<p>This study investigated, within the context of Brain Reserve Capacity (BRC) theory, whether repeated concussions resulted in residual deficits in cognitive and academic functioning of early adolescent rugby players relative to non-contact sports controls.</p>

<p>This study investigated, within the context of Brain Reserve Capacity (BRC) theory, whether repeated concussions resulted in residual deficits in cognitive and academic functioning of early adolescent rugby players relative to non-contact sports controls.</p>

Philosophiae Doctor - PhD / This study investigated, within the context of Brain Reserve Capacity (BRC) theory, whether repeated concussions resulted in residual deficits in cognitive and academic functioning of early adolescent rugby players relative to non-contact sports controls. / South Africa

Background: Cognitive reserve (CR) has been associated with better cognitive function and lower risk of dementia in older people, yet it remains unclear whether sex moderates the association between CR and cognition. This study aims to identify whether sex influences both the relationships between brain-cognition and how CR proxies moderate the brain-cognition relationship. Materials and Methods: Complete data on the measures of CR, education, occupation, and cognition were available for 189 healthy individuals aged 60 to 71 years (105 men and 84 women). Multiple linear regression models were used to investigate the potential effect of sex and CR proxies on the association between the brain and cognition measures. Results: The results highlighted differences in speed/attention for males compared to females at high education and high occupational complexity. No significant sex differences in brain measures were observed in meanPutamen, meanCaudate, and meanHippocampal volume. Conclusion/Significance: Traditional reserve contributors are influenced by gender and may be a result of different social determinants among men and women. Both sex-specific risk and protective factors for cognitive decline trajectories are critical for advancing knowledge for individualized interventions.

Les capacités de réserve cérébrale représentent les capacités de résilience du cerveau face à différents processus lésionnels comme ceux induit par la maladie d’Alzheimer (MA). Plusieurs hypothèses ont été proposées afin d’expliquer ce mécanisme. Les deux concepts les plus validés sont celui de la réserve cérébrale faisant référence au substrat cérébral et la réserve cognitive faisant référence à la fonctionnalité cérébrale. Les données épidémiologiques puis d’imagerie ont permis d’identifier des expériences de vie associées à de meilleures capacités de réserve cérébrale. Le niveau d’études, la profession exercée et la pratique d’activités de loisirs sont ainsi des indicateurs du niveau potentiel de capacités de réserve d’un sujet. L’objectif général de ce travail de thèse avait pour objectif de mieux caractériser les relations entre ces indicateurs et les capacités de réserve cérébrale. Tout d’abord, nous nous sommes intéressés à la relation entre ces expériences de vie et le substrat cérébral à partir de données d’imagerie recueillies au cours du suivi de la cohorte des 3 cités à Bordeaux. Nous avons ainsi montré que seul le niveau d’éducation était associé à des différences de volume de substance grise et de substance blanche entre les sujets de haut et bas niveau d’études. Par ailleurs, les sujets de haut niveau d’études présentaient une progression moindre en hypersignaux de substance blanche au cours du suivi de la cohorte des 3 cités indépendamment de la présence de facteurs de risque cardiovasculaires. Le niveau d’étude semble avoir une place non négligeable dans la constitution des capacités de réserve cérébrale et aussi bien dans la réserve cognitive que la réserve cérébrale. Ceci explique le rôle protecteur du niveau d’études vis à vis du risque de démence. Dans un 3ème travail, nous nous sommes servis de cette relation particulière entre le niveau d’études et la démence afin d’illustrer comment un facteur social peut être un facteur de maladie chronique liée à l’âge et comment en le modifiant, on peut éventuellement modifier la survenue de cette maladie. Enfin, nous nous sommes intéressés à la relation entre la pratique d’activités de loisirs au cours de la retraite et le risque de démence à partir des données de la cohorte Paquid. La pratique de jeux de société est associée à un moindre risque de démence mais le lien fort avec la cognition ne permet pas d’éliminer une causalité inverse. Cependant, un engagement plus important dans la pratique d’activités de loisirs au cours de la retraite est associé à un risque moindre de démence équivalent à des sujets ayant toujours pratiqué des activités. Des essais d’intervention permettraient de confirmer l’effet bénéfique de la pratique d’activités de loisirs sur la cognition et le risque de démence. / The brain reserve capacities represent the resilience of the brain to cope against different pathological processes such as Alzheimer's disease (AD). Several hypotheses have been proposed to explain this mechanism. The two most validated concepts are the brain reserve referring to brain volume and cognitive reserve referring to brain function. Epidemiological and imaging data helped identify life experiences associated with better brain reserve capacities. Thus, the education, occupation and practice of leisure activities are recognized as proxies of the brain reserve capacities. The main objective of this thesis aimed to better characterize the relationship between these proxies and the brain reserve capacities. First, from brain imaging data collected during follow-up of the three cities cohort of Bordeaux, we examined the relationship between life experiences and brain volume. Only the education was associated with differences in gray and white matter volume. In addition, highly educated subjects had a smaller progression of white matter hyperintensities during the follow-up of this cohort independently of the presence of cardiovascular risk factors. Education seems to have a significant role in the formation of brain reserve capacities in both cognitive reserve and brain reserve. This explains the protective role of educational level against dementia. In a third work, we used this particular relationship between educational level and dementia to illustrate how a social factor can be a factor of a chronic disease related to aging and can modify the occurrence of this disease. Finally, we are interested in the relationship between the practice of leisure activities during retirement and the risk of dementia from data of Paquid study. The practice of playing board games is associated with a lower risk of dementia, but the strong link with cognition not eliminates reverse causality. However, a greater engagement in the practice of leisure activities in retirement is associated with a lower risk of dementia similar to subjects who always practiced activities. Intervention trials could confirm the beneficial effect of the practice of leisure activities on cognition and dementia risk.

OBJECTIVE: The brain reserve hypothesis was examined in a secondary analysis of cross-sectional data from a community-based sample of 468 older adults residing in Charlotte County, Florida. The objective of the analysis was to determine the association between head circumference and eight cognitive outcomes and to assess any potential effect modification of existing associations by Apolipoprotein E (APOE) genotype. METHODS: Cognitive outcomes include scores from the Modified Mini-Mental State Exam (3MS), the Hopkins Verbal Learning Test-Revised (HVLT-R), Stroop Color-Word Test, Trail-Making Test A and B, and a word-stem completion task measuring implicit memory. Descriptive statistics were calculated for each variable. Head circumference and dependent cognitive outcomes were modeled as dichotomous variables using logistic regression, adjusting for gender, age, education, income, height, and Spot The Word test score, a measure of pre-morbid IQ. For dichotomized test scores, poor outcomes (cases) were defined as having scores in the lowest quintile; the remaining top four quintiles were considered non-cases. RESULTS: small head circumference was significantly associated with low 3MS scores [OR(95%CI): 2.97 (1.12, 7.89), p=0.03], after adjustment for age, income and pre-morbid IQ. The association remained statistically significant after adjustment for gender and education as well. After adjustment, head circumference was not found to be statistically significantly associated with any other cognitive outcome. No effect modification was found by APOE genotype or years of education. CONCLUSION: This analysis confirms previous findings that exposure to low head circumference significantly impacts cognition in late life.

Rugby is a popular sport in South Africa, and has been played by young boys from as early as seven years old (South African Rugby Union [SARU], 2011). Despite various physical health benefits, it carries a high risk for injury, especially head injury, and consequently has a high incidence of concussion (Alexander, 2009; Laubscher, 2006; Shuttleworth-Edwards, Smith & Radloff, 2008). It is common for 12 to 13 per cent of adolescent rugby players to report mild traumatic brain injury or concussion per season (Laubscher, 2006; Shuttleworth-Edwards et al., 2008). The true incidence is however considered to be higher, even as high as 70.4% (Shuttleworth-Edwards et al., 2008). Concussion, otherwise known as mild traumatic brain injury (mTBI) is described as a traumatically induced alteration in mental status, or traumatically induced cerebral dysfunction (Kraus, McArthur, Silvermand & Jayaraman, 1996) which may, or may not involve loss of consciousness (Quality Standards Subcommittee, American Academy of Neurology [AAN], 1997). The nature of concussion has traditionally been considered to be transient, and symptoms are usually resolved within a few days or weeks (Kirkwood et al., 2008; Taylor et al., 2010). However, when concussions are not fully resolved prior to players returning to the game, they may be vulnerable to second impact syndrome. This syndrome causes herniation and brain oedema, which may result in death (Patel, 2005), as has been reported in South African press (Alexander, 2009; South African Press Association [SAPA], 2012). Even without second impact syndrome, repeated concussions may render the brain neurocognitively vulnerable, leading to an array of short- and long-term cognitive symptoms (Alexander, 2009; Shuttleworth-Edwards et al., 2008). Short-term problems include difficulties with attention, focus and concentration; following multi-step instruction, engaging in mental problem-solving; verbal expression, receiving and processing verbal and visual information; maintaining effective levels of mental and physical energy; controlling mood; suppressing impulsive behaviours; initiating and maintaining productive interpersonal relationships with peers; engaging in meaningful conversation and participating in group activities (Jantz & Coulter, 2007). Short-term cognitive impairments due to repeated concussion have also been found, and include amongst the former symptoms, also problems with delayed memory, learning, social functioning, and abstract thinking (Anderson, Brown, Newitt & Hoile, 2011; Laubscher, 2006). Long-term sequelae follow when children did not return to their baseline level of functioning after three months (Kirkwood et al., 2008; Taylor et al., 2010). Long-term sequelae include problems with memory, visuo-motor processing, executive functioning, learning and abstract thinking (Anderson, 2002; Anderson et al., 2010; Horton et al., 2010; Lezak et.al., 2004; Shuttleworth-Edwards & Radloff, 2008). As mTBI is traditionally thought to be of transient nature, researchers tend to investigate moderate to severe TBI, rather than mTBI (Alexander, 2009; Anderson et al., 2010; Patel, 2005). This could easily lead to important facts about mTBI being missed or not acknowledged. Nevertheless, recent investigations are uncovering facts about mTBI that could transform the way in which we understand mTBI, providing increasing evidence that mTBI is more serious than widely believed (Blakemore, 2012; Maxwell, 2011; Toleda et al., 2012). However, there remains a lack of research investigating mTBI from a single cause. Considering the above information, the current study provides unique information about mTBI. It specifically investigated the long-term effects of mTBI on adolescents from a homogenous cause, which makes results more comparable. The importance of this study is highlighted in the face of evidence for the long-term effects of multiple concussions, that were sustained during school rugby, on academic achievement (Alexander, 2009; Laubscher, 2006).In the light of grey areas in existing research, the aim of this current study was to investigate whether there is a significant difference in academic achievement within and between two groups of adolescents that had either played rugby and sustained multiple concussions, or had not played rugby nor sustained any concussions, when measured at four points in time over six years. A retrospective data-analysis was performed on matched, controlled, prospective longitudinal data, which was obtained from a study that evaluated the impact of repeated mTBI on the cognitive and academic functioning of early adolescent rugby players over time (Alexander, 2009). This study elaborates on a subset of the previous data, adding the gr. 12 results for academic aggregate scores, to the previously reported academic dataset. Participants were selected from Alexander‟s study (2009), and had either played rugby and obtained two or more concussions (Rugby/Concussed (RC- group); n=17), or did not play rugby nor sustained any concussions (Non-rugby/Non-concussed (NRC-group); n=13). Academic aggregate scores from baseline (gr. 7) through gr. 12 were analysed using quantitative statistical measures. A normal probability plot determined that the data was distributed normally. Descriptive statistics were reported, where after repeated measures ANOVA‟s were conducted to determine the statistical significance of differences in academic scores between and within the groups over time. These results indicated that the NRC-group displayed statistically significant increase in academic achievement over time (p = .000), whereas the RC-group did not display any significant differences, despite displaying a downwards trend in achievement. The difference between the two groups was measured at its highest in gr. 12 (p = .003), indicating that the NRC-group performed statistically significantly better than the RC-group over time. However, a Pearson‟s correlation test revealed that the estimated IQ (Vocabulary subscale of the WISC-III) (Wechsler, 1991) had a positive correlation on academic achievement [r(34) = .54, p < .05)]. To control for the effect that this correlation had on the academic results, an ANCOVA was conducted. This analysis indicated a statistically significant difference in academic achievement between the two groups in gr.12 (p = .004), with a large effect size (d = 1.41), implicating practical significance. Findings consequently confirmed our hypothesis. The significant increase in academic achievement observed within the NRC-group over time, is consistent with what could be expected if the brain is allowed to develop normally without disruption such as mTBI (Blakemore, 2012; Horton et al., 2010). The finding that the RC group did not display statistically significant intra-group differences in academic achievement when measured over time, but that academic achievement followed a downward trend, is difficult to substantiate in the literature. The few research studies on the effect of cumulative concussion on young athletes do not isolate academic achievement as a variable (Iverson et al., 2004; Shuttleworth-Edwards et al., 2008). Further research into intra-group differences in this specific area of enquiry and population group is therefore necessary. Normal cognitive and brain development, maintains that the brain develops in a posterior to anterior direction, and the prefrontal regions which are vulnerable to concussion, develop last (Anderson, 2010; Blakemore, 2012; Lezak, 2004). Whereas the primary motor and sensory areas and areas for receptive and expressive language are fully developed by the age of ten years, the prefrontal brain areas that are responsible for more complex and abstract thought repertoires only start maturing in early adolescence and this development continues up to the age of 24 and even into the early 30s (Toleda et al., 2012). Injury to the developing brain at this critical stage of maturation may adversely affect the development of cognitive skills, preventing the child from acquiring the effective cognitive strategies needed for normal academic functioning and adequate academic achievement after TBI (Horton et al., 2010). However, if there is no insult to the brain, cognitive functions are expected to develop normally as a result of synaptic pruning and increased white-matter volume in the prefrontal cortex (Blakemore, 2012), making it likely that the maturation of these abilities will lead to greater cognitive and academic ability (Blakemore & Choudhury, 2006), such as seen for the NRC-group in this study. Limitations for this study include a small sample size and the testing of only one variable. It is therefore recommended that future studies include more variables, and aim at creating a larger, randomized sample size, possibly providing a more representative pool of participants to study this phenomenon in South African context. It is also advised that future studies consider using neuropsychological measures to test cognitive functioning. As previous studies have indicated specific impairment in executive functioning after TBI, it may be worth researching the effect of concussion on executive functioning more thoroughly (Anderson, 2002; Anderson et al., 2010; Horton et al., 2010). Further it may be valuable to consider using functional MRI studies to broaden existing knowledge about the interaction between pathophysiology and cognitive functioning This study also highly recommends that schools and rugby clubs catering for child and adolescent players reconsider the importance of implementing proper return to play protocols after players obtain concussions. / Thesis (MA (Clinical Psychology))--North-West University, Potchefstroom Campus, 2013

Rugby is a popular sport in South Africa, and has been played by young boys from as early as seven years old (South African Rugby Union [SARU], 2011). Despite various physical health benefits, it carries a high risk for injury, especially head injury, and consequently has a high incidence of concussion (Alexander, 2009; Laubscher, 2006; Shuttleworth-Edwards, Smith & Radloff, 2008). It is common for 12 to 13 per cent of adolescent rugby players to report mild traumatic brain injury or concussion per season (Laubscher, 2006; Shuttleworth-Edwards et al., 2008). The true incidence is however considered to be higher, even as high as 70.4% (Shuttleworth-Edwards et al., 2008). Concussion, otherwise known as mild traumatic brain injury (mTBI) is described as a traumatically induced alteration in mental status, or traumatically induced cerebral dysfunction (Kraus, McArthur, Silvermand & Jayaraman, 1996) which may, or may not involve loss of consciousness (Quality Standards Subcommittee, American Academy of Neurology [AAN], 1997). The nature of concussion has traditionally been considered to be transient, and symptoms are usually resolved within a few days or weeks (Kirkwood et al., 2008; Taylor et al., 2010). However, when concussions are not fully resolved prior to players returning to the game, they may be vulnerable to second impact syndrome. This syndrome causes herniation and brain oedema, which may result in death (Patel, 2005), as has been reported in South African press (Alexander, 2009; South African Press Association [SAPA], 2012). Even without second impact syndrome, repeated concussions may render the brain neurocognitively vulnerable, leading to an array of short- and long-term cognitive symptoms (Alexander, 2009; Shuttleworth-Edwards et al., 2008). Short-term problems include difficulties with attention, focus and concentration; following multi-step instruction, engaging in mental problem-solving; verbal expression, receiving and processing verbal and visual information; maintaining effective levels of mental and physical energy; controlling mood; suppressing impulsive behaviours; initiating and maintaining productive interpersonal relationships with peers; engaging in meaningful conversation and participating in group activities (Jantz & Coulter, 2007). Short-term cognitive impairments due to repeated concussion have also been found, and include amongst the former symptoms, also problems with delayed memory, learning, social functioning, and abstract thinking (Anderson, Brown, Newitt & Hoile, 2011; Laubscher, 2006). Long-term sequelae follow when children did not return to their baseline level of functioning after three months (Kirkwood et al., 2008; Taylor et al., 2010). Long-term sequelae include problems with memory, visuo-motor processing, executive functioning, learning and abstract thinking (Anderson, 2002; Anderson et al., 2010; Horton et al., 2010; Lezak et.al., 2004; Shuttleworth-Edwards & Radloff, 2008). As mTBI is traditionally thought to be of transient nature, researchers tend to investigate moderate to severe TBI, rather than mTBI (Alexander, 2009; Anderson et al., 2010; Patel, 2005). This could easily lead to important facts about mTBI being missed or not acknowledged. Nevertheless, recent investigations are uncovering facts about mTBI that could transform the way in which we understand mTBI, providing increasing evidence that mTBI is more serious than widely believed (Blakemore, 2012; Maxwell, 2011; Toleda et al., 2012). However, there remains a lack of research investigating mTBI from a single cause. Considering the above information, the current study provides unique information about mTBI. It specifically investigated the long-term effects of mTBI on adolescents from a homogenous cause, which makes results more comparable. The importance of this study is highlighted in the face of evidence for the long-term effects of multiple concussions, that were sustained during school rugby, on academic achievement (Alexander, 2009; Laubscher, 2006).In the light of grey areas in existing research, the aim of this current study was to investigate whether there is a significant difference in academic achievement within and between two groups of adolescents that had either played rugby and sustained multiple concussions, or had not played rugby nor sustained any concussions, when measured at four points in time over six years. A retrospective data-analysis was performed on matched, controlled, prospective longitudinal data, which was obtained from a study that evaluated the impact of repeated mTBI on the cognitive and academic functioning of early adolescent rugby players over time (Alexander, 2009). This study elaborates on a subset of the previous data, adding the gr. 12 results for academic aggregate scores, to the previously reported academic dataset. Participants were selected from Alexander‟s study (2009), and had either played rugby and obtained two or more concussions (Rugby/Concussed (RC- group); n=17), or did not play rugby nor sustained any concussions (Non-rugby/Non-concussed (NRC-group); n=13). Academic aggregate scores from baseline (gr. 7) through gr. 12 were analysed using quantitative statistical measures. A normal probability plot determined that the data was distributed normally. Descriptive statistics were reported, where after repeated measures ANOVA‟s were conducted to determine the statistical significance of differences in academic scores between and within the groups over time. These results indicated that the NRC-group displayed statistically significant increase in academic achievement over time (p = .000), whereas the RC-group did not display any significant differences, despite displaying a downwards trend in achievement. The difference between the two groups was measured at its highest in gr. 12 (p = .003), indicating that the NRC-group performed statistically significantly better than the RC-group over time. However, a Pearson‟s correlation test revealed that the estimated IQ (Vocabulary subscale of the WISC-III) (Wechsler, 1991) had a positive correlation on academic achievement [r(34) = .54, p < .05)]. To control for the effect that this correlation had on the academic results, an ANCOVA was conducted. This analysis indicated a statistically significant difference in academic achievement between the two groups in gr.12 (p = .004), with a large effect size (d = 1.41), implicating practical significance. Findings consequently confirmed our hypothesis. The significant increase in academic achievement observed within the NRC-group over time, is consistent with what could be expected if the brain is allowed to develop normally without disruption such as mTBI (Blakemore, 2012; Horton et al., 2010). The finding that the RC group did not display statistically significant intra-group differences in academic achievement when measured over time, but that academic achievement followed a downward trend, is difficult to substantiate in the literature. The few research studies on the effect of cumulative concussion on young athletes do not isolate academic achievement as a variable (Iverson et al., 2004; Shuttleworth-Edwards et al., 2008). Further research into intra-group differences in this specific area of enquiry and population group is therefore necessary. Normal cognitive and brain development, maintains that the brain develops in a posterior to anterior direction, and the prefrontal regions which are vulnerable to concussion, develop last (Anderson, 2010; Blakemore, 2012; Lezak, 2004). Whereas the primary motor and sensory areas and areas for receptive and expressive language are fully developed by the age of ten years, the prefrontal brain areas that are responsible for more complex and abstract thought repertoires only start maturing in early adolescence and this development continues up to the age of 24 and even into the early 30s (Toleda et al., 2012). Injury to the developing brain at this critical stage of maturation may adversely affect the development of cognitive skills, preventing the child from acquiring the effective cognitive strategies needed for normal academic functioning and adequate academic achievement after TBI (Horton et al., 2010). However, if there is no insult to the brain, cognitive functions are expected to develop normally as a result of synaptic pruning and increased white-matter volume in the prefrontal cortex (Blakemore, 2012), making it likely that the maturation of these abilities will lead to greater cognitive and academic ability (Blakemore & Choudhury, 2006), such as seen for the NRC-group in this study. Limitations for this study include a small sample size and the testing of only one variable. It is therefore recommended that future studies include more variables, and aim at creating a larger, randomized sample size, possibly providing a more representative pool of participants to study this phenomenon in South African context. It is also advised that future studies consider using neuropsychological measures to test cognitive functioning. As previous studies have indicated specific impairment in executive functioning after TBI, it may be worth researching the effect of concussion on executive functioning more thoroughly (Anderson, 2002; Anderson et al., 2010; Horton et al., 2010). Further it may be valuable to consider using functional MRI studies to broaden existing knowledge about the interaction between pathophysiology and cognitive functioning This study also highly recommends that schools and rugby clubs catering for child and adolescent players reconsider the importance of implementing proper return to play protocols after players obtain concussions. / Thesis (MA (Clinical Psychology))--North-West University, Potchefstroom Campus, 2013

De nombreux changements cérébraux s’opèrent au cours du vieillissement normal, et ce, tant au niveau structurel que fonctionnel. Ces changements résultent le plus souvent en une certaine détérioration du fonctionnement cognitif et se répercutent ainsi sur la qualité de vie des personnes âgées. Il appert toutefois que certaines personnes se voient relativement épargnées et parviennent à maintenir un niveau de fonctionnement cognitif comparativement élevé en dépit de l’avancement en âge. En présence d’une accélération du vieillissement populationnel, il devient donc criant de comprendre les facteurs et les mécanismes neurobiologiques qui contribueraient à cette meilleure résilience face aux effets du vieillissement sur le cerveau et la cognition. Bien que plusieurs modèles aient tenté de rendre compte de ce phénomène, la compréhension des mécanismes qui le sous-tendent demeure à ce jour relativement lacunaire. L’objectif principal de cette thèse était ainsi d’exposer les corrélats neurobiologiques associés à une meilleure résilience face aux effets du vieillissement normal sur le cerveau et la cognition, et ce, par l’entremise de mesures d’imagerie par résonance magnétique structurelle (IRM) et fonctionnelle (IRMf). Cette thèse contient quatre articles. L’intention du premier article (Chapitre II) était de faire une synthèse des connaissances quant aux mécanismes impliqués dans la résilience contre les effets délétères du vieillissement normal sur la cognition. Dans cette revue de la littérature, nous nous sommes intéressés à deux des principaux modèles visant à rendre compte de ce phénomène de protection : la réserve cérébrale et la réserve cognitive. L’examen de la littérature empirique amène à la conclusion qu’une meilleure résilience pourrait reposer sur des différences cérébrales à la fois structurelles et fonctionnelles, et donc que les deux modèles proposés pourraient amener une contribution indépendante au phénomène de résilience. Par ailleurs, nous soulevons l’hypothèse que les corrélats de la résilience s’apparentent grandement aux différences cérébrales associées aux entraînements cognitifs. Les trois articles suivants sont des articles empiriques qui s’intéressent à la mémoire de travail, une fonction qui décline avec l’âge, mais qui montre d’importantes différences interindividuelles. L’objectif du second article (Chapitre III) était d’investiguer la relation entre la scolarité, un indicateur de réserve (reserve proxy) bien établi, et le volume régional de la substance grise, ainsi qu’entre la scolarité et les activations cérébrales lors d’une tâche de mémoire de travail chez des participants âgés et cognitivement sains. Les résultats indiquent qu’un nombre d’années de scolarité plus élevé est à la fois associé à une moindre perte de volume liée à l’âge dans les régions frontales et pariétales, ainsi qu’à une plus grande activation liée à l’âge dans certaines régions préfrontales faisant partie du réseau de la mémoire de travail. La troisième étude (Chapitre IV) visait à examiner si les différences cérébrales fonctionnelles associées à la scolarité sont compatibles avec des mécanismes d’efficacité ou de flexibilité neuronale. Elle avait ensuite pour objectif d’examiner l’effet « protecteur » de ces différences fonctionnelles sur le plan de la performance en mémoire de travail. Les résultats suggèrent que les deux mécanismes posés seraient associés à une meilleure préservation de la mémoire de travail face aux effets de l’âge, mais que leur implication respective dépendrait du niveau d’exigence de la tâche. Enfin, l’objectif de la quatrième étude (Chapitre V) était de tester, dans un premier temps, la relation entre l’engagement dans un style de vie stimulant et le maintien de l’intégrité de la substance blanche. Puis, l’étude visait dans un second temps à examiner si une plus grande intégrité de la substance blanche diminuait l’impact de l’âge sur la mémoire de travail. Les résultats de l’étude suggèrent qu’un style de vie plus stimulant serait associé à un moindre volume de lésions de la substance blanche liées à l’âge et que ce moindre volume de lésions de la substance blanche serait en retour associé à de meilleures performances en mémoire de travail. / Many changes to the brain occur in normal aging, both structurally and functionally. These changes most often result in a certain deterioration of cognitive functioning, and thus affect the quality of life of the elderly. It appears, however, that some people are relatively spared and manage to maintain a comparatively high level of cognitive functioning despite advancing in age. In the presence of an acceleration of population aging, there is a striking need to understand the factors and neurobiological mechanisms that may contribute to this better resilience to the effects of aging on the brain and on cognition. Although several models have attempted to account for this phenomenon, the understanding of the mechanisms that underpin it is still relatively incomplete. The main objective of this thesis was to expose the neurobiological correlates associated with a better resilience to the effects of normal aging on the brain and cognition, through structural (MRI) and functional magnetic resonance imaging (fMRI). This thesis contains four articles. The intention of the first article (Chapter II) was to synthesize knowledge about the mechanisms involved in the resilience against the degenerative effects of normal aging on cognition. In this literature review, we were interested in the two main models attempting to account for this protective phenomenon: the brain reserve and cognitive reserve. Examining the empirical literature leads to the conclusion that better resilience might be based on both structural and functional brain differences, and therefore that the two proposed models could make an independent contribution to the resilience phenomenon. In addition, we hypothesize that the correlates of resilience are very similar to the brain differences associated with cognitive training. The following three articles are empirical articles which focus on working memory, a function that declines with age but shows significant inter-individual differences. The purpose of the second article (Chapter III) was to investigate the relationship between education, a well-established reserve proxy, and the regional volume of gray matter, as well as between education and brain activation during a working memory task in cognitively healthy elderly participants. Results indicate that higher years of education are associated with lower age-related loss of volume in the frontal and parietal areas, as well as greater age-related activation in some prefrontal regions that are part of the working memory network. The third study (Chapter IV) sought to examine whether functional education-related brain differences are compatible with neuronal efficiency and/or flexibility mechanisms. It further aimed to examine the "protective" effect of these functional differences on working memory performance. The results suggest that the two proposed mechanisms would be associated with a better preservation of working memory in the face of the effects of age, but that their respective involvement would depend on the level of the task requirement. Finally, the objective of the fourth study (Chapter V) was to first test the relationship between engagement in a stimulating lifestyle and white matter integrity maintaining. Then, the study aimed to examine whether greater white matter integrity decreased the impact of age on working memory. The results of the study suggest that a more stimulating lifestyle would be associated with a lesser age-related white matter lesions volume, and that this smaller white matter lesions volume would in turn be associated with better working memory performance.