Compared with age-, sex-, and SES-matched controls, children with ADHD manifested consistent and broad executive dysfunction across a wide variety of measures emphasizing four EF components: Response inhibition, response preparation (including efficiency and consistency of response speed), verbal and spatial working memory, and planning. The present results are consistent with prior studies and meta-analyses of executive control skills in ADHD (Willcutt et al., 2005
). Girls and boys with ADHD showed similar
patterns of deficit on tasks involving working memory and response preparation; however, they manifested different
patterns of executive dysfunction on tasks related to response inhibition and planning. These sex differences in performance cannot be explained by sex differences in comorbidities (which were controlled), ADHD subtypes (which were proportionate among boys and girls), or differential medication use.
These results were not fully consistent with our hypotheses, in which we expected boys with ADHD to be impaired on all components of EF and girls to be impaired only on “higher order” EFs (but not on the more “basic,” earlier-developing EFs). The data suggest that, after controlling for comorbidities and sex differences in ADHD subtype among children ages 8–13, girls with ADHD were impaired (relative to control girls) across all EF components—including basic and higher order skills. In contrast, boys with ADHD were impaired on the more basic functions (response inhibition, response preparation), as well as working memory, but not on planning/shifting. Thus, the outstanding deficit among girls with ADHD in this age range involves strategic planning. This planning deficit was also observed when comparing girls with ADHD directly to age-matched boys with ADHD. Additionally, within response inhibition, girls with ADHD showed elevated motor overflow, while boys with ADHD showed greater impairment during conscious, effortful response inhibition. This dissociation may be related to the pattern of anomalous supplementary motor complex (reduced in boys and girls) and prefrontal (reduced in boys but not girls) development.
While the present findings are consistent with past literature highlighting deficits in response inhibition among children with ADHD (Barkley, 1997
; Willcutt et al., 2005
; Wodka et al., 2007
), they suggest that impaired inhibitory control manifests differently in boys and girls with the disorder. Compared with age- sex-, and SES-matched controls, both girls and boys with ADHD showed deficits in response inhibition; however, among boys, the deficits were consistently observed on tasks requiring the ability to consciously
withhold an incorrect response (i.e., boys, but not girls with ADHD were more impaired than sex-matched controls on go/no-go commission rate and Color Word Interference Inhibition). Conversely, girls (but not boys) with ADHD showed significantly increased motor overflow (relative to sex-matched controls) on a structured motor examination. Of note, this relative
difference in overflow rate (relative to sex-matched controls) between boys and girls with ADHD was driven by sex differences in overflow among controls (i.e., control boys having significantly more overflow than control girls), whereas boys and girls with ADHD did not differ in total overflow.
Three prior studies from our research group have highlighted earlier development of motor skills among girls. Using the PANESS (Denckla, 1985
) in typically developing children ages 7–14, Gidley Larson and colleagues (2007)
reported that girls were faster and more proficient than age-matched boys on timed movements and showed fewer subtle signs, including overflow and dysrhythmia, accompanying gaits and stations. Roeder and colleagues (2008)
examined age-related changes in right (vs. left) hand superiority among 144 right-handed children. While right-hand superiority (right-left gap) decreased significantly with age on four of six timed tasks, two tasks also showed gender differences; thus, girls demonstrated a more rapid “equalization” of speed than boys, suggesting that diminishing right-left speed differences may indicate cerebral and corpus callosum maturation. Cole and colleagues (2008)
examined effects of age and sex on motor performance in children ages 7–15 with and without ADHD. Typically developing girls and boys, and girls with ADHD showed age-related reductions in motor subtle signs, whereas boys with ADHD did not. Together, these findings suggest that girls with and without ADHD may have different trajectories of development than boys—arguing for examination of sex-specific patterns of motor development in children with ADHD. In this regard, the results are consistent with prior studies highlighting the relative superiority of typically developing girls in this age range on tasks of motor skill development (Gidley Larson et al., 2007
; Roeder et al., 2008
). The present findings are in contrast, however, with prior studies showing greater overflow among boys (compared with girls) with ADHD (Cole et al., 2008
), perhaps due to the reduced age range in the present study or to oversampling for subtypes less common for gender to obtain subtype equalization in the present study. The current data, which are more balanced with regard to sex distribution, also contrast with previous findings from our research unit (in a sample of primarily boys with and without ADHD) in which overflow movement was found to predict impairment on response inhibition on tasks in which conscious inhibition of an incorrect response is required (Mostofsky et al., 2003
Unlike the patterns observed on tasks of response inhibition, the observed deficits on tasks of planning/switching among children with ADHD were driven by notably impaired performance among girls with ADHD. That is, girls (but not boys) with ADHD showed robust deficits (compared with sex-matched controls) on tasks of planning/switching. Further, girls with ADHD were also impaired (relative to boys with ADHD) on spontaneous strategy use. This specific (relative) deficit in strategic planning among girls with ADHD may have emerged because we oversampled for girls with ADHD Combined subtype (and for boys with ADHD Inattentive subtype) in order to eliminate the subtype confound in sex comparisons within the participants with ADHD. This emphasis on matching for subtype distribution in boys and girls with ADHD may have also contributed to the striking absence of differences between ADHD subtypes on any
component of EF, and may have contributed to the different pattern of findings than has been observed in previous studies that found girls and boys with ADHD to have similar profiles of executive dysfunction in preteen and teen years (Seidman et al., 2005
Taken together, these findings highlight the need to examine sex as a potential moderating variable when considering executive control “deficits” in children with ADHD. Because boys and girls are at different points in their biological development during the elementary school years, absolute
skill difference between boys and girls with ADHD may not be the most salient variable (Mahone and Wodka, 2008
). Rather, skill deficits are most salient when considered relative
to one's age- and sex-matched peers. Because typically developing girls are more neurobiologically mature than typically developing boys prior to adolescence, girls with ADHD may not need to perform as poorly as boys with ADHD in order to be deficient (relative to their sex-matched peers). This consideration is important for clinicians when making decisions about diagnosis and treatment among girls with ADHD.
The current study also suggests that core elements of executive control (response inhibition, response preparation, and working memory) are affected in children with ADHD, regardless of sex. Both boys and girls with ADHD performed poorly (compared with sex-matched controls) on EF components involving response preparation and working memory, and had similar profiles of dysfunction on the tasks used to assess each component. Overall, girls and boys with ADHD were equally impaired on response inhibition, though their profiles of dysfunction on the individual tasks were different. Research in both girls and boys with ADHD has highlighted deficits in these areas (de Jong et al., 2009
; Rommelse et al., 2008
; Wodka, Mostofsky, et al., 2008
), and problems with response inhibition, working memory, and response preparation appear to be core deficits in both girls and boys with ADHD. Additionally, the failure to identify deficits among boys with ADHD on higher order planning/shifting components of EF may be due, in part, to the age range of our sample (8–13 years). Taking into account the later development among boys, it may be that boys with ADHD will continue to “grow into” their planning/shifting deficits later in their teenage years and beyond, highlighting the need for continued assessment of these skills throughout adolescence. A related question is whether the earlier developing girls with ADHD “grow out” of some of their deficits in later adolescence, or whether these difficulties remain into young adulthood.
Strengths of the current study include multiple measures of different EF components, strict diagnostic procedures for ADHD, and control of subtypes and comorbidities. Nevertheless, these methods of controlling extraneous variance in the sample may also have contributed to the slightly above average IQ level of the sample. The increased IQ range may have attenuated some of the deficits on these performance-based measures of EF (Mahone et al., 2002
). Additionally, the current findings may also be less robust than might be observed among samples of children with ADHD who present with more comorbidities or in larger sample sizes. In particular, the absence of subtype effects in children with ADHD may partially reflect limited power due to sample size. Finally, different patterns of EF deficits in boys and girls with ADHD may be evident when other EF measures are utilized, such as rating scales of adaptive functioning.
The current findings highlight the multidimensional nature of EF and the need to use multiple, diverse methods of assessment when considering the construct. Our approach, which relied on a priori assumptions of these components in determining EF structure, assumed a degree of separability and independence among response inhibition, working memory, planning, and response preparation. In fact, these component skills are likely inter-related, and many tests “load” on several EF factors. For example, in a confirmatory factor analysis, Miyake, Friedman, Rettinger, Shah, & Hegarty, 2001
evaluated the extent to which three components of EF, shifting, working memory, and inhibition, represented unitary or separable constructs. Correlations among the three variables were moderate, suggesting they are separable, but related.
Future research should continue the use of multiple measures of each EF component, allowing for the clarification of latent variables underlying the construct in both boys and girls with ADHD. The use of fMRI may be particularly fruitful in helping to understand how girls with ADHD (who have relatively spared prefrontal cortex development) recruit prefrontal regions for “top-down” control of behavior, including the inhibition of hyperactivity and more conscious response control; and whether this recruitment occurs at the expense of higher-order planning skills, by draining resources away from the cognitive tasks.