Using Multi-Voxel Pattern Analysis to investigate subtypes of Conduct Disorder

Lead Research Organisation: University of Birmingham
Department Name: Sch of Biosciences

Abstract

Severe antisocial behaviour in childhood (Conduct Disorder (CD)) is a risk factor for negative life outcomes, such as criminality, unemployment, and poor physical and mental health. Children with CD are a heterogeneous group, and Magnetic Resonance Imaging (MRI) techniques have already been used to characterise distinct CD subtypes. This thesis will use a multivariate pattern classification method (Support Vector Machine (SVM)) to classify subtypes with greater accuracy than can be achieved with univariate analyses, including the use of SVM to combine structural and functional imaging data in the same analysis.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
BB/M01116X/1 01/10/2015 31/03/2024
1644241 Studentship BB/M01116X/1 05/10/2015 30/09/2019 Ruth Pauli
 
Description We investigated differences in parenting, grey matter volume and facial emotion recognition abilities between youths with conduct disorder and high levels of callous-unemotional traits (CD/HCU) versus youths with conduct disorder and low levels of callous-unemotional traits (CD/LCU). Youths with CD/HCU exhibit poor empathy, lack of guilt and remorse and shallow affect relative to those with CD/LCU. CD/HCU is a risk factor for psychopathy in adulthood.

Using univariate statistics and machine learning classifiers, we demonstrated that differences in parenting behaviours can be used to distinguish between youths with CD/HCU, youths with CD/LCU and typically developing youths with accuracies of 58 - 75%. Youths with CD/HCU were characterised by low levels of positive parenting (e.g., lack of involvement) as well as high levels of negative parenting (e.g. inconsistent discipline). Youths with CD/LCU were characterised primarily by high levels of negative parenting.

Using the same techniques, we investigated differences in grey matter volume in CD/HCU versus CD/LCU. Youths with CD/HCU had reduced grey matter volumes in key brain areas relative to typically developing youths, but did not differ significantly from youths with CD/LCU. Areas that distinguished CD/HCU from typical development included the insula, operculum and orbitofrontal cortex. Youths with CD/HCU, youths with CD/LCU and typically developing youths were also distinguished from each other at above-chance levels based on regional differences in grey matter volume, although classifier performance was poorer than when using parenting data as a predictor.

Finally, we investigated differences in facial emotion recognition abilities. Youths with CD/HCU and youths with CD/LCU did not differ in their ability to recognise any of the six basic emotions (anger, disgust, fear, surprise, happiness or sadness), but youths with both subtypes performed significantly more poorly than typically developing youths. Using a machine learning classifier, youths with CD/HCU were distinguished from typically developing youths with 63% accuracy. This compared to 61% accuracy for youths with CD/LCU. However, as at the group level, youths with CD/HCU and youths with CD/LCU could not be distinguished from each other based on emotion recognition abilities.
Exploitation Route I will continue to work on these data after finishing the PhD, including applying for grants based on these findings.

Our findings highlight the importance of considering the parenting environment in conduct disorder, in addition to neurobiological differences. They add to a growing body of literature that can help to inform parenting interventions for youths with CD/HCU.
Sectors Education,Healthcare,Other

 
Description We have engaged with teachers and students at local secondary schools to present our research and provide information on emotional dysfunction in behavioural disorders. I have also participated in a Channel 4 documentary about antisocial behaviour, related to the PhD research. I am applying for further finding based on the results of the PhD research.
First Year Of Impact 2017
Sector Education
Impact Types Cultural

 
Description 'Nice Kids can Spit' event at Birmingham Public Library 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact I presented a lie detector test an event for children and families at Birmingham public library. Several children who attended later participated in our research.
Year(s) Of Engagement Activity 2016
 
Description BBC world service radio interview 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact I was interviewed on BBC world service radio about a paper that I co-authored as part of my PhD.
Year(s) Of Engagement Activity 2019
 
Description Brain Awareness week 2016 and 2019 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact I presented a 'lie detector' test and other outreach activities related to my research at a local science museum for children and families.
Year(s) Of Engagement Activity 2016,2019
 
Description British Neuroscience Festival 2017 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact I organised craft activities for children and families related to our research and to neuroscience more generally.
Year(s) Of Engagement Activity 2017
 
Description What Makes a Murderer Channel 4 documentary 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Media (as a channel to the public)
Results and Impact I participated in a channel 4 documentary about antisocial behaviour, filmed in part at the Centre for Human Brain Health, University of Birmingham.
Year(s) Of Engagement Activity 2019