Impulsivity is a hallmark of two of the three DSM-IV ADHD subtypes and is associated with myriad adverse outcomes. Limited research, however, is available concerning the mechanisms and processes that contribute to impulsive responding by children with ADHD. The current study tested predictions from two competing models of ADHD--working memory (WM) and behavioral inhibition (BI)--to examine the extent to which ADHD-related impulsive responding was attributable to model-specific mechanisms and processes. Children with ADHD (n = 21) and typically developing children (n = 20) completed laboratory tasks that provided WM (domain-general central executive [CE], phonological/visuospatial storage/rehearsal) and BI indices (stop-signal reaction time [SSRT], stop-signal delay, mean reaction time). These indices were examined as potential mediators of ADHD-related impulsive responding on two objective and diverse laboratory tasks used commonly to assess impulsive responding (CPT: continuous performance test; VMTS: visual match-to-sample). Bias-corrected, bootstrapped mediation analyses revealed that CE processes significantly attenuated between-group impulsivity differences, such that the initial large-magnitude impulsivity differences were no longer significant on either task after accounting for ADHD-related CE deficits. In contrast, SSRT partially mediated ADHD-related impulsive responding on the CPT but not VMTS. This partial attenuation was no longer significant after accounting for shared variance between CE and SSRT; CE continued to attenuate the ADHD-impulsivity relationship after accounting for SSRT. These findings add to the growing literature implicating CE deficits in core ADHD behavioral and functional impairments, and suggest that cognitive interventions targeting CE rather than storage/rehearsal or BI processes may hold greater promise for alleviating ADHD-related impairments.