Funding: Collaborative Research in Computational Neuroscience (CRCNS)

NSF Directorate for Computer and Information Science and Engineering (CISE)

The National Institute of Neurological Disorders and Stroke

Description. Dr. Tim Kiemel and Dr. Sean Carver worked on the problem of how multisensory information is incorporated into the postural control scheme. Generally, present postural control models are under-constrained, meaning that more than one model can account for the same experimental results. Two types of models were used to solve this problem: descriptive and mechanistic. Descriptive models use well-known mathematical techniques (e.g., ARMA) to to reproduce experimental trajectories, such as changes in center of mass position over time. Developing a descriptive model has the advantage of clearly identifying the dynamic characteristics of postural sway that must be accounted for by any “mechanistic” model. Mechanistic models differ from descriptive models in that they explicitly refer to underlying mechanisms or subsystems (e.g., vestibular) that are known to contribute to postural control. Mechanistic models often incorporate control theory concepts such as optimal control or proportional-integral-derivative (PID) control. A mechanistic model is ultimately the goal because of the link to underlying mechanisms of: 1) Estimation, a process which estimates center of mass position and velocity from sensory information and; 2) Control, a process which controls the center of mass once this estimate is known. However, we argue that before any mechanistic model is considered feasible it must be able to account for the dynamic characteristics identified by a descriptive model. For example, we have tested simple control theory models and found that they do not reproduce the dynamic characteristics of quiet stance. This led us to develop a modification of a optimal-control model that accounts for the characteristics required by our descriptive model.

We argue that these analysis techniques could not only contribute to our basic understanding of sensorimotor integration, but could also prove extremely useful to focus rehabilitative efforts more precisely on the basis of a balance deficit or disorder.


Kiemel T, Elahi A, Jeka J (2008) Identification of the plant for upright stance in humans: multiple movement patterns from a single neural strategy. Journal of Neurophysiology, 100(6), 3394-3406. 

Carver S, Kiemel T, Jeka JJ (2006) Modeling the dynamics of sensory reweighting: Rates depend upon weights. Biological Cybernetics, 95(2), 123-134.

Carver S, Kiemel T, van der Kooij H, Jeka JJ (2005) Comparing internal models of the dynamics of the visual environment. Biological Cybernetics, 92, 147-163.

Kiemel T, Oie KS, Jeka JJ (2002) Multisensory fusion and the stochastic structure of postural sway. Biological Cybernetics:87, 262-277. 

Oie K, Jeka JJ, Kiemel T (2001). Human multisensory fusion: Detecting nonlinearity with small changes in the sensory environment. Neuroscience Letters, 315(3): 113-116.

Oie K, Kiemel T, Jeka JJ (2001) Multisensory re-weighting in response to small amplitude environmental motion.  In J Duysens, BCM Smits-Engelsman, H Kingma (Eds), Control of Posture and Gait, (pp 302-306). 

Kiemel T, Oie K, Jeka JJ (2001) Control theory and the stochastic structure of postural sway. In J Duysens, BCM Smits-Engelsman, H Kingma (Eds), Control of Posture and Gait, (pp 867-870).

Schöner, G.S., Dijkstra, T.M.H., & Jeka, J.J. (1998). Action-perception patterns emerge from coupling and adaptation. Ecological Psychology, 10(3-4), 323-346

Jeka, J.J., Oie, K.S., Schöner, G.S., Dijkstra, T.M.H., & Henson, E. (1998). Position and velocity coupling of postural sway to somatosensory drive. Journal of Neurophysiology. 79, 1661-1674.

Jeka, J.J., Schöner, G.S., Dijkstra, T.M.H., Ribeiro, P. & Lackner, J.R. (1997). Coupling of fingertip somatosensory information to head and body sway. Experimental Brain Research, 113, 475-483.

Keywords: Modeling, multisensory integration, posture.