Attention deficit /hyperactivity disorder (ADHD) is a complex neurodevelopmental disorder which frequently co-occurs with behavioural problems as well as learning disorders. It is strongly associated with underachievement during the school years and frequently has adverse effects on mental health. Three papers in this issue provide new insights into the condition: Martel proposes its re-conceptualisation as a disorder of emotion regulation, Chan et al. investigate ADHD using a cognitive approach and Sonuga-Barke and colleagues focus on the aetiology of conduct problems in ADHD from the perspective of gene-environment interaction. Together these papers underline the need to view ADHD from a multi-disciplinary perspective. More generally, they exemplify the changing face of research in developmental psychopathology: ADHD is a dimensional trait rather than a categorical disorder; like all other disorders of childhood, it needs to be viewed within a developmental perspective; and the impact of ADHD on behaviour and adult functioning depends upon a complex interaction of multiple genes acting through varied environments.

In a comprehensive review, Martel lays out the groundwork for viewing ADHD, not as a problem of cognitive control, but as a problem with the regulation of behaviour. The basic argument is that emotions are not only affected by behaviour but also they regulate behaviour. In short, emotions are involved in modifying internal feeling states including motivation (Eisenberg & Spinrad, 2009). It follows that emotion regulation is crucial to the achievement of personal goals. Related research suggests that the relationships between ADHD, personality and temperament traits need also to be taken account. Overall, children with disruptive behaviour are characterized by negative emotionality and neuroticism, extraversion, and low agreeableness; Martel hypothesizes that these traits may interact with aspects of emotion control to predict behavioural outcomes.

If Martel is right, and ADHD is not well characterized at the cognitive level of explanation, then it is not surprising that attempts to understand clinical symptoms of inattention and distractibility using laboratory methods have been rather unsuccessful. However, in this issue Chan et al. report an interesting new lead from an experimental study of selective attention, cast within the Load Theory of Visual Selection (Lavie, Hirst, de Frocker, & Viding, 2004). The paradigm used is the ‘flanker task’ in which participants have to respond as quickly as they can when they detect a target letter in a display of letters varying in size. At the same time, an irrelevant stimulus (which participants ignore) appears on a proportion of trials in varied locations (to the left, right or central to fixation). Selective attention is assessed by the flanker interference effect (measured by subtracting the reaction time (RT) to identify the letter when the flanker is present from the identification RT when the flanker is absent). Within this theory, if perceptual demands are low (smaller display sizes), attention capacity is available to allow automatic processing of distracting stimuli and hence the flanker interference effect is large. On the other hand, when perceptual demands are high (larger display sizes), this exhausts available attentional resources so that processing of irrelevant stimuli is blocked and hence, the flanker interference effect is small.

With evidence that ADHD is associated with abnormalities in frontal, striatal, and parietal right hemisphere systems as a back-drop, a novel finding of the Chan study is that abnormalities of selective attention in ADHD are a function of the spatial location of targets: specifically, children with ADHD compared to age- and IQ-matched controls, showed left-sided inattention. Although there were some methodological limitations of the study, confidence in the findings is provided by convergent evidence from a re-analysis of data from a similar study by Huang-Pollock, Nigg, & Carr (2005). The authors note that the functional impact of abnormal spatial asymmetries remains unclear; this is an issue to which we will return, for the moment noting that the children with ADHD in this study were significantly impaired in spelling relative to controls.

Sonuga-Barke and colleagues’ paper returns us to the issue of behavioural outcomes in ADHD. This is an important study reporting that gene variations interact with environmental risk and protective factors to determine the co-morbidity of ADHD with conduct and emotional problems. The study was of a large sample of boys with diagnoses of ADHD-combined subtype, recruited from 12 specialist ADHD clinics. The focus was on three of the genes known to be involved in the aetiology of ADHD: the dopamine transporter gene (DAT1), the dopamine receptor genes (DRD4), and the serotonin transporter gene (5HTT). Three gene variants (alleles) were examined for each gene.

It is already well-established that positive emotions expressed by mothers about their children (notably a high degree of warmth) reduce the likelihood of co-morbid conduct problems. To investigate the possibility that genotypic variations moderate sensitivity to maternal expressed emotion (MEE), the investigators collected data on MEE during a one-hour family interview, as well as teacher and parent ratings of conduct and emotional problems. They proceeded to assess the degree of association between gene variants and behavioural outcomes; here we focus on conduct problems in ADHD.

For all 3 genes, children whose mothers expressed positive emotions about them (positive MEE group) showed fewer conduct problems. However, for two of the genes (DRD4 and 5HTT) there were complex interactions such that whether an individual was sensitive to positive MEE depended on genotype. The researchers went on to create an index of ‘insensitivity to positive MEE’. Essentially this indicated, for each individual, whether they possessed 0, 1 or 2 risk alleles. Further analysis revealed that probands with an insensitivity index of 2 (possessing both genotypes) were less likely to benefit from positive MEE. In short, two gene variants appear to moderate sensitivity to MEE; if an individual possesses both of these variants they are at increased risk of conduct disorder.

Together these findings underline the complexities of risk and resilience in the development of ADHD: different genotypes appear to both reduce the negative effects of low positive MEE and the positive effects of high positive MEE. The pattern of findings differs from the synergistic interplay between genetic and environmental factors reported elsewhere in the literature (e.g. Rutter, Moffitt, & Caspi, 2006); in the present study, different genotypes were associated with opposite effects, depending on family environment.

Reflecting on these papers, what are the implications for future research and practice? First, there is an urgent need for longitudinal studies that trace the developmental pathways between genes, through the brain to behavioural outcomes. In this regard, it is important to focus on dimensions of risk and to test alternative causal models (Hulme & Snowling, 2009). Martel proposes three such models underlining the importance of mapping longitudinal trajectories. One specific causal hypothesis is that, while control processes underpin the development of inattentive symptoms, positive and negative emotionality might be linked to the symptom domain of hyperactivity and impulsivity. Arguably, if negative emotionality predisposes a child to hyperactive and impulsive behaviour, and positive maternal emotion can moderate the risk of developing conduct problems, then parenting styles characterized by high degrees of negative expressed emotion may represent a target treatment for some children with ADHD (Sonuga-Barke et al., 2009).

Second, there is pressure to develop multi-causal models of ADHD (and other disorders too). When considering ADHD, it is useful to bear in mind the distinction between ‘cool’ and ‘hot’ executive functions (Zelazo & Muller, 2000; Castellanos, Sonuga-Barke, Millham, & Tannock, 2006). According to this view, cognitive deficits that are observed in ADHD are subserved by different brain circuits from those underlying impairments of motivation and emotion. However, the downstream effects of each can be expected to interact at the level of behaviour to produce considerable variability among affected individuals in their long-term outcome. Future progress ultimately requires a closer integration of these two perspectives. Moreover, from the perspective of reading research, I would like to see the role of educational underachievement assuming a more central position in research studies. The co-morbidity between attention and reading disorders is high (in particular the association between dyslexia and the inattentive subtype) and there is evidence that genetic factors may underpin this association. In general, children with ‘diagnosed’ reading disability will be excluded from research samples, but children with quite weak reading skills may remain, such as those with reading standard scores between 80 and 90 who are likely to be struggling to access the curriculum. Moreover, while it is reasonably well-established that phonological skills and word-level decoding processes are intact in ADHD, less research has looked at reading comprehension or writing processes (both of which require a high degree of executive control).

Why does all this matter? Aside from theoretical concerns, children who have literacy problems are extremely vulnerable in our educational system; clinical experience suggests some may take out their frustration by becoming the class clown and others, who feel demoralised, may display emotional or behavioural difficulties at home. Importantly here, the paper by Fergusson, Boden and Horwood in this issue carries the salutary message that children with situation-specific conduct problems (either at home or at school) are at as a high a risk of adverse adult outcomes as children with generalised conduct disorders. It would be interesting to know whether reading problems contribute to this risk. More generally, practitioners need to consider the impact of a child’s learning disorder on the manifestation and management of the other behaviours associated with ADHD, and the role of parent support in this regard. In the recent Independent Review of Dyslexia for the Department for Children, Schools and Families in England, Rose (2009) advises teachers to identify children who are failing to learn to read and write early and not to await a ‘diagnosis’ of dyslexia. At the same time, the review urges educators to take account of co-occurring difficulties, including those associated with ADHD. To do so has the potential to bear important fruits, not only for children and families but also for research in the field of developmental psychopathology. On the other side of the coin, reading researchers have much to learn about the role of gene-environment interaction in producing variations in reading disorders.

On a final note, in November 2009, this journal will publish its first Virtual Issue: The Changing Concepts of Dyslexia: Nature, Treatment and Co-morbidity.


  1. Top of page
  2. References