# Modeling treated recovery from aphasia

> **NIH NIH P50** · UNIVERSITY OF SOUTH CAROLINA AT COLUMBIA · 2020 · $609,647

## Abstract

Summary: Project 1
Stroke is the leading cause of serious adult disability in the United States. One of the most devastating
impairments resulting from stroke is aphasia, a language impairment caused by left hemisphere damage
involving cortical language areas. It is generally accepted that behavioral aphasia treatment is effective.
Nevertheless, different patients experience very different degrees of benefit from aphasia treatment. Despite
considerable differences in the response to aphasia treatment, the relationship between patient factors and
treatment response is poorly understood and very few reliable prognostic indicators have been identified. This
is a major problem, as both time and resources are wasted when clinicians do not know what patients are likely
to respond to treatment, or which treatment best fits individual patients. The purpose of the current project is to
develop a model that includes biographical and cognitive/linguistic factors to predict patients' response to
aphasia treatment. Aphasia severity is one of the few factors that has been identified as a reliable predictor of
performance in treatment; it is generally accepted that more severe aphasia is associated with poorer
treatment outcomes. However, aphasia severity is a multidimensional construct and patients with similar
overall severity scores might demonstrate very different language impairment profiles. To better understand
how language impairment relates to treatment outcomes, the dual stream model (DS model;1) will be
consulted. Specifically, we will test whether measures of proportional damage to the cortical areas that
comprise the DS model improve prediction of aphasia treatment response, beyond biographical and
cognitive/linguistic factors. Although the DS model is a functional model grounded in neuroanatomy, we expect
measures of speech and language that assess processes supported by the two major components of the DS
model – the dorsal and ventral streams – might be redundant with measures of cortical damage. To
understand whether our predictive model can be generalized across different kinds of treatment foci, each
patient will undergo treatment devoted to phonological stimulation and a separate treatment phase focusing on
semantic stimulation. Ultimately, the goal here is to construct a predictive model that will be made available on-
line so that clinicians can enter test scores from individual patients to predict how likely a given patient is to
respond to treatment, as well as the focus of that treatment.
There is a great need for prognostic indicators of aphasia treatment response. At the completion of our
research, we will understand why some patients respond better to aphasia treatment than others. We have
selected treatment approaches that are routinely used in clinical practice, allowing for immediate translation of
the findings directly into patient management. The current project will yield a vast dataset that will be made
publicly available allow...

## Key facts

- **NIH application ID:** 9889936
- **Project number:** 5P50DC014664-05
- **Recipient organization:** UNIVERSITY OF SOUTH CAROLINA AT COLUMBIA
- **Principal Investigator:** JULIUS FRIDRIKSSON
- **Activity code:** P50 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $609,647
- **Award type:** 5
- **Project period:** — → —

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9889936

## Citation

> US National Institutes of Health, RePORTER application 9889936, Modeling treated recovery from aphasia (5P50DC014664-05). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/9889936. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*