9
Behavioral Studies

The behavior of living organisms is a visible manifestation of activity of the central nervous system. Thus, the study of behavior is a central feature of contemporary neuroscience research in animals. In some studies, the research emphasizes behavior itself, and the primary goal is to characterize behavior and its environmental determinants. In others, the behavior of an animal may be correlated with measurement of brain electric or chemical activity to understand brain mechanisms underlying behavior. Behavioral measures are also used often to detect or measure changes in brain function that may be produced by disease, neural injury, genetic modification, or exposure to various agents and treatments.

The purposes of this chapter are to address several general issues that arise in behavioral studies and to give more detailed consideration to a few specific aspects of neuroscience research in which the measurement of behavior is a central feature.

USE OF APPETITIVE AND AVERSIVE STIMULI

Terminology

Stimuli that can be labeled appetitive (attractive or pleasant) or aversive (noxious or unpleasant) are often used in behavioral research. Such stimuli may include food pellets, sweet or bitter tasting solutions, loud noises, drugs, or electric shock. Because use of such stimuli, especially aversive stimuli, is sometimes a source of concern in behavioral studies, this section begins with a brief discussion of the ways in which behavioral neuroscientists commonly describe



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 123
9 Behavioral Studies The behavior of living organisms is a visible manifestation of activity of the central nervous system. Thus, the study of behavior is a central feature of contemporary neuroscience research in animals. In some studies, the research emphasizes behavior itself, and the primary goal is to characterize behavior and its environmental determinants. In others, the behavior of an animal may be correlated with measurement of brain electric or chemical activity to understand brain mechanisms underlying behavior. Behavioral measures are also used often to detect or measure changes in brain function that may be produced by disease, neural injury, genetic modification, or exposure to various agents and treatments. The purposes of this chapter are to address several general issues that arise in behavioral studies and to give more detailed consideration to a few specific aspects of neuroscience research in which the measurement of behavior is a central feature. USE OF APPETITIVE AND AVERSIVE STIMULI Terminology Stimuli that can be labeled appetitive (attractive or pleasant) or aversive (noxious or unpleasant) are often used in behavioral research. Such stimuli may include food pellets, sweet or bitter tasting solutions, loud noises, drugs, or electric shock. Because use of such stimuli, especially aversive stimuli, is sometimes a source of concern in behavioral studies, this section begins with a brief discussion of the ways in which behavioral neuroscientists commonly describe

OCR for page 123
and categorize these events. In general, appetitive stimuli are ones that an organism will voluntarily make contact with or approach, and aversive stimuli are ones that an organism will try to escape or avoid. Central to those definitions is the idea that the labeling of a stimulus as appetitive or aversive is based on an organism’s behavior, not on physical features of the stimuli themselves. Indeed, the same stimulus may be appetitive in some situations or to some individuals, but aversive in other situations or to other individuals. For example, in certain behavioral procedures, rats and monkeys have been shown to engage repeatedly in behaviors that produce exposure to electric shock, an event commonly assumed to be an aversive (Brown and Cunningham, 1981; Cunningham and Niehus, 1997; Cunningham et al., 1993; Kelleher and Morse, 1968). Thus, under these experimental conditions, electric shock would be labeled an appetitive stimulus, not an aversive one. Similarly counterintuitive examples can be found in the literature on behavioral effects of abused drugs. For example, the same dose of alcohol that produces a conditioned place aversion in rats will produce a conditioned place preference in mice (Cunningham et al., 1993). Moreover, a drug’s ability to produce a conditioned preference may be completely reversed (to conditioned aversion) simply by changing the temporal relationship between drug injection and the associated stimulus (e.g., Cunningham et al., 1997; Fudala and Iwamoto, 1990). It has also been shown that injection of an abused drug may concurrently induce preference for a paired spatial location, but aversion for a paired flavor solution in the same animal (e.g., Reicher and Holman, 1977). All of these examples illustrate that decisions about whether a given stimulus should be considered appetitive or aversive cannot be based solely on its physical properties, but must be informed by expert knowledge of its behavioral effects in various contexts. Importantly, those effects may vary significantly as a function of the species, genotype, sex, and past experience of each animal. In more technical terms, the stimuli under consideration here are often referred to as either reinforcers or punishers, depending on their effects in behavioral procedures in which the response-contingent presentation or removal of a stimulus produces either increase or decrease in the rate of the target response. Stimuli that increase the rate of a contingent behavior are called reinforcers, whereas events that decrease the rate of a contingent behavior are called punishers. Both reinforcement and punishment may involve either the presentation or the removal of a stimulus (Domjan, 1998). Typically, the response-contingent presentation of an appetitive stimulus produces an increase in responding (positive reinforcement) and the response-contingent presentation of an aversive stimulus produces a decrease in responding (positive punishment). In contrast, the response-contingent removal or omission of an aversive stimulus produces an increase in responding (reinforcement based on escape or avoidance), whereas the response-contingent removal or omission of an appetitive stimulus produces a decrease in responding (punishment based on omission training).

OCR for page 123
Although the foregoing definitions indicate that evaluation of behavioral changes produced by response-contingent delivery or removal of such events is critical for applying the labels, the manner in which the stimuli are used in experiments may or may not involve a response contingency. For example, in studies that use instrumental learning or operant conditioning, there will be an explicit, experimenter-defined relationship between some feature of the animal’s behavior (such as whether a lever is pressed) and the delivery (or withholding) of the stimulus. In contrast, studies that use classical or Pavlovian conditioning typically do not involve a response-outcome contingency; rather, the emphasis is typically on the predictive relationship between some other stimulus (such as a light or tone) and delivery of the appetitive or aversive stimulus (Rescorla, 1988). Because brain mechanisms underlying the different types of learning may differ, the decision to present appetitive and aversive events in a response-contingent or response-noncontingent manner should be based on the scientific goals of the study. Rationales for Using Appetitive and Aversive Stimuli Appetitive or aversive stimuli are typically used to motivate an animal to perform a particular behavior. However, the scientific reasons for producing that behavior can vary widely, and the overall purpose of the study will be an important consideration in the selection of the appetitive or aversive stimuli. For example, a considerable body of neuroscience research using appetitive and aversive stimuli focuses on understanding the neurobiology of basic motivational processes, such as those involved in feeding, drinking, foraging, mating, drug addiction, aggression, fear (anxiety and phobias), and the avoidance of pain or discomfort. In such cases, the selection of a particular motivational stimulus (such as salt water or a sexually receptive conspecific) is typically dictated by the specific motivational or behavioral system under study (such as sodium appetite or copulation). In other cases, however, investigators may have more leeway in the selection of motivational stimuli. For example, investigators interested in the general neural mechanisms underlying simple learning (such as classical and operant conditioning), cognition, or memory may be able to use a range of stimuli, both appetitive and aversive, to achieve their scientific aims. Similarly, investigators who simply wish to establish a reliable behavioral baseline for studying motor, sensory, attentional, or perceptual processes or for assessing the effect of various manipulations may also have some flexibility in their choice of motivational stimuli. Relevant considerations might include whether the stimulus has similar effects among species or among individuals within a species. Another consideration is the degree of variability in the efficacy of the stimulus among individuals or of repeated exposures to the stimulus in the same individual. For example, because of rapid satiation, a food rich in calories will be a poor choice as a reinforcer in a procedure that requires the animal to respond repeatedly for

OCR for page 123
food over a period of several hours. Of course, the choice of motivational stimuli in such experiments will also be guided by appropriate consideration of their potential to cause pain or distress. This issue is addressed in more detail in the next section. Animal Care and Use Concerns As in other types of research with laboratory animals, investigators conducting behavioral research must consider the recommendations in the Guide when making decisions about the choice of appetitive and aversive stimuli. In particular, consideration must be given to Principle IV of the US Government Principles (IRAC, 1985): Proper use of animals, including the avoidance or minimization of discomfort, distress and pain when consistent with sound scientific practices, is imperative. Unless the contrary is established, investigators should consider that procedures that cause pain or distress in human beings may cause pain or distress in other animals. Neuroscientists proposing to use appetitive or aversive stimuli should provide a clear and complete description of the characteristics of the stimulation (such as unit amount, concentration or intensity, duration, and total number) and a scientific rationale for their use in their animal-use protocols. Due consideration must be given to the immediate consequences of acute exposure to these stimuli (for example, do they cause more than momentary or slight pain or distress?) and to possible detrimental effects of long-term or repeated exposure (for example, development of dental caries after prolonged exposure to sugared foods or fluids). Consideration must also be given to possible adverse consequences of restricted access to food or fluids that may be required to provide an appropriate motivational state for the appetitive stimulus (see Chapter 3). As noted earlier, selection of the specific motivational stimuli in a task may be influenced by limitations imposed by the recording techniques; for example, an event that produces little or no movement in an animal may be preferred in sensitive physiologic recording procedures. In some situations, choice of a motivational stimulus and its characteristics will be guided by previous research showing that variability in response to it is low, thus reducing the number of animals that must be used in the procedure. It is also possible that the characteristics of the event must be adjusted individually for each animal to maximize its efficacy or minimize its detrimental effects. General strategies used by the IACUC, veterinary staff, and members of the research team to evaluate the choice of appetitive and aversive stimuli should mirror those described in previous chapters. In the case of aversive/punishing stimuli with the potential to cause pain and distress, the evaluation process described in Chapter 2 (“Pain and Distress”) can be used. As noted earlier, generally

OCR for page 123
acceptable levels of noxious stimulation are those that are well tolerated and do not result in maladaptive behaviors. Use of aversive stimuli at intensities or durations that approach or exceed the animal’s pain tolerance level should generally be avoided in behavioral procedures, unless a scientific justification is provided. As discussed previously, it is important to note that the appearance of escape and avoidance behaviors may occur well before the intensity of a stimulus reaches the pain tolerance level. In such cases, these behaviors would be considered appropriate adaptive responses. It is only when the animal’s behavior is dominated by escape-avoidance attempts that the behavior becomes maladaptive, signaling unacceptable levels of pain (NRC, 1992). At first glance, one might assume that avoidance or minimization of discomfort, distress, and pain is more problematic when aversive stimuli are used to motivate behavior than when appetitive stimuli are used. However, that is not necessarily true, especially when one considers that the efficacy of some appetitive foods and fluids depends on the introduction of a restricted schedule of access to food or water (see Chapter 3). Thus, in some situations, an aversive stimulus that does not require prior induction of a “need” state (such as contact with mild electric shock or placement in a pool of water) may actually produce less overall discomfort and distress than the combination of an appetitive stimulus with food or fluid restriction. At the same time, however, one must recognize that detection and measurement of “distress” in animals remains problematic (NRC, 2000) (see “Pain and Distress” in Chapter 2). In some cases, an investigator’s choice of a particular appetitive or aversive stimulus will be determined by scientific reasons. For example, the choice of aversive stimulation such as exposure to electric shock or a predator could be justified by a specific scientific interest in understanding the brain mechanisms underlying behaviors motivated by fear or anxiety. In other situations, however, the scientific question may not directly dictate the choice of one type of stimulus over another. For example, the scientific goals of investigators interested in the neural bases of learning and memory or the mechanisms underlying a specific type of motor behavior might be accomplished by using a broad range of appetitive or aversive events. In situations where the scientific rationale for the choice of a particular motivational stimulus is not compelling or the IACUC is unsure whether one stimulus produces more or less overall discomfort or distress than another (e.g., mild electric shock versus a food pellet combined with food restriction), a useful strategy may be to allow the research to begin using the investigator’s preferred stimulus, but to agree in advance to a joint plan for rigorous monitoring and periodic re-evaluation by the IACUC. If apparent pain or distress is higher than expected or other adverse consequences are noted, stimulus parameters can be refined or the stimulus choice changed with approval by the IACUC. If no problems arise during the monitoring phase, the protocol may continue as originally proposed.

OCR for page 123
An example of the issues that must be considered when evaluating the selection of a behavioral task can be provided by comparing the features of three different procedures commonly used to study spatial learning and memory in rodents: the radial arm maze (Olton and Samuelson, 1976), the Morris water maze (Morris, 1981), and the Barnes circular platform maze (Barnes, 1979). A growing interest in understanding the cognitive decline that accompanies aging together with the recent increase in the number of mouse models carrying genetic mutations thought to affect brain function has encouraged many investigators to use one or more of these tasks to assess “cognitive” function. Although many of the behavioral and brain mechanisms involved in solving these tasks are thought to overlap, the motivational basis for performance differs significantly in each task. For example, the radial arm maze procedure usually involves food or water restriction to motivate animals to seek reinforcers placed at the end of each maze arm, requiring the IACUC to consider the issues discussed previously in Chapter 3 (“Food and Fluid Regulation”). In contrast, the Morris water maze involves immersion in water at or a few degrees above room temperature to motivate animals to swim to a hidden or visible platform. Because exposure to water has the potential for evoking a stress response, the time an animal is in the water should be minimized. Moreover, consideration must be given to drying the animal and providing access to an appropriate heat source (unregulated heating pads and heat lamps should be avoided as they can develop hot spots and cause thermal burns) after water exposure to prevent hypothermia. In the Barnes circular platform maze, the animal is typically placed on a large open platform in a well-illuminated room. The behavior of finding the hole that leads to the darkened escape tunnel located beneath the platform is presumably motivated by the animal’s natural aversion to bright open spaces. Some investigators have suggested that this task produces less stress in rats than tasks involving water immersion or food restriction (e.g., McLay et al., 1998). However, the procedures used in several recent studies suggest that additional aversive stimulation (e.g., intense lights, loud sounds, air stream from an overhead fan) may be required to adequately motivate mice to perform in the Barnes maze (Pompl et al., 1999; Inman-Wood et al., 2000; King and Arendash, 2002; Zhang et al., 2002). Thus, at least in mice, this task has the potential to evoke a stress response that may be similar to or greater than that evoked by the other two tasks. In the case of lesioned or genetically modified animals, the choice of task may be further complicated by sensory-motor impairments that could increase the likelihood of stress or serious injury (e.g., drowning in the water maze, falling off the edge of a Barnes maze). As suggested above, when there is uncertainty about which task produces the least amount of discomfort or distress while still meeting the investigator’s scientific goals, the IACUC’s best strategy may be to work with the investigator to develop a thorough plan for monitoring the impact

OCR for page 123
of the procedure in conjunction with frequent re-evaluations by the IACUC until the consequences of the procedure are better understood and shown to be acceptable. BEHAVIORAL SCREENING TESTS Behavioral screening tests are used in pharmacology, genetics, and health surveillance (health surveillance through evaluation of animal behavior is discussed in “Using Animal Behavior to Monitor Animal Health” in Chapter 2). Behavioral screening tests differ from hypothesis-driven experiments in that screening tests assess multiple behavioral measures because there is little information to indicate what important effects might be observed. Screening also is used if limitation of time and resources require a test that can be administered quickly to a number of animals. Screening tests are usually directed at broad functional domains, such as motor coordination, emotion, or sensory functions. Neurobehavioral screens were developed more than 25 years ago to study pharmaceutical and chemical agents (Kulig et al., 1996; Moser, 2000b; Ross et al., 1998). Similar methods are used to screen for genetic mutants (Crabbe et al., 1999b; Sarter et al., 1992a,b; Warburton, 2002). In reviewing the history of behavioral screening, Warburton (2002) considers the advantages of quantitative methods versus the simplicity of observational screening methods. Observational methods are especially appealing to those with little experience in behavioral science, who may not focus on the possible limitations of observational methods: subjective interpretation, higher variability of baseline behavior, and observational variation among observers. Screening results are most useful if one can demonstrate between-observer reliability, establish standardized protocols, and validate the screen with “gold-standard” procedures. Behavioral Screening in Pharmacology and Toxicology Unlike research protocols for pharmacology and toxicology, drug screening usually implies that the effects of the test compound are not well known. Screening studies can be justified by the need to detect a chemical’s ability to cause health problems in humans or animals (such as the abuse liability of a new pharmaceutical or the neurotoxicity of an industrial product) or to determine whether its effects warrant more detailed investigations of its potential as a treatment for behavioral or neurologic disorders. Screening tests also are used when a drug’s pharmacokinetics are not well known and observations are required over an unknown time to determine whether an organism’s response to the drug changes during chronic exposure and whether such exposure can lead to physical dependence. The IACUC must be aware that regulatory agencies (such as the Environmental Protection Agency or the Food and Drug Administration) some-

OCR for page 123
times require investigators to use specific test methods and experimental designs (Weisenburger, 2001). Behavior has proved to be a convenient experimental variable in screening because it is noninvasive and because alterations in many physiologic systems can be reflected in changes in behavior. The functional observational battery (FOB; see Table 9-1) (Moser, 2000a) is a systematic neurologic examination for rodents involving a neurologic examination with numerous behavioral measures. It provides more extensive behavioral measures than the mouse ethogram discussed in Chapter 3 (“Genetically Modified Animals”). The FOB procedure has also been adapted for use with weanling rats (see Table 9-2) (Bushnell et al., 2002; Moser, 2000a). Scoring of the FOB is semiquantitative, and the FOB should be administered and scored by an experienced technician. When a skilled technician is not available, or when handling the animals might be dangerous to animals or staff, observation of an animal in its home cage can be useful, particularly if a quantitative rating scale is used to document the appearance of abnormal behaviors. Better quantification is obtained with commercially available equipment, such as photocell arrays, than through direct observation. The equipment is placed outside a rodent’s home cage to measure activity, such as locomotion and rearing, and this avoids the necessity of handling the animal and the possibility that handling may cause changes in the animal’s behavior (Evans et al., 1986; Lessenich et al., 2001). Additional methods used in screening for neurotoxicity are reviewed by Weisenburger (2001). Screening methods for nonhuman primates can be considered along a scale of intrusiveness into the nonhuman primate’s living space. Nonintrusive procedures are used to minimize handling the animals. Behavioral activity level, diurnal rhythms, etc., can be monitored with photocell arrays surrounding the home cage (Evans et al., 1989). A food-pickup test can also be used while a primate remains in its home cage. Small pieces of food (such as raisins and peanuts) are systematically placed on a tray and then moved to within the nonhuman primate’s reach. The observer measures the time taken to extract the food and the accuracy in terms of the number of attempts required to retrieve all of it (Evans et al., 1989; Merigan et al., 1982). That provides evidence of visuomotor coordination and appetite. Finally, if the experiment permits removing the nonhuman primate from its home cage to a special test apparatus (see “Restraint” in Chapter 3), video cameras can be used to remotely monitor nonhuman primates while they are in the special test apparatus (Ro et al., 1998). A battery of operant conditioning techniques have also been employed to assess neurologic changes caused by a drug or chemical in nonhuman primates (Schulze et al., 1988). This operant screening test is called the Operant Test Battery (OTB; see Table 9-3), and was developed at the National Center for Toxicological Research (NCTR). Additional studies have shown that the OTB can also be used to assess neurological effects in humans and rats (Paule, 2000, 2001).

OCR for page 123
TABLE 9-1 Functional Observational Battery for Adult Rats Endpoint Measurement/Scale Pupil Response present/absent Abnormal Body Posture present/absent Piloerection present/absent Forelimb and Hindlimb Grip Strength kg of force Landing Foot Splay cm/mm Body Weight g Body Temperature °C Open-Field Rearing number Gait Score (also description of gait) 1 to 5 Ataxia Score 1 to 5 Aerial Righting Response 1 to 4 Home-Cage Activity 1 to 5 Open-Field Activity 1 to 6 Arousal 1 to 5 Ease of Removal 1 to 5 Handling Reactivity 1 to 4 Tremorigenic Score 1 to 4 Salivation 1 to 3 Lacrimation 1 to 3 Urination/Defecation 1 to 5 Tail-pinch Response 1 to 5 Click Response 1 to 5 Touch Response 1 to 5 Approach Response 1 to 5   SOURCE: Moser, 2000b. TABLE 9-2 Functional Observational Battery for Pre- and Post-weanling Rats Endpoint Measurement/Scale Body Weight g Open-Field Rearing number Gait Score (and description of gait) 1 to 3 Forelimb Grabbing 1 to 4 Surface Righting Response 1 to 4 Open-Field Activity 1 to 6 Arousal 1 to 5 Handling Reactivity 1 to 4 Tremorigenic Score 1 to 3 Urination/Defecation 1 to 5 Lacrimation 1 to 3 Salivation 1 to 3 Tail-Pinch Response 1 to 5 Click Response 1 to 5   SOURCE: Moser, 2000b.

OCR for page 123
TABLE 9-3 NCTR Operant Test Battery Function Name of Test Motivation Progressive Ratio Task Discrimination Color and Position Discrimination Task Timing Temporal Response Differentiation Task Short-term Memory Delayed Matching-to-Sample Task Learning Repeated Acquisition Task   SOURCE: Paule, 2000. An important consideration for the IACUC, researcher, and veterinarian is the selection of the animal species to be used for behavioral screening in pharmacology and toxicology (Luft and Bode, 2002). Available data on kinetics and metabolism should be taken into consideration in identifying a species whose behavior will best predict effects in humans. Generally speaking, rodents are good models for behavioral screening in studies of neurotoxicity and neuropharmacology (Luft and Bode, 2002). Some behavioral-toxicology experiments involve dosing that produces deleterious effects. The protocol should provide a contingency plan for conditions in which side effects will be alleviated or that require an animal’s removal from an experiment (see “Animal Care and Use Concerns Associated with Toxicity or Long-lasting Drug Effects” in Chapter 8). Behavioral Screening of Genetically Modified Animals General Considerations Once a general health assessment of a newly developed strain of genetically modified animals is completed (see “Genetically Modified Animals” in Chapter 3), behavioral phenotyping should proceed as soon as sufficient numbers of transgenic animals are available to identify sensory, motor, or motivational deficits that may compromise animal well-being. Sensory and motor assessments should be completed before assessment of more complex behaviors—such as learning and memory, aggression, mating, and parental behaviors—because sensory and motor deficits may confound the interpretation of other behavioral assessments. Behavioral tests assess the effects of altering, adding, or removing a gene (and gene product) on behavior, not the effects of the normal gene on behavior (Nelson, 1997). Behavioral phenotyping can also be confounded by impairments that are secondary to the missing or inserted gene; for example, knocking out a gene may cause the compensatory overexpression of a second gene and any changes in behavior could be the result of the overexpression of the second gene. Those possible problems can be overcome in the same way as in other types of

OCR for page 123
ablation studies: by collecting converging evidence with a variety of pharmacologic, lesion, and genetic manipulations. Because mammalian genome mapping currently focuses on mice (Mus musculus), standardized behavioral testing of mice should be adopted (Brown et al., 2000; Crawley, 2000). Altered behaviors of knockout mice are often sufficiently obvious or unusual that they catch the attention of animal-care personnel, who then notify the investigators. Dramatic behaviors that include increased aggression, altered maternal care, decreased sexual behaviors, seizures, and impaired motor coordination and sensory abilities are commonly reported for knockout mice (e.g., Barlow et al., 1996; Brown et al., 1996; Brown et al., 2000; Chen et al., 1994; Crawley, 2000; Nelson et al., 1995; Saudou et al., 1994). Presumably, knockout mice may have more subtle behavioral changes that have not yet been discovered, even among mutants with no obvious behavioral phenotypes. Some of the behaviors probably will be revealed only if the animals are housed in conditions that are ecologically relevant with respect to space and social organization (Cabib et al., 2000; Pfaff, 2001; Potts et al., 1991). Behavioral performance is compared among wild-type (+/+), heterozygous (+/–), and homozygous (–/–) mice in which the gene product is produced normally, produced at reduced levels, or missing, respectively. The comparison of +/+ and –/– littermates of an F2 recombinant generation is probably the minimal acceptable control in determining the behavioral effects of knocking out a gene or genes (Morris et al., 1996). In the past, many knockout strains were generated by using stem cells from one strain and blastocysts from another strain (see “Knockout and Knockin Mutants” in Chapter 3 for review). Therefore, behavioral differences shown by knockout mice may reflect strain effects rather than the effects of the absence of the missing gene (Broadbent et al., 2002; Gerlai, 1996; Threadgill et al., 1995). Given the potentially important effect of background genotype on ability to detect effects of targeted mutations (Crabbe et al., 1999a; Lariviere et al., 2001), behavioral neuroscientists should attend to the genetic background of the transgenic animals under study to ascertain that proper controls for strain differences are used. Another limitation of the interpretation of behavioral data from knockout mice is the possibility that compensatory or redundant mechanisms might be activated when a gene is missing. For example, mice lacking the gene for the neuronal isoform of nitric oxide synthase (nNOS–/–) have a 20% increase in the expression of the endothelial isoform of nitric oxide synthase (Burnett et al., 1997). A compensatory mechanism may spare behavioral function and cloud interpretation of the normal contribution of the gene to behavior. Knockout mice are almost always raised by their natural mothers, which are missing one or more genes that may directly or indirectly affect maternal behavior. Thus, any changes in behavior observed in the knockout offspring may reflect the absence of the missing gene or reflect alterations in maternal care. Cross-fostering of matched-size litters can be used to untangle these influences.

OCR for page 123
distress in human beings may cause pain or distress in other animals.” Thus, the level of aversive stimulation applied to an animal generally should not exceed that tolerated by a human being. Not uncommonly, a member of the IACUC will experience the aversive stimulus for him/herself to better understand what the animal will undergo. Restriction of access to food or water is often used in behavioral neuroscience and neurophysiologic recording paradigms to motivate animals to execute desired behavioral tasks. This process merits attention to specific considerations that are addressed in Chapter 3 (“Food and Water Regulation”). In brief, investigators should provide a sound rationale for using appetitive or an aversive procedures. If access to liquid or solid food is to be restricted, the proposed level of dietary control should be justified, and appropriate monitoring and record keeping procedures should be described in the animal-use protocol. The procedures can be based on the literature or on an investigator’s own experience and should include criteria for determining intervention endpoints for removal of an animal from a particular conditioning paradigm. The goal of the monitoring procedures is not only to keep animals in a highly motivated state but also to maintain their health and welfare. Accordingly, records often include details regarding the animal’s performance on the behavioral task and various physiologic indexes. Documentation and Record Keeping Because of the potential health implications of food or fluid restriction, the health status of animals used should be well documented if food or water availability is restricted. Representative animal records might include weight or assessment of hydration status, general appearance or disposition, performance during the behavioral-task session, volume of fluid consumed (earned plus supplemented), dietary supplements or treats that were given, and experimental manipulations that were performed or treatments that were administered. For some species, the welfare of the animal may be further assured by monitoring its behavior in the home cage. Investigators should weigh animals according to suggested guidelines (NIH, 2002) and be alert to changes in mood, behavior, or appearance that indicate a potential medical concern. Individual animals may respond adversely to the weighing process; in such cases, judicious adjustment of weighing frequency or modifying the means of obtaining weights to better accommodate the individual animal may be necessary, and alternative methods of monitoring hydration may be advisable, for example, skin turgor, moistness of feces, and general appearance and demeanor (see “Methods for Assessment of Proper Nutrition and Hydration” in Chapter 3). There is some likelihood of weight loss during different phases of training (NIH, 2002). An animal’s use in a food- or fluid-restricted behavioral experiment should be assessed with veterinary input if persistent weight loss occurs. No

OCR for page 123
single physiologic measurement will always provide a reliable index of an animal’s well-being throughout the course of a behavioral neurophysiology experiment, but regular monitoring of several measurements (such as of food and water intake, weight, urine and feces, fur and skin, and behavior) usually permits adequate noninvasive evaluation. Each animal has different needs for food and fluids, so flexible criteria are preferable to rigid prescriptions of how much food or fluid animals should receive daily. Because different animals respond to food or fluid restriction challenges with different physiologic and behavioral accommodations, monitoring of each animal is essential, and adjustment of the restriction protocol is sometimes necessary (Toth and Gardiner, 2000). That is especially important during the initial stages of learning a new behavioral task. Emphasizing the role of professional judgment in these types of experiments, Toth and Gardiner (2000) recommend that: If task performance is not adequately supporting minimal intakes, the experimenter should re-evaluate and perhaps simplify the training strategy to facilitate the animal’s ability to learn and master the task. Standard clinical tests will reveal serious pathologic conditions, but the more insidious, gradual deterioration of an animal’s status can be recognized and treated only if there is regular observation and the implementation of professional judgment. Perhaps the greatest challenge in the maintenance of awake, behaving animals is the determination of their overall status. An animal’s overall behavior in its cage is a sensitive indicator of its psychologic and physical status (NIH, 1991). Investigators, veterinary personnel, and when available, behavior experts share in the responsibility of observing behavior, general appearance, and demeanor throughout an experimental regimen. Handlers of animals that are used in behavioral experiments should be knowledgeable and skilled in the interpretation of behavior such that changes that could indicate underlying health and well-being problems are readily identified and reported (Bayne, 2000). To this end, each animal should serve as its own behavior control, with baseline observations made prior to the initiation of the study. MOOD-DISORDER MODELS There has been considerable debate about the validity of animal models of human affective disorders. At a minimum, a good animal model of an affective disorder should meet many or all of the following criteria (Redei et al., 2001): strong behavioral similarities with the human disorder, a cause similar to the cause of the human disorder, similar pathophysiology, and similar treatments. Several animal models of affective disorder have been developed, especially for depression. In these models, depressive behavior may be caused by genetic manipulation, environmental perturbations or stressors, or drug treatments (Redei et

OCR for page 123
al., 2001). As is the case in all behavioral research, care must be taken to assess performance in these model systems in a valid and reproducible manner. Depression The so-called Porsolt swim test is the most commonly used test for assessment of depression in animal models (Porsolt, 2000; Porsolt et al., 1977). Other tests and procedures include the tail-suspension test, anhedonia (such as with consumption of sucrose solution), learned helplessness, chronic mild stress, olfactory bulbectomy, differential reinforcement of low rate of responding behavior, and conditioned place preference (Porsolt, 2000; Redei et al., 2001; Vaugeois et al., 1997; Willner, 1997). In all those tests, treatment with antidepressants that are effective in treating humans with depression reverses the depressed behavioral responses. It is generally accepted, however, that the Porsolt swim test (behavioral despair) and tail-suspension tests model human depression most closely (Crawley, 2000; Porsolt, 2000). In the Porsolt test, rodents are placed in a container of water at least 30 cm deep (to prevent an animal from touching the bottom of the container with their tail) and at least 15 cm from the top of the container (to prevent escape). To avoid temperature-related stress responses, the water temperature should be 24–30°C. Rodents placed in water generally swim, but if manipulated with some drugs or brain lesions, they stop swimming and float. Floating is considered a measure of depression because the animals appear to stop trying (learned helplessness or behavioral despair) and because drugs that are effective antidepressants in humans decrease floating time (Crawley, 1999). Genetically modified animals may require special attention; any rodent that fails to swim or float should be removed from the water immediately. However, even if transgenic animals can remain afloat, locomotor difficulties can interact with performance in the Porsolt swim test and cause nondepressed transgenic animals to appear depressed. The tail-suspension test avoids the problems of locomotion somewhat and avoids the hypothermia and stress associated with forced swimming (Vaugeois et al., 1997). Animals are suspended by their tails and the amount of “immobility” is measured by a force-strain gauge that records all their movements (Steru et al., 1985). Longer periods of immobility are associated with higher depressive scores. The immobility can be reversed with antidepressant treatment (Vaugeois et al., 1997). Reduced ingestion of a sucrose solution is another reliable indicator of depression-like-behavior in rodents (Gittos and Papp, 2001; Stock et al., 2000). This test avoids some of the problems of locomotion and coordination of the Porsolt test, but if the targeted gene affects metabolism or food intake, its reliability for depressed behavior may be impaired. In all those behavioral tests of depression, proposed procedures for monitoring, record keeping, and humane intervention should be described in the associated animal-use protocol and approved by the IACUC.

OCR for page 123
Anxiety Assessment of anxiety was described earlier in this chapter (“Behavioral Screening of Genetically Modified Animals, Assessment of Anxiety”). Alcohol and Drug Addiction Several testing paradigms assess responsiveness of rodents to drugs that have abuse potential, including paradigms involving self-administration of alcohol, cocaine, morphine, or nicotine (Crawley, 1999; Grahame and Cunningham, 1995). Self-administration is typically achieved by requiring the rodent to press a lever or display a place preference. Tolerance, dependence, and withdrawal symptoms can be studied. With this approach, transgenic mice may have locomotor or coordination difficulties that interfere with self-administration (McClearn and Vandenbergh, 2000). Additional information is provided in Chapter 8 (“Addictive Agents”). Animal Care and Use Concerns The primary goal of the preceding behavioral assays is the induction of stress or aversive states. It is important for the investigator to determine the earliest possible or least severe endpoint when the manipulation has adverse effects on an animal. Any behavioral test that subjects animals to water has the potential for evoking a stress response. Therefore, it is important that the time that the animal is in the water be minimized and that animals be monitored closely to avoid unnecessary stress. Animals should be dried thoroughly after the swim test, and it is advisable to place their cages on a heating pad for several minutes. The use of unregulated heating pads or heat lamps should be avoided as they can develop hot spots and cause thermal burns. Continuous monitoring is also important for automated tasks, such as tasks that use roto-rods, platforms, or other devices in which animals may be injured. Because of the likelihood of multiple testing, excellent record keeping is imperative. BEHAVIORAL STRESSORS Some neuroscience research involves exposing animals to behavioral stressors. These manipulations can be social (such as involving social separation or mixing of unfamiliar animals) or nonsocial (such as exposing animals to novel environments or restricting behavioral activity). This research focuses on three avenues of investigation. The first is aimed at understanding the effects of exposure to behavioral stressors on aspects of neural function or conversely understanding how neural manipulation affects responses to behavioral stressors (von Borrell, 1995). For example, a pregnant monkey

OCR for page 123
might be exposed to various behavioral stressors, such as noise and unfamiliar surroundings, and neurochemicals associated with the stress response would be measured in her offspring to determine the effects of prenatal stress on the development of stress responsiveness in young animals (Schneider et al., 1998). The second is aimed at understanding the neural substrates or correlates of particular behaviors or aspects of temperament, including social recognition, affiliation, pair bonding and attachment, parental behavior, social dominance, aggression, predation, play, and fearfulness (Amaral, 2002; Kavaliers and Choleris, 2001; Siegel et al., 1999; Young, 2002). In those studies, animals may have lesions, be genetically modified (mouse knockouts), or be electrically or chemically stimulated, and the resulting behaviors can be observed; or neural function may be measured during or after the performance of the behaviors of interest. The third category consists of pharmacologic studies to determine the efficacy of various compounds in reducing aggression, anxiety, or fearfulness (Mench and Shea-Moore, 1995). The purpose of those studies is usually to identify compounds that may be useful in human or veterinary clinical medicine, but pharmacologic testing can also be used for studies of underlying mechanisms of behavior: the behavior of interest is stimulated in some way, usually by staging an aggressive encounter or placing an animal in a fear-inducing situation, and compound efficacy is then evaluated with behavioral measures. Social Disruption Social disruption can be used as an experimental technique in neuroscience and behavioral research, but it can also be an inadvertent confounder of the research. Experimental designs that purposefully incorporate social disruption, do so through the temporary removal and reintroduction of offspring or of group or pair-mates, longer-term or repeated reorganization of social groups by removal of group members or by introduction of unfamiliar animals to groups or to one another, or even the merging of different groups of animals. Abnormal social conditions can also be created by placing animals in atypically small or large social groups, by forming groups of atypical composition (such as all-male groups or groups comprising only animals of similar age), or by crowding them. In addition to the study goals described above, this technique has recently been used to study coronary artery atherosclerosis, heart rate reactivity, and the effects of exercise in conjunction with social disruption on coronary heart disease (Kaplan et al., 1982, 1993; Manuck et al., 1983a,b; Williams et al., 1991, 2003). The effects of social separation (such as individual housing) or social isolation on an animal’s behavioral profile have been documented in various species. The impact of social separation or isolation can depend on the species or strain of animal, the age at which an animal is removed from conspecifics, the duration of the separation, and the completeness of the separation (with respect to visual, auditory, or olfactory cues from other animals). In nonhuman primates, the lack

OCR for page 123
of physical contact appears to be the most important cause of abnormal behavior, both in infants and in adult animals (Bayne and Novak, 1998). Animals that are isolated to disrupt the infant-parent bond often display acute responses to indicate stress. Distress vocalization, changes in general activity and heart rate, as well as elevated cortisol/corticosterone concentrations can occur and are adaptive under normal circumstances. However, if the separation is prolonged, as during experiments where the effects of infant-parent bond disruption are being studied, it becomes distressful and can lead to maladaptive behaviors as the infant animal matures. Self-injurious behaviors, stereotypic behaviors, extreme timidity or aggressiveness, and inability to mate or provide adequate care to offspring are maladaptive behaviors that might result from the social disruption (NRC, 1992). Kittens separated from their mothers at an early age tend to be more aggressive and nervous as adults (Seitz, 1959), and social play is critical for a kitten’s development (O’Farrell and Neville, 1994). Puppies that are not adequately socialized to other dogs or people may be excessively fearful or aggressive (O’Farrell, 1996). Wolfle (1990) has described a puppy-socialization program and behavioral scoring method specifically for use in the research environment. Monkeys reared in partial or total social isolation develop a syndrome of behavioral abnormalities that includes rocking, huddling, self-clasping, and excessive self-orality (Cross and Harlow, 1965; Harlow and Harlow, 1965). As the animals age, stereotypic patterns emerge, such as repetitive locomotor patterns, floating limbs, and eye poke or salute. The isolation syndrome is also manifested in the development of abnormal social relationships (Mason, 1968). A restricted social environment can also affect adult animals. For example, long-term (2-year) individual housing of adult nonhuman primates has been shown to alter social behavior (Taylor et al., 1998). Unless the research focuses on social restriction or veterinary concerns develop, infant animals should be reared in a social environment with mother and peers, with mother only, or with peers only to reduce or prevent psychopathologic conditions (Bayne and Novak, 1998). Similarly, when the research, health, and safety of the animals allow it, adult social animals should be maintained in a social environment (for example, pair- or group-housed). Animal Care and Use Concerns The primary animal care and use concern associated with social disruption is the distress that leads to the display of maladaptive behaviors. When studies involve the use of social disruption, the animal-use protocol should include humane endpoints for removal of the animal from the study. Determining endpoints that are predictive of severe distress is a matter of professional judgment and should evolve through discussions between the IACUC, veterinarian, and PI.

OCR for page 123
It is important to recognize that the display of the maladaptive behavior affects not only the isolated animal but can also have unintended affects on the dam (in studies of infant-parent bonds), the potential offspring of the animal, and the conspecifics that may be forced into the animal’s social group. In some species, such as nonhuman primates, dams also show a response to separation from their infants. Their behavioral and physiologic reactions appear to be similar to those of the infant, although less persistent and intense (NRC, 1992), and steps should be taken to minimize this distress if it is an unintended byproduct of the experiment. The offspring of animals in social disruption experiments may also be impacted by the maladaptive behavior of its dam. For example, female rhesus macaques that are isolate-reared can be neglectful or abusive of their infants (Suomi, 1978). In that situation, it may be appropriate to provide additional support to the offspring or protect it from injury. In some cases, social disruption causes aggression toward conspecifics. For example, social restriction of male mice will lead to intermale fighting (Brain, 1975). Similar findings have been observed in gerbils, hamsters, and rats (Karim and Arslan, 2000; Payne, 1973; Wechkin and Breuer, 1974). Isolation-reared rhesus monkeys are hyperaggressive and do not develop normal social relationships with other monkeys (Anderson and Mason, 1974; Mason, 1961); this aggression can be directed to other animals or be self-directed (Gluck et al., 1973). Steps should be taken to prevent injury in these cases. For instance in nonhuman primates, this may require housing the aggressive animal separately (AWR 3.81(a)(1)) or the use of screen barriers within cages to permit side-by-side contact, but prevent agonistic encounters. Induced Aggression or Predation Several common models are used in studies whose primary intent is to induce aggression or predatory behavior (Mench and Shea-Moore, 1995): Isolation-induced aggression. This involves isolating a male mouse or rat for several weeks and then staging a brief encounter (usually 5–10 minutes) with an unfamiliar group-housed male. Encounters may be staged either in the isolate’s cage or in a neutral arena. If drugs are administered, they may be administered either to the isolate or to both animals. Because cues from the introduced animal can affect the outcome of the encounter, introduced mice are sometimes rendered anosmic before testing to make them less responsive to social stimulation (Stowers et al., 2002). Naturalistic paradigms. These studies aggression by placing animals in circumstances that approximate the situations that they might encounter in the wild, where they have to compete for resources, defend territories, or integrate into new social groups. Examples are introducing an ‘intruder” animal into the cage or enclosure of a group of resident animals (Blanchard et al.,

OCR for page 123
1975), mixing two social groups by removing a partition between their cages (Zwirner et al., 1975), and requiring animals to compete for access to food by displacing one another from a tunnel (Miczek, 1974). Isolation of mice is not necessary to study aggression; pair-housing of a male with a female promotes consistent aggressive behavior when the male is tested in a resident-intruder situation (Fish et al., 2002). Animals may also simply be observed in their normal social groups, either in the laboratory or in the wild; this process is facilitated by the use of osmotic minipumps to deliver neuromodulators or hormones and radiotransmitters for remote collection of physiologic data. Aggression modified by drugs. Using an “intruder” paradigm, it has been shown that drugs, such as alcohol and allopregnanolone (a positive modulator of the GABAA receptor) can increase the expression of aggressive behavior in mice (Fish et al., 2002). In contrast, other drugs, such as 5-HT1B agonists (for example, anpirtoline) will inhibit the expression of aggression (de Almeida and Miczek, 2002). Predatory aggression. This involves introducing prey species to animals, especially introducing rodents to cats and mice to rats (the muricide model). If the object of the research is to understand or influence the full predatory sequence or if the sequence ensues so rapidly after initial attack that intervention is not possible, death of the prey animal is often the endpoint. Because pain and injury to both the prey animal and predator are significant welfare issues with these kinds of studies, methods to protect the prey animal from physical attack or modeling elements of the predation sequence should be considered (Novak et al., 1998b). It may not even be necessary to use live prey. The number of times an animal serves as prey should be limited. The use of wild caught animals may be preferred due to their potentially greater experience and skill in predatory avoidance (Novak et al., 1998b). In those instances where the prey animal dies, the study should be designed to expedite the predation sequence and to minimize the pain and distress experienced by the prey animal (Huntingford, 1992). Any situation in which unfamiliar animals are mixed or established social groups are perturbed has the potential to result in aggression, whether or not aggression is central to the aims of a study. The effects of the aggression on the recipient animal will depend on the intensity, duration, and potential for injury associated with the aggression and hence on the species being studied, the ages and sexes of the animals, and their past social experiences. If aggression is incidental to the goal of the study, many methods can be used to reduce the potential for injury, including gradual introduction of animals, allowing partial contact (for example, visual, auditory, olfactory, or tactile) before mixing and providing refuge areas to which introduced animals can escape from aggressors (Bayne and Novak, 1998). Naturalistic approaches to inducing aggression or predation may not only minimize injury but also provide information that is more reflective of the range and types of behaviors shown by animals under more ecologically relevant circumstances (Kavaliers and Choleris, 2001; Mench and Shea-Moore, 1995).

OCR for page 123
Even when aggression is a desired outcome of a study, attention should be given to minimizing injury and distress (Anonymous, 2002; Bayne and Novak, 1998; Ellwood, 1991; Huntingford, 1984). Ways of doing that include minimizing the numbers of animals used; decreasing the length of an encounter to the shortest time necessary to collect the required information, which may involve continuous observation with intervention to stop aggression at predetermined points; using artificial “model” animals rather than real animals as the recipients of aggression or the initiators of predatory encounters; placing introduced animals behind protective screens (for example, Habib et al., 2000) or barriers (Perrigo et al., 1989); and allowing the introduced or subordinate animal to control the intensity of aggression by providing refuge areas. Each of those strategies has limitations, and their usefulness will depend on the species being studied and the purpose of the study. Animals that are severely injured during an encounter should be removed as soon as possible and treated or euthanized. The use of specific animals as targets of prolonged aggression should be well justified. Environmental Deprivation Animals may be exposed to nonsocial behavioral stressors to determine their effects on neural and neuroendocrine function. For example, animals may be restrained for brief or for sustained periods by being held, tethered, chaired, or immobilized by other restraint devices or placed in small enclosures or wrappings that restrict movement. Restraint may be repeated at intervals to cause intermittent stress. The animal-welfare issues associated with restraint are discussed in Chapter 3 (“Physical Restraint”). In other studies, the behavior of animals is restricted by placing them in barren environments that provide few opportunities for normal behaviors or by restricting sensory input. One or more sensory modalities (touch, audition, vision, and olfaction) may be restricted, or animals may even be kept in complete sensory isolation. The goal of such studies is generally to determine the effects of restricted environmental input on neural development. Restricted sensory or behavioral input often leads to the development of severely abnormal behaviors. Whether these effects are reversible depends on the species, the duration of restriction, and the age at which the animals are restricted. Consideration should be given to the impact of this type of research using long-lived animals due to the protracted and resilient behavior changes invoked. Environmental Stimulation Stress can be induced by exposing animals to novel or extremely complex environments. The emphasis is usually on neural development, generally with a focus on fear and exploratory behaviors. Fear and exploration may be assessed with a standard battery of tests, some of which are described earlier in this

OCR for page 123
chapter. Extreme novelty or complexity can have adverse physiologic and behavioral effects. However, moderate novelty and species-appropriate complexity actually have generally beneficial effects, such as enhancing neural development, learning and spatial ability, and stress competence. This is reflected in the AWRs (3.81), which mandates an appropriate plan for environmental enhancement adequate to promote the psychological well-being of nonhuman primates. The purposeful use of environmental stimulation for experimental reasons should be distinguished from incidental, but no less stressful, stimuli that may occur in an animal facility and impact ongoing research. In either case, young animals are more susceptible to a prolonged effect of environmental stimulation and thus the use of long-lived species in this research should be well justified if the intention is to maintain the animals in the colony for extended periods of time. Animal Care and Use Concerns The goal of many studies involving behavioral stressors is the induction of stress responses. Exposure to intense, repeated, or prolonged stressors can have a variety of adverse effects, including suppression of reproduction, immune dysfunction, cardiovascular and gastrointestinal impairment, and persistent disruption of neuroendocrine function (Moberg and Mench, 2000). One consequence of exposure to behavioral stressors may be the development of abnormal behaviors, including self-mutilation, mutilation of other animals (such as tail-biting in pigs and cannibalism), and stereotypic behaviors (such as bar-chewing or route-tracing). Causative factors of abnormal behaviors include social isolation, rearing in a barren environment or lack of sensory stimulation, and excessive environmental or social stimulation. Once developed, the behaviors tend to persist even when the original eliciting stimulus is removed, so the animals in question may have special husbandry and care requirements. Minimizing the duration, frequency, and intensity of stressors can minimize the effects. When young animals are separated from their dams, parents, or broader social groups for experimental purposes, provisions must be made to care for the animals, both physically and behaviorally. In some cases, partial socialization (either temporally or physically limited contact) with peers or compatible species may be possible to mitigate the immediate stress imposed by the socially restricted environment and to improve the long-term behavioral health of the experimental animals. Alternatively, separation may be delayed until the animals are older to limit the effect of restriction. Novak et al. (1998a) suggest that young animals be monitored closely and evaluated regularly if they are separated, thus enabling more informed management decisions to address the animals’ well-being. Animal-use protocols for research involving behavioral stressors should include a thorough description of the potential animal-welfare issues associated with each stressor and a detailed plan for monitoring, record keeping, and deter-

OCR for page 123
mining when to end a test early to avoid unnecessary pain and/or distress. If little or nothing is known about the possible outcomes of exposure to a particular behavioral stressor, IACUC review and approval of the protocol may involve a requirement to conduct pilot studies, mandatory oversight of initial testing by veterinary staff, or provision of regular progress reports as a condition of continuing approval.