Physiology &Behavior, Vol. 32, pp. 25-30. Copyright ©Pergamon Press Ltd., 1984. Printed in the U.S.A. 0031-9384/84 $3.00 + .00 Memory for Feeding Time: Possible Dependence on Coupled Circadian Oscillators ALAN M. ROSENWASSER, RODNEY J. PELCHAT AND NORMAN T. ADLER Department of Psychology, University of Pennsylvania, Philadelphia, PA 19104 Received 27 May 1983 ROSENWASSER, A. M., R. J. PELCHAT AND N. T. ADLER. Memory for feeding time: Possible dependence on coupled circadian oscillators. PHYSIOL BEHAV 32(1) 25-30, 1984.--Rats maintained on limited-access daily feeding schedules develop food-anticipatory activity rhythms which coexist with the photic circadian activity rhythm. These food anticipatory rhythms aplbear to depend upon a food-entrainable circadian oscillator which is separate and distinct from the light-entrainable circadian oscillator system. This study explored the long-term behavior of the putative food-entrainable oscillator in the presence and in the absence of a feeding schedule, and under light-dark cycles and constant light. The results suggest that a food-entrainable oscillator can show persisting self-sustained oscillations in the absence of a feeding schedule, and that the food- and light-entrainable circadian oscillators may show varying degrees of coupling, depending upon feeding conditions. Such a flexible coupling arrangement may allow the oscillator system to function as a "con- tinuously consulted clock" in the adaptive temporal coordination of behavior with stable and unstable environmental periodicities. Circadian rhythms Wheel running Food restriction Entrainment RICHTER [20] demonstrated that rats fed during a restricted period at the same time each day show "'anticipatory" wheel-running activity preceding each daily access to food. Bolles and coworkers [1,3] later showed that food anticipatory activity rhythms (FARs) depend upon a circa- dian timing system: FARs were seen under 24 hour (circa- dian), but not 19 or 29 hour (non-circadian) feeding schedules. Subsequent work on FARs has attempted to further elucidate the circadian properties of these rhythms, and the nature of the oscillator(s) which underlies them. This work has shown that circadian feeding schedules are capable of synchronizing food anticipatory rhythms in a number of species, and in a variety of behavioral and physiological pa- rameters (cf., [12,27]; also see [4] for a review). Despite the circadian properties of FARs, the oscillator underlying them seems to be both functionally and anatomically distinct from the oscillator(s) responsible for photically influenced activity rhythms. FARs may be ob- served concurrently with light-entrained or free-running "photic" activity rhythms, even when the two rhythms have different periods [5, 7, 8, 9]. In addition, FARs are not dis- rupted by lesions of the suprachiasmatic nuclei (SCN), a putative circadian "master oscillator" for photically influ- enced circadian rhythms (see [21] for a review). Such lesions abolish photic activity rhythms, while sparing FARs, in the same animal [5, 13, 18, 22, 25, 26]. The sparing of FARs by SCN lesions provides the strongest evidence, in the rat, for non-SCN circadian oscillators, and supports the view that overall circadian organization depends upon a multioscillator circadian timing system (see discussion in [14], pp. 185-186). However, like other circadian rhythms, FARs occur only within a limited circadian "range of entrainment" (that is, when the period of the feeding schedule is close to 24 hours) [1, 3, 5, 22, 25]. Furthermore, under certain conditions, FARs appear to be at least somewhat self-sustaining: when a feeding schedule is terminated and the animal is then main- tained under continuous food deprivation, daily increases in activity are seen for five or more days at the projected time of the previously scheduled feeding. On the other hand, when the feeding schedule is replaced by continuous ad lib food access, persistence of FARs is more variable, generally less robust, and damps out more rapidly than under depriva- tion [5,26] (but see [7] for an exception). While even the temporary persistence of FARs in the absence of a feeding schedule provides strong evidence for an endogenous circadian oscillator underlying FARs, it is not yet known whether the temporary persistence and gradual damping of FARs reflects the properties of the underlying oscillator. In other words, the damping of the overt FAR may result from the damping of the oscillator, or may be due to the uncoupling of the overt rhythm from a persisting un- derlying oscillator. Indeed, there are several mechanisms by which such un- coupling may occur. First, the behavioral expression of the ~A preliminary report of these data was presented at the annual meeting of the Eastern Psychological Association, Baltimore, MD, April 1982 (A. M. Rosenwasser and N. T. Adler, Long-term effects of circadian feeding schedules). These studies were supported in part by NIH grant HD 04522 (to NTA), NSF grant BNS 81-20816 (to NTA) and NIH Postdoctoral Fellowship AM 06394 (to AMR). 25