Project Summary
The goal of this proposal is to investigate the sources of variability in lexical retrieval deficits in individuals with
left hemisphere stroke-induced lesions and how these deficits may be uniquely impacted depending on
semantic context. We will achieve this by examining the brain dynamics of lexical retrieval and behavior in
individuals with left prefrontal cortex (LPFC) versus left posterior temporal cortex (pLTC) lesions compared to
matched controls. Lexical retrieval deficits are observed to some extent in all individuals with stroke-induced
aphasia. Lexical activation and selection are key processes enabling us to retrieve words as we speak. Lexical
activation occurs through spreading activation from semantic to lexical representations and is facilitated in
semantically-related contexts. Lexical selection is the selection of the target word from activated semantically-
related alternatives and can be slowed in semantically-related contexts. Lexical retrieval stages can be
differently affected depending on the type of semantic context. Taxonomically-related words (e.g., wolf and dog)
have been linked with semantic interference. However, thematically-related words (e.g., dog and leash) have
been linked with facilitation. Lexical retrieval is often impaired in individuals with aphasia, but the driving forces
behind lexical retrieval deficits and how these may be differentially impacted depending on the type of semantic
context are not well understood. Separating individuals who have a greater activation versus selection deficit
from individuals who have a greater selection versus activation deficit may be a more informative way to
characterize lexical retrieval deficits and may help predict the impact of different semantic contexts in a more
individualized manner. Here, we propose that activation deficits, corresponding to a deficit in lexical activation,
are associated with pLTC lesions as individuals with lesions in the pLTC have been shown to struggle with
recognizing the correct word even when presented to them as an option to choose from. Contrastively, we
propose that selection deficits, involving a deficit in the ability to suppress co-activated lexical items during
selection, are associated with LPFC lesions as the LPFC has been proposed to play a role in overcoming
semantic interference. Aim 1 will dissociate individuals with lexical activation versus selection retrieval deficits
by using pre-established behavioral paradigms allowing this dissociation. Aim 2 will investigate the impact of
taxonomic and thematic relationships on lexical retrieval subprocesses in individuals with pLTC and LPFC
lesions through the use of electrophysiological techniques. The results of this study will provide a key step
towards understanding how different lexical retrieval deficits interact with semantic contexts, and the associated
brain dynamics in stroke-induced aphasia, providing a knowledge basis for the future development of clinical
assessment tools for lexical retrieval.