[Population Modeling] Introduction

[Michael Thomas]

Hi everyone,

My name is Michael Thomas. I'm a professor of cognitive neuroscience at
Birkbeck, University of London, UK. The focus of my research lab is to use
multi-disciplinary methods to understand the brain and cognitive bases of
cognitive variability, including behavioural, brain imaging, computational,
and genetic methods. More details can be found here:
http://www.psyc.bbk.ac.uk/research/DNL/. I'm interested in using population
modelling methods to understand the causes of variations in trajectories of
cognitive development between children, from developmental disorders to
giftedness. Here's a brief overview:

Since the late 1980s, researchers have used computational modelling as a
tool to study cognitive development. Artificial neural networks are a
commonly employed architecture to investigate how children can demonstrate
transitions between qualitatively different stages of behaviour, through
the interaction of experience-dependent learning mechanisms with structured
learning environments. Researchers have usually applied models to specific
mechanisms and behavioural domains. One more recent computational approach
has been to simulate cognitive development in large populations of children
and to include intrinsic (neurocomputational) and extrinsic (environmental)
factors whose interaction produces variability in developmental
trajectories across the whole population.

I have used this approach within several contexts. These include studying
the causes of delay in developmental trajectories of language (Thomas &
Knowland, 2014), investigating how socio-economic status effects may
influence language development (Thomas, Forrester & Ronald, 2013), and
investigating how population-wide individual differences can interact
within pathological mechanisms in neurogenetic disorders such as autism
(Thomas, Knowland & Karmiloff-Smith, 2011; Thomas et al., 2015).

Within a machine learning framework, artificial neural network
architectures can be complemented with genetic algorithms, to stipulate a
genetic level of description. The framework permits a consideration of the
relationship between population variability at multiple levels of
description, including genes, brain structure, brain function, and
behaviour. Examples of this work include simulation of genome-wide
association analyses (Thomas, Forrester & Ronald, 2016), and simulation of
developmental changes in the heritability of intelligence with age and
their relation to differential rates of change of brain structure (Thomas,
in press).

Finally, we have used this multi-scale framework within a purely machine
learning context to evaluate whether it may offer new methods for transfer
learning (Stamate, Magoulas & Thomas, 2015).


Thomas, M. S. C. & Knowland, V. C. P. (2014). Modelling mechanisms of
persisting and resolving delay in language development. Journal of Speech,
Language, and Hearing Research, 57(2), 467-483. doi:
10.1044/2013_JSLHR-L-12-0254
Thomas, M. S. C., Forrester, N. A., & Ronald, A. (2013). Modeling
socioeconomic status effects on language development. Developmental
Psychology, 49(12), 2325-43. doi: 10.1037/a0032301
Thomas, M. S. C., Knowland, V. C. P., & Karmiloff-Smith, A. (2011).
Mechanisms of developmental regression in autism and the broader phenotype:
A neural network modeling approach. Psychological Review, 118(4),
637-654. doi: 10.1037/a0025234
Thomas, M. S. C., Davis, R., Karmiloff-Smith, A., Knowland, V. C. P., &
Charman, T. (2016). The over-pruning hypothesis of autism. Developmental
Science, 9(2), 284-305. doi: 10.1111/desc.12303
Thomas, M. S. C., Forrester, N. A., & Ronald, A. (2016). Multi-scale
modeling of gene-behavior associations in an artificial neural network
model of cognitive development. Cognitive Science, 40(1), 51-99. DOI:
10.1111/cogs.12230.
Thomas, M. S. C. (in press). Do more intelligent brains retain heightened
plasticity for longer in development? A computational investigation.
Developmental Cognitive Neuroscience. doi:10.1016/j.dcn.2016.04.002.
http://www.sciencedirect.com/science/article/pii/S1878929315301043
Stamate, C., Magoulas, G. D., & Thomas, M. S. C. (2015). Transfer learning
approaches for financial applications. In R. Everson, E. Keedwell & D.
Walker (Eds.), Proceedings of the UK Workshop on Computational
Intelligence, University of Exeter 7th - 9th September 2015.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://simtk.org/pipermail/popmodwkgrpimag-news/attachments/20160516/2d5b59c2/attachment.html>