Extrapolating behaviours – picking the right analogue

Palaeobehaviour is perhaps the hardest area of research in palaeontology, and is certainly one with the most speculation involved. If you know where and how to look, there are all kinds of evidence available in the ichnological and osteological record but of course the real issue is interpreting it correctly and tying the evidence to the right taxon in the right way. Sure you have a nice hadrosaur femur with some tooth marks, but from what? A theropod or a crocodile? And did it kill it, or just find it already dead? Was there only one animal feeding or more? Even with a good fossil record to let you know what other animals were out there, and how many, and what the environment was like, there are generally so many possibilities it can be very hard to make even vaguely reasonable generalisations about behaviour based on a single specimen or even a good collection.

This is not to say that we cannot or should not try to make such inferences, but I think they have often tended to be published with overmuch emphasis on the speculation and not enough on the reasoning behind those ideas. Obviously a balance has to be struck: if you do not indulge in a little extrapolation you will end up with so many caveats, you can’t make any inferences at all (well we have the tooth here from species X buried in the skeleton, but it could be a pathological tooth from species Y, or an unknown other species, or it might have ended up there by accident). One must look at the (often limited) data available, put it in context, get rid of the most unlikely scenarios and then try to apply some basic ecology to the problem, be it from the Extant Phylogenetic Bracket, or analogy with ecologically similar species.

A good example is the predatory behaviour of theropods, something I am working on a lot right now. We have two real choices in living animals to use as proxies for making inferences about their behaviour and neither are especially satisfactory which causes various problems with their use. We can use the EPB and try to reconstruct their behaviour based on their nearest living relatives, but those are crocodilians and birds. It is a bit unreasonable to think that large terrestrial predators like allosaurs would have hunted like the semi-aquatic ambush predators and piscivores that are crocodiles, or like the fully flying predators and scavengers of hawks and vultures. (There are actually a couple of very relevant exceptions to this, but if you know of them, keep it under your hat for now, that is coming in another post). Equally unreasonably in some ways, we can turn to nice extant analogues (large, terrestrial predators that hunt large prey) such as big cats, wolves and hyenas but then we have the problem that they are still rather different animals to multi-ton bipedal Mesozoic reptiles. It is a very hard balance to strike and no matter what you do, someone will be unhappy (as I can testify from reviewers comments).

The biggest handicap to this work however often seems to be misplaced limitations on those behaviours that are inferred from other groups. If you work under the assumption that all raptorial birds are aerial predators, or that all crocodiles are aquatic ambush predators then inferences and extrapolations you make from this will be flawed – whether you try to apply them as part of the EPB or even just to extinct raptorial birds and crocs. In this regard the obvious comment is that people need to be careful when looking at behaviours both those of the extinct animals they want to learn about, but also the living animals they want to use as analogues. Picking one lone example is bad, but just as bad is picking a whole clade and blanketing them under a single heading. I’ve seen plenty of both in the literature and online and while clearly there are times when this is not a problem, or even a good idea, there are I think, far more where it is not. Information online for researchers is increasingly accessible and the excuses available for not checking out just a few relevant species are reducing rapidly. It’s not hard to find basic information on the predatory habits of gharials, dwarf crocodiles and caiman so read it and apply it.

Share this Post

5 Responses to “Extrapolating behaviours – picking the right analogue”


  1. 1 Tom 09/12/2009 at 6:21 am

    Good post. Very helpful; as I am currently writing a novel which contains dinosaurs and I want to betray their behavior as accurately as possible and up-to-date with our current understandings of these creatures.

    • 2 David Hone 09/12/2009 at 8:49 am

      No problem, glad it’s of interest!

      BTW were you the one who left a comment about the correct names for various groups of dinosaurs a while back? I found it lost in my archives and never answered it, sorry. You had mentioned it was urgent at the time so I guessed it was too late – though knowing the editorial process of books, I guess it’s not too bad. If this is still an issue let me know and i’ll dig it up and write something, or you can just e-mail me, or stick something on AAB.

  2. 3 Tom 13/12/2009 at 5:54 am

    I actually don’t think that was me.
    But thanks anyway. Glad to know that you’d go to that effort if I did ask. lol.

    I’ll be sure to comment if you post regarding anything dinosaur behaviors I could feature in my novel, and I need more info.

    • 4 David Hone 14/12/2009 at 8:54 am

      Well you should dig out my posts on theropod behaviour from a few months back where I do talk about this is some detail which is based around a paper i published this year on that very subject.


Comments are currently closed.



@Dave_Hone on Twitter

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 455 other followers


%d bloggers like this: