An Observational Examination of Circumlocution Ability in Acute Left-Hemisphere Stroke Survivors as a Possible Estimative Behavioral Measure of Very Mild Aphasia
Related Experiment Video
Updated: May 26, 2026
Dual-Task Stroop Paradigm for Detecting Cognitive Deficits in High-Functioning Stroke Patients
Published on: December 16, 2022
Victoria Tilton-Bolowsky1, Ashley Raman2
1Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD; Department of Biobehavioral Sciences, Teachers College, Columbia University, New York, NY.
Objective:
To evaluate whether measuring circumlocution ability may help identify very mild aphasia (VMA).
Design:
A retrospective, cross-sectional analysis of a subset of data from an observational longitudinal study.
Settings:
Two urban hospitals, research laboratory office rooms, participants' homes, or on Zoom.
Participants:
We recruited stroke survivors and controls via convenience sampling and word-of-mouth. For eligibility, stroke survivors required a formal diagnosis of a primary acute ischemic left cerebrum stroke, normal/corrected-to-normal hearing and vision, and needed to be proficient English speakers. Controls had to have an unremarkable MRI and/or report no history of stroke, traumatic brain injury, or other neurologic conditions. Our sample consisted of 3 groups: 12 stroke survivors with mild to severe aphasia per the Western Aphasia Battery-Revised (WAB-R), 15 stroke survivors without aphasia per the WAB-R, and 15 controls. Ten of the 15 stroke survivors without aphasia per WAB-R had subjective and/or objective evidence of VMA.
Interventions:
Not applicable.
Main Outcome Measures:
The total number of informative features and the average number and type of informative features participants produced per item description during a task created to measure participants' circumlocution ability.
Results:
There were significant differences in the total number of informative features produced between all three groups, F(2, 39)=14.75, P<.001. There was a significant interaction between group and informative feature type, F(5.11, 99.55)=5.76, P<.001. Adding our circumlocution measure to a predictive model accounted for more of the variance in discourse informativeness than the base model with naming ability only, ∆F(1, 22)=15.86, P<.001, ∆Adjusted R2=0.22.
Conclusions:
Measuring circumlocution ability, which can be done using a simple naming task, may help identify VMA. However, larger samples of acute stroke survivors will be needed to evaluate diagnostic potential.
