Introduction: Enhancing Care

This chapter focuses on behavioral training, specifically positive reinforcement training (PRT) techniques, as it applies to health care programs for laboratory animals. Training laboratory animals to voluntarily participate in necessary veterinary, husbandry, and research procedures is an important refinement (refinement is one of the 3Rs [Russell and Burch 1959]) that can substantially improve the well-being, welfare, and wellness of captive animals, enhancing the quality and utility of the animals as research models, and therefore the reliability and validity of many types of research (Graham et al. 2012).

In our opinion, one of the primary goals of state-of-the-art, comprehensive training for a health care program is to provide the animals with opportunities to voluntarily participate in their own care (Schapiro et al. 2014). Examples of this might include the animal presenting a leg for venipuncture (Graham et al. 2012), placing its face in the mask of a nebulizer (Gresswell and Goodman 2011), sitting in one place for an acupuncture or laser treatment (Magden et al. 2013, 2016), and choosing which of two medications it prefers when given a choice (Schapiro et al. 2014). In these, and similar, scenarios, the animals are given opportunities to make meaningful choices concerning their health care; they can choose to sit still for acupuncture or they can walk away. One of the strongest indications that this approach is useful for the animals would be if they continued to perform the target behavior (sitting still) in the absence of external reinforcement (food or clicks), because this could suggest that the behavior is being performed to achieve internal reinforcement (the treatments make them feel better), rather than merely for food. To us, this represents the evolution of animal care to the next level, where the animals themselves are determining the efficacy of the treatments they choose to receive.

Training the Animals

In many instances, training is simply another word for teaching. Trainers teach animals to perform particular tasks by providing reinforcement immediately following the performance of the task. Animals learn that they receive reinforcement when they perform the target behavior, altering (increasing) the probability that they will perform that behavior again in the future. Psychologists have been studying learning for many years, in many different ways (Domjan 2015), and for the purposes of this chapter, we focus on one type of learning, operant conditioning, with an emphasis on one type of operant conditioning, PRT. Virtually all animals learn via operant conditioning, whether it is in a formalized laboratory training session (a dog being trained to provide a blood sample or a rat pressing a lever in a Skinner box) or under naturalistic circumstances (a chimpanzee learning to crack palm nuts with stone tools or a blue jay eating a distasteful monarch butterfly). Although punishment is an important factor in discussions of many types of “natural” learning, it has no place in applied operant conditioning for laboratory animals, especially in relation to training for health care purposes.

Training for many health care–related behaviors typically involves at least two processes. The first is the training of the animals to perform the target behaviors. This is usually a fairly straightforward process, as described in the next section. A complementary process, one that is best accomplished concurrently with training, is the desensitization (pairing of a positive reward with a potentially negative situation) of the animals to the personnel, tools, and apparatus that tend to be associated with health care behaviors. These may include people, such as the veterinarian and the veterinary technician, and/or items, such as carts, syringes, lancets, glucometers, vacuum devices, and stethoscopes, which, based on previous experiences, may have negative connotations for the animals.

Additional discussions related to both basic and applied aspects of the psychology of animal (and human) learning can be found in many psychology text books, including Domjan (2015).

Training the Trainers

Just as there are differences in animals’ interest and motivation in participating in training, there are differences in caregivers’ interest and aptitude for training. Patience and consistency are two extremely important traits that are characteristic of good laboratory animal trainers. An effective trainer must utilize PRT techniques properly and be extremely patient, able to deal with the regressions that are typical in most PRT programs. Additionally, trainers must be consistent with the stimuli that they provide for the animals (these are the commands that are given to the subjects) and the reinforcers that they use. In order to maximize the success of the training program, consistency must be maintained within and across trainers. Trainers must also be sensitive to circumstances in which they are potentially being trained by the animals, rather than vice versa (see the example in the “Urine” training section below). While we know a few things about the aspects of temperament that make certain animals good trainees (Coleman et al. 2005), we know considerably less about the aspects of temperament that make certain people good trainers.

In general, training programs seem to work well when (1) personnel whose job titles (and skill sets) identify them as trainers are responsible for training new behaviors and (2) personnel whose job titles include more generalized animal care responsibilities maintain basic behaviors that have already been trained. Such a distribution of effort can be useful for getting the most out of the people involved in the implementation of the training program.

It is important to include veterinary technicians and/or similar workers who may be involved in the performance of the final target behavior in all phases of the training process. In some facilities, veterinary technicians may be viewed negatively by the animals due to the typical animal-related duties they perform (dosing, blood sampling, restraint, etc.). The animals may be less willing to perform the target behaviors in the presence of these “scary” individuals. By involving veterinary technicians in the early, shaping phases of the training process, the vet techs become less scary, ultimately transforming them from negative stimuli to neutral, or even positive stimuli. This is likely to occur as the vet techs participate in more and more neutral and/or positive interactions with the animals and only a small proportion of “negative” interactions.

Brief Review of Studies of the Training Process

As mentioned above, most PRT procedures are based on the stimulus–response–reinforcement contingency. Several publications are available that chronicle many aspects of the training process (Bloomsmith et al. 1994; Gillis et al. 2012). Overall, laboratory animals (primarily, but not exclusively, nonhuman primates) have been successfully trained to voluntarily perform a variety of health care–related behaviors, including moving between enclosures (Veeder et al. 2009), presenting for injections (Perlman et al. 2004), presenting for venipuncture (Lockhard et al. 2013), presenting for capillary blood samples (Reamer et al. 2014), and stationing for acupuncture (Magden et al. 2013). Table 32.1 provides a list of additional examples and citations for health care–related behaviors trained using PRT techniques.

Table 32.1

Some Health Care–Related Behaviors That Have Been Trained Using PRT Techniques.

Health Care Issues Amenable to Training Solutions

Many health care issues are amenable to training solutions (Laule and Whittaker 2007; Whittaker and Laule 2012). In most instances, solutions to complicated health care situations can be derived from appropriate desensitization procedures and the combination of sequences of multiple, fairly basic trained behaviors. For example, an animal can be trained to voluntarily participate in a capillary blood sampling procedure (Reamer et al. 2014) by training (shaping) a sequence of behaviors, including approach, sit, stay, hold, and present a digit.

Targeting, Shifting, and Stationing

Among the most basic of behaviors to be trained for health care purposes are shifting, targeting, and stationing. Shifting (sometimes called gating) refers to training the animals to move from one part of their enclosure to another part on command (Bloomsmith et al. 1998; Veeder et al. 2009). This effectively allows the entire enclosure to be cleaned and works best when the animals live in “suites” of at least two “rooms.” Animals can be shifted to room B when room A is being cleaned and can then return to room A so that room B can be cleaned. When training animals to shift, it is important to remember the stimulus–response–reinforcement contingency and to provide the animals with a verbal stimulus (shift) and then reinforce them (with food, e.g., all or part of their morning ration) when they move as requested. In order for play cages or exercise pens to be valuable components of animal care programs (Griffis et al. 2013), the animals must reliably shift back and forth between the home cage and the play cage.

Training animals to target (Figure 32.1) and station is also a basic requisite for a variety of health care–related tasks. Training subjects to target involves teaching the animal to touch (with hands, feet, nose, etc.) an object (the target), such as a paddle, a stick, or a piece of PVC, when the stimulus (command) is given (Schapiro et al. 2003; Fernstrom et al. 2009). The target can be either stationary or moved to various positions around the enclosure. A target that can be moved allows the trainer to position the animal at a variety of different places within the enclosure and is typically quite helpful when training animals to shift, separate, or enter transport apparatus. Training subjects to station involves teaching the animal to stay in a desired position (typically at its target) (Schapiro et al. 2001) for progressively longer durations until the desired time frame is achieved. Animals that are reliably trained to station for periods ranging from several seconds to many minutes are (1) less likely to interfere with the training of group mates, (2) less likely to act aggressively toward people, (3) typically appropriately positioned for the performance of health care behaviors (presentation of body parts, nebulizer, acupuncture, venipuncture, etc.), and perhaps most importantly, (4) demonstrating that they are willing participants in the procedures that follow. For many of the health care–related behaviors discussed in the sections that follow, it is essential that the subject is trained to station reliably.

Figure 32.1

(See color insert.) Squirrel monkeys “targeting” (touching) the PVC tube held by the trainer.

Diagnosis

Behaviors that have been trained using PRT techniques can play an extremely important role in the diagnosis of health care–related problems. As mentioned above, the relatively straightforward behaviors of stationing and presenting body parts allow veterinarians, technicians, and researchers to closely and frequently observe the animals for wounds or illness. Animals can also be trained for a variety of types of diagnostic sample collections; this is among the most useful training applications for health care purposes. In addition to the attainment of body fluids for diagnostic purposes, animals can be trained to allow unrestrained images (e.g., x-ray) to be taken, and to allow the interrogation of implanted monitoring units.

Monitoring Units

For those animals with acknowledged health issues, monitoring units can be surgically implanted in the animals and regularly interrogated, using the associated “reading” device. For instance, implantable loop recorders, identical to those used in humans, can be used to monitor cardiac patterns in great apes (Lammey et al. 2011; Magden et al. 2016). These devices record and store deviations from normal cardiac function, with the data downloaded approximately every 3 weeks. To download the data, animals are trained to station in a position in relation to the cage mesh that allows the technician to place the reader in close proximity to the implanted recorder (chimpanzees station with their back to the cage mesh [Magden et al. 2016]) (Figure 32.2a and b). Data can typically be transferred in less than 1 minute, leaving the implanted device ready to record data for the next 3 weeks. After remaining stationed for the download, the animal receives reinforcement for having done so and then can resume its normal activity.

Figure 32.2

(a) Chimpanzee with an implantable loop recorder presenting its back so that the loop recorder can be “read.” (b) Implantable loop recorder being read.

Sample Collection

Many different types of samples can be collected from an animal that is (1) trained to station and present body parts at the enclosure interface and (2) desensitized to the collection device or personnel (syringe, blood sleeve, lancet, swab, vet tech, etc.). These include many types of body fluids, hair samples, skin scrapings, body temperatures, and images. The cooperative handling work with macaques of Graham and colleagues (2012) is particularly relevant here.

Body Fluids

Animals can be trained to voluntarily provide a variety of body fluids for health-related purposes. Blood is obviously the most meaningful and relevant fluid to collect, but saliva, mucous (nasal discharge for assessment of MRSA) (Figure 32.3), vaginal fluid, semen, and wound discharge can also be readily obtained from trained animals.

Figure 32.3

Chimpanzee presenting for a voluntary nasal swab.

Saliva

The most common technique for collecting saliva samples is to train animals to chew on an object (dental rope, cotton swab, gauze pad, etc.) and then retrieve the object that has been salivated on from the animals. Saliva samples have been collected from marmosets (Kaplan et al. 2012), chimpanzees (Kutsukake et al. 2009), and rhesus monkeys (Lutz et al. 2000), and have been used to assess levels of cortisol or the presence of several zoonotic pathogens (Evans et al. 2015). Obviously, obtaining saliva samples from dogs involves relatively little training (Borah et al. 2014), given a dog’s propensity to salivate. Pigs also can be trained to provide saliva samples (Decorte et al. 2014).

Urine

Animals (marmosets [Anzenberger and Gossweiler 1993], vervets [Kelley and Bramblett 1981], and chimpanzees [Laule and Desmond 1998; Anestis 2005; Perlman et al. 2010]) can be trained to urinate on command, providing samples for assessment of pregnancy-related hormones, glucose levels, and cortisol levels. Unlike most other applications of PRT techniques, where shaping (reinforcement of successive approximations) is typically used to train the target behavior, training an animal to urinate requires that the trainer “capture” the target behavior. This means that the trainer establishes conditions in which it becomes increasingly likely that the animal will perform the target behavior (providing the animal with a lot of fluid) while repeatedly giving the animal the stimulus (the command “urinate”). When the animal finally does urinate, the reinforcer (a very high-value reinforcer the first time, a jackpot) is immediately provided. While one might think that initially providing the animal with large quantities of fluids might facilitate the training of urination, this can be counterproductive, as the trainer may end up training the animal to “hold it in,” rather than to urinate (the longer they hold it in, the more of the fluid they receive). This is one (of a number of) potential circumstances in which the animals may end up training the trainer, rather than the other way around. Since midstream urine samples are usually the most desirable for analysis, some sort of apparatus (a cup and/or a wand, rather than a pan beneath the cage) (Laule et al. 1996) is typically involved in the training and collection process.

Blood

The value of blood samples as diagnostic tools is well known and multifaceted. While blood samples can be obtained from restrained animals, there are many reasons to acquire voluntary blood samples from unrestrained animals, especially when these samples can be collected at “cage side.” Such samples should provide data that is relatively less likely to be confounded with factors such as (1) the stress associated with manual restraint and (2) the potential for interactions with the anesthetic agents associated with chemical restraint (Schapiro et al. 2005; Lambeth et al. 2006). Voluntary blood samples can be obtained using a variety of techniques. The quantity of blood required often helps determine the blood collection technique. The same animal can be trained for multiple techniques, all of which require that considerable effort be devoted to a systematic maintenance program. Voluntary venipuncture and blood collection is one of the few behaviors that is typically trained and maintained by trainers, rather than by animal care technicians.

Capillary

Small (80–200 μL) voluntary blood samples for the assessment of blood glucose levels and/or for use with the i-STAT system (Abbott) can be collected from animals that are trained to present a finger or toe for sampling (Figure 32.4a and b). The training for this process is straightforward; in fact, many chimpanzees among those that are trained regularly for a variety of other behaviors will perform this behavior the first time they are asked, prior to any specific training for capillary blood sampling (Reamer et al. 2014). Training for a capillary sample typically includes sequentially reinforcing stationed animals for allowing touching of the finger or toe, cleansing with a disinfectant wipe, skin puncture with a lancet, cleansing with another wipe, gentle squeezing, and finally, collection of the blood sample. When a 200 μL sample is required, a vacuum device (Innovac Quick-Draw^®) can be used to prevent clotting and/or hemolysis of the sample. When a single drop of blood is required for use in a glucometer, no vacuum device is needed.

Figure 32.4

(See color insert.) (a) Using the lancet to prick the finger of a chimpanzee for a voluntary capillary blood sample. (b) Voluntary capillary blood sample from the toe of a chimpanzee for blood glucose analysis.

Venous

Voluntary blood samples, for analyses (complete blood counts [CBCs], chemistries, viral analyses, etc.) that require larger volumes of blood, can be collected from animals that are trained to voluntarily “enter” and remain stationary in a device that is specially designed to facilitate unrestrained blood samples (PVC sleeve, fabric sling, etc.) (Coleman et al. 2008; Stracke et al. 2011). This can be particularly important when a subject is ill and an anesthetic episode to collect a blood sample could further compromise the animal’s health. In the case of the sleeve that has been successfully utilized with nonhuman primates (chimpanzees and rhesus macaques [Coleman et al. 2008]) (Figure 32.5), the animals are trained to hold onto the bolt at the end of the sleeve, exposing the cephalic vein for venipuncture through the opening cut in the sleeve. The target behavior of holding onto the bolt serves several purposes in this process: (1) it is a behavior that is incompatible with grabbing the humans, (2) it ideally positions the cephalic vein for venipuncture, (3) it indicates the animal’s willingness to voluntarily provide the sample, and (4) it provides an early warning signal in the event that an animal is going to pull its arm out of the sleeve. The trainer typically touches the animal’s hand during the venipuncture process, not as a method of restraint, but to assess the animal’s positioning, to determine its degree of relaxation, and to provide the animal with contact comfort and reassurance during the process.

Figure 32.5

(See color insert.) Chimpanzee providing a voluntary venous blood sample using the sleeve.

As has been mentioned for other behaviors, the training process for voluntary venipuncture using the sleeve involves both training the target behavior and desensitization of the subject to the apparatus, personnel, and process (the sleeve, the vet tech or veterinarians, the occlusion of the vein, etc.). Once trained, multiple tubes of blood using a Vacutainer^® system and/or a catheter can be quickly and voluntarily obtained from subjects (Lambeth et al. 2005). Importantly, certain CBC, chemistry, and immunological parameters may differ between voluntary and nonvoluntary blood samples obtained from chimpanzees (Lambeth et al. 2005; Schapiro and Lambeth 2007).

Chimpanzees, among other species, can also be trained to voluntarily accept intravenous infusions using the same sleeve and similar training techniques (Pavonetti et al., unpublished data).

Imaging

Animals can be trained to station to facilitate the collection of images for diagnostic and research purposes. X-rays and ultrasounds are the most common diagnostic images collected. Pregnancy has been monitored via unrestrained ultrasound in a variety of zoo animals (e.g., snow leopards [Broder et al. 2008] and bonobos [Drews et al. 2011]). Extremities have been x-rayed in zoo-housed gorillas (Laule, personal communication). Dogs have been desensitized to many of the sensory aspects (noise, vibrations, tight quarters, etc.) associated with MRI scanning and have been trained to station and remain motionless in the scanner, allowing the collection of function magnetic resonance imaging (fMRI) images (Berns et al. 2012, 2013).

Treatments

PRT techniques can facilitate not only diagnostic procedures with laboratory animals, but also treatment procedures. Animals that have been successfully trained to voluntarily target and station, in the absence of restraint, can be treated in a variety of different ways, including topically, orally, via injection, with acupuncture, and/or with medicinal lasers. Preventative medicine procedures can also be performed, such as toothbrushing and skin moisturizing treatments.

There are many conceptual similarities between training animals for diagnostic and treatment behaviors: (1) voluntary participation in these behaviors is extremely important for an animal that is being diagnosed and treated because it is ill, as restraint may cause additional problems for the animal; (2) animals must be trained to perform the targeted behavior and desensitized to the relevant apparatus, personnel, and so forth; and (3) a comprehensive maintenance program must be established in order to maximize the probability that the target behavior will be performed when necessary.

Topical

It is extremely advantageous to have animals that are trained to present eyes, ears, mouths, digits, tails, wounds, and so forth, for topical treatment (disinfectants, sugar slurries, antibiotic creams, eye or ear drops, etc.). Socially living animals may occasionally be wounded by group mates, and individually housed animals may occasionally injure themselves; any captive animal may occasionally contract a minor infection that requires topical treatment. Having animals that are trained to voluntarily present body parts to veterinary personnel for examination and handling can facilitate the treatment of these injuries without the need for anesthesia. This can be particularly important for minor health issues among socially housed animals, as the process of anesthetizing, removing, treating, and reintroducing the animal can often create issues that are worse than the original problem. It would be less than ideal if the removal and treatment of an animal for a minor ear infection resulted in a fight upon the animal’s reintroduction to the group in which the animal lost a finger or toe.

Present for Injection

There are many advantages to training animals to voluntarily present a body area for an injection (Figure 32.6). Regular subcutaneous injections for diabetic animals, intramuscular anesthetic injections for animals prior to physical examinations, vaccinations, and intramuscular antibiotic treatments for ill animals are all important applications of training animals to present for an injection. Although intuitively it may seem otherwise, training an animal to present for injection is relatively straightforward. However, animals often regress (fall below the criterion for successful training) after a real injection (especially of a dissociative anesthetic). The behavior will typically recover (return to the criterion of successful training), a process that can be significantly accelerated by an effective maintenance program (only a very small proportion of training episodes should actually result in a real injection). Typically, when an animal voluntarily presents for an injection, a positive reinforcer (food or juice) can be presented immediately after the injection, establishing a positive context for the process. Obviously, when an anesthetic is injected, food reinforcement is not possible, highlighting the importance of the conditioned reinforcer (clicker or whistle) in the training of this valuable health care–related behavior.

Figure 32.6

Rhesus monkey voluntarily presenting for an injection.

Present for injection is a behavior for which it is extremely important to desensitize the animals to the apparatus and personnel involved. This is especially true in situations where the veterinary technician has previously been involved in relatively few positive interactions with the animals (the presence of the vet tech may reliably indicate to the animal that something “aversive” is about to happen). It is reasonably straightforward for the trainer to train an animal to present for a “mock” injection; however, when the time comes for the real injection to be administered by the vet tech, the animal is likely to become noncooperative. Therefore, it is critical that the vet tech is involved in the early shaping and desensitization stages of present for injection training, so that the animal can learn that the vet tech is not a reliable indicator of an aversive interaction or outcome.

For some species, injectable antibiotic treatments are superior to oral antibiotic medications. If an animal is sick enough to require an antibiotic treatment, then the possibility exists that restraining the animal to provide the treatment may further compromise the animal’s health. If the animal is trained to voluntarily present for antibiotic injections, then a significant reduction in stress can be achieved, increasing the probability of successful treatment and recovery. However, it is important to note that levels of voluntary participation for these behaviors are often lower when animals are sick.

Although not directly health care related, training animals to voluntarily present for an anesthetic injection can influence at least CBC and chemistry values in subsequently collected blood samples (Lambeth et al. 2006). Clinical parameters in venous blood samples obtained when animals voluntarily presented for an injection of anesthetic differed significantly from clinical parameters obtained when animals were nonvoluntarily anesthetized. These findings applied to both within-subject and between-subject analyses of a fairly large data set (Lambeth et al. 2006), suggesting that the process of administering the anesthesia and/or the anesthesia itself could be a potential confounder in experiments that analyze parameters from venous blood samples. Voluntary presentation for anesthetic injections may minimize the confounding effects of these factors.

Nebulizer

Chimpanzees have been trained to accept the use of a nebulizer to treat respiratory problems (Gresswell and Goodman 2011; Haller et al., in preparation) (Figure 32.7). The target behavior that is trained is the placing of the animal’s face (more specifically its mouth and nose) into the nebulizer mask to receive the aerosolized treatment. In essence, the animal makes a “kissy face” to position itself appropriately. Nebulizer treatments can then be administered for up to 20 minutes, several times a day. It is not uncommon for the subject’s social partners to learn the behavior via observation, and even to “request” it (Haller et al., personal observation).

Figure 32.7

Chimpanzee voluntarily placing its face in the nebulizer mask.

Acupuncture

At this point, it is important to mention again that stationing is among the most fundamental trained behaviors that underlie many of the health care–related behaviors mentioned in this chapter. Animals that are trained to position themselves appropriately at a station are demonstrating that they are willing subjects, ready to voluntarily perform any number of health care–related (and other types of) behaviors when the appropriate stimulus is presented.

Chimpanzees have been trained to allow the insertion of acupuncture needles as an adjunct therapy for the symptoms associated with osteoarthritis (Magden et al. 2013, 2016) (Figure 32.8). Animals are initially stationed and desensitized to the acupuncture needles prior to remaining still for the approximately 10-minute sessions. Initially, subjects receive primary (grapes) and conditioned (clicks) reinforcers for remaining stationed, but if the acupuncture is truly relieving their arthritis symptoms, the animals will eventually station and accept treatment without the reinforcement of clicks or grapes. This suggests that the symptom relief experienced as a function of the acupuncture treatment may become the primary reinforcer for the behavior, a potential example of one of the primary goals of any animal training program”“the voluntary participation of the animal in its own care, because the treatment is making the animal feel better. This can be interpreted as an example of self-medication, a concept discussed in more detail below.

Figure 32.8

(See color insert.) Chimpanzee voluntarily presenting for a combination of acupuncture and laser therapy.

Medicinal Laser

Medicinal lasers can be used (alone or in conjunction with acupuncture) in a manner similar to that of acupuncture, as an adjunct therapy for osteoarthritis. Additionally, medicinal lasers can be used to treat injuries and/or infections. Treatment with a medicinal laser requires that the animal station and tolerate the handheld laser and its flashing light (Magden et al. 2016). One advantage of laser treatments compared with acupuncture is that animals typically have to remain stationed for much shorter periods (~2 minutes) for a laser treatment than for acupuncture (~10 minutes). External reinforcers may also become extraneous for animals receiving laser treatments when the internal reinforcement of symptom relief takes place (Schapiro et al., personal observation).

Medications: Choice Procedure

Providing animals with the opportunity to choose between two medications for treatment of an ailment is an innovative approach that can truly put the animal’s health care under its own control, allowing the animal to demonstrate to care personnel which of two treatments it finds preferable. This can effectively open new lines of communication between the animals and those charged with their care and should be the next step in the continuing evolution of animal health care programs.

As an example, arthritic chimpanzees have been allowed to choose between meloxicam and ibuprofen as the treatment for their arthritis symptoms (Schapiro et al. 2014). In a small study that involved a number of necessary control conditions, chimpanzees received ibuprofen daily in colored Gatorade for 2 months, after which they received meloxicam daily in a different color of Gatorade for an identical period. The conditions were then reversed to make it an ABBA design. Finally, after a total of 8 months of being presented with one or the other medication, the animals were presented with both medications simultaneously and allowed to choose the one that they wanted. Although only a small number of subjects were involved in this study, all subjects preferred one medication (meloxicam) over the other, and perhaps more importantly, their behavior varied, depending on the medication they received or chose. In general, subjects engaged in more species-appropriate behavior when they received their preferred medication (Schapiro et al. 2014). A similar choice procedure can be used in many circumstances, for many conditions, which are typically treated by the administration of a compound, allowing for not only choices between medications, but also titration of doses. As mentioned above, such choice procedures (1) allow animals to control how they are treated, (2) open lines of communication between animals and their caregivers that can truly benefit the animals, and (3) represent the next step in the evolution of laboratory animal care.

The fact that chimpanzees self-medicate under natural conditions (Huffman 1997) is one of the reasons that these types of medication choice procedures are potentially useful for this species. If wild chimpanzees are feeling ill, they will search out the specific plant that makes them feel better and consume it and, shortly thereafter, feel better (Huffman 1997). While this may sound fairly straightforward, it represents a complicated chain of thought processes indicative of the complex cognitive abilities of chimpanzees. The choice procedure was designed with this in mind, to provide the animals with a naturalistic opportunity to use their inherent abilities to address the important situations that they are experiencing.

Cost–Benefit Analyses

Many of the benefits associated with training animals to perform health care–related behaviors have been discussed in the preceding sections. However, there are costs associated with training as well. Depending on the behavior (and the animal), a substantial amount of time may need to be invested in initially training an animal to perform a complicated behavior on command. However, overall, an effective PRT program is extremely likely to provide benefits for the people and for the animals that far outweigh these initial costs. Time and money that are invested at the beginning of a training program typically result in substantial time and money savings later in the process, when animals are trained to perform the behavior at the time that it is needed. For example, mangabeys were trained using PRT to shift between rooms in their enclosures for cleaning purposes (Veeder et al. 2009). An initial investment of 26.5 hours in the training process was recouped in just 35 days, as caregivers saved 23 minutes each time they shifted the animals (twice each day). Since those first 35 days, caregivers have experienced a savings of 46 minutes every day for several years.

In addition to straight cost–benefit analyses involving hours or dollars invested compared with hours or dollars saved, there are additional benefits that are likely to accrue as a function of working with trained animals. In certain cases, diagnostic and treatment options may only be available to animals that have been trained using PRT techniques. Specifically, the nebulizer as a treatment for respiratory issues (Gresswell and Goodman 2011) is only practical to use with animals that have been trained to insert their mouth and nose into the nebulizer mask. Anesthetizing the animals for multiple treatments each day would not be practical. Similarly, acupuncture and laser treatments (Magden et al. 2013, 2016) are only practical for animals that will voluntarily station for their 2- to 10-minute sessions.

Further benefits of training laboratory animals include better definition of the animals as biomedical models. Although not strictly a health care issue, the data collected from animals that are trained for health care–related and research-related behaviors may be “superior” to data collected from untrained animals (Lambeth et al. 2006; Graham et al. 2012). One of the easiest ways to think about this involves decreases in interindividual variation in the data obtained from subjects that are voluntarily providing samples. Each animal is trained similarly, and individual differences in stress that may exist when untrained samples are collected are likely to diminish. Voluntary samples are definitely a refinement in the way we care for laboratory animals, and if interindividual variation is significantly diminished, then a reduction in the number of subjects required for a research project may also be achieved.

Emerging Technologies

A number of emerging technologies can be utilized more efficiently with animals that have been trained. Implanted or worn devices, including telemetry systems (Lopez et al. 2014), loop recorders (Lammey et al. 2011; Magden et al. 2016), and activity monitors (Mann et al. 2005), which can be interrogated when animals are trained to station near the “reader,” provide noninvasive techniques for collecting vital health (and research) data. Animals with vascular access ports can be trained to present their ports (Graham et al. 2012) and do not have to be anesthetized for sample collection or compound administration. Animals can be acclimated to wear jackets and undershirts (Kelly et al. 2014; Field et al. 2015), critical components of management systems that involve tethered catheterization. These are all examples of technologies whose utility can be enhanced by training animals to perform the behaviors that facilitate the use of these devices.

Conclusion

The use of PRT techniques to provide animals with opportunities to voluntarily participate in health care–related behaviors is an important component of laboratory animal care programs, especially those that involve nonhuman primates, dogs, and pigs. Building on relatively simple behaviors, animals can be trained to perform fairly complex behaviors that facilitate preventative medicine, as well as the diagnosis and treatment of health issues. Importantly, over time, animals that are initially trained to perform basic behaviors learn to learn, expediting the training of later, more complicated behaviors. Significant benefits accrue to people and animals when resources are invested in training animals to perform behaviors that facilitate health care, including, among numerous other behaviors, presenting for an injection, stationing for an acupuncture session, allowing a capillary blood sample, and urinating on command. Providing laboratory animals with opportunities to participate in their own care is one example of taking captive care to the next level, the goal of all progressive captive management programs.

Acknowledgments

We were supported in part by Cooperative Agreement NIH U42 OD-011197. We thank the animal care staff, the behavioral management team (especially the trainers), and the executive leadership at the Keeling Center for their dedication, commitment, and support for the ongoing enhancement of the care and management programs in the nonhuman primate areas. Kristina Adams of Animal Welfare Information Center provided help with references.

References

• Adams, K. M., Navarro, A. M., Hutchinson, E. K.et al.2004. A canine socialization and training program at the National Institutes of Health. Lab Animal33(1):32–36. [PubMed: 14752529]
• Anestis, S. F.2005. Behavioral style, dominance rank, and urinary cortisol in young chimpanzees (Pan troglodytes). Behaviour142:1245–1268.
• Anzenberger, G., and Gossweiler, H. 1993. How to obtain individual urine samples from undisturbed marmoset families. American Journal of Primatology31:223–230. [PubMed: 31937001]
• Baker, K. C., Bloomsmith, M. A., Neu, K.et al.2010. Positive reinforcement training as enrichment for singly housed rhesus macaques (Macaca mulatta). Animal Welfare19:307–313. [PMC free article: PMC4423822] [PubMed: 25960611]
• Behringer, V., Stevens, J. M. G., Hohmann, G.et al.2014. Testing the effect of medical positive reinforcement training on salivary cortisol levels in bonobos and orangutans. PLoS One9:e108664. [PMC free article: PMC4177400] [PubMed: 25250566]
• Berns, G. S., Brooks, A. M., and Spivak, M. 2012. Functional MRI in awake unrestrained dogs. PLoS One7(5):e38027. [PMC free article: PMC3350478] [PubMed: 22606363]
• Berns, G. S., Brooks, A., and Spivak, M.2013. Replicability and heterogeneity of awake unrestrained canine fMRI responses. PLoS One8(12):e81698. [PMC free article: PMC3852264] [PubMed: 24324719]
• Bliss-Moreau, E., Theil, J. H., and Moadab, G. 2013. Efficient cooperative restraint training with rhesus macaques. Journal of Applied Animal Welfare Science12:98–117. [PMC free article: PMC3692558] [PubMed: 23544752]
• Bloomsmith, M. A., Laule, G. E., Alford, P. L.et al.1994. Using training to moderate chimpanzee aggression during feeding. Zoo Biology13:557–566.
• Bloomsmith, M. A., Stone, A. M., and Laule, G. E.1998. Positive reinforcement training to enhance the voluntary movement of group-housed chimpanzees within their enclosures. Zoo Biology17:333–341.
• Borah, B. M., Halter, T. J., Xie, B.et al.2014. Kinetics of canine dental calculus crystallization: An in vitro study on the influence of inorganic components of canine saliva. Journal of Colloid and Interface Science425:20–22. [PubMed: 24776659]
• Bridges, J. P., Haller, R. L., Buchl, S. J.et al.2015. Establishing a behavioral management program for geriatric chimpanzees. American Journal of Primatology, 77(Suppl. 1):111.
• Broder, J. M., MacFadden, A. J., Cosens, L. M.et al.2008. Use of positive reinforcement conditioning to monitor pregnancy in an unanesthetized snow leopard (Uncia uncia) via transabdominal ultrasound. Zoo Biology27:78–85. [PubMed: 19360606]
• Bunyak, S. C., Harvey, N. C., Rhine, R. J.et al.1982. Venipuncture and vaginal swabbing in an enclosure occupied by a mixed-sex group of stumptailed macaques (Macaca arctoides). American Journal of Primatology2:201–204. [PubMed: 31995904]
• Coleman, K., Pranger, L., Maier, A.et al.2008. Training rhesus macaques for venipuncture using positive reinforcement techniques: A comparison with chimpanzees. Journal of the American Association for Laboratory Animal Science47:37–41. [PMC free article: PMC2652623] [PubMed: 18210997]
• Coleman, K., Tully, L. A., and McMillan, J. L.2005. Temperament correlates with training success in adult rhesus macaques. American Journal of Primatology65:63–71. [PubMed: 15645460]
• Council of Europe. 2010. Directive 2010/63/EU of the European Parliament and of the council of 22 September 2010 on the protection of animals used for scientific purposes. Brussels: Council of Europe. http://register​.consilium​.europa.eu/pdf/en​/10/st06/st06106-re01.en10.pdf.
• Cross, N., Pines, M. K., and Rogers, L. J.2004. Saliva sampling to assess cortisol levels in unrestrained common marmosets and the effect of behavioral stress. American Journal of Primatology62:107–114. [PubMed: 14983468]
• Crouthamel, B., and Sackett, G. 2004. Oral medication administration: Training monkeys to take juice from a syringe. Laboratory Primate Newsletter43:14–15.
• Decorte, I., Van Breedam, W., Van der Stede, Y.et al.2014. Detection of total and PRRSV-specific antibodies in oral fluids collected with different rope types from PRRSV-vaccinated and experimentally infected pigs. BMC Veterinary Research10:134. [PMC free article: PMC4072892] [PubMed: 24938323]
• Domjan, M.2015. The Principles of Learning and Behavior. 7th ed. Boston: Cengage Learning.
• Drews, B., Harmann, L. M., Beehler, L. L.et al.2011. Ultrasonographic monitoring of fetal development in unrestrained bonobos (Pan paniscus) at the Milwaukee County Zoo. Zoo Biology30:241–253. [PubMed: 20073051]
• Evans, T. S., Barry, P. A., Gilardi, K. V.et al.2015. Optimization of a novel non-invasive oral sampling technique for zoonotic pathogen surveillance in nonhuman primates. PLOS Neglected Tropical Diseases9:e0003813. [PMC free article: PMC4457869] [PubMed: 26046911]
• Fernstrom, A. L., Fredlund, H., Spangberg, M.et al.2009. Positive reinforcement training in rhesus macaques”“Training progress as a result of training frequency. American Journal of Primatology71:373–379. [PubMed: 19195008]
• Field, A. E., Jones, C. L., Kelly, R., III, et al.2015. Measurement of fecal corticosterone metabolites as a predictor of the habituation of rhesus macaques (Macaca mulatta) to jacketing. Journal of the American Association for Laboratory Animal Science54:59–65. [PMC free article: PMC4311743] [PubMed: 25651092]
• Gillis, T. E., Janes, A. C., Kaufman, M. J.2012. Positive reinforcement training in squirrel monkeys using clicker training. American Journal of Primatology74:712–720. [PMC free article: PMC3412074] [PubMed: 22553135]
• Graham, M. L., Rieke, E. F., Mutch, L. A.et al.2012. Successful implementation of cooperative handling eliminates the need for restraint in a complex non-human primate disease model. Journal of Medical Primatology41:89–106. [PMC free article: PMC3309152] [PubMed: 22150842]
• Gresswell, C., and Goodman, G. 2011. Case study: Training a chimpanzee (Pan troglodytes) to use a nebulizer to aid the treatment of airsacculitis. Zoo Biology30:570–578. [PubMed: 21506143]
• Griffis, C. M., Martin, A. L., Perlman, J. E.et al.2013. Play caging benefits the behavior of singly housed laboratory rhesus macaques (Macaca mulatta). Journal of the American Association for Laboratory Animal Science52:534–540. [PMC free article: PMC3784657] [PubMed: 24041207]
• Gruen, M. E., Thomson, A. E., Clary, G. P.et al.2013. Conditioning laboratory cats to handling and transport. Lab Animal42:385–389. [PMC free article: PMC4410384] [PubMed: 24051643]
• Hannibal, D., Minier, D., Capitanio, J.et al.2013. Effect of temperament on the behavioral conditioning of individual rhesus monkeys (Macaca mulatta). American Journal of Primatology75(Suppl. 1):66.
• Hopper, L. M., Lambeth, S. P., Schapiro, S. J.et al.2015. The importance of witnessed agency in chimpanzee social learning of tool use. Behavioural Processes112(SI):120–129. [PMC free article: PMC4675325] [PubMed: 25444770]
• Huffman, M. A.1997. Current evidence for self-medication in primates: A multidisciplinary perspective. Yearbook of Physical Anthropology40:171–200.
• Institute of Medicine, Committee on the Use of Chimpanzees in Biomedical and Behavioral Research. 2011. Chimpanzees in Biomedical and Behavioral Research: Assessing the Necessity. Washington, DC: National Academies Press. [PubMed: 22514816]
• Kaplan, G., Pines, M. K., and Rogers, L. J.2012. Stress and stress reduction in common marmosets. Applied Animal Behaviour Science137(SI):175–182.
• Kelley, T. M., and Bramblett, C. A.1981. Urine collection from vervet monkeys by instrumental conditioning. American Journal of Primatology1:93–97. [PubMed: 31995942]
• Kelly, R., III, Carlson, A., and Kern, S. J.2014. Evaluation of the use of primate undershirts as a refinement practice for jacketed rhesus macaques (Macaca mulatta). Journal of the American Association for Laboratory Animal Science53:267–272. [PMC free article: PMC4128564] [PubMed: 24827568]
• Kutsukake, N., Ikeda, K., Honma, S.et al.2009. Validation of salivary cortisol and testosterone assays in chimpanzees by liquid chromatography-tandem mass spectrometry. American Journal of Primatology71:696–706. [PubMed: 19452511]
• Lambeth, S. P., Hau, J., Perlman, J. E.et al.2006. Positive reinforcement training affects hematologic and serum chemistry values in captive chimpanzees (Pan troglodytes). American Journal of Primatology68:245–256. [PubMed: 16477594]
• Lambeth, S. P., Perlman, J. E., Thiele, E.et al.2005. Changes in hematology and blood chemistry parameters in captive chimpanzees (Pan troglodytes) as a function of blood sampling technique: Trained vs. anesthetized samples. American Journal of Primatology66(Suppl. 1):182–183.
• Lambeth, S. P., Schapiro, S. J., Bernacky, B. J.et al.2013. Establishing ”quality of life’ parameters using behavioural guidelines for humane euthanasia of captive non-human primates. Animal Welfare22:429–435. [PMC free article: PMC4260970] [PubMed: 25505822]
• Lammey, M. L., Jackson, R., Ely, J. J.et al.2011. Use of an implantable loop recorder in the investigation of arrhythmias in adult captive chimpanzees (Pan troglodytes). Comparative Medicine61:71–75. [PMC free article: PMC3060426] [PubMed: 21819684]
• Laule, G., Bloomsmith, M. A., and Schapiro, S. J.2003. The use of positive reinforcement training techniques to enhance the care, management, and welfare of primates in the laboratory. Journal of Applied Animal Welfare Science6:163–173. [PubMed: 14612265]
• Laule, G., and Desmond, T.1998. Positive reinforcement as an enrichment strategy. In Second Nature: Environmental Enrichment for Captive Animals, ed. D.Shepherdson, J.Mellen, and M.Hutchins, 302–313. Washington, DC: Smithsonian Institution Press.
• Laule, G., Thurston, R., Alford, P.et al.1996. Training to reliably obtain blood and urine samples from a young diabetic chimpanzee (Pan troglodytes). Zoo Biology15:587–591.
• Laule, G., and Whittaker, M. 2007. Enhancing nonhuman primate care and welfare through the use of positive reinforcement training. Journal of Applied Animal Welfare Science10:31–38. [PubMed: 17484676]
• Lockhard, J., Wilson, K., and Lanman, C.2013. The effects of operant training on blood collection for domestic cats. Applied Animal Behaviour Science143:128–134.
• Lopez, J. E., Vinet-Oliphant, H., Wilder, H. K.2014. Real-time monitoring of disease progression in rhesus macaques infected with Borrelia turicatae by tick bite. Journal of Infectious Diseases210:1639–1648. [PMC free article: PMC4326309] [PubMed: 24879799]
• Luttrell, L., Acker, L., Urben, M.et al.1994. Training a large troop of rhesus macaques to cooperate during catching: Analysis of the time investment. Animal Welfare3:135–140.
• Lutz, C. K., Tiefenbacher, S., Jorgensen, M. J.et al.2000. Techniques for collecting saliva from awake, unrestrained, adult monkeys for cortisol assay. American Journal of Primatology52:93–99. [PubMed: 11051444]
• Magden, E. R., Haller, R., Thiele, E.et al.2013. Acupuncture as an effective adjunct therapy for osteoarthritis in chimpanzees (Pan troglodytes). Journal of the American Association for Laboratory Animal Science52:475–480. [PMC free article: PMC3725933] [PubMed: 23849446]
• Magden, E. R., Sleeper, M. M., Buchl, S. J.et al.2016. Use of an implantable loop recorder in a chimpanzee (Pan troglodytes) to monitor cardiac arrhythmias and assess the effects of acupuncture and laser therapy. Comparative Medicine66:52–58. [PMC free article: PMC4752036] [PubMed: 26884410]
• Mann, T. M., Williams, K. E., Pearce, P. C.et al.2005. A novel method for activity monitoring in small non-human primates. Laboratory Animals39:169–177. [PubMed: 15901360]
• McKinley, J., Buchanan-Smith, H., Bassett, L.et al.2003. Training common marmosets (Callithrix jacchus) to cooperate during routine laboratory animal procedures: Ease of training and time investment. Journal of Applied Animal Welfare Science6:209–220. [PubMed: 14612269]
• Meunier, L. D.2006. Selection, acclimation, training, and preparation of dogs for the research setting. ILAR Journal47:326–347. [PubMed: 16963813]
• Mikkelsen, L. F., Bentsen, H., and Gerlach, B.2003. The restraint platform”“Refinement of long term restraining of dogs for experimental procedures. Scandinavian Journal of Laboratory Animal Science30:217–219.
• NRC (National Research Council). 2011. Guide for the Care and Use of Laboratory Animals. 8th ed. Washington, DC: National Academies Press. [PubMed: 21595115]
• Perlman, J. E., Bloomsmith, M. A., Whittaker, M. A.et al.2012. Implementing positive reinforcement animal training programs at primate laboratories. Applied Animal Behaviour Science137:114–126.
• Perlman, J. E., Bowsher, T. R., Braccini, S. N.et al.2003. Using positive reinforcement training techniques to facilitate the collection of semen in chimpanzees (Pan troglodytes). American Journal of Primatology60(Suppl. 1):77–78.
• Perlman, J. E., Horner, V., Bloomsmith, M. A.et al.2010. Positive reinforcement training, welfare, and chimpanzee social learning. In The Chimpanzee Mind, ed. E. V.Lonsdorf, S. R.Ross, and T.Matsuzawa, 320–331. Chicago: University of Chicago Press.
• Perlman, J. E., Thiele, E., Whittaker, M. A.et al.2004. Training chimpanzees to accept subcutaneous injections using positive reinforcement training techniques. American Journal of Primatology62(Suppl. 1):96.
• Phillippi-Falkenstein, K., and Clarke, M. 1991. Procedure for training corral-living rhesus monkeys for fecal and blood-sample collection. Laboratory Animal Science42:83–85. [PubMed: 1316520]
• Priest, G.1991. Training a diabetic drill (Mandrillus leucophaeus) to accept insulin injections and venipuncture. Laboratory Primate Newsletter30(1):1–4.
• Reamer, L., Malling, R., Thiele, E.et al.2014. The use of positive reinforcement training in managing type 2 diabetes in captive chimpanzees (Pan troglodytes): Factors affecting initial training success. Zoo Biology33:212–220. [PMC free article: PMC5038595] [PubMed: 24706518]
• Reinhardt, V.1991. Training adult male rhesus monkeys to actively cooperate during in- homecage venipuncture. Animal Technology42:11–17.
• Reinhardt, V.1992. Transport-cage training of caged rhesus macaques. Animal Technology43:57–61.
• Reinhardt, V.1997. Training nonhuman primates to cooperate during blood collection: A review. Animal Technology48:55–73.
• Reinhardt, V., and Cowley, D. 1990. In-homecage blood collection from conscious stumptailed macaques. Animal Welfare1:249–255.
• Rogge, J., Sherenco, K., Malling, R.et al.2013. A comparison of positive reinforcement training techniques in owl and squirrel monkeys: Time required to train to reliability. Journal of Applied Animal Welfare Science16:211–220. [PMC free article: PMC4652797] [PubMed: 23795685]
• Russell, W. M. S., and Burch, R. L.1959. The Principles of Humane Experimental Technique. London: Methuen.
• Schapiro, S. J., Bloomsmith, M. A., and Laule, G. 2003. Positive reinforcement training as a technique to alter nonhuman primate behavior: Quantitative assessments of effectiveness. Journal of Applied Animal Welfare Science6:175–187. [PubMed: 14612266]
• Schapiro, S. J., and Lambeth, S. P.2007. Control, choice, and assessments of the value of behavioral management to nonhuman primates in captivity. Journal of Applied Animal Welfare Science10:39–47. [PubMed: 17484677]
• Schapiro, S. J., Perlman, J. E., and Boudreau, B. A.2001. Manipulating the affiliative interactions of group housed rhesus macaques using positive reinforcement training techniques. American Journal of Primatology55:137–149. [PubMed: 11746277]
• Schapiro, S. J., Perlman, J. E., Thiele, E.et al.2005. Training nonhuman primates to perform behaviors useful in biomedical research. Lab Animal34(5):37–42. [PubMed: 15861157]
• Schapiro, S. J., Reamer, L. A., Mareno, M. C.et al.2014. Providing chimpanzees with opportunities to voluntarily participate in their own care: Choice of medications. American Journal of Primatology76(Suppl. 1):73.
• Skinner, B. F.1938. The Behaviour of Organisms: An Experimental Analysis. New York: Appleton-Century-Crofts.
• Smith, T. E., McCallister, J. M., Gordon, S. J.et al. 2004. Quantitative data on training New World primates to urinate. American Journal of Primatology64:83–93. [PubMed: 15356860]
• Stracke, J., Bert, B., Fink, H.et al.2011. Assessment of stress in laboratory beagle dogs constrained by a Pavlov sling. Alternatives to Animal Experimentation28:317–325. [PubMed: 22130486]
• Tornqvist, H., Kujala, M. V., Somppi, S.et al.2013. Visual event-related potentials of dogs: A non-invasive electroencephalography study. Animal Cognition16:973–982. [PubMed: 23572066]
• Turkkan, J. S.1990. New methodology for measuring blood pressure in awake baboons with use of behavioral training techniques. Journal of Medical Primatology19:455–466. [PubMed: 2213854]
• VandeVoort, C. A., Neville, L. E., Tollner, T. L.et al.1993. Noninvasive semen collection from an adult orangutan. Zoo Biology12:257–265.
• Veeder, C. L., Bloomsmith, M. A., McMillan, J. L.et al.2009. Positive reinforcement training to enhance the voluntary movement of group-housed sooty mangabeys (Cercocebus atys atys). Journal of the American Association for Laboratory Animal Science48:192–195. [PMC free article: PMC2679662] [PubMed: 19383217]
• Videan, E. N., Fritz, J., Murphy, J.et al.2005. Training captive chimpanzees to cooperate for an anesthetic injection. Lab Animal34(5):43–48. [PubMed: 15861158]
• Wergard, E.-M., Temrin, H., Forkman, B.et al.2015. Training pair-housed rhesus macaques (Macaca mulatta) using a combination of negative and positive reinforcement. Behavioural Processes113:50–59. [PubMed: 25555746]
• Westlund, K.2014Training is enrichment”“And beyond. Applied Animal Behaviour Science152:1–6.
• Whittaker, M., and Laule, G. 2012. Training techniques to enhance the care and welfare of nonhuman primates. Veterinary Clinics of North America Exotic Animal Practice15:445–454. [PubMed: 22998961]