At my invitation, Wiseman and Smith carried out 4 videotaped experiments with a dog called Jaytee, with whom I have carried out more than 100 videotaped experiments (Sheldrake, 1999b). My experiments showed that Jaytee usually waited by the window for a far higher proportion of the time when his owner was coming home than when she was not. This occurred even when his owner, Pam Smart, came at non-routine, randomly-selected times and travelled by unfamiliar vehicles such as taxis. This pattern was already clearly apparent months before Wiseman et al. carried out their tests.
The abstract is only short, but that doesn't mean that you have to cram as much detail into it as possible. What you want is to grab the reader's attention with the first statement, add a few of the most important details, then leave them with the overall message of the manuscript in the last sentence. In this way it is similar to a news article. Have a look at some articles in a newspaper. Most often the first sentence contains the crucial information about the story and then the details follow after that. This is a good model for your abstract; after all, what you're writing is news for the scientific community.