Theoretical predictions strongly support decision accuracy as a major driver of ecological specialization?

[raid517]

OK I am probably going to feel dumb asking this but I have been given this paper to read: http://www.pnas.org/content/early/2009/03/20/0807247106.full.pdf#page=1&view=FitH . However I'm not confident that I fully understand it. In particular I don't feel I fully understand the two central premises of the paper, which are:

 

a) We demonstrate that specialists make more accurate resource-use decisions than generalists when the consequences of using a non-host are neutral or positive but not very positive.

 

and

 

b) Pronounced unsuitability of non-host resources in fact promotes higher decision accuracy in generalists.

 

The paper then goes on to explain the following:

 

The predictions of the model are shown in Fig. 1 and can be summarized very succinctly. Extreme specialists outperform extreme generalists cognitively when, and only when, the fitness consequences of using a non-host resource are neutral or slightly

positive (specialist advantage has all but disappeared when non-host choice is approximately equal to half of the reward for

host choice). If choosing a non-host resource has negative fitness consequences, generalists always outperform specialists.

 

I can only say I find the above statements confusing. Firstly what does "out perform cognitively" really mean? How can one organism be judged to be outperforming another on a purely cognitive basis, and how does this aid the process of the evolution of of specialists and generalist species?

 

Secondly decision accuracy and speed are (presumably, given this paper) enhanced when a specialist encounters a non-host resource that is only minimally rewarding?

 

But why would a species adapt to use only a minimally rewarding - non-host resource when there are plenty of other more abundant and better resources it could chose from? Would not the reverse be true, that specialisation may occur in circumstances when a species encounters a maximally rewarding resource? For example say a species adapted to use a single resource that met all of their dietary and nutritional needs, why would the presence of other less rewarding resources drive evolution in the direction of specialisation? Is this paper saying that although a species could live off of a wide variety of resources within it's environment, that it benefits somehow from not exploiting these?

 

Lastly the second conclusion is even more confusing than the first, in that it states:

 

If choosing a non-host resource has negative fitness consequences, generalists always outperform specialists.

 

To my undoubtedly small mind, I can't figure out what this means at all. In this case how why would a generalist choose to exploit a non-host resource that has negative fitness consequences for doing so? What exactly is a negative fitness consequence in this case? Could anyone please break this down into plain English for me?

 

I should perhaps point out that this is just a model and that the researchers plan an extended field trip later in the summer to test the model out. To this end I would be happy if anyone was able to offer some meaningful criticisms of the paper. One criticism I did encounter was that neural network modelling on its own has fallen out favour somewhat over the last decade in biology, as it is now thought that other more recent mathematical/computer modeling techniques provide a more accurate picture of how the processes of generalisation and specialisation occur. But clearly until I understand what the paper is saying, can't give very many other useful criticisms of my own.

 

Any help would therefore be very much appreciated.