Because it is frightened, not because it is angry or spiteful. Separation anxiety is a fear response to being left, and the destruction, howling and soiling are symptoms of panic — not a decision your dog made about you.
Veterinary behaviour describes canine separation anxiety by three signs that occur in the owner's actual or virtual absence: destructiveness, inappropriate elimination and distress vocalisation. Supporting signs include hypersalivation, anxiety in response to pre-departure cues, rearranging of household items, and hyperattachment during ordinary interaction.
The word virtual matters. The dog does not need you to leave the building — for some dogs a closed door between you is enough, which is why owners who say "he is fine, he just hates the bathroom door" are often describing the same condition at an earlier stage.
Destruction concentrates at exit points: windows, doors, the places a body would leave from. A dog chewing the doorframe is not redecorating. It is trying to follow you.
Fear has a specific definition in veterinary behaviour: an emotion that induces an animal to avoid situations it perceives as dangerous. The emphasis on perception is deliberate — the fact that an owner does not find the event frightening does not make the animal's fear less real. The pet's perception is its reality, and that is what it acts on.
Your front door closing is not dangerous. Your dog's nervous system does not need it to be.
The most persuasive evidence for the "he knew what he did" theory is the greeting: the crouch, the averted eyes, the slow tail. Alexandra Horowitz tested it directly, manipulating whether dogs had actually disobeyed and whether owners believed they had.
What humans read as guilt tracked the appearance of a cross or scolding human — not the misdeed. The study is careful about what it does not claim: it does not show dogs cannot feel guilt, only that the look follows your demeanour rather than the four hours you were gone.
So the greeting is evidence about the last thirty seconds. It is not a confession.
Estimates disagree, and the disagreement is informative. Large general-population studies reviewed in JAVMA found prevalence of 17.2% and 13%, with the authors noting that owner under-recognition likely produced an underestimate. AVSAB puts it at roughly 15–20% of dogs in the United States.
Against those, a study of 13,715 dogs across 264 breeds found separation-related behaviour in 5% — well behind noise sensitivity at 32% and general fearfulness at 29%.
The spread is the detection problem showing up in the data. Studies that depend on owners noticing find fewer cases, because a silent dog that stops eating leaves nothing behind to notice. The figure with no ambiguity is the specialist caseload: 20–40% of dogs seen by a behaviour specialist are diagnosed with it.
Why Your Dog Panics When You Leave is a 53-page guide in nine lessons, with every claim sourced in the margin of the page it sits on: finding your dog's threshold and protecting it, the training absence from one second upward, what crates and second dogs and bark collars actually do, and where medication fits. It comes with the Absence Ladder spreadsheet you log each departure in, and the owners' community.
$45.95 today, then $4.00 a month to keep access. Cancel anytime.
Get the guide ›