University of Sussex
Browse
Canalization_paper_revised_12_03_22_resubmitted.pdf (538.3 kB)

Canalization and plasticity in psychopathology

Download (538.3 kB)
journal contribution
posted on 2023-06-10, 05:53 authored by R L Carhart-Harris, S Chandaria, D E Erritzoe, A Gazzaley, M Girn, H Kettner, P A M Mediano, D J Nutt, F E Rosas, L Roseman, C Timmermann, B Weiss, R Zeifman, K J Friston
This theoretical article revives a classical bridging construct, canalization, to describe a new model of a general factor of psychopathology. To achieve this, we have distinguished between two types of plasticity, an early one that we call 'TEMP' for 'Temperature or Entropy Mediated Plasticity', and another, we call 'canalization', which is close to Hebbian plasticity. These two forms of plasticity can be most easily distinguished by their relationship to 'precision' or inverse variance; TEMP relates to increased model variance or decreased precision, whereas the opposite is true for canalization. TEMP also subsumes increased learning rate, (Ising) temperature and entropy. Dictionary definitions of 'plasticity' describe it as the property of being easily shaped or molded; TEMP is the better match for this. Importantly, we propose that 'pathological' phenotypes develop via mechanisms of canalization or increased model precision, as a defensive response to adversity and associated distress or dysphoria. Our model states that canalization entrenches in psychopathology, narrowing the phenotypic state-space as the agent develops expertise in their pathology. We suggest that TEMP - combined with gently guiding psychological support - can counter canalization. We address questions of whether and when canalization is adaptive versus maladaptive, furnish our model with references to basic and human neuroscience, and offer concrete experiments and measures to test its main hypotheses and implications.

History

Publication status

  • Published

File Version

  • Accepted version

Journal

Neuropharmacology

ISSN

0028-3908

Publisher

Elsevier

Volume

226

Page range

e109398 1-14

Event location

England

Department affiliated with

  • Informatics Publications

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2023-01-09

First Compliant Deposit (FCD) Date

2023-01-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC