Skip to main content

Brain scans find differences in 'badly behaved' teen boys

Thursday 16 June 2016

"'Striking' structural differences seen in study which compared brain scans of young men with antisocial behavioural problems with their healthy peers," The Guardian reports.

The results suggest these behavioural problems could have a neurological dimension.

Researchers used brain scanning techniques to compare the brain structure of groups of male children and adolescents with conduct disorder with matched healthy controls.

Conduct disorder is a type of personality disorder characterised by violent and disruptive behaviours that go way beyond typical naughty childish "acting up" or "teenage rebellion".

The study mapped the thickness of the brain's outer layer, comparing the thickness at different points, both within the groups and between the groups. They found boys who developed conduct disorder before the age of 10 had similarities in overlapping areas of outer brain thickness. This differed from boys without conduct disorder, and those who developed it in adolescence.

The study suggests that changes in brain development may contribute to conduct disorder, but it doesn't tell us the root cause of the condition. Crucially, we don't know what caused the differences in brain structure between the groups, or whether the same results would be found in larger samples.

An important point to bear in mind is that the structures of the brain have a high degree of plasticity, as they can change in response to external factors. So, hopefully, even if there is such a thing as a "conduct disorder" type of brain, this doesn't mean it cannot change.

Where did the story come from?

The study was carried out by researchers from the University of Southampton, University of Cambridge, University of Rome, Martinos Centre for Biomedical Imaging in Boston, Harvard Medical School, Gent University, Columbia University, the University of Bologna and the Medical Research Council.

It was funded by the Wellcome Trust, Medical Research Council, and Southampton and Cambridge Universities.

The study was published in the peer-reviewed Journal of Child Psychology and Psychiatry on an open-access basis, so you can read the paper for free online.

The best coverage was in The Guardian, which explained the techniques used, as well as the results and their limitations. The Daily Mail also gave a good overview.

The Daily Mirror and Daily Telegraph over-simplified the study, saying it had found certain areas in the brains of children with conduct disorder to be thicker, while the picture was more complex than that. The Mirror claimed researchers had "identified the roots of serious anti-social behaviour", which is not the case.

The Mirror's use of the term "delinquents" is also questionable and rather old-fashioned, conjuring up images of mods and rockers fighting on Brighton beach.

What kind of research was this?

This was a case control study, which used brain imaging (specifically MRI scans) to compare the brain structure of children and adolescents aged 13 to 21 with conduct disorder, with a group of the same age and sex (all male).

Case-control studies can show links between factors (such as brain structure and behaviour), but cannot show that one causes another.

What did the research involve?

Researchers recruited 95 boys and young men aged 13 to 21 from pupil referral units and youth offending services, who were interviewed and found to fit the diagnosis of conduct disorder. They also recruited 57 boys and young men of the same age from mainstream schools, without conduct disorder.

All the boys has MRI brain scans. Researchers analysed the scans to look for variations and similarities in the thickness of the outer layer of the brain – the cortex – within and between the groups.

The study was done in two phases, with separate scanners and different groups of participants at Cambridge and Southampton universities, to check that the results of the first phase could be repeated.

People who had been diagnosed with serious mental or physical illness, or with the developmental disorder autism, were not included in the study. As well as comparing brain scans between people with and without conduct disorder, the researchers looked at people who'd had child-onset conduct disorder (before the age of 10) and adult-onset (after age 10).

When carrying out the analysis, they adjusted their figures to take account of the following potential confounders:

  • age
  • IQ
  • overall brain size
  • whether the person also had attention deficit hyperactivity disorder (ADHD)

What were the basic results?

Boys and young men with conduct disorder dating from childhood had a distinct pattern of cortical thickness, showing variations in thickness in all four areas of the cortex, including the frontal, parietal, temporal and occipital cortices.

These patterns were not seen in boys and young men without conduct disorder, or with conduct disorder starting in adolescence. Those with adolescent-onset conduct disorder showed fewer correlations in cortical thickness, compared to those without conduct disorder.

The results held true after adjusting for confounding factors, and were similar in both the Cambridge and Southampton studies, which used different groups of participants.

How did the researchers interpret the results?

The researchers say their results suggest that both child- and adolescent-onset conduct disorder "are associated with changes in the synchronised development of the brain". They say this shows that "neurobiological factors" are important contributors to the development of conduct disorder, whether in childhood or adolescence. They suggest that brain scans might be of use in testing treatments for conduct disorder in the future.

They say their findings are "among the first" to show "marked differences in brain structure" between the child-onset and adolescent-onset forms of conduct disorder, and that this suggests the age at which the disorder begins is important.

They warn that "the biological underpinnings of interregional correlations in cortical thickness are not well understood," so any suggestions about why the brain develops differently in people with conduct disorder are speculative.


This interesting study raises a lot of questions about the way the brain develops in childhood and adolescence, and whether its development is different in those with conduct disorder. However, it doesn't give us answers as to why this might happen.

The results suggest there are differences in the development of these children's brains, which may play a part in their condition. However, as with all observational studies, we can't tell from the study whether these brain differences are the cause of the conduct disorder.

The study also showed that substance abuse and deprivation were more common among boys with child-onset conduct disorder, suggesting they might also play a part.

The study only looked at boys, so we don't know whether the findings would apply to girls with conduct disorder. It's important to be aware that the results only showed areas of overlap between cortical thickness at certain areas of the brain in boys with this disorder, not a defined "map" of brain structure in this condition, so (for example) brain scans could not be used at this stage to diagnose conduct disorder.

Conduct disorder is a challenging condition for parents and schools to manage. Until the causes are better understood, it will be difficult to find useful treatments. Studies such as these are a starting point for finding out more about what causes conduct disorder.