Algorithmic Human Development

The Algorithmic Human Development group is interdisciplinary, using findings from neuroscience and psychology to study the computational and mathematical aspects of:

  • Cognitive-emotional interaction and maturation in the brain, particularly within the context of experience-dependent early neural development and attachment theory
  • Developmental trajectories resulting from suboptimal early experience, including psychopathologies such as borderline personality and dissociative disorder
  • Psychotherapeutic techniques, in particular the newly introduced Self-Attachment therapy, which aim to increase self-agency and capacities for emotion regulation

News

  • Virtual reality could help treat depression (UCL)
  • Recent study finds evidence for epigenetic inheritance of trauma in Holocaust survivors (Scientific American, Guardian)

Research Themes

Self-Attachment Therapy
Self-attachment is a new, integrative, self-administrable psychotherapy, which has the structure of an algorithm. It uses ideas from attachment theory, developmental neuroscience, fMRI studies of bond-making and neural plasticity along with simulation and imagery techniques in an attempt to promote optimal neural pathways and contain pathological subcortical emotional circuits.

Modelling psychotherapeutic processes
Of general interest are the computational and neural underpinnings of psychotherapeutic techniques that aim to increase self-agency and capacities for emotion regulation, including mentalization and mindfulness.

Transgenerational trauma
We investigate the inter-generational transmission of trauma, including (for example) the long term impact of the Mongol invasions in the Middle East. We are particularly interested in the effect of such transmission on the wider capacity for the emergence of cooperative behaviour within an agent population.




Modelling attachment types and behavioral prototypes in artificial neural networks
We attempt to model the early formation of secure, avoidant, ambivalent and disorganised attachment schemas at the neural level, in circuits involved in implicit memory, emotion, fear, reward, social cognition and habitual behaviour.

Cognitive-emotional interaction
We study the computational aspects of cognitive-emotional interaction and maturation in the brain, with a particular focus on the effects of attachment-based implicit emotional schemas on cognitive processing.

Psychopathology
Insecure early attachment has been linked to a range of psychiatric conditions, including borderline personality and dissociative disorders. We examine the computational mechanisms underlying such disturbances, along with their putative foundations in social cognition.

Publications

Title Publication Authors Date
A Neural Model of Empathic States in Attachment-Based Psychotherapy To Appear in Computational Psychiatry (CPSY) David Cittern, Abbas Edalat 2017
An Immersive Virtual Reality Mobile Platform for Self-Attachment Proceedings of the Society for the Study of Artificial Intelligence and Simulation of Behaviour (AISB) Annual Convention David Cittern, Abbas Edalat, Ibrahim Ghaznavi 2017
Self-Attachment: A Holistic Approach to Computational Psychiatry Computational Neurology and Psychiatry, (Eds.) Peter Erdi, Basabdatta Sen Bhattacharya, Amy Cochran, Springer Series in Bio/Neuroinformatics (ed.: Nikola Kasabov) Abbas Edalat 2016
Introduction to Self-Attachment and its Neural Basis Proceedings of the International Joint Conference on Neural Networks (IJCNN) Abbas Edalat 2015
Towards a Neural Model of Bonding in Self-Attachment Proceedings of the International Joint Conference on Neural Networks (IJCNN) David Cittern, Abbas Edalat 2015
Reinforcement Learning for Nash Equilibrium Generation (Extended version) Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS) David Cittern, Abbas Edalat 2015
An Arousal-Based Neural Model of Infant Attachment Proceedings of the IEEE Symposium Series on Computational Intelligence: Cognitive Algorithms, Mind, and Brain (IEEE SSCI) David Cittern, Abbas Edalat 2014
A Neural Model of Mentalization/Mindfulness based Psychotherapy Proceedings of the International Joint Conference on Neural Networks (IJCNN) Abbas Edalat, Zheng Lin 2014
Capacity of strong attractor patterns to model behavioural and cognitive prototypes (Supplementary material) Proceedings of Advances in Neural Information Processing Systems (NIPS) Abbas Edalat 2013
Strong Attractors of Hopfield Neural Networks to Model Attachment Types and Behavioural Patterns Proceedings of the International Joint Conference on Neural Networks (IJCNN) Abbas Edalat, Federico Mancinelli 2013
Trauma Hypothesis: The enduring legacy of the Mongol Catastrophe on the Political, Social and Scientific History of Iran The Persian version appeared in Bukhara magazine, no. 77-78, January 2011 (1389) pages 227-263 Abbas Edalat 2011

Talks

Title Venue Speaker Date
Self-attachment: an integrative and holistic psychotherapeutic technique 9th International Congress for Psychotherapy (Asian chapter), Tehran Abbas Edalat 18/05/2016
Self-Attachment: proposed method for treating chronic anxiety and depression (in Farsi) Iranian Psychological Association workshop on self-attachment, Tehran Abbas Edalat 03-04/01/2015
Self-Attachment: A New and Integrative Psychotherapy Institute of Psychiatry, King's College London Abbas Edalat 02/05/2013
Introducing Self-Attachment Imperial College Business School Abbas Edalat 10/01/2013

Theses and Reports

Title Type Authors Date
Self-Attachment: A Self-Administrable Intervention for Chronic Anxiety and Depression (Persian version) Department of Computing Technical Report 2017/3 Abbas Edalat 2017
Computational Models of Attachment and Self-Attachment PhD Thesis David Cittern 2017

Members

Prof. Abbas Edalat

Head of Group

ae at ic.ac.uk

Homepage

Ibrahim Ghaznavi

PhD Student

David Cittern

PhD Graduate

david.cittern10 at ic.ac.uk

Zheng Lin

MSc Graduate

Contact Us

Email

ae at ic.ac.uk

Address

Department of Computing
Imperial College London
180 Queen's Gate
London SW7 2AZ