REFINEMENT OF A LIMIT-CYCLE OSCILLATOR MODEL OF THE EFFECTS OF LIGHT ON THE HUMAN CIRCADIAN PACEMAKER

In 1990, Kronauer proposed a mathematical model of the effects of ligh t on the human circadian pacemaker. Although this model predicted many general features of the response of the human circadian pacemaker to light exposure, additional data now available enable us to refine the original model. We first refined the original model by incorporating t he results of a dose response curve to light into the model's predicte d relationship between light intensity and the strength of the drive o nto the pacemaker. Data from three bright light phase resetting experi ments were then used to refine the amplitude recovery characteristics of the model. Finally, the model was tested and further refined using data from an extensive phase resetting experiment in which a 3-cycle b right light stimulus was presented against a background of dim light. In order to describe the results of the four resetting experiments, th e following major refinements to the original model were necessary: (i ) the relationship between light intensity (I) and drive onto the pace maker was reduced from I-1/3 to I-0.23 for light levels between 150 an d 10000 lux; (ii) the van der Pol oscillator from the original model w as replaced with a higher-order limit cycle oscillator so that amplitu de recovery is slower near the singularity and faster near the limit c ycle; (iii) a direct effect of light on circadian period (tau(x)) was incorporated into the model such that as I increases, tau(x) decreases , which is in accordance with ''Aschoff's rule''. This refined model g enerates the following testable predictions: it should be difficult to enhance normal circadian amplitude via bright light; near the critica l point of a type 0 phase response curve (PRC) the slope should be ste eper than it is in a type 1 PRC; and circadian period measured during forced desynchrony should be directly affected by ambient light intens ity. (C) 1998 Academic Press.