Working and short-term memory in children with ADHD: an examination of prefrontal cortical functioning using functional Near-Infrared Spectroscopy (fNIRS)

ABSTRACT

Working memory impairments are an oft-reported deficit among children with ADHD, and complementary neuroimaging studies implicate reductions in prefrontal cortex (PFC) structure and function as a neurobiological explanation. Most imaging studies, however, rely on costly, movement-intolerant, and/or invasive methods to examine cortical differences. This is the first study to use a newer neuroimaging tool that overcomes these limitations, functional Near Infrared Spectroscopy (fNIRS), to investigate hypothesized prefrontal differences. Children (aged 8–12) with ADHD (N = 22) and typically developing (N = 18) children completed phonological working memory (PHWM) and short-term memory (PHSTM) tasks. Children with ADHD evinced poorer performance on both tasks, with greater differences observed in PHWM (Hedges’ g = 0.67) relative to PHSTM (g = 0.39). fNIRS revealed reduced hemodynamic response among children with ADHD in the dorsolateral PFC while completing the PHWM task, but not within the anterior or posterior PFC. No between-group fNIRS differences were observed during the PHSTM task. Findings suggest that children with ADHD exhibit an inadequate hemodynamic response in a region of the brain that underlies PHWM abilities. The study also highlights the use of fNIRS as a cost-effective, noninvasive neuroimaging technique to localize/quantify neural activation patterns associated with executive functions.

KEYWORDS:

• Attention-deficit/hyperactivity disorder (ADHD)
• working memory
• short-term memory
• functional near-infrared spectroscopy
• executive function

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ A fourth WM component has also been proposed in recent years. The episodic buffer is purported to bind visuospatial and phonological information prior to central executive processing. However, evidence for its validity and purpose is mixed (A. D. Baddeley & Hitch, 2019) and is particularly weak among school-aged children (Gray et al., 2017).

² Although sophisticated methods for reducing movement artifacts exist, these statistical correction techniques can improve the signal to noise ratio for small-scale movements emitted within an fMRI scanner but cannot address the larger-scale movements associated with improved performance among children with ADHD (Lloyd-Fox et al., 2010).

³ The examiner remained in the assessment room but out of the child’s view to ensure that children did not spin in their chair or engage in other behaviors that could impact optode placement. The experimenter did not interact with the child or provide performance-related feedback during the tasks.

⁴ Although stimuli were presented visually on a computer screen, the task included herein is still considered a phonological task. Read material, such as the letters and numbers that comprise the stimuli during the present task, is orthographically converted to a phonological code in order to extract meaning from visual stimuli (A. Baddeley, 2007). Once encoded, this newly converted phonological information is stored temporarily in the PHSTM subsystem, whereupon the CE processes held information. Therefore, this task places demands on the domain-specific phonological system, rather than the visuospatial system, and should be considered a PHWM task (see L. M. Friedman et al., 2017 for a review).

Additional information

Funding

This study was conducted without external funding.