The Big Picture: Evaluation and Management of Aggressive Behavior in Captive Animals
Aggressive behaviors are one major reason that zoo staff seeks assistance from veterinarians. A wide range of behaviors may be considered aggressive and assistance is often sought only after the frequency, duration, or intensity of the relevant behaviors have become intolerable to husbandry staff or are considered significantly atypical for the species. In many cases, the behavior has escalated to the point where the animal is consistently injuring itself or other animals, or threatening human safety.
Behavior, and why behavior develops or changes, can be examined and understood with a variety of sciences or models (e.g., neuroendocrine, genetic, physiology, etc.). These different approaches lead to different hypotheses about why the behavior is happening and different treatment or management strategies. Three major models directly relevant to clinical veterinarians include medical, ethological (natural history), and applied behavior analysis. These approaches are not mutually exclusive and when utilized together enhance the development and success of humane behavior change strategies.
Veterinarians are trained to consider medical causes for behavior. For example, increased aggression (e.g., biting when approached) can be a sign of disease such as pain, a brain tumor, or abnormal neurochemistry. Neurophysiologic research has demonstrated that some problem behaviors may be due to differences in brain activity at a molecular level.1,5 The development of some problem behaviors is likely a combination of both biological predisposition and development events. For instance, animals raised in barren environments or under a great deal of stress during the early weeks, months or years of development have fewer neurons in the brain, decreased dendritic branching and spine density, and reduced synaptic connectivity compared to animals raised in enriched environments.1,5 Many captive animals are reared in environments dissimilar to natural environments and this may contribute to neurophysiologic abnormalities, though this is currently speculative.
Whatever the cause, it is likely that some portion of a given population has neurophysiologic dysfunction that limits the individual’s ability to learn and behave normally in common environmental situations. In addition, some animals may be normal but less able to cope with the physical, mental, and behavioral limitations of a given captive environment, either temporarily or long term. In these cases, using appropriate psycho-pharmaceutical medications such as serotonin reuptake inhibitors (e.g., fluoxetine) or benzodiazepines (e.g., alprazolam, diazepam) may be effective in reducing aggression.2 Antipsychotics, such as haloperidol, produce inconsistent results for the treatment of aggression and may increase aggression in some cases. Antipsychotics lead to overall suppression of behavior and their high incidence of side effects make them inappropriate for long-term therapy.2 The effectiveness of any psychotropic medication can be challenging to predict. In addition, many of these medications take a long time to show efficacy. This needs to be taken into account when effectively planning interventions. It is inappropriate to expect medication to compensate for a poor environment. In addition to medications, increasing the animal’s behavioral control and choice through appropriate enrichment and effective positive reinforcement training is generally critical to successful management plans, including those in which medication is called for and effective.
In the zoological field, we often consider ethological correlates for behavior and behavior change, i.e., the behavioral adaptations that have evolved to fit the animals’ ecological niche. For example, increased aggression can be understood as an inherited modal action pattern such as a territorial defense chain elicited by breeding season cues, or secondary to social, hierarchical influences within a group. To complicate matters, many captive animals have not been reared in normal social groups and may not have learned to respond to other animals’ cues typically, leading to increased or inappropriate levels of aggression.
Animals may display aggressive behaviors which are typical for the species, but which are disruptive to captive management goals or displayed at a greater intensity than desired for success in captivity.7 These behaviors are often relevant to the reproductive cycle. In these cases, medications that specifically reduce reproductive hormones may reduce aggressive behaviors; examples include GnRH agonists (e.g., deslorelin or leuprolide), or progestins to suppress ovarian cycling (e.g., melengestrol acetate, megestrol acetate, medroxyprogesterone acetate). In addition, a thorough review of the individual (and where applicable, the group) behavior pattern is appropriate to better understand and document social interactions and other environmental stimuli that are associated with aggressive behaviors. For example, a subordinate (versus the more easily identified aggressor) may be precipitating aggressive events through inappropriate responses to social cues. Understanding the natural history of a species is important when conducting a thorough evaluation of the animal’s environment, as well as ensuring that staff expectations for behavior are appropriate.
In addition to these more familiar models, behavior analysis is critical for understanding how a specific behavior emitted by an individual animal is learned and maintained, due to interaction with the environment in which it occurs.4 Behavior analysis is a trans-species science that investigates the universal laws of behavior change due to experience, i.e., learning. Applied behavior analysis (ABA), the behavior change technology derived from behavior analysis, takes the individual animal’s learning history and current environmental conditions into account and investigates the purpose (i.e., function) the behavior serves for the animal. In the above example, we can hypothesize that the increased aggressive behavior is the result of learning, i.e., the behavior was reinforced in the past. Even complex, severe aggressive behaviors are responsive to this approach.7
From the ABA perspective, understanding and changing behavior results from identifying the discriminative stimuli that set the occasion for the behavior (i.e., setting events, motivating operations, and discriminative stimuli), and the consequences that give the behavior strength (frequency, duration, intensity, etc.). The focus of the behavior change plan is to modify the environment to set the occasion for appropriate alternative behaviors and reinforce them when they occur. With ABA, we change the environment to change the animal’s behavior.
Reviews of behavior analysis science for veterinarians exist and a functional assessment and intervention design (FAID) worksheet for evaluating problem behavior and developing appropriate behavior support plans is also available.3,4,6,7 The FAID worksheet provides a standardized approach to cases (similar to a SOAP format) and prompts a complete, individualized evaluation of the problem behavior situation and development of a plan specific to that individual in that environment. A summary of five major questions related to behavior is provided (Table 1). This model provides a powerful, systematic method of behavior evaluation and intervention. The ABA approach is under-utilized by most veterinarians and husbandry staff due to lack of training and general under-estimation of the importance of prior learning and current conditions as a major factor governing actual behavior in individual animals.
Veterinarians should utilize all three models when diagnosing and managing aggressive behavior displayed by animals so that relevant medical conditions, ethological, and learning variables are all evaluated.
Table 1. Summary questions to prompt investigation of the environmental factors related to problem behavior
Purpose of Question
What does the behavior look like?
Reduce use of labels and focus on actual, observable behaviors in preparation for identifying relevant environmental stimuli that predict and maintain the behaviors. If multiple disruptive behaviors are present, each should be evaluated individually.
Labeling the animal: The animal is aggressive.
Behavior A: The animal lunges at the door and bites the air.
Behavior B: The animal pushes into the other male and bites his neck.
What conditions predict when the behavior will occur (when is the behavior most likely)?
Identify the relevant environmental stimuli that cue or set the stage for the behavior.
Behavior A happens when keepers approach the door with food.
Behavior B happens when browse has been added to the enclosure.
What does the animal get from or get away from by doing the behavior?
Identify the relevant environmental stimuli that are reinforcing (maintaining) the behavior.
After behavior A happens, food is left in the enclosure and the keeper leaves the area.
After behavior B happens, the other animal leaves the area and the first animal has access to the browse.
Under what conditions does the animal not exhibit the behavior (when is the behavior least likely)?
Identify the environment when the animal is most successful. This step helps staff realize there are environments in which the animal is successful and the problem behavior is not occurring.
Behavior A is least likely to happen if a keeper approaches without food.
Behavior B is least likely to happen when the females, as well as other males, are present or the two males are fed hay.
What can the animal do instead?
Identify another behavior the animal can do in place of the problem behavior. These are behaviors that can be trained (reinforced). In some cases, reinforcing another behavior is not possible and management of environmental stimuli to not cue the problem behavior is appropriate.
Instead of doing behavior A, the animal can stand with his head near the water bowl when food is added to the enclosure.
Instead of doing behavior B, the animal can eat browse in a separate part of the exhibit, or the two males are not fed browse when they are being housed without the females.
1. Cabib, S. 2006. The neurobiology of stereotypy II: the role of stress. In: Mason, G., and J. Rushen (eds.). Stereotypic Animal Behavior, Fundamentals and Applications to Welfare, 2nd ed. CAB International, Oxfordshire, UK. Pp. 227–255.
2. Crowell-Davis, S.L., and T. Murray. 2006. Veterinary Psychopharmacology. Blackwell Publishing, Ames, Iowa.
3. Friedman, S.G. 2007. A framework for solving behavior problems: functional assessment and intervention planning. J. Exotic Pet Med. 16: 6–10.
4. Friedman, S.G., and L.I. Haug. 2010. From parrots to pigs to pythons: universal principles and procedures of learning. In: Tynes, V. V. (ed.). Behavior of Exotic Pets. Wiley-Blackwell, Ames, Iowa. Pp. 190–205.
5. Lewis, M.H., M.F. Presti, J.B. Lewis, and C.A. Turner. 2006. The neurobiology of stereotypy I: environmental complexity. In: Mason, G., and J. Rushen (eds.). Stereotypic Animal Behavior, Fundamentals and Applications to Welfare, 2nd ed. CAB International, Oxfordshire, UK. Pp. 190–226.
6. O’Neill, R.E., R.H. Horner, R.W. Albin, J.R. Sprague, K. Storey, and J.S. Newton. 1997. Functional Assessment and Program Development for Problem Behavior: A Practical Handbook, 2nd ed. Cengage Learning, Pacific Grove, California.
7. Šusta, F. 2010. Reducing aggressive behavior of kiang—Tibetan wild ass (Equus kiang hodereri) male in Prague zoo. ABMA Wellspring 10,11: 22–24.