Humans don’t have to directly experience trauma to be deeply affected; seeing someone else go through a distressing event can often be equally stressful. The vicarious experience is often sufficiently stressful to produce negative long-term effects resulting in PTSD, depression and other mood disorders. For humans, seeing stress is often the same as feeling stress.
To study the biological basis of vicariously acquired stress disorders in humans we need a good animal model but our models are based on physical stress. The first question then becomes, is there an analogous response to vicarious stress in animals. To answer the question researchers subjected a group of mice to the physical stress of social defeat while another group witnessed it.
One group of mice were put in the same cage with a larger (40g vs 28g) aggressive mouse (CD-1) and the smaller mouse was subjected to physical stress (PS) via social defeat while the emotional stress (ES) group was able to observe from an adjacent cage.
With the aggressive CD-1 mouse in a cage, the smaller PS mouse (C57BL/6J) is placed into the same container. CD-1 mouse then attacks PS mouse. Under natural conditions the weaker mouse would quickly retreat, in the confines of the cage he is under the mercy of the dominant-aggressive CD-1 mouse. The socially defeated PS will avoid social contact with the aggressor, avoid moving and adopt a defensive posture. After 10 minutes ES mouse is moved to a new cage away from CD-1 and PS and PS is moved to cage adjacent to CD-1. This was done for 10 days but each day the ES mouse witnessed the defeat of a new PS mouse by a new CD-1 .

A) Weight-loss associated stress in PS and ES cohorts. It takes 5 days post treatment for weight to normalize.
B) Acute (40 minutes after first day) and Chronic (24h after last day) levels of corticosterone. Emotional and physical stress builds up over time.
Assays
Forced swim test – mice are forced to swim in a water-filled cylinder. They eventually stop trying to scape and become immobile. Treatment with antidepressants reduces immobility and extends time spent in escape behavior. It is used as a model of depression.
Elevated Plus Maze – measures effect of anxiety on exploratory behavior. A cross “+” platform 1 meter off the floor with safety walls on one axis and no walls on the other. Anxiolytic treatment increases time spent in open arms.
Sucrose Preference – examines tendency for self reward and insensitivity to reward. Mice are given option of water and sucrose spiked water. Insensitivity to reward is associated with depressive-like states.[I’ve noticed similar behavior in aversively/dominance trained dog]
Social Interaction – a measure of social anxiety that uses the animal’s tendency to interact and explore a target mouse behind a divider; the test is run 2X, once without a target mouse. Physical stress reduces the tendency to interact and increases social avoidance; anxiolytic treatment increases the time spent interacting. The canine equivalent would be a dog investigating another dog behind a fence.
Results and Discussion
The mice underwent short-term (24 hours after last day of stress) and long-term (30 days after last day) testing. For all graphs: short-term right, long-term left.

661 changes in gene expression that are common to ES and PS. Other studies report gene expression alteration in hippocampus and amygdala.
The results indicate that seeing stress results in physiological, behavioral and genetic changes that are comparable – in the short and long term – to the effects of physical stress and social defeat.
In the long term (30 days after last exposure) ES and PS mice showed social avoidance, elevated corticosterone, higher levels of anxiety as measured by EMP and increased immobility (depression-like behavior) in the forced swim test.
The ES group showed no loss in reward sensitivity (sucrose preference) at the end of the stress test but 30 days later they had reduced sucrose preference similar to the mice that had undergone physical stress. This indicates the effects of stress continue to build and develop in the absence of exposure to stress. And a reason why so punishment/balanced trainers complain about their dogs being unmotivated by rewards – it’s anhedonia.
These findings add supporting evidence to the neurological explanation for abnormal behavior associated with PTSD, depression and other disorders. It also makes the habit of some dog trainers who punish stress-related and stress-induced behavior seem ever more futile, counter-productive and cruel. It also addresses a common complaint among balance/aversive trainers as to why their dogs don’t seem motivated by reinforcement.
Does this indicate a biological origin for animal empathy? That question is outside of the paper’s scope and it is not explored by the authors. However if an animal can be stressed by witnessing trauma in another then it’s possible that empathy is a way to alleviate its own stress. .
Assuming that dogs (like humans and mice) are also affected by witnessing (olfactory/auditory/visual) stress, what does this mean for shelter dogs? Should shelters take action to minimize the risk of anxious/aggressive dogs from affecting others? I think it’s something that deserves study.
And If you haven’t figured it out by now; dominance/aversive/alpha training is based on social defeat.
Related Links
Social Defeat Changes Young Brains (Psychology Today)
REFERENCES
Warren BL, Vialou VF, Iñiguez SD, Alcantara LF, Wright KN, Feng J, Kennedy PJ, Laplant Q, Shen L, Nestler EJ, & Bolaños-Guzmán CA (2013). Neurobiological sequelae of witnessing stressful events in adult mice. Biological psychiatry, 73 (1), 7-14 PMID: 22795644
Pingback: Seeing stress is just as stressful as experiencing it first hand? - Pet Forums Community
Pingback: Stress Can Hurt… Literally | Science of Dogs
Reblogged this on Dissecting Behavior and commented:
I agree with the author, these kinds of findings present tremendous implications for shelter dogs and questions for future research. Thank you Science of Dogs.
Thank YOU for SPARCS conference — lost a lot of lab time but it was worth it.
Pingback: Seeing Stress Means Feeling Stress. | Canid Science Library
Is a really interesting article. I just have a issue I’d like to discuss. Most of the time, experimenters measuring stress related behavior do it through inescapable aversive events, which lead to learned helplessness, and we (dog trainers and public in general) generalize the findings to ‘stress’ in general. I think that’s a mistake, since there can be more ways in which we induce stress and probably it would’t lead us to the same findings. One thing is stress, in general, other thing stress induced by inescapable aversive events. I’d love to see your thought about this.
Stress is not only measured with inescapable aversive events. There is a rich literature in behavioural adaptations, role of instrumental learning, GxE, coping styles, etc.
Reblogged this on Lady Chauncey Barkington III's Finishing School For Dogs and Other Beasts and commented:
More on why coercion in animal training just doesn’t make for a highly conductive learning experience.
Any studies done on how long it takes for the stress effect to wear down?
Not here or this type of stressor. However I wrote on restraint stress and many of the papers cited do track stress indicators.
https://dogbehaviorscience.wordpress.com/2012/08/06/click-for-health/