What drives task performance in fluency tasks in people who had COVID-19?

Abstract

Background

Some people who had COVID-19 have been shown to perform below a normative sample on fluency tasks. Unveiling the factors driving performance in this population can explain their underlying impairments. In this article we assess (1) whether people who had COVID-19 are impaired in animal or letter fluency relative to a normative sample; and (2) whether performance (total correct word count) can be explained by demographic factors, common COVID-19 symptoms, number of switches, mean cluster size, and word properties of fluency tasks.

Methods

Eighty-four Spanish-speaking people who had COVID-19 responded to an animal and a letter fluency task, 10–35 days after hospital discharge or self-quarantining. We obtained demographic, common symptom/factors, and calculated number of switches, mean cluster size and eight word properties for each correct word. A comparison of correct words with a normative sample was used to address Aim 1, and Random forests/Conditional inference trees for Aim 2. The last method is not affected by multicollinearity.

Results

People who had COVID-19 were not impaired in fluency tasks compared to normative data collected before the pandemic. Number of switches predicted total number of correct words in both fluency tasks. In addition, frequency, age of acquisition, and familiarity predicted animal fluency scores; and concreteness predicted letter fluency scores. No other measure showed as important.

Conclusion

Number of switches and word properties predicted the performance in fluency tasks of people who had COVID-19. Concreteness was not expected to predict letter fluency and may indicate a specific linguistic pattern in people who had COVID-19.