To have a sense of how long it took they would still need imput from their legs. Standing still can not be counted. Only movement counts.
It seems reasonable that an insect brain could have a basic connection between time and it's own movement forward. At least, that seems no more of a stretch than counting.
If they are counting, then how high can they count? What if the food source is ten thousand steps away? What number system so they use? Is it base ten, or binary? Do they count steps, or leg cycles or body lengths or do they simply count seconds (and we're back to time)?
At the very least, I don't think the conclusion the media reported that these researchers arrived at, can be derived from the bounds of the experiment.