Biol Cybern DOI 10.1007/s00422-010-0416-4 ORIGINAL PAPER Intermittent control: a computational theory of human control Peter Gawthrop · Ian Loram · Martin Lakie · Henrik Gollee Received: 17 June 2010 / Accepted: 12 October 2010 © Springer-Verlag 2011 Abstract The paradigm of continuous control using inter- nal models has advanced understanding of human motor control. However, this paradigm ignores some aspects of human control, including intermittent feedback, serial bal- listic control, triggered responses and refractory periods. It is shown that event-driven intermittent control provides a framework to explain the behaviour of the human oper- ator under a wider range of conditions than continuous control. Continuous control is included as a special case, but sampling, system matched hold, an intermittent pre- dictor and an event trigger allow serial open-loop trajecto- ries using intermittent feedback. The implementation here may be described as “continuous observation, intermittent action”. Beyond explaining unimodal regulation distribu- tions in common with continuous control, these features naturally explain refractoriness and bimodal stabilisation distributions observed in double stimulus tracking experi- ments and quiet standing, respectively. Moreover, given that human control systems contain significant time delays, a biological-cybernetic rationale favours intermittent over con- P. Gawthrop (B ) · H. Gollee School of Engineering, University of Glasgow, Glasgow G12 8QQ, UK e-mail: Peter.Gawthrop@glasgow.ac.uk H. Gollee e-mail: Henrik.Gollee@glasgow.ac.uk I. Loram Institute for Biomedical Research into Human Movement and Health, Manchester Metropolitan University, John Dalton Building, Oxford Road, Manchester M1 5GD, UK e-mail: I.Loram@mmu.ac.uk M. Lakie School of Sport and Exercise Sciences, The University of Birmingham, Birmingham, Edgbaston B15 2TT, UK e-mail: M.D.Lakie@bham.ac.uk tinuous control: intermittent predictive control is computa- tionally less demanding than continuous predictive control. A standard continuous-time predictive control model of the human operator is used as the underlying design method for an event-driven intermittent controller. It is shown that when event thresholds are small and sampling is regular, the inter- mittent controller can masquerade as the underlying contin- uous-time controller and thus, under these conditions, the continuous-time and intermittent controller cannot be distin- guished. This explains why the intermittent control hypoth- esis is consistent with the continuous control hypothesis for certain experimental conditions. Keywords Intermittent control · Predictive control · Optimal control · Human operator · Human balancing 1 Introduction As discussed by Shadmehr and Wise (2005) and Todorov and Jordan (2002), computational level theories (in the sense of Marr (1982)) try to explain mathematically why a phys- iological system behaves as it does and provide a comput- able algorithm to explain the behaviour. This paper presents a computational-level theory of human control systems in terms of intermittent control which is, in essence, a sequence of open-loop trajectories determined by intermittent feed- back. In particular, it is explained why a continuous-time theory is not sufficient to explain some experiments; equally importantly, is is shown that intermittent control appears to be continuous-time control in some circumstances. The human control system contains time delays. It is shown that inter- mittent control provides a conceptually and computationally simple solution to the control of time-delay systems. 123