Exploring the use of multiplex network science to guide selection of treatment stimuli for word finding disorders in individuals with Primary Progressive Aphasia

Abstract

Primary progressive aphasia (PPA) is a neurodegenerative language disorder gradually affecting individuals before the age of 65 years. The hallmark of PPA onset is the progressive and isolated impairment of word retrieval and/or comprehension. Other cognitive skills such as memory, visuospatial, reasoning and behaviour-related skills generally remain undisturbed during the initial stage of the disorder. There are 3 identified variants of PPA: semantic PPA (svPPA), nonfluent PPA (nfvPPA) and logopenic PPA (lvPPA). Each subtype of PPA is classified based on specific speech and language characteristics of the variant. Word finding difficulties are present in all three variants of PPA, but each variant shows different profiles of impairment in lexical processes. Due to lvPPA being the most recently discovered, the processes of lexical retrieval have been underexplored in this clinical population. The aim of this thesis is to apply the principles of network science to explore the influence of lexicon structure on lexical processing in individuals with lvPPA. The two network variables selected for investigation are degree (i.e., the number of words immediately linked to a given word, by phonological or semantic similarity) and closeness centrality (i.e., how closely connected a word is to all other words in the network and, therefore, how central its position is in the network). This thesis consists of three experimental studies (Chapters 2-4) that aim to address the identified gaps.