Past Projects

Four principal domains of research investigation, conceptually interrelated and methodologically interdependent, emerge from a review of HGARC research progress during the past few years. These four categories of research activity, focused on both aphasia theory and aphasia therapy, clearly serve as conceptual threads which tie the Aphasia Research Center of the past to the Aphasia Research Center of the future.

On-Line Measures of Language Processing

Members of the Harold Goodglass Aphasia Research Center were among the first to champion the use of on-line measures that tapped into the unfolding of language processing. These techniques are now in the toolkit of research aphasiologists around the world. Baker was one of the first to use reaction time as a measure of processing in aphasic individuals, and this technique remains a staple in his exploration of cognitive factors influencing communication with the C-ViC program. Blumstein and Milberg used semantic priming tasks to examine the speed of lexical access. Because Broca’s aphasics were highly sensitive to the predictability of semantic relatedness whereas Wernicke’s aphasics were not, they concluded that Broca’s aphasics were more likely than Wernicke’s aphasics to rely upon heuristic strategies in the processing of language. They are carrying on this research approach, using their finely-tuned techniques to examine the nature of the temporal parameters of language processing. Additional examples of on-line measures of language processing come from the work of Zurif and Prather, who employed both list-priming and cross-modal priming techniques as measures of the time-course of lexical activation and syntactic dependencies, demonstrating that for semantic priming Broca’s aphasics are delayed in lexical activation and Wernicke’s aphasics maintain lexical activation longer than normal. Using cross-modal priming they argued that Broca’s aphasia involves impairment of real-time processing of traces resulting from constituent movement.

Neural Bases of Language Comprehension

Zurif, Grodzinsky, Goodglass, Wingfield, Blumstein, Milberg, Brownell and their co-investigators all delved into aspects of comprehension after brain damage. Grodzinsky focused on a revised model of the functional neuroanatomy of language. He asked aphasic patients to make grammaticality judgments on a set of complex syntactic structures involving violations of constraints on syntactic movement. His results support a view of receptive grammatical mechanisms in the left cortex, which is functionally restrictive but neuroanatomically widely distributed. Zurif examined the interface between meaning and grammatical role. He found that there is a processing cost associated with the binding of meaning and grammatical role for a word whose meaning implies repetition but its grammatical role does not, (such as the word “jumped” in the sentence, “The horse jumped for an hour.”). Wernicke’s aphasics appear to have an impairment in allocating the necessary resources to accomplish this type of binding. Goodglass and Wingfield used the word-onset gating technique to examine the role of prosody in lexical access. Based on their findings they argue that not only are factors such as word frequency, word onset cohort size, and phonological neighborhood density important for lexical access, but the full prosodic pattern of a word is represented in the mental lexicon and aids lexical access. Brownell examined the role of the right hemisphere in comprehending another’s communicative intent in discourse. He found that the ability to distinguish lies from jokes was strongly correlated with measures of the ability to attribute correctly second order beliefs. The fragility of RHD patients’ understanding of second-order mental states underlies a portion of their difficulties in discourse comprehension. He also noted that the underlying impairment is not restricted to right hemisphere dysfunction.

Neural Bases of Language Production

Continuing his work in naming, Goodglass published, with Wingfield, a monograph on anomia, and several journal articles demonstrating that word initial phonology is important in the ability of aphasic persons to access a word in the mental lexicon. Albert and his colleagues continued to examine word-finding difficulties in older adults, finding an age-related decline in action naming as well as in object naming, as had been previously demonstrated by this group. Blumstein and Alexander re-examined speech production in the foreign accent syndrome in terms of features of consonant production, vowel production, and prosody. They concluded that the constellation of phonetic features associated with foreign accent syndrome is distinct from the features of speech production of Broca’s aphasics. Blumstein is currently working on the basis of phonemic paraphasias — are their origins in impaired articulatory programming or in incorrect phonological selection? Grodzinsky, continuing to explore grammar, published a major paper on tense and agreement in agrammatic production demonstrating that agrammatism consists of an impaired tense node and that all syntactic production involving structures higher than the tense node are compromised. Brownell pursued his investigations of communication impairments secondary to right hemisphere damage (RHD). He examined how RHD subjects use social and linguistic information in choosing formal versus informal terms of reference for an absent third person. Brownell’s results suggested that a decreased use of knowledge shared between a speaker and addressee disrupts RHD patients’ discourse and contributes to their aberrant interpersonal behavior. Brownell also examined factors that influence the production of requests for behavior. He found that the RHD subjects produced less explanatory supportive material than control subjects, and they tended not to vary the amount of explanatory material as a function of the request scenario. The discourse strategies observed were likely due to deficits both in pragmatic awareness and in planning utterances.

Aphasia Therapy and Predictors of Outcome

The Harold Goodglass Aphasia Research Center has always recognized as one of its mandates to translate basic research into practical benefits for persons with aphasia: new approaches to therapy, assessment tools, predictors of outcome. Albert is continuing his studies of the potential benefit accruing to persons with aphasia by means of pharmacological intervention. These studies are directly dependent on earlier HGARC research which he, Helm-Estabrooks and their colleagues carried-out on the role of perseveration in aphasia and the potential for treating aphasia by treating perseveration. Naeser and Palumbo published several papers demonstrating that extent of lesion in specific subcortical regions is highly predictive of recovery of speech and auditory comprehension. Naeser and Palumbo reported on the finding that although there are visible changes in lesion borders on CT scan after 5 years post-stroke making the lesion appear larger, the patients showed continued recovery of some language functions. They also examined the relationship between lesion patterns and outcome following treatment with Baker’s C-ViC program. The study reported that presence of lesion in specific areas, the supplementary motor area and Wernicke’s area, resulted in poor outcome for the treatment program. Baker examined cognitive factors that predict communicative success in a computerized alternative communication system, and found that short-term visual recognition memory is intact in globally aphasic individuals, but that subtle disruptions of the semantic network may underlie difficulty that some patients had with learning the C-ViC treatment program.

Arthur Wingfield, Brandeis University

Dr. Wingfield’s project in the Brandeis University Memory and Cognition Laboratory examines questions relating to speech production and its implications for memory retrieval. In one approach to this question, they explore various types of naming deficits fololwing left hemisphere focal brain damage, primarily stroke. The ongoing projects are investigating the effect of different forms of brain injury on language comprehension and production. By examining language abilities in people with specific brain injuries, a better understanding of how these areas function in non-injured adults is gained.

Aphasic Comprehension of Raising and Passive Constructions
Ken Wexler, Massachusetts Institute of Technology

This project investigates the grammatical comprehension of passives and raising constructions in aphasic patients to better understand how language therapy should be designed. This research attempts to better characterize the linguistic difficulties stemming from neurological damage related to aphasia.
Please visit Dr. Wexler’s lab website.

Metaphor Training Project
Hiram Brownell, Boston College and Kristine Lundgren, Boston University

This project addresses an important gap in the clinical literature: the dearth of tools for remediation of the communication deficits associated with brain damage caused by stroke in the right cerebral hemisphere and by traumatic brain injury. Many patients with these types of injuries exhibit a range of communication impairments with non literal language that impact their lives in negative ways. The protocol is designed to evaluate and remediate: 1) difficulty generating appropriate associations to words; 2) difficulty evaluating connotative shared meaning; and 3) difficulty selecting from among alternative interpretations. The protocol examines what specifically changes during training and the impact of the intensity of training on the duration and generalizability of gains. In addition, the protocol assesses the fading of treatment gains over time and will explore approaches to slowing or preventing that decline.

Neural Modeling and Imaging of Speech

Frank Guenther, Boston University

The primary goal of this research project is the continued development, testing, and refinement of a computational model describing the neural processes underlying speech. The project involves several coordinated fMRI studies and modeling projects, each aimed at addressing a different aspect of speech production or perception. Computer simulations of the neural model are compared to the experimental results to test and refine the model. In the past year we have worked on fMRI data analysis of an fMRI study involving speaking rate manipulation and a study involving within-category vs. cross-category auditory perturbation during speech production. Please visit Dr. Guenther’s lab website.

Sequencing and Initiation in Speech Production

Frank Guenther, Boston University

The overall goal of this project is to develop and experimentally test a neural model of the brain interactions underlying the production of speech sound sequences. In particular, we will focus on several brain regions thought to be involved in motor sequence production, including the lateral prefrontal cortex, ventral premotor cortex, supplementary motor area (SMA), pre-SMA, basal ganglia, cerebellum, and thalamus. Each of these brain regions was modeled in the GODIVA model, which was published in Journal of Cognitive Neuroscience in the past year. We have also begun data collection on an fMRI experiment involving the learning of novel syllables and another investigating the representation of syllabic frames in the brain. Please visit Dr. Guenther’s lab website.

Neural Networks and Language Recovery in Aphasia from Stroke” FMRI Studies
Margaret Naeser, Boston VA Healthcare System and Boston University

The major goal of this four-year research project is to utilized functional magnetic resonance imagihg (FMRI)to investigate brain reorganization for language behavior in chronic stroke patients with aphasia. Patients are studies with fMRI paradigms for Overt Propositional Speech fMRI; and Nonverbal Semantic Decision FMRI before and after a series of real or sham TMS treatments.
Please visit Dr. Naeser’s lab website.