This question involves a period before historical documents, and so you might be better off asking it of /r/askanthropology. In 1974-1975, I developed a matrix program (using Fortran, which is also prehistoric!), to evaluate Upper Paleolithic cave art (and then a subsequent attempt to evaluate mobilary art from a site in the Pyrenees). The majority of animals represented were herd herbivores including aurochs and horse. These were usually found deeper in caves. Bears, which were also deep within caves, were relatively rare and were the most likely to be "killed": that is, shown with what appeared to be blood flowing from mouth and nose. Predators including lions and owls were often shown at the entrance of caves. Deer, ibex and other herbivores were rarer and more difficult to reach a conclusion about placement. Depictions of humans were extremely rare, although hand prints and silhouettes were common.

Ultimately, my matrix program failed, because while it showed some relationships that exceeded the standard deviation from the mean, interpreting or understanding what those deviations meant placed me in the same subjective, interpretative, place that had proven the ruin of so many others who tried to understand what the art intended. Without records, we fly blindly into the past.

edit: by way of sources, I was always intrigued by the work of Alexander Marshack (1918-2004) and his "The Roots of Civilization" (1972) because of his attempt to see indications of lunar calendar marks - groups of seven and 28 in notches associated with the art. Modern work tends to focus on the idea that the sudden development of art in the Upper Paleolithic indicates a shift in human culture that is the most meaningful thing we can conclude from the art - not what it specifically means, but rather that its appearance at that time suggests by way of a new approach to culture and creativity. The wiki site is suspect, but it includes some more recent references that you might find useful.