Can you interrupt the behavior?

If not:

These episodes could be partial seizures (taking place in only parts of the brain, not the whole brain). If so, likely located in the occipital lobe (vision, if occipital lobe impaired symptoms are distorted vision/ hallucination)

http://en.wikipedia.org/wiki/Complex_partial_seizure

If occipital lobe:

http://en.wikipedia.org/wiki/Occipital_lobe

Last paragraph in that entry:

"Epilepsy and the occipital lobe

Recent studies have shown that specific neurological findings have had an impact on idiopathic occipital lobe epilepsies[4] Occipital lobe seizures are triggered by a flash, or a visual image that contains multiple colors. These are called flicker stimulation (usually through TV) these seizures are referred to as photo-sensitivity seizures. Patients who have experienced occipital seizures described their seizure as seeing bright colors, and having severe blurred vision (vomiting was also apparent in some patients). Occipital seizure are triggered mainly during the day, through television, video games or any flicker stimulatory system.[5]Occipital seizures originate from an epileptic focus confined within the occipital lobes. They may be spontaneous or triggered by external visual stimuli. Occipital lobe epilepsies are etiologically idiopathic, symptomatic, or cryptogenic.[6]Symptomatic occipital seizures can start at any age, as well as any stage after or during the course of the underlying causative disorder. Idiopathic occipital epilepsy usually starts in childhood.[7] Occipital epilepsies account for approximately 5% to 10% of all epilepsies.[8]"

It has been hypothesized that migraines and seizures are closely related, and many migraines get triggered by the same things that trigger occipital lobe seizures.

...if it is indeed epilepsy with partial seizures:

Seizure meds are inexpensive, and acc to the literature, most dogs respond. I had one foster dog myself who was an epileptic, and he has been fine on the meds for years, as do the others that I watched in rescue. I would not do an MRI, unless there are plenty of funds, usually nothing can be seen. If anything diagnostic, then an EEG. Acc. to Dodman, most but not all dogs with seizures show the patterns of epilepsy in an EEG. None was done with my foster dog.
I'd take her to her vet again and discuss this possibility.

Occipital lobe seizures is a guess based on what you describe. More on the triggers for partial seizures in the occipital lobe: http://en.wikipedia.org/wiki/Photosensitive_epilepsy

In your log of her episodes, I'd include factors such as
exposure to TV, computer games (the moving dots), fluorescent lights incl. energy saving bulbs.
I doubt adorable Ms Moo goes to a disco with a disco ball!

And, if the setup of your house permits it, keep her as far away as possible from these stimuli for a week or so, with a closed door in between, relax, and see how she does.

This is something for her vet. In specialty veterinary medicine, epilepsy seems to be in the expertise of both veterinary behaviorists (especially if they have access to the apparatus for an EEG) and neurologists. But most dogs with seizures are treated by their regular vets.

I found this list for resources on epilepsy:
http://www.berner.org/pages/neurological_disorders/canine_epilepsy_links.php

I looked for yahoo groups, and there seem to be 2 in the US and one group in the UK. They can be great or horrible....

Clinical use of an herbal-derived compound (Huperzine A) to treat putative complex partial seizures in a dog.
Schneider BM, Dodman NH, Faissler D, Ogata N.
Clinical Sciences Department, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA, USA.
Epilepsy Behav. 2009 Aug;15(4):529-34. doi: 10.1016/j.yebeh.2009.06.011. Epub 2009 Jul 17.

A Bernese mountain dog was diagnosed with complex partial seizures that were supported by electroencephalographic findings. Clinical signs of the problem included "star gazing," fly snapping, licking, vacuous chewing, and ongoing anxiety. Treatment with Huperzine A, a compound isolated from Chinese club moss with NMDA receptor blocking activity, anticholinesterase activity, and anticonvulsant properties, produced useful suppression of the abnormal behavior for more than months. A relapse occurred when the dog was treated with tramadol for joint pain and the improvement that had been made was not recaptured with Huperzine A. At this stage, phenobarbital therapy was instituted and the dog improved greatly. The role of Huperzine A in controlling seizures is discussed.

http://www.ncbi.nlm.nih.gov/pubmed/19616481