A convenient, noninvasive method of continuously monitoring motor activity in humans involves the use of Actiwatch recorders (Philips-Respironics, Bend, OR, USA; part number U198-0301-00). These unobtrusive watch-size recording devices are typically worn on the wrist, and continuously collect activity–rest data, providing information about 24-h motor activity rhythms as well as insights into the quality of sleep. These devices comprise an accelerometer and a 32-Hz microprocessor, which selects peak activity intensity during each second to sum up activity counts per sample. Using a 1-min epoch setting Actiwatches can continuously record data for 45 days, which can then be readily downloaded to a personal computer using a dedicated data reader (Philips-Respironics; part number 198-0150-00). Actiware 5.0 software can then be used to determine the total level of activity in an individual, and importantly, the mean level of activity during specific times of the day (e.g., day time or night time). Equally important, the software can analyze bouts of activities that occur during the night and provide insights into sleep quality based on parameters such as sleep latency, sleep fragmentation index, and number of wake bouts. Note, Actical recorders are also commercially available (Philips-Respironics; part number 198-0210-03). They are very similar to Actiwatch recorders except they average the activity intensity (instead of taking the peak activity intensity) during each second to sum up activity counts per sample. They are used with Actical 2.1 software and are primarily used in studies that focus on energy expenditure studies rather than sleep.

Rhesus macaques, like humans, are diurnal and also show consolidated sleep patterns. Consequently, we have made effective use of the same Actiwatch devices to study 24-h activity–rest cycles in rhesus macaques (12–15). Attaching the Actiwatch to the animal’s wrist is not a viable option and so instead we usually place it inside an aluminum protective case (Philips-Respironics; part number 198-0232-00 M); this case is compatible with the nylon and aluminum primate collars available from Primate Products, Inc. (Immokalee, FL, USA) and can be unobtrusively worn around the animal’s neck. For downloading the stored activity data, the animal is briefly sedated using ketamine (10 mg/kg body weight, i.m.) and the collar/recorder removed. Alternatively, if the animal is trained to enter a small transfer cage, which restricts head movement, then the Actiwatch can be removed without resorting to the use of sedation.

An alternative strategy to using the protective case and collar, which we have also found to be very effective, is to place the Actiwatch directly inside a custom-made small pocket at the rear of a nylon mesh primate vest (Lomir Biomedical, Inc., Malone, NY, USA). These protective vests are worn by our animals when they have been fitted with an indwelling subclavian or jugular vein catheter and connected to a swivel-based remote infusion/sampling system (Lomir Biomedical, Inc.). This minimally invasive long-term blood sampling system (Fig. 9.1), when combined with the Actiwatch recorder, enables 24-h hormone rhythms to be monitored with reference to the light–dark cycle as well as with reference to the animal’s endogenous activity–rest cycle (Fig. 9.2). As already indicated, many hormones show a pronounced 24-h release pattern. Consequently, many serial blood samples are needed to clearly disclose changes in the hormone secretion pattern, and often these samples need to be collected when an individual is asleep. The situation is even more complicated when attempting to disclose plasma profiles in hormones, such as luteinizing hormone (LH), which is released episodically or in a pulsatile manner. For example, the first major endocrine event associated with the onset of puberty in primates is an evening increase in the amplitude of LH pulses. Further subtle differences occur across the menstrual cycle of adults (16), in which the follicular phase is associated with high-frequency low-amplitude pulses whereas the luteal phase is associated with low-frequency high-amplitude pulses (Fig. 9.3). By remotely collecting small (<0.3 ml) blood samples every 10 min for 24 h, it is possible to establish detailed pulsatile LH release profiles for each individual, without perturbing its sleep–wake cycle or behavior.

The actograms generated by the Actiware 5.0 software are typically double-plotted (Fig. 9.2). This means that each line of data represents two consecutive days worth of activity; the second day’s data are duplicated at the start of the next line, etc. This double-plotted form of activity depiction helps with the visualization of the activity rhythm, especially when it shows daily drift (e.g., when it is uncoupled from external environmental cues and is free-running). When humans or monkeys are maintained under fixed photoperiods, such as 12 L:12D (i.e., 12 h of light and 12 h of darkness per day), analysis of the actograms reveals the intensity of total daily activity, as well as activity during the day and activity during the night. Such data are valuable in establishing whether an individual is showing normal activity behavior. For example, using a night-vision camera we have established that rhesus macaques typically wake up a few minutes before the lights come on in the morning at 7:00 h; that is, they anticipate when dawn will occur. They then show a major peak of activity in the late morning, just after their morning feed at ∼8:00 h. Finally, after the lights turn off in the evening (19:00 h), the animals continue to move around for a few minutes but generally close their eyes soon afterward and cease to show major episodes of motor activity. An animal that is sick typically does not show such a biphasic diurnal activity profile; it may show an attenuated level of activity during the daytime or excessive activity at night. Consequently, an additional use for the Actiwatch is to gain insights into the general health of an animal. A way to further refine the nocturnal activity is to perform a sleep analysis. However, it should be emphasized that this component of the Actiware 5.0 software has been validated for human studies only, and so strictly speaking, when applying it to the analysis of nonhuman primate activity one cannot refer to it as “sleep analysis” but simply as “nocturnal activity analysis.” Nevertheless, the algorithm has been designed to provide information about “sleep fragmentation” and “sleep latency,” which are used to establish whether a human subject is experiencing perturbed sleep, or in the case of nonhuman primates, whether the animal is showing unusual activity during its normal nocturnal rest period. Note that the epoch setting of the Actiwatch needs to be set to 1-min bins, or short, in order for this function to work. A practical application of this analysis is that if an individual is not sleeping well because of some pharmacological intervention or illness, then performance in a cognitive task may be impaired – indirectly because of perturbed sleep–wake or activity–rest cycles (14). Note also that the absolute level of activity detected by the Actiwatches may be influenced by whether an animal is caged single or pair-caged. As part of a psychological enrichment strategy, rhesus macaques at ONPRC are typically group housed or pair-caged, which promotes beneficial grooming behavior. On the other hand, when animals are caged individually, as when part of a remote blood sampling study, pair caging is not recommended because the protective catheter tubing from the two animals can become entangled. So instead, we cage the animals individually but install a semi-open partition between adjacent cages; this enables neighboring animals to perform mutual grooming without invading each other’s cage floor space. Animals kept under different housing/caging conditions may show difference in their absolute activity levels, and this needs to be taken into account when comparing data from animals that have been housed differently.