Mechanisms underlying rapid forgetting in health and Alzheimer’s disease

Working memory is known to allow retaining and manipulation of information over short periods of time. It forms an essential part of cognitive processes and over the years different researchers came up with models trying to explain the mechanisms underlying how we hold onto important information. From a clinical perspective, it is even more essential to explore what underlies forgetting over such short periods of time given that patients presenting in the clinic with memory problems can suffer from transient innocuous to severe neurodegenerative disease associated with forgetting. In a first chapter, classical and modern models of working memory are explored and mechanisms of guiding attention in working memory are introduced. Their implications and relationship to anatomical structures are discussed in order to set the scene for the following chapters in which their relevance for the detection of significant impairment due to neurodegeneration is tested. In particular, one domain of working memory, rarely explored in the past, is examined: Though previous studies on working memory focused on encoding and maintenance, only few researchers have yet looked into retrieval the literature on which is in particular explored in the introduction. The following three empirical chapters explore in a total of nine experiments how retrieval cues impact on retrieval from visuospatial working memory, how retrieval of information already in a privileged state is protected from interference and finally how information retrieved from working memory can be tested in the absence of a cue. This new uncued free full report allows for the distinction between errors that happen due to forgetting or confusion of one or several features of objects that are to be remembered. In the second half of this thesis, essential findings in working memory come to be applied in patient populations. Firstly, the effects of rivastigmine on working memory performance and focussing attention are under investigation in people suffering from Parkinson’s disease dementia. Secondly, a new gamified tablet-based task for the use in people at risk of developing Alzheimer’s disease in the future is introduced and tested in different age groups In conclusion, this thesis shows that retrieval cues themselves can impair retrieval from visuospatial short-term memory and that most recently presented information is protected from these detrimental effects. Moreover, one can test working memory in the absence of a probe which allows to assess the fate of all features bound into objects individually. Finally, rivastigmine is shown to improve working memory performance in patients suffering from Parkinson’s disease dementia and different age groups perform similar in a portable tablet task that is aimed at detecting pathologic but not age-related changes in working memory.