The Role of Executive Functions in Reading Comprehension by Adolescents with ADHD: The Case of Domain-General Versus Domain-Specific Functions

Abstract

The present study explored whether poor executive functions (EF) underlying the difficulties of individuals with ADHD in reading comprehension (RC) are general or specific to the textual content they regulate. To address this question, adolescents with and without ADHD answered questions following the reading of texts and completed a series of tasks that assessed cognitive flexibility, updating and inhibition skills, using (a) non-verbal items, to assess a domain-general EF, (b) single written words, to assess verbal-specific EF, and (c) illustrated scenarios, to assess discourse-specific EF. Findings demonstrated that verbal and/or discourse measurements exhibited significant mediating effects beyond the contribution made by non-verbal measurements. When EF measurements of the same modality were computed as a single composite score, only the discourse measurement accounted for a significant mediating contribution. These findings suggest a domain-specific, or even a discourse-specific role of EF in RC of adolescents with ADHD, and bear implications for educational practice.

Notes

1 We preferred paired over triple mediation analyses to reduce the amount of multicollinearity existing between these highly correlated mediators of the same executive function. A high level of multicollinearity conceals the estimations of mediators’ partial relationships with the outcome variable (Hayes, 2013).

2 The time limit of 1700ms was determined based on a pilot study that explored a response window that was long enough to allow participants to produce a response while still short enough to prevent them from correcting their responses during production.

3 The four texts were used for additional reading goal settings (e.g., entertainment, mind wondering), as part of a larger research project. For this study, we included only the two texts that participants read for the purpose of answering questions after reading.

4 The same finding was also received when using the original Stroop measurement, i.e., the difference in RT between the incongruent and congruent trials.

Additional information

Funding

This work was supported by the Israel Science Foundation (ISF) under Grant no. 1692/19.