Memory binding is a cognitive process that enables complex objects to be stored or retrieved coherently during perception, learning, or action. Binding functions are aimed at reducing the misattribution of the features of objects in crowded and changing sensory contexts, ensuring accurate representation in visual working memory. Binding is a relatively new concept in working memory research. However, as an integrative function it provides a rich context in which to investigate the mechanisms underlying memory deterioration. In this PhD project, a range of experimental temporary binding paradigms were used to investigate whether some of the memory impairments observed in patients with Alzheimer’s Disease could be accounted for by deficits in this memory function. A set of neuropsychological tasks were used to investigate binding operations across memory domains (i.e., verbal and nonverbal), sensory modalities (i.e., visual and auditory), types of information (e.g., objects and colours), and retrieval processes (i.e., recognition and recall) in healthy individuals, Alzheimer’s Disease patients and other clinical populations. The results suggest that the efficiency of short-term memory to store bound complex events depends on the nature of the information presented (e.g., type of information bound into objects) (Chapter 2). Short-term memory seems to be equipped with relatively separate mechanisms to store integrated objects and individual features (Chapter 4). It was also observed that the binding properties of short-term memory apply to healthy young and older people, and are functions which are preserved in the elderly (Chapter 3). In two additional experimental chapters (5 and 6) the preserved binding abilities of older people were compared with temporary binding in Alzheimer’s Disease. The latter group showed a very large impairment in binding that was distinct from their impairments in memory for individual features. These findings suggest that memory binding tasks could reliably separate the cognitive changes in normal ageing from those linked with Alzheimer’ Disease. Moreover, the results of Chapter 7 suggested that memory binding tasks may detect memory changes in people that will develop Alzheimer’ Disease (i.e., asymptomatic carriers of the gene defect E280A of the Preseniline-1 gene) almost 10 years before the average age of onset. These results are relevant to our understanding of short-term memory and to the memory models currently available. Finally, it is suggested that the constructs of memory binding may increase the sensitivity of current assessment procedures for people at risk of developing Alzheimer’s Disease.

The thesis focuses on memory binding from the perspective of cognitive psychology, neuropsychology and presents structural correlates of memory. It reflects current research and concepts such as Subjective Cognitive Disorder (SCD) and mild cognitive impairment (MCI) and advocates previously created units for the continuum of Alzheimer's disease. The aim of the work is to verify the functionality of the Czech version of the Buschke Memory binding test for clinical practice. In the Empirical section it focuses on the MBT memory binding capability test and performance comparison in three groups of the sample population (without cognitive deficit, SCD and MCI). Statistical testing (ANCOVA) found a significant difference in performance in selected scores between the three groups. The results showed that memory binding performance is impaired as the deficit depth progresses. In our sample, no significant performance difference could be found between the cognitively normal group and the SCD group. The references are cited accoding to APA 7th edition manual. Key words Alzheimer's disease, neuropsychology, cognition, memory binding

INTRODUÇÃO: O teste de memória integrativa (TMI) de curto prazo, nas modalidades de recordação espontânea (RE) e reconhecimento visual (RV), tem se mostrado promissor para apoiar o diagnóstico precoce da doença de Alzheimer (DA), com grande potencial para aplicação clínica. Até o momento, não foram encontrados estudos sobre o TMI junto à população idosa brasileira, nem tampouco pesquisas utilizando a tarefa de RV em pacientes com a variante comportamental da demência frontotemporal (DFTvc). OBJETIVOS: A presente pesquisa comparou o desempenho de pacientes com DA, pacientes com DFTvc e Controles nas duas modalidades de TMI e examinou a acurácia diagnóstica dos testes para a detecção da DA. MÉTODOS: Foram avaliados participantes de três centros de pesquisa, USP São Paulo, USP Ribeirão Preto e Universidade Federal de Minas Gerais, assim como do Instituto Paulista de Geriatria e Gerontologia. A amostra foi composta por 85 indivíduos, subdivididos em 32 Controles, 35 pacientes com DA e 18 pacientes com DFTvc. Foram geradas curvas ROC para calcular a acura?cia diagno?stica dos testes contrastando os grupos diagno?sticos. Realizou-se análises de regressão logística, de correlação e análises de interação entre diagnóstico e condição experimental (características integradas x não-integradas). RESULTADOS: No TMI com RE, Controles apresentaram resultados semelhantes aos pacientes com DFTvc e os pacientes com DA mostraram resultado significativamente menor em relac?a?o aos dois primeiros grupos (Controles = DFTvc > DA). No TMI com RV, Controles mostraram escores significativamente maiores que os pacientes com DFTvc e DA (Controles > DA = DFTvc). A acurácia diagnóstica do teste de RE para DA foi de 0,853, com sensibilidade e especificidade de 84,4% e 80%, respectivamente. A acurácia diagnóstica do teste de RV para DA foi de 0,809, com sensibilidade e especificidade de 65,4% e 72,2%, respectivamente. As modalidades do TMI mostraram correlação significativa com outros testes neuropsicológicos. A tarefa de RV apresentou correlação maior com testes relacionados as funções executivas e de atencão, enquanto que a tarefa de RE mostrou correlação com testes que avaliam memória episódica e funcões executivas. CONCLUSÕES: O presente estudo demonstrou a aplicabilidade do TMI na população brasileira. As tarefas apresentaram características adequadas para a prática clínica: são fáceis e rápidas de aplicar, necessitam de pouco tempo de aplicação, e podem contribuir para o diagnóstico da DA. Especialmente a tarefe de RE, que foi capaz de diferenciar DA de DFTvc / INTRODUCTION: The free recall (FR) and visual recognition (VR) modalities of the memory binding test (MBT) are promising tools to support the early diagnosis of Alzheimer\'s disease (AD). Until the present moment, no studies have been conducted with these tests among the elderly population in Brazil, and there are no studies with the VR modality of the MBT involving behavioural variant frontetemporal dementia (bvFTD) patients. OBJECTIVES: the present research examined the diagnostic accuracy of the two modalities of the MBT for the detection of AD, and compared the performance of patients with AD, bvFTD and controls. METHODS: Participants of three research centres were assessed, USP São Paulo, USP Ribeirão Preto and Federal University of Minas Gerais, and from the Paulista Institute of Geriatrics and Gerontology. The sample comprised 85 participants: 32 controls, 35 AD and 18 bvFTD patients. ROC analyses were used to examine the diagnostic accuracy of the MBT contrasting the different diagnostic groups. Logistic regression analysis, correlations were calculated. ANOVA analyses tested the interaction between diagnosis and experimental condition (features bound and unbound). RESULTS: In the FR task, controls and bvFTD showed similar results, and AD patients showed worse performance (controls = bvFTD > AD). In the VR task, controls showed better performance than the clinical groups (controls > AD = bvFTD). The diagnostic accuracy of the FR task for AD was 0.853, with 84.4% of sensitivity and 80% of specificity. The diagnosis accuracy of the VR task for AD was 0.809, with 65.4% of sensitivity and 72.2% of specificity. The MBT showed significant correlation with other neuropsychological tests. The VR task correlated with tests associated with executive functions and attention, while the FR task correlated with episodic memory and executive functions. CONCLUSIONS: The present study showed the applicability of the MBT for the Brazilian population. The tasks presented adequate characteristics for clinical use: they are quick to administer and can contribute to the diagnosis of AD. Especially the FR task, that could differentiate AD from bvFTD

INTRODUÇÃO: O teste de memória integrativa (TMI) de curto prazo, nas modalidades de recordação espontânea (RE) e reconhecimento visual (RV), tem se mostrado promissor para apoiar o diagnóstico precoce da doença de Alzheimer (DA), com grande potencial para aplicação clínica. Até o momento, não foram encontrados estudos sobre o TMI junto à população idosa brasileira, nem tampouco pesquisas utilizando a tarefa de RV em pacientes com a variante comportamental da demência frontotemporal (DFTvc). OBJETIVOS: A presente pesquisa comparou o desempenho de pacientes com DA, pacientes com DFTvc e Controles nas duas modalidades de TMI e examinou a acurácia diagnóstica dos testes para a detecção da DA. MÉTODOS: Foram avaliados participantes de três centros de pesquisa, USP São Paulo, USP Ribeirão Preto e Universidade Federal de Minas Gerais, assim como do Instituto Paulista de Geriatria e Gerontologia. A amostra foi composta por 85 indivíduos, subdivididos em 32 Controles, 35 pacientes com DA e 18 pacientes com DFTvc. Foram geradas curvas ROC para calcular a acura?cia diagno?stica dos testes contrastando os grupos diagno?sticos. Realizou-se análises de regressão logística, de correlação e análises de interação entre diagnóstico e condição experimental (características integradas x não-integradas). RESULTADOS: No TMI com RE, Controles apresentaram resultados semelhantes aos pacientes com DFTvc e os pacientes com DA mostraram resultado significativamente menor em relac?a?o aos dois primeiros grupos (Controles = DFTvc > DA). No TMI com RV, Controles mostraram escores significativamente maiores que os pacientes com DFTvc e DA (Controles > DA = DFTvc). A acurácia diagnóstica do teste de RE para DA foi de 0,853, com sensibilidade e especificidade de 84,4% e 80%, respectivamente. A acurácia diagnóstica do teste de RV para DA foi de 0,809, com sensibilidade e especificidade de 65,4% e 72,2%, respectivamente. As modalidades do TMI mostraram correlação significativa com outros testes neuropsicológicos. A tarefa de RV apresentou correlação maior com testes relacionados as funções executivas e de atencão, enquanto que a tarefa de RE mostrou correlação com testes que avaliam memória episódica e funcões executivas. CONCLUSÕES: O presente estudo demonstrou a aplicabilidade do TMI na população brasileira. As tarefas apresentaram características adequadas para a prática clínica: são fáceis e rápidas de aplicar, necessitam de pouco tempo de aplicação, e podem contribuir para o diagnóstico da DA. Especialmente a tarefe de RE, que foi capaz de diferenciar DA de DFTvc / INTRODUCTION: The free recall (FR) and visual recognition (VR) modalities of the memory binding test (MBT) are promising tools to support the early diagnosis of Alzheimer\'s disease (AD). Until the present moment, no studies have been conducted with these tests among the elderly population in Brazil, and there are no studies with the VR modality of the MBT involving behavioural variant frontetemporal dementia (bvFTD) patients. OBJECTIVES: the present research examined the diagnostic accuracy of the two modalities of the MBT for the detection of AD, and compared the performance of patients with AD, bvFTD and controls. METHODS: Participants of three research centres were assessed, USP São Paulo, USP Ribeirão Preto and Federal University of Minas Gerais, and from the Paulista Institute of Geriatrics and Gerontology. The sample comprised 85 participants: 32 controls, 35 AD and 18 bvFTD patients. ROC analyses were used to examine the diagnostic accuracy of the MBT contrasting the different diagnostic groups. Logistic regression analysis, correlations were calculated. ANOVA analyses tested the interaction between diagnosis and experimental condition (features bound and unbound). RESULTS: In the FR task, controls and bvFTD showed similar results, and AD patients showed worse performance (controls = bvFTD > AD). In the VR task, controls showed better performance than the clinical groups (controls > AD = bvFTD). The diagnostic accuracy of the FR task for AD was 0.853, with 84.4% of sensitivity and 80% of specificity. The diagnosis accuracy of the VR task for AD was 0.809, with 65.4% of sensitivity and 72.2% of specificity. The MBT showed significant correlation with other neuropsychological tests. The VR task correlated with tests associated with executive functions and attention, while the FR task correlated with episodic memory and executive functions. CONCLUSIONS: The present study showed the applicability of the MBT for the Brazilian population. The tasks presented adequate characteristics for clinical use: they are quick to administer and can contribute to the diagnosis of AD. Especially the FR task, that could differentiate AD from bvFTD

Submitted by Viviane Lima da Cunha (viviane@biblioteca.ufpb.br) on 2016-09-14T16:15:04Z No. of bitstreams: 1 arquivototal.pdf: 2412011 bytes, checksum: adf2606f46d74fde5f51bff921362506 (MD5) / Made available in DSpace on 2016-09-14T16:15:04Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 2412011 bytes, checksum: adf2606f46d74fde5f51bff921362506 (MD5) Previous issue date: 2016-02-29 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / Human aging causes different decline in many cognitive function. Episodic processing, among all memory systems, shows to be the most affected along aging. The episodic memory is a subsystem of memory responsible for receiving and storing information about certain episodes or events, temporally dated, high lighting a spatiotemporal relationship between them. One way of assessing this construct would be using behavioral criteria for identifying information about what happening, where and when, which are the basic components of “What-Where-When” paradigm, or else, by proposing to replace the temporal component by context, “What-Where-Which” paradigm. Thus, the objective of this research was to analyze the effect of aging on episodic memory, assessed by two experimental paradigms, to study the correlation of these tests with classic measures of episodic memory. We used a sample with 70 volunteers, divided in two group: younger (N = 35, M = 22,74; SD= 2,99) and older (N = 35, M = 62,54; SD = 4,82) adults. The evaluation protocol was composed of: (i) test based on the paradigm "What-Where-When"; (ii) test based on the paradigm "What-Where-Which"; (iii) classic episodic memory tests; and (iv) a neuropsychological battery, the latter being applied only to the elderly group. The results showed that different episodic memory task apparently measuring something in common. Comparisons between group showed effective impairment associated with aging for all aspects evaluated by the test "What-Where-When", and the integration aspect was the component that suffered the greatest decline associated with age [F(1, 68)= 53,86; p < .001; η² = .442]. Other results showed that Free Recall also presented decline [F(1, 68)= 18,60; p < .001; η² = .215]. However, it was not found significant differences between groups for the task based on the paradigm "What-Where-Which". The findings of this study show that task-based paradigm "What-Where-When" and "What-Where-Which" are promising instruments for measurement of episodic memory. However, apparently the temporal aspect was more sensitive to detect possible changes related to age. / O envelhecimento humano provoca declínios em diversas funções cognitivas. O processamento episódico, dentro todos os sistemas de memória, demonstra ser um dos mais afetados durante a velhice. A memória episódica é um subsistema mnemônico responsável por receber e armazenar informações sobre determinados episódios ou eventos temporalmente datados, existindo uma relação espaço-temporal entre eles. Uma forma de avaliação desse construto seria através de critérios comportamentais para identificação de informações sobre o que aconteceu, onde e quando, sendo esses os componentes básicos do paradigma “Que-Onde-Quando”, ou então, por meio de outra proposta que substitui o componente temporal pelo contextual “Que-Onde-Qual contexto”. Diante disso, o presente trabalho objetivou analisar o efeito da idade sobre a memória de integração, mensurada por dois paradigmas experimentais, e estudar os correlatos destas tarefas com medidas clássicas de memória episódica. Para tal utilizou-se uma amostra de 70 voluntários, dividida em dois grupos: Adultos jovens - GA (N = 35, M = 22,74; DP= 2,99) e Idosos – GI (N = 35, M = 62,54; DP = 4,82). O protocolo de avaliação utilizado foi composto por: (i) Teste baseado no paradigma “Que-Onde-Quando”; (ii) Teste baseado no paradigma “Que-Onde-Qual contexto”; (iii) Testes clássicos de memória episódica; e (iv) uma bateria neuropsicológica, sendo essa última aplicada apenas ao grupo de idosos. Os resultados mostraram que as diferentes medidas de memória episódica utilizadas aparentemente estão medindo algo em comum. As comparações efetivamente mostraram haver comprometimento, associado ao envelhecimento, de todos os aspectos avaliados pelo teste baseado no paradigma “Que-Onde-Quando”, sendo a “integração” o componente que sofreu maior declínio proporcional associado ao aumento da idade [F(1, 68)= 53,86; p < .001; η² = .442]. Achados similares a esse também foram verificados para tarefa de Recordação Livre Imediata (testes clássico de memória episódica) [F(1, 68)= 18,60; p < .001; η² = .215]. Entretanto, não foi registrada diferença significativa entre os grupos para a tarefa baseada no paradigma “Que-Onde-Qual contexto”. Os achados do presente estudo evidenciaram que tarefas baseadas no paradigma “Que-Onde-Quando” e “Que-Onde-Qual contexto” são promissores instrumentos válidos de mensuração da memória episódica. Porém, aparentemente a proposta relacionada ao aspecto temporal mostrou-se mais sensível para detecção de possíveis alterações relacionadas à idade.