A two-year-old Border Terrier kept whipping his head around to look behind him, as if something had moved. Sometimes he snapped at the air. Then he would hide, stare into space, and briefly seem to switch off.
A one-year-old poodle mix licked his lips, stared, and snapped at things nobody else could see. A four-year-old mixed-breed dog became anxious and unusually clingy, made repetitive mouth movements, and drooled clear saliva from the right corner of his mouth only, enough to leave the fur on that lip visibly wet.
None of them convulsed. All three had epileptic activity on their brain recordings.
Episodes That Looked Like Fear, Not Fits
The case series was published in August in Frontiers in Veterinary Science by Adriana Czerwik, Daniela Farke and Martin Jurgen Schmidt of the Small Animal Clinic at Justus Liebig University in Giessen, Germany, working with Agnieszka Olszewska of a specialist veterinary center in Cologne.
Episodes lasted a few minutes. For the terrier and the poodle mix, they happened almost daily. For the mixed-breed dog, they came every one to two weeks. Owners could not interrupt or distract the dogs while an episode was underway, and there were no warning signs beforehand or grogginess afterward.
Clinical and neurological examinations were unremarkable in all three. Bloodwork and cerebrospinal fluid analysis were normal. Brain MRI found nothing. A blood panel testing for the neuronal antibodies associated with autoimmune encephalitis in humans came back negative in every dog, which does not rule out an autoimmune process but makes it less likely.
The only abnormal finding in the entire workup was on the EEG.
What the Wireless Recordings Picked Up
Each dog wore a wireless EEG system for between 1.5 and four hours, with a sedative used only to place the electrodes. The recordings were reviewed by one veterinary neurologist and then re-analyzed blind by a second.
All three showed interictal epileptiform discharges, the between-seizure electrical signature of an epileptic brain. In the Border Terrier, the discharges pointed to both temporal lobes and the right parietal region. In the poodle mix, they localized to the left temporal lobe. In the mixed-breed dog, they came from the left parietal region.
In humans, the temporal lobe is the most common site of focal epilepsy, and temporal lobe seizures are the ones most associated with fear, confusion, repetitive mouth movements, and autonomic signs rather than convulsions. The dogs' behavioral and autonomic episodes mapped onto that pattern closely enough that the authors raised temporal lobe involvement as a possibility.
Human focal seizures are classified by whether awareness is preserved, and people having them may stare blankly, appear confused, or smack their lips. Veterinary neurology deliberately avoids the human concept of an aura, because a dog cannot report a subjective warning sensation. That gap is part of why these episodes are so easily filed under behavior.
The team also flagged the limits of their own method. There is no standardized electrode system for dogs comparable to the human 10-20 montage, and skull shape and muscle thickness vary enormously across breeds, so anatomical localization in dogs stays presumptive.
One Dog Kept Having Episodes, and Fluoxetine Did Not Fix It
All three started phenobarbital and all three improved. The Border Terrier's episodes stopped entirely after four months. The mixed-breed dog achieved complete seizure freedom, still holding at a phone follow-up three months later.
The poodle mix was the exception. His episodes worsened again after three months despite dose increases and drug levels in the therapeutic range. Levetiracetam was added, which reduced intensity but not frequency. Roughly a third of dogs with idiopathic epilepsy develop drug-resistant epilepsy, a figure close to the human one.
Because the fly-catching behavior persisted, the team added fluoxetine, following an earlier multicenter evaluation of 24 dogs suggesting that fly-catching syndrome may respond better to an antidepressant than to phenobarbital. In this dog, it produced a brief, modest drop in intensity that faded within a month. His owners chose to stay with the phenobarbital and levetiracetam combination, which had taken him from almost daily episodes to roughly one a week.
Why This Is a Case Series and Not a Diagnostic Rule
Three dogs cannot establish how often behavioral episodes in dogs turn out to be seizures. The authors are explicit about what they did not capture. No actual episode occurred during any EEG recording, so the findings are between-seizure discharges rather than proof of a seizure caught on tape.
They also acknowledge that residual effects of the sedative used to place electrodes cannot be entirely excluded from the early part of each recording, and that they did not quantify discharge frequency.
The practical argument is still worth something. Veterinary consensus documents on epilepsy classification in companion animals already recognize that seizures can be focal and non-convulsive, and a recent large survey that set out to characterize canine seizure semiology identified episodes of impaired awareness as one of four distinct categories. Those are exactly the events most likely to be logged as anxiety, a phobia or a quirk.
The authors also point to a separate report describing four related Labrador retrievers with panic-attack-like episodes tied to a gene variant, though those dogs had visible MRI abnormalities that the Giessen dogs did not. Taken together, the reports make the same argument from different angles: behavior can be the main event in a seizure rather than a side note.
Dog owners noticing repeated episodes of staring, air snapping, sudden fear or one-sided drooling should describe them to a veterinarian, ideally with video. Nothing here means such behaviors are usually epileptic, and this series does not change treatment guidance.
Key Questions Answered
What did the dogs actually do? They had recurring episodes of fear, restlessness, hiding, staring into space, lip licking, air snapping and, in one dog, one-sided drooling. Two showed signs of impaired consciousness.
What is a focal impaired awareness seizure? It is a seizure that starts in one part of the brain and alters consciousness without necessarily producing convulsions. In humans, it can involve fear, confusion, staring, and repetitive movements.
What did the EEG show? Interictal epileptiform discharges in all three dogs, localized to temporal or parietal regions, depending on the animal.
Did treatment work? Two dogs became free of episodes on phenobarbital. The third improved but continued having episodes despite added levetiracetam and fluoxetine.
Does this prove behavioral episodes in dogs are seizures? No. This is three dogs; no episode was recorded during EEG, and the authors present the findings as supportive rather than definitive.
What should owners do? Report repeated unexplained episodes to a veterinarian, with video if possible, rather than assuming the behavior is anxiety or a habit.
