Volume 3, Issue 25, July 7, 2005

Advancing the Neuroscience of ADHD

Author: Dr. Siegfried Othmer

Biological Psychiatry 57, #11, June 1, 2005

I had hoped to review the latest issue of Biological Psychiatry, focusing on ADHD, for our readers, but the project is too much for one issue of the newsletter. So we will take it in smaller bites.

Joseph Biederman’s overview: ADHD is seen as “multi-factorial” and “clinically heterogeneous.” Predictors of persistence of ADHD into adulthood include: family history, psychiatric comorbidity such as Oppositional-Defiant Disorder and Conduct Disorder, and psychosocial adversity. Important risk factors for ADHD: pregnancy and delivery complications; maternal smoking during pregnancy, and adverse family environment. Risk has been found to increment for every additional indicator of family adversity: severe marital discord, low social class, large family size, paternal criminality, maternal mental disorder, and foster care placement. (This dependency, incidentally, is shared with major depression.) The obvious importance of such environmental variables notwithstanding, overall heritability of ADHD averages some 77% across studies. In fact, familial factors may be considered as predictors of adaptive functioning and emotional health in general, rather than of ADHD specifically.

"…the more overt symptoms of impulsivity/hyperactivity tend to wane early in life, whereas the more covert symptom of inattention tends to persist over time." Even with persistence of the disorder there is a wide dispersion in outcomes with respect to emotional, educational, and social adjustments, with some 20% functioning poorly in all three domains, and some 20% functioning well in all three domains. The role of early treatment in affecting later outcome remains obscure.

Major comorbidities with ADHD of the antisocial disorders, the mood disorders, anxiety disorders, and drug or alcohol dependency imply a complex genetic heterogeneity for ADHD. For ADHD persisting into adulthood, comorbidities for antisocial disorders, mood disorders, and alcohol/drug dependency all fall in the range of 30% for males, with anxiety at 50%. Whereas anxiety and learning disabilities appear to be largely independent of ADHD, the conduct disorders, childhood bipolar disorder, and even depression appear to be more substantially related to ADHD. Learning disorders are comorbid at the 25% level for male children, and that is only fractionally worse than prevalence in the population at large. Oppositional-defiant disorder, on the other hand, is comorbid at the 60% level for males, 30% for females.

Biederman repeats the canard that the Feingold diet has been shown ineffective, when in fact the research showed that it is relevant to small subset of the ADHD population, a subset that undoubtedly regards itself as non-negligible. But it is true that the Feingold hypothesis did not lead to a successful model of ADHD, which was Biederman’s interest here. Of course that also became clear to Feingold himself quite early on.

Dysfunctions in ADHD may be localized mainly to fronto-subcortical pathways, and are predominantly associated with dopaminergic and noradrenergic systems, with the effective medications serving as reuptake inhibitors. The activation these neuromodulator systems is deemed to enhance inhibitory control over sub-cortical nuclei, in particular the caudate, putamen, and globus pallidus. Because of the variation in symptom expression, the neurologic networks associated with ADHD are deemed to be highly heterogeneous. “…it is not yet clear whether the prefrontal anomalies in ADHD are secondary to “lesions” of the pre-frontal cortex or to brain areas with prefrontal projections.”

It is of historical interest indeed that the “lesion” model of ADHD has survived even into a 2005 publication. Bruce Pennington in his commentary also speaks to the lingering objective, operative until recently, of finding a “single underlying core deficit” for not only ADHD but also for dyslexia and the autistic spectrum. “..it is becoming increasingly clear that a single cognitive deficit will not suffice for any of these disorders.”

One such proposed core deficit is “Executive Function” (EF). But no specific EF has been shown to have sensitivity or specificity high enough to support the proposition that EF is the core deficit in all of ADHD. The most discriminating test was found to be the “stop signal reaction time, for which only 50% of ADHD children exhibited a deficit, i.e. they scored below the 10th percentile of controls. Following close behind are CPT reaction time variability, and performance on the Stroop Color Word test. Although nearly 80% of ADHD children show such a deficit on one EF subtest or another, that holds true for 50% of controls. So there is substantial overlap in the distributions of ADHD and controls. Hence EF discriminant tests may exhibit good sensitivity but poor specificity. The hope remains that EF might be useful as a discriminant for a particular subtype of ADHD, with other subtypes being organized around other specific deficits.

Pursuing that line of thought, another subgroup can be identified as a “motivational deficit” subtype, involving perhaps inadequate or delayed reward signaling. Since that is under the management of the dopamine system, it has been referred to as the DA model. Yet another hypothesis involves energetic dysfunction in the regulation of activation of signal processing. This model can be used to explain the increased reaction time variability, particularly at the higher inter-stimulus intervals. This is referred to as the cognitive-energetic or CE model. It remains to be seen how much of ADHD can be understood in terms of a single such failure mechanism or of a combination of the above.

Yet another model that has been proposed focuses on cognitive inhibition, suggesting that an active process must be engaged to organize the contents of working memory. This must be distinguished from behavioral inhibition such as would be tested by the stop reaction time test, as well as from the resistance to interference such as would be measured in the Stroop. Rather, cognitive inhibition involves such things as active forgetting and thought suppression.

I have always liked Pennington’s writing, and once again he does not disappoint: “So, like most other psychopathologies, understanding ADHD requires us first to understand how self-regulation develops normally and then to understand how it goes wrong….”
What is refreshing about such a developmental focus as a point of departure is that it gets out of the bind of assuming causal homogeneity in ADHD. One has the impression that with the increasing refinement of models based on traditional neuropsychological categories the ideal of a single overarching model of ADHD is being eclipsed in practice by a focus on the identification of subtypes that give a better basis for moving forward.

Perhaps it will be only a matter of time before the whole construct of ADHD as a useful clinical entity will survive only in advertisements for stimulant medication. The problem is that whenever we back off to a more encompassing perspective such as “disorders of self-regulation” we hoover up much more than ADHD. So what would such a model look like? We would posit that the core issue in disorders of disregulation lies in the timing and frequency domains. The first nexus of that construct with neuropsychological categories would then be found at the most basic network level, for example Posner’s attentional networks, in terms of which specific EF deficits could be discussed. Other circuits subserve the integration of the affective and motivational domain with cognitive function.

The clinical question then becomes one of whether we have the tools to address the specific deficits that may be identified in a particular individual. The question of whether such deficits can be incorporated in a monolithic model of ADHD then becomes almost irrelevant and hence frivolous. Consider that in this entire Journal issue there was not one paper that addressed the success with which stimulant medication remediated the particular specific deficits that are associated with the various candidate subtypes. The entire project appears to be one of contributing to the understanding of ADHD, of improving the identification and discrimination of ADHD (which is seen as the encompassing entity here by assumption), with the general efficacy of stimulants in relief of ADHD being implicitly understood.

One sees at work here a determined incrementalism, a collective captivity to a historical model that it is impossible for any one contributor to break out of. So I am once again confirmed in the belief that the mainstream is incapable of getting to where we are incrementally. We ourselves would not have gotten there but for the relentless pull of a very effective technique. And even when the technique almost beckons us to a generalized model, we tend to get mired in particularity.
To be continued…

Siegfried Othmer

The Stop Signal Reaction Time (SSRT) Test:

This is a type of continuous performance test in which two stimuli---each calling for a specific response---are presented at regular intervals. Since each calls for a reaction, a tendency to respond develops. A small percentage of the time (say 25%) a signal is given that the response is to be withheld. This signal is given with varying latency with respect to the stimulus presentation. With sufficient data a curve can be obtained which shows the dropoff in response inhibition at some latency. The advantage of this test is that it measures only the time interval related to decision-making and response inhibition, avoiding the added variability from the execution of a motor output.

Dr. Siegfried Othmer