February 17, 2020

expert reaction to study looking at antisocial behaviour and brain structure

A study, published in The Lancet Psychiatry, reports that life-course-persistent antisocial behaviour may be associated with differences in brain structure.

This roundup accompanies an SMC Briefing.

Prof Huw Williams, Associate Professor of Clinical Neuropsychology, University of Exeter, said:

“This is a valuable and insightful contribution to the debate on what drives crime. It is well known that crime tends to spike in the teenage years. And many grow out of delinquent styles of behaviour. There is, clearly, a strong theme that the brain systems for controlling mood and behaviour is somewhat different in those who go on to offend beyond the adolescent period.  The authors note how genetic factors are important in determining the nature of these systems. But also, how early adversity may shape them. This is important for policy makers to take on board. This reinforces the need to help children and young people who have trouble ‘self-regulating’ to get help at the earliest opportunity to reduce risk of escalation of behaviour.  Maybe in schools, to help manage behaviour that could lead to school exclusion.

“In other work there is evidence that Traumatic Brain Injury (TBI) and ADHD are important factors in increased risk of crime. In this study there was an assessment of ‘Head Injury’ (and whether people were hospitalised). It is not clear when and how head injury was assessed, especially as to its severity, during development. Although there was questioning of people about any head injury at age 45 years of age. Brain trauma in childhood and adolescence may have been a factor, therefore, I would have liked to see more fully accounted for. We now know that relatively Mild TBIs can have an effect on young peoples’ brains in the longer term. Especially if repeated. By age 45, people may not recall such events – or see them as serious enough to report. That being said, this is a fascinating, and well conducted study which helps us piece together the neuroscience of crime.”

Dr Graeme Fairchild, Reader in Developmental Psychopathology, University of Bath, said:

“Overall, I commend the authors for a heroic effort – retaining so many members of the Dunedin cohort since birth and linking together prospective longitudinal data on antisocial behaviour across the lifespan with structural brain imaging data at age 45 is a major achievement.  The authors are correct in saying that previous neuroimaging studies of antisocial behaviour have mostly been small in size (n = 40-80), and they have also been cross-sectional (only looked at the study participants at a single time).  This study is much larger, and the participants have been tracked their whole lives, and are very well-characterised in terms of confounding variables.  It is therefore an important contribution to knowledge about the brain basis of antisocial behaviour.

“However, their brain imaging assessment was also cross-sectional and the limitation of collecting brain imaging data at age 45 is that we cannot tell whether these differences in brain structure were present in early life and led to lifelong patterns of antisocial behaviour, or whether they partly or mainly reflect the consequences of lifestyle differences (e.g. drug or alcohol use, smoking, poor diet) between the life course persistent group and the other groups.  It would have been helpful for the authors to show that the results are not entirely explained by lower IQ in the life-course persistent group – perhaps by analysing data from a subset of these individuals who had average IQ. 

“The authors’ comment that the brain areas showing cortical thickness differences in the life-course persistent and adolescence-limited (AL) groups were very different from each other did not seem to be supported by the data.  Some of the effects appear to be almost in the same part of the temporal cortex – one area is slightly further forward in the brain (temporal pole) whereas the ones that were different in the AL group are slightly further back, but still in the temporal lobe.  This isn’t very convincing evidence that completely different brain networks are implicated in each subtype of antisocial behaviour – they actually seem to be quite similar.  This fits with the idea that we have previously put forward, that the childhood-onset and adolescence-onset forms of antisocial behaviour have more in common with each other, than they are different – and that both forms of antisocial behaviour are linked to brain abnormalities.  I would be interested to see what the findings for subcortical areas of the brain show – as the analyses in the present study only investigate the outer layer of the brain, whereas previous studies have shown that subcortical structures such as the amygdala and basal ganglia may also be important in the aetiology of antisocial behaviour. 

“Overall, the findings suggest that the life course persistent form of antisocial behaviour is more strongly related to brain abnormalities, and despite the caveats mentioned above, is more likely to be a neurodevelopmental disorder (similar to ADHD or autism, in that the brain may develop abnormally in early life).”

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“There are some potentially interesting results here, and, as a statistician, I’m generally happy with the statistical approach. What it all means is not easy to interpret, because of limitations in what they were able to study. The problem is that the researchers were able to scan the participants’ brains just once, when they were 45 years old. At that age, they found good statistical evidence of differences in the brain surface area measurements between those classified as displaying antisocial behaviour across much of their lives, and those showing only low antisocial behaviour. The evidence of differences between those displaying antisocial behaviour across their life-course, and those only displaying these behaviours during adolescence, is also pretty strong. (The picture for differences in cortical thickness isn’t quite so clear-cut statistically.)

“But the study can’t tell us from what age those brain differences were there, because the participants’ brains were scanned only at age 45. One possibility is that the differences arose at some time long after early life, perhaps even as a result of factors related to the anti-social behaviours, such as addictions. In that case they can’t be the cause of life-long anti-social behaviour, because they happened after the anti-social behaviours began. Another possibility is that the differences were there right from a very early age for some reason, perhaps genetic. In that case, the different brains might have led to different patterns of anti-social behaviour throughout life. Or they may not have – it remains possible that the brain differences are caused by some other genetic or early-life factors, and that those other factors are what led to the persistent antisocial behaviour, not the brain differences at all. Or the pattern of cause and effect might be some complicated combination of all of these explanations. The researchers make it clear in their report that this can’t be sorted out with brain scans at only one relatively late age.

“It’s true that these research findings are consistent with the hypothesis that life-course-persistent antisocial behaviour arises as a result of abnormal brain development. But observations being consistent with a hypothesis doesn’t mean that the hypothesis must be true, only that it can’t yet be ruled out. The findings are consistent with a lot of other explanations as well, and much more research would be needed to distinguish these.”

Prof Derek Hill, Professor of Medical Imaging, University College London (UCL), said:

“The question being asked by these researchers is: are the brains of people who develop anti-social behaviour as adolescents, and in whom that continues into adulthood different from the brains of people who are never develop anti-social behaviour, or in whom antisocial behaviour is only temporary, and confined to adolescence. If so, then a brain scan can tell the difference between someone who has temporary anti-social behaviour as an adolescent – but who will recover from it – and the individual who will have life-long antisocial behaviour with high risk of incarceration and poor physical and mental health.  

“This research studies the nearly 700 people who have been part of a ‘birth cohort’ of individuals in New Zealand who have been part of a long-term observational research study since they were born. These individuals had brain scans as adults (age 45) that were analysed to measure the size and shape of various brain structures. The authors report a difference in the brains of those with persistent antisocial behaviour – in particular brains that have a smaller surface area and thinner grey matter in certain key regions.

“It is important to note that, while the research involves nearly 700 individuals, only 80 of the people they studied at long term anti-social behaviour and 151 had antisocial behaviour limited to adolescence.  The relatively small number of people in these groups means that there is a reasonable chance that these results are a random finding that cannot be replicated. Furthermore, there are a number of challenges in doing a study of this sort, including difficulty in objectively defining ‘anti-social behaviour’, or other factors that might account for the difference in brains (such as substance abuse being higher in the anti-social behaviour group). These results must therefore, as the authors themselves state, be considered preliminary.

“Furthermore, while the authors found a possible difference between individuals with persistent anti-social behaviour and those without – they haven’t developed a ‘test’ for persistent anti-social behaviour. They also haven’t answered the question as to whether there would be detectable differences in the brains of those with persistent anti-social behaviour it they had their brains scanned as teenagers rather than adults. So, this research is a long way away from providing a way of identifying which anti-social teenagers are likely to be a long-term cost to society, and which will, on their own, live normal adult lives.”

Prof John Stein, Trustee and scientific advisor to the Inst. of Food, Brain and Behaviour, said:

“Carlisi et al’s study is the largest and most persuasive so far to show that people with a 45 year history of antisocial behaviour since childhood, demonstrate significant reductions in the size of the cortical areas particularly involved in sociability than people with no such history.  

“However it still leaves a burning question unsettled.  What causes the reduction?  Most twin and adoption studies concur that heredity accounts for about half of variability between individuals in sociability, so that environmental influences such as childhood abuse or poor nutrition probably play equally important roles.   Because it is potentially reversible, the devastating effects of poor nutrition on the developing brain, particularly lack of the omega 3 long chain fatty acid, docosahexaenoicacid (DHA), should be recognized in future studies.”

 ‘Associations between life-course-persistent antisocial behavior and brain structure in a population-representative longitudinal birth cohort’ by Carlisi et al. was published in The Lancet Psychiatry at 23:30 UK time on Monday 17 February.  

Declared interests

Dr Graeme Fairchild: “No conflicts”

Dr Derek Hill: “No conflicts”

Prof Kevin McConway: “Prof McConway is a member of the SMC Advisory Committee, but his quote above is in his capacity as a professional statistician.”

Prof Huw Williams: “No conflicts”

Prof John Stein: “I am a trustee and scientific advisor to the Inst. of Food, Brain and Behaviour, views are my own, not necessarily those of IFBB.”