Aphasia is a language disorder that affects approximately 1 million US stroke survivors and results in impaired communication across modalities, often with detrimental effects on autonomy and quality of life. While the extent to which these domains are affected vary from person to person, a common symptom found in many individuals with aphasia is sentence comprehension impairment, which are most notable with complex (noncanonical) sentences that require establishing a long-distance relationship between non-adjacent sentence constituents. Of interest to the work presented here are the ways in which auditory sentence input is processed and interpreted. Evidence in the literature implicates lexical-level processing impairments, such as delayed and/or reduced activation of lexico-semantic information, as contributors to syntactic processing disruptions and sentence comprehension deficits. By limiting timely lexical activation during sentence processing, these impairments can also preclude successful lexical encoding and amplify the processing costs of similarity-based interference during syntactic retrieval (i.e., dependency linking). A series of studies are presented to explore three different approaches (processing time, attentional engagement, and semantic pre-activation) to facilitate lexical access and strengthen concomitant encoding of a target word in a sentence. The effects of these approaches on real-time lexical and syntactic processing during complex (object-relative) sentence processing was examined using eye-tracking-while-listening methodology in neurotypical college-age adults (YNC), older adults (age-matched controls, AMC) and individuals with aphasia (IWA). In each of these approaches, I ask whether enhancing the processing of a target lexical item during sentence processing will 1) facilitate its lexical access and encoding, 2) increase its availability for downstream syntactic retrieval operations, and 3) improve sentence comprehension. This novel area of exploration has the potential to improve our understanding of the interaction between lexical and syntactic levels within sentence processing and the cognitive and linguistic factors promoting lexical processing within a sentence context.

Language is a complex process and a unique human ability. However, this intricate process can become impaired, leading to significant challenges in comprehension and expression. This dissertation focuses on the brain basis of language impairments across two distinct populations: adults with post-stroke aphasia and children with Developmental Language Disorder (DLD). These populations are etiologically distinct; adults with aphasia have acquired language impairments in an established language and neural system, while children with DLD have language impairments in a developing language and neural system. The dissertation consists of four studies that investigate the brain functions supporting language, focusing on cerebral blood flow (CBF) in both groups and functional connectivity in children with DLD, to better understand the relationship between these measures and language abilities. Chapter 1 emphasizes the importance of characterizing baseline CBF values in individuals with aphasia. This study demonstrates how regions distal to the site of neural damage exhibit reduced CBF, which contributes to resultant language abilities, highlighting the critical need to properly characterize CBF patterns in clinical populations. In Chapter 2, the focus shifts to children with DLD. The neural underpinnings of DLD are not well understood, thus this chapter synthesizes over 50 years of behavioral and neural research in DLD to identify key brain regions of interest (ROIs) and language deficits that characterize the disorder. Chapter 3 builds on these identified ROIs, exploring brain function in children with DLD. Results from this study provide preliminary evidence of a relationship between altered brain function in children with DLD and language impairments. However, Chapter 3 utilized standardized measures of language abilities, which may not capture nuanced differences in real-time language processing. To address this, Chapter 4 employs an online priming paradigm to investigate real-time syntactic dependency linking in children with DLD at regular and slow rates of speech. Results show that children with DLD possess intact linguistic knowledge, but real-time temporal constraints during sentence processing lead to comprehension breakdowns. These studies provide evidence of the impact of altered brain function on language comprehension in language-impaired populations and illustrate the importance of properly characterizing the neural underpinnings of language processing.

This dissertation investigates the factors that contribute to the sentence comprehension deficits exhibited by individuals with a particular acquired language disorder, Broca’s aphasia. Individuals with Broca’s aphasia (IWBA) have no difficulty comprehending sentences that conform to canonical (S-V-O) word order. However, difficulty arises for non-canonical sentences that contain syntactic dependencies. There are several accounts that attempt to explain these deficits. The focus of these accounts differs, some focus on syntax, others on lexical level deficits, and others on reduced cognitive resources. The current dissertation offers and focuses on a more recent account of the sentence processing deficits in IWBA, the Intervener Hypothesis (IH).

The IH suggests that difficulty comprehending sentences that contain syntactic dependencies arises when comprehension requires the crossing of an intervening noun phrase (NP), or intervener. The intervener causes similarity-based interference (SBI) because it is a possible element in the syntactic dependency. This dissertation tests the IH using sentences that contain a syntactic dependency inherent to unaccusative verbs.

Chapter 2 outlines pertinent linguistic concepts. Chapter 3 provides an overview of aphasia. Chapter 4 reviews psycholinguistic literature focusing on verb processing. Chapter 5 describes the sentence comprehension deficits observed in aphasia and theoretical accounts of these deficits. Chapter 6 details the account this dissertation focuses on, the IH. Chapter 7 investigates the time-course of lexical reactivation of unaccusative verbs during sentence processing. The results revealed that IWBA exhibit a time-course of lexical reactivation similar to neurologically unimpaired age-matched control (AMC) individuals indicating that IWBA are sensitive to the unaccusative property. Chapter 8 evaluates the IH by comparing final comprehension of sentences that contain unaccusative verbs in sentence constructions that contain an intervener and ones that do not. For IWBA, chance comprehension was found for sentences that contained an intervener suggesting SBI interferes with comprehension. Chapter 9 further investigates the pattern of comprehension observed in Chapter 8 using a real-time methodology, and also includes the first semantic manipulation of an intervener. The results revealed that SBI is observed during real time sentence processing of sentences that contain an unaccusative verb and an intervener, and that animacy reduced observed SBI.

This dissertation explores how listeners extract meaning from personal and reflexive pronouns in spoken language. To be understood, words like her and herself must be linked to a prior element in the speech stream (or antecedent). This process draws on syntactic knowledge and verbal working memory processes. I present two original research studies investigating moment-to-moment sentence processing and final comprehension of sentences containing proforms in children and adults with and without language impairments.

Chapter 2 provides an overview of syntactic constraints and psycholinguistic evidence from unimpaired listeners in addition to an introduction to the language impaired populations. Chapter 3 reviews psycholinguistic evidence of similarity-based interference in verbal working memory during unimpaired language processing before presenting a recent account – the Intervener Hypothesis – that points to similarity-based interference as a source of processing and comprehension difficulty in language impaired populations.

Chapter 4 presents a series of experiments in children with typical language development (TLD) and children with a developmental language impairment – specific language impairment (SLI). The processing patterns of the two participants groups reveal qualitative differences and children with SLI demonstrate increased comprehension difficulty relative to TLD children. Chapter 5 presents a series of experiments in adults with an acquired language impairment – agrammatic Broca’s aphasia – and age-matched control participants. Results reveal increased processing and comprehension difficulty of sentences containing pronouns relative to reflexives for individuals with Broca’s aphasia.

The processing patterns demonstrated by unimpaired child and adult participants reveal a number of striking similarities. In addition, the processing patterns of children with SLI mirror those of adults with Broca’s aphasia. Although the etiology of language impairments differs in these groups, children with SLI and adults with Broca’s aphasia share many of the same language processing difficulties. As such, I evaluate the potential of a single theoretical framework (the Intervener Hypothesis) to account for the patterns observed in each group. In addition, I evaluate how processing and comprehension patterns change across the lifespan in participants with unimpaired language and the extent to which processing and comprehension patterns overlap.

Aphasia is a condition resulting from stroke that impairs individuals' communicative abilities. There are currently one million individuals living with aphasia in the US and within this population, many have sentence comprehension impairments. This dissertation aims to investigate (1) approaches with which sentence comprehension impairment can be mitigated in individuals with aphasia, and (2) whether the integrity of connectivity between frontal and temporo-parietal regions is predictive of individuals’ sensitivity to the proposed mitigation approaches. Overall, the data in the current dissertation revealed that reducing the interference level via semantic cues can benefit the real-time sentence processing in individuals with aphasia. Moreover, there is an emerging pattern for the functional role of the anterior segment of the arcuate fasciculus and frontal aslant tract in the left hemisphere regarding real-time sensitivity to semantic cues during sentence comprehension.