Based in Sydney, Australia, Foundry is a blog by Rebecca Thao. Her posts explore modern architecture through photos and quotes by influential architects, engineers, and artists.

How Fear Becomes Aggression in Dogs and Humans

Ask most emotion psychologists which emotion is associated with aggressive behavior—defined as behavior intended to cause harm to another—and they will quickly name anger. We associate anger with a strong urge to hit something or say cruel things, which are forms of aggression.

However, ask most dog behaviorists what emotion is associated with aggressive behavior in pet dogs, and they will probably name fear. If they are predisposed or conditioned through experience to be fearful, some dogs will resort to behaving aggressively when they are afraid. This common behavioral problem is often referred to as fear aggression. I have lots of books on my Kindle about it.

So people and animals act aggressively out of anger, fear, or both. This type of ambiguity bothers me.

What does it mean for theories of emotion? Are fear and anger really distinct affective states, and if so, how do they both lead to aggression? Secondly, I want to understand how fear emerges seemingly out of nowhere and sometimes escalates into debilitating aggression, as it did with my previous dog Ronan (see my last post). Fear aggression is responsible for a huge number of dogs being surrendered to shelters or euthanized, and most people (including myself) don’t see it coming until it’s too late.

Tidy Theories vs. the Real World

A major hallmark of an emotion, according to Most Emotion Science People (MESP; a definitely not made-up entity for whom I will now speak), is that emotions exist to direct our behavior and cognitions. They motivate us to do particular things—laugh, cry, play, run away, find solutions to pressing problems, swipe right on Tinder. Many debates in the field of affective science are trying to decide which emotions humans (and other animals) share, and how specifically they change our brains, bodies, and behavior.

Before I bring this back to fear and aggression, I’ll review, and oversimplify, two major perspectives endorsed by MESP. Because MESP are particularly proud of the S in the acronym, we have a longstanding debate in the field that is basically our version of “nature versus nurture” (see also: every other subfield in psychology and biology). This is a debate about whether humans and other animals have innate emotion systems with particular neural underpinnings and associated behaviors, or whether we only feel like fear and anger and happiness are discrete states because we have been socialized to think so. In reality, according to some, we are not programmed to experience specific emotions; they are neat, made-up categories we squeeze our emotions into.

Perspectives in emotion research that fall closer to this “nurture” end of the nature-nurture spectrum are called dimensional models of emotion. According to dimensional models (1, 2, 3), all emotions you could ever experience are points in a 2- or 3-dimensional affect space (see diagram below). The most commonly hypothesized emotion dimensions are called arousal (meaning how active and high-energy the state is) and valence (meaning how good or bad the feeling is). Happiness, for instance, is really just a point in the 2D space that is medium-high on the arousal dimension and on the positive end of the valence dimension.

Dimensional models of emotion suggest that all the feeling states we experience as being discrete emotions, including fear and anger, vary primarily in how aroused (alert/excited) and good vs. bad they feel. Image from Wikimedia

Dimensional models of emotion suggest that all the feeling states we experience as being discrete emotions, including fear and anger, vary primarily in how aroused (alert/excited) and good vs. bad they feel. Image from Wikimedia

Anger is high arousal/negative valence. Fear is also high-arousal/negative valence. According to dimensional models, these supposedly “discrete” emotions are in fact overlapping points in the 2D space. So according to some dimensional models, anger and fear aren’t categorically different emotions and it makes sense that they might lead to similar behaviors, like aggression. As far as the fear aggression question goes, dimensional models have an easy explanation—the states we call fear and anger are really indistinguishable from each other at a biological level, so they should lead to similar behaviors, like aggression.

 I’m not convinced that we should reduce emotions to points in a continuous 2D (or even 3D) space. After all, fear and anger feel different and motivate us towards different outcomes (4). They are triggered by different events (5, 6, 7), have different neurobiology (8,9,10), and have different effects on the body (11). Their expressions (12) and behavioral tendencies are generally different, meaning you and I can easily tell the difference between someone who's angry and someone who's afraid (fear aggression being our problematic exception at the moment).

Putting the Animal Back in the Animus

The reasons I just mentioned are why I lean slightly more towards the “nature” end of the nature-nurture spectrum, towards evolutionarily focused emotion theories. These theories tend to pay more attention to (or at least acknowledge the existence of) the rich emotional lives of animals, which I obviously dig.

There are many flavors of these evolutionary theories, but the general idea is that emotions are evolved adaptive states that respond to different environmental pressures. Just as sneezing handles the problem of dust going up your nose while sleeping handles the problem of needing to consolidate memory and restore bodily resources (sorry, sleep researchers, I’m sure I butchered that), different emotions like joy and fear and anger and love evolved to respond to unique problems. It is nonsensical to put categorically different behaviors like sleep and sneezing on the same continuum; similarly, some theorists would say it is nonsensical to put fear and anger on a continuum.

According to these basic or evolutionary theories of emotion (13, 14, 15), different emotions are elicited by the organism perceiving different environmental cues, such as danger or a potential mate. Different emotions also have distinct behavioral tendencies that address the problems signaled by the environmental cues. Disgust, for instance, causes behaviors that get the organism away from rotten or poisonous objects. [For my emotion researchers: let’s cast aside the outdated portrayal of basic emotion theories as saying either 1) there is no room for learning or variability in how emotions are expressed and experienced and 2) there must be fMRI-able modules in the brain generating each emotion.]

Let’s see how these evolutionary theories distinguish between the emotions of interest, fear and anger.

  • Fear emerges in response to threats of pain, the presence of predators, or because something is simply unfamiliar. Fear is traditionally thought of triggering fight or flight responses to help the organism escape danger, as well as a wide-eyed facial expression (No really, click that link).
  • Evolutionary theorists propose that anger, or rage, is triggered when an organism perceives that one of its goals has been thwarted, its movement is restricted (more on this later), or there is competition for resources (16). As a side note, it is likely that other species besides millennials can be hangry, because the state of hunger signals limited resources, a reliable trigger of anger. Anger, as we have established, triggers aggressive behavioral tendencies like biting, punching, and uncontrollable Facebook rants.

To complicate matters, anger (as it is traditionally defined, at least) is not the only root cause of aggression. This complexity is reflected in the human and non-human animal literature, which is rich in theories of aggression (17-22). Another example of contexts that can trigger aggression is pain or distress (23). Aggression can also become a learned response. This means that a person or animal can learn to use aggression in times that don't fit the classic "anger" contexts.

Animals and people often learn that aggressive behavior gets them what they want, such as status, resources, or a feeling of safety. Behaving aggressively, be it bullying other kids in school or biting someone who tries to pet you, gets desired results. Really effectively. The bully gets to feel powerful and the scared dog gets the strange person to leave her alone. A general theme I’ve found across theories of aggression is this distinction between what I’ll call angry aggression and learned aggression. Angry aggression is straightforward from an emotion theorist's point-of-view, but learned aggression complicates the picture, because it can originate from a variety of emotional/motivational states, such as fear.

The Origins of Fear Aggression

Fear-based aggression is a case of learned aggression that happens when an animal (human or otherwise) experiences intense fear combined with an inability to escape it with fight-or-flight-style behavior (8, 24). Imagine you are alone on a walk in the woods and a slow-moving, knife-wielding murderer emerges from behind a tree. Without much deliberation, you realize that you can achieve your high-priority goal of staying alive by fleeing. Your fear motivates escape, and escaping, in turn, reduces fear as you reach safety. This avoidant (25) fear-based behavior is self-reinforcing (it is technically a negative reinforcer) because it felt good to reduce your feeling of fear, so you are likely to do it the next time you feel fear. In the future, you may be even more likely to run away at the first signs of danger, since it was such a rewarding behavior previously.

This learned avoidance becomes a common coping behavior for some fearful dogs. Maybe you know a dog who hides under the bed in fear when the vacuum comes out, or a person who refuses to enter a pet shop because SNAKES. Although we don’t want the dogs or people in our lives to feel unnecessary fear like this, an avoidant behavior is preferable to the approach fear-based behavior: aggression.

Now imagine a second version of the alone-in-the-woods scenario. Imagine the murderer is quick on his feet. You run as fast as you can to a nearby abandoned cabin, and take refuge. But he finds you. Trapped, your only option is to fight back. Fortunately, you take a lot of kick boxing classes at the gym. Thus your fear led you to aggress, and fear aggression is born.

Unfortunately, the modern dog’s life is designed to force a fearful dog to face his fears rather than to flee or hide. The stranger-shy dog is trapped in a small apartment while her owner invites friends over for coffee—and to make things worse, her owner pulls her over by the collar to “say hi” to the strangers entering her home. The dog who has been previously attacked by another dog is trapped on his leash during walks, where he is forced to walk by other dogs on leashes. These dogs cannot escape, but their fear is real. So some of them take matters into their own hands (metaphorically speaking, obviously).

The first time a dog growls, snaps, or bites out of fear, it is often a last-ditch resort. But man, does it work. The scary animal or person backs off, and this serves as a reward, making aggressive responses more likely when the dog feels fear in the future. The real problem is fear aggression is so self-reinforcing—it almost always makes the scary thing go away—that dogs like Ronan start to use it in instrumental, preventative ways rather than as reactions to truly threatening stimuli. 

In Ronan’s case, rather than just barking and snapping when a strange human or dog caused him pain, he began using this extreme behavioral solution when he felt the slightest level of anxiety about the presence of a stranger. Eventually, even when he seemingly had no reason to feel afraid, he became conditioned to bark and lunge aggressively as soon as he saw a person, even from far away. 

Furthermore, what initially looked fearful--in the beginning, Ronan would yelp, ears back submissively, as he "punched" a scary person's face with his muzzle--evolved into angry-looking aggression. Every time the aggression is reinforced by causing the fear-eliciting thing/person to go away, it becomes more aggressive, and perhaps could be better categorized as anger (17). This transition from fear to angry aggression is especially clear if the person or animal chooses to aggress even when they have the option to flee. 

Not only do physical restraints limit a person or animal’s ability to escape threatening situations, but physical restraint is itself a trigger for aggression (of the “angry” variety). Watch this video of a baby getting frustrated and angry about being unable to move to see what I mean. Leash aggression is the perfect example of the physical restraint -> frustration -> aggression chain of events, which ends with a dog who isn’t normally aggressive becoming so on leash. I think this restraint-based aggression just compounds fear aggression when a fearful dog is trapped on leash.

What Does All This Mean for Dog Behavior?

The number one dog-related point I wanted to highlight in this post is the mere existence of fear aggression. All good dog behaviorists are keenly aware of fear as the root for many undesirable dog behaviors, but in my experience, many lay people see an aggressive, dangerous dog, and think the dog is mean, dominant, mad, and needs a strong hand (don't even talk to me about the Dog Whisperer). Many solutions people try—yanking the leash, using punishment-based training approaches, forcing the dog into the very situation she fears—will only make things worse. If a child refused to get into the pool because she was afraid of drowning, you would not push her in or punish her until she complied. That would exacerbate the fear she associates with the pool. By recognizing fear as the root of many (obviously not all) aggressive behaviors, you instantly shift your treatment approach. I’ll point you to resources for handling fear aggression in dogs in these links (also see the books I linked to in the second paragraph), because I am certainly not qualified to be dispensing training advice, especially for such a serious issue. Talk to a dog behaviorist who relies on positive reinforcement techniques.

What about dogs who seem shy, timid, or fearful, but haven’t displayed aggression? I think the animal psychology literature makes it clear that you should never force this dog to be around the thing he fears. Don’t bring a visibly shy or stressed dog to a children’s birthday party and hold her down while the kids pet her, don’t tease her with the vacuum, don’t drag her on leash towards the slobbering, friendly Labrador to force her into a head-on greeting. If your dog has an avoidant strategy like running away, continue to give him that option as you work on counter-conditioning and other fear reduction training techniques under the guidance of an expert. A dog who does not feel like she can escape when she is afraid may develop learned helplessness (which is closely tied to depression, 26), but she also has the potential to resort to aggression. In fact, one popular approach to dealing with fearfulness in dogs (Behavioral Adjustment Training) helps people teach their dogs they can always escape to safety (25).

An important personality dimension on which animals differ is their predisposition towards responding fearfully, and this is true for people as well as dogs. Ronan, our dog who had severe fear aggression towards strangers, was generally afraid of everything unfamiliar. After Halloween one year some neighbors left a jack-o-lantern rotting by the side of an apartment building, and Ronan growled suspiciously at its squashed remains (pun intended) every time we walked by it. Lola, our current puppy, is largely unafraid and resilient, and I don’t have to worry that a phobia will pop up after one negative encounter with an object. Knowing where your dog lies on this fearfulness dimension tells you how careful you should be in introducing her to new experiences, as emotional learning is rapid and difficult to undo, regardless of species (27).

What About Human Psychology?

Perhaps fear is at the root of a lot more aggressive human behaviors than we acknowledge. My mom always told me growing up that bullies act out of insecurity, and what is insecurity if not social fear? She may have been on to something in identifying fear, anxiety, and insecurity as the roots of many social and physical aggressions (for a non-maternal reference, see 28). The current political zeitgeist includes vocal hostility towards certain minority members, and is thought by some to be rooted in fear.

When we identify a cruel or malicious behavior as originating from fear rather than anger or contempt, we should shift the intervention accordingly. Responding to hostility with more hostility will always, always backfire, because your hostile reaction increases how threatened the aggressor feels. People who are afraid need to be made to feel safe and need an alternate behavioral solution for remedying their fear besides aggression.

On the theoretical side, I think these ambiguous behaviors that seem rooted in both fear and anger make a strong case that different emotion systems can work simultaneously (9). I think the old notion that someone is EITHER afraid or angry might need to be discarded. 

Fear aggression cases are also an excellent illustration of how innate emotion programs, like fear, can combine with learned behaviors (aggression) to complicate the more old-school version of basic emotion theories, in which the stimulus->emotion->behavioral response/expression sequence is set in stone. In other words, I think emotions are like computer programs that come preinstalled in our systems, but through experience and learning we can make updates to them, turn off some of their settings, or hack them for our own uses.

We see this type of learned modulation of emotional behavior in people all the time, and not just with fear. Sighing is a vocalization associated with affective states, like sadness or stress, and it appears to serve biological functions (29). But many people learn to exaggerate this vocal expression, sighing dramatically to influence others, like a significant other who said he does feel like seeing that movie you so desperately want to see. I've seen dogs (both of mine included) learn to make their whines for food, blankets, toys, etc. more and more exaggerated as the behavior continues to get them what they want. Does this mean there was no adaptive origin of the behavioral tendency? No. But learning and experience are powerful forces in shaping emotion-motivated behaviors.

Emotions motivate animals to act on their environments to achieve a particular end goal. In these fear-driven situations, the goal is to feel safe and avoid pain. People and dogs alike will sometimes resort to extreme behaviors to get there.


  1. Posner, J., Russell, J. A., & Peterson, B. S. (2005). The circumplex model of affect: An integrative approach to affective neuroscience, cognitive development, and psychopathology. Development and psychopathology,17(03), 715-734.
  2. Tellegen, A., Watson, D., & Clark, L. A. (1999). On the dimensional and hierarchical structure of affect. Psychological Science10(4), 297-303.
  3. Russell, J. A., & Barrett, L. F. (1999). Core affect, prototypical emotional episodes, and other things called emotion: dissecting the elephant. Journal of personality and social psychology76(5), 805.
  4. Frijda, N. H., Kuipers, P., & Ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of personality and social psychology57(2), 212.
  5. Roseman, I. J., Spindel, M. S., & Jose, P. E. (1990). Appraisals of emotion-eliciting events: Testing a theory of discrete emotions. Journal of Personality and Social Psychology59(5), 899.
  6. Boissy, A. (1995). Fear and fearfulness in animals. Quarterly Review of Biology, 165-191.
  7. Kudryavtseva, N. N. (2000). An experimental approach to the study of learned aggression. Aggressive Behavior26(3), 241-256.
  8. Moyer, K. E. (2013). Biological bases of aggressive behavior. Biological Foundations of Emotion3, 219.
  9. Panksepp, J., & Biven, L. (2012). The archaeology of mind: neuroevolutionary origins of human emotions (Norton series on interpersonal neurobiology). WW Norton & Company.
  10. Landers, M. S., & Sullivan, R. M. (2012). The development and neurobiology of infant attachment and fear. Developmental neuroscience34(2-3), 101-114.
  11. Stemmler, G., Aue, T., & Wacker, J. (2007). Anger and fear: Separable effects of emotion and motivational direction on somatovisceral responses.International Journal of Psychophysiology66(2), 141-153.
  12. Kohler, C. G., Turner, T., Stolar, N. M., Bilker, W. B., Brensinger, C. M., Gur, R. E., & Gur, R. C. (2004). Differences in facial expressions of four universal emotions. Psychiatry research128(3), 235-244.
  13. Ekman, P. (1992). An argument for basic emotions. Cognition & emotion6(3-4), 169-200.
  14. Panksepp, J. (1998). Affective neuroscience: The foundations of human and animal emotions. Oxford university press.
  15. Keltner, D., Haidt, J., & Shiota, M. N. (2006). Social functionalism and the evolution of emotions. Evolution and social psychology, 115-142.
  16. Panksepp, J., & Biven, L. (2012). The archaeology of mind: neuroevolutionary origins of human emotions (Norton series on interpersonal neurobiology). WW Norton & Company.
  17. Kudryavtseva, N. N. (2000). An experimental approach to the study of learned aggression. Aggressive Behavior26(3), 241-256.
  18. Feshbach, S. (1964). The function of aggression and the regulation of aggressive drive. Psychological review71(4), 257.
  19. Attili, G., & Hinde, R. A. (1986). Categories of aggression and their motivational heterogeneity. Ethology and Sociobiology7(1), 17-27.
  20. Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. Mcgraw-Hill Book Company.
  21. Tinklenberg, J. R., & Ochberg, F. M. (1981). Patterns of adolescent violence: A California sample. Biobehavioral Aspects of Aggression., 121-140.
  22. Moyer, K. E. (1968). Kinds of aggression and their physiological basis.Communications in Behavioral Biology2(2), 65-87.
  23. Ulrich, R. (1966). Pain as a cause of aggression. American Zoologist6(4), 643-662.
  24. Lorenz, K. (2002). On aggression. Psychology Press.
  25. Tortora, D. F. (1983). Safety training: the elimination of avoidance-motivated aggression in dogs. Journal of Experimental Psychology: General112(2), 176.
  26. Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A theory-based subtype of depression. Psychological review,96(2), 358.
  27. Brown, E. J., Heimberg, R. G., & Juster, H. R. (1995). Social phobia subtype and avoidant personality disorder: Effect on severity of social phobia, impairment, and outcome of cognitive behavioral treatment. Behavior Therapy26(3), 467-486.
  28. Donnellan, M. B., Trzesniewski, K. H., Robins, R. W., Moffitt, T. E., & Caspi, A. (2005). Low self-esteem is related to aggression, antisocial behavior, and delinquency. Psychological science16(4), 328-335.
  29. Vlemincx, E., Van Diest, I., & Van den Bergh, O. (2016). A sigh of relief or a sigh to relieve: The psychological and physiological relief effect of deep breaths. Physiology & Behavior.



    A Unified Theory of Yawning

    About this blog