars propose that a key element to boosting the benefits from training programs is enhancing or emphasizing self-regulatory factors, such as knowledge about memory, beliefs about ability, or motivational factors. Research supporting this claim evidence adds that programs that enhance trainees’ confidence in their abilities improve memory performance and that multifactorial programs are more effective than strategy-training-only programs. Setting performance goals and receiving feedback are two self-regulatory factors known to relate to memory performance that may sometimes be included in some training programs. However, previous research has not directly compared the effectiveness of strategy instruction with and without goal-setting and performance feedback elements. This was the purpose of the present research: We compared strategy instruction with goal-setting and positively-framed feedback across three assessments of memory performance, strategy use, and task commitment. Participants were 48 university students who were randomly assigned to two conditions: All participants watched a brief memory strategy video, but participants in the Strat+GFB condition set goals for their memory performance and received positively-framed objective performance feedback and participants in the StratOnly condition did not. Research assistants conducted the experimental procedures individually with participants in 1-hour-long Zoom video calls. Primary outcome measures (memory performance, strategy use, and task commitment) were assessed three times, once before and twice after strategy instruction, with or without goal-setting and feedback between each test, depending on condition assignment. Shopping list recall tests were used to assess memory performance, and the number of to-be-recalled stimuli increased at each testing occasion as 15 additional items were added at each trial. Participants recalled more items, but a smaller percentage of the items, over time, and this pattern was not different for the experimental conditions. Additionally, number of strategies used, as self-reported on a retrospective checklist, increased from before to after strategy instruction. Importantly, the Strat+GFB condition maintained levels of commitment to the memory tests across the three trials, whereas the StratOnly condition reported drops in their task commitment. Study results emphasize that including self-regulatory factors, such as setting goals and receiving feedback, may increase commitment to a task, however those benefits may not immediately translate to better memory performance when training and testing is part of a brief, single experimental session. We suggest that future research evaluate a multiday memory intervention with the addition of goal-setting and feedback. Results of this study suggest that including goal-setting and feedback as part of a training program may benefit trainees’ commitment, which we speculate could aid individuals in maintaining persistent effort despite challenges and ultimately lead to better performance over a longer term.

The low literacy levels of the majority of first-year students at tertiary institutions in South Africa have been a major cause of concern. Various attempts have been made to assist students to develop their academic literacy levels – especially academic reading. However, most of these attempts are solely cognitive-oriented, even though there has been increasing acknowledgement of the relationship between socio-affective factors and students’ academic reading abilities. The purpose of this research was to explore a socio-affective approach to improving the reading abilities of first-year students at the University of Pretoria (UP). The following questions guided the research: (1) Is there a significant relationship between socio-affective factors and students’ academic reading abilities? (2) Which socio-affective factors best predict tertiary students’ academic reading abilities? (3) How can knowledge of socio-affective factors be used to design a more effective reading intervention? (4) How effective is a reading intervention programme that incorporates socio-affective factors? A mixed methods design was used for the study which was conducted in four phases. The first phase consisted of an exploratory study in the form of a questionnaire survey that elicited information on first-year UP students’ reading background, socio-affective reading levels and the use of reading strategies in relation to their reading proficiency levels, as determined by the Test for Academic Literacy Levels (TALL). ANOVA tests were used for the analysis of TALL results while a Cumulative Logit (regression) analysis was conducted to determine the socio-affective factors that best predict these students’ reading ability. ANOVA tests showed a robust relationship between students’ social and affective reading background on one hand, and their reading proficiency levels on the other. The regression analysis showed that self-efficacy was the best predictor of students’ reading ability, followed by intrinsic motivation. Based on the empirical results, and an adapted model of Guthrie and Wigfield (2000), an intervention programme that served as enrichment to the existing Academic Reading module, and aimed at improving the reading abilities of students by focussing on socio-affective issues in particular, was designed (as phase 2) and implemented (as phase 3) of the study. Two control groups and two intervention groups of At Risk and Low Risk students were used for the study. Questionnaires on affective reading levels and strategy use were administered before and after the intervention. In phase four, quantitative analysis using t-tests (independent and paired t-tests) with effect sizes were performed on the pre- and post-intervention questionnaire responses. Results showed significant improvements in affective levels for reading in the intervention groups compared to the control groups. In addition, qualitative data were collected via interviews on the socio-affective teaching techniques used for the intervention, and analysed qualitatively using content analysis. The results of the qualitative study were used to support the quantitative findings in terms of the measure in which the teaching approach contributed to the improvement in students’ socio-affective levels in reading, which according to research, correlates with students’ reading ability. Based on the findings, recommendations are made at the classroom and institutional levels. The significance of the study in terms of enriching theory and designing innovative support to improve students’ reading ability serve as a conclusion to the thesis. / Thesis (DPhil)--University of Pretoria, 2012. / Unit for Academic Literacy / Unrestricted

Wir können Mathematikleistungen über fluide Intelligenz (FI) vorhersagen. Der Einfluss von FI auf kognitive Prozesse und neuronale Mechanismen, die mathematischen Fähigkeiten in verschiedenen Teildisziplinen zugrunde liegen, ist jedoch wenig verstanden. Vorliegende Arbeit spezifiziert FI-bezogene Unterschiede in diesen Prozessen und Mechanismen beim Lösen von Geometrie-, Arithmetik- und Algebra-Aufgaben. Mithilfe eines multimethodalen Ansatzes beleuchtet sie das Zusammenspiel zwischen FI, Leistung und Faktoren wie Aufgabenkomplexität, Lernen und Strategiewahl, die kognitive Prozesse und Anforderungen beim Problemlösen beeinflussen. Leistungsunterschiede wurden durch Messung von Reaktionszeiten und Fehlerraten, Strategien durch Augenbewegungsanalyse erfasst. Als Indikator kortikaler Aktivität diente die ereigniskorrelierte (De-)Synchronisation (ERD/ERS) im Alpha-Band. Um kognitive Prozesse zu unterscheiden, haben wir die ERD/ERS im Theta-Band und den Alpha-Unterbändern einbezogen. Beim Lösen unvertrauter geometrischer Analogien zeichnete sich hohe FI durch verstärkte Verarbeitung visuell-räumlicher Informationen zum Repräsentieren von Merkmalszusammenhängen aus. Schüler mit hoher FI passten ihre Strategiewahl den Anforderungen flexibler an. Erstmals konnten wir durch trialweise Identifikation von Strategien FI-bezogene Unterschiede in der neuronalen Effizienz der Strategieausführung feststellen. Beim Lösen vertrauter arithmetischer und algebraischer Terme zeigten sich bei Schülern mit hoher im Vergleich zu Schülern mit durchschnittlicher FI geringere Anforderungen zur Aktualisierung numerischer Repräsentationen und eine bessere Leistung in komplexen Aufgaben. Weitere Analysen legen nahe, dass Schüler mit hoher FI Zusammenhänge in der Aufgabenstruktur besser erkennen und passende Routinen abrufen können. Die Fähigkeit Zusammenhangsrepräsentationen zu bilden könnte demnach ein Schlüsselaspekt zur Erklärung FI-abhängiger Unterschiede in mathematischen Fähigkeiten sein. / Fluid intelligence (FI) is a strong predictor of mathematical performance. However, the impact of FI on cognitive processes and neural mechanisms underlying differences in mathematical abilities across different subdivisions is not well understood. The present work specifies FI-related differences in these processes and mechanisms for students solving geometric, arithmetic, and algebraic problems. We chose a multi-methodological approach to shed light on the interplay between FI, performance, and factors such as task complexity, learning, and strategy selection that influence cognitive processes and task demands in problem-solving. We measured response times and error rates to evaluate performance, eye movements to identify solution strategies, and the event-related (de-)synchronization (ERD/ERS) in the broad alpha band as indicator of general cortical activity. Further, we considered the ERD/ERS in the theta band and the alpha sub-bands to distinguish between associated cognitive processes. For unfamiliar geometric analogy tasks, students with high FI built relational representations based on a more intense processing of spatial information. Strategy analyses revealed a more adaptive strategy choice in response to increasing task demands compared to students with average FI. Further, we conducted the first study identifying strategies and related cortical activity trial-wise and thereby identified FI-related differences in the neural efficiency of strategy execution. For solving familiar arithmetic and algebraic problems, high compared to average FI was associated with lower demands on the updating of numbers leading to a better performance in complex tasks. Further analyses suggest that students with high FI had an advantage to identify the relational structure of the problems and to retrieve routines that match this structure. Thus, the ability to build relational representations might be one key aspect explaining FI-related difference in mathematical abilities.

The syllabi for the subject English in both Swedish compulsory and upper secondary school state in the core content for English that it should provide the opportunity to learn how to use linguistic strategies in speech, i.e. oral communication strategies. However, we as teachers are not informed by these documents what oral communication strategies are and which ones are to be preferred. For this reason, we as future teachers of English, posed the following research questions: What are oral communication strategies according to the literature, and how are these assessed in terms of being positive and negative strategies? According to research what factors correlate with strategy use, and what are the potential pedagogical implications for the Swedish school context? To answer these questions, we have read and analyzed fourteen different empirical studies regarding communication strategies. The first question was answered by analyzing the empirical studies and relevant theory. We found various definitions in our studies, stemming from different theoretical perspectives. However, they all define oral communication strategies as serving the purpose of furthering interaction. Furthermore, by comparing the definitions in the Swedish syllabi for English and the Common European Framework of Reference for Languages to our literature, we were able to answer the second part of the first question. The literature shows that there is a preference for achievement strategies over avoidance or reduction strategies. To answer the first part of the second question, some studies indicated a positive correlation between strategy use and the level of proficiency. Regarding pedagogical implications, some studies indicate that explicit strategy training has a positive effect on oral performance. Drawing on the results of these studies and the theoretical framework provided, we conclude that achievement strategies are to be preferred and that they should be taught explicitly.

No description available.