If we knew exactly what Recuva was doing them we might be able to explain this. It may be a number of things, for instance during the scan counting all the files found both in the directory and in the deep scan (large number), and then discounting those in the deep scan which already exist in the directory (smaller number).
Maybe, that could be an explanation. However the first scan I did was not deep, and found nothing at all. Have since done 3 deep scans, all the same results as above. And still why is it bringing back the old stuff that was recorded over (the whole shoot in tact) and not a single picture from either of the newer shoots?