# Information complexity of query algorithms?

Information complexity has been a very useful tool in communication complexity, mainly used to lower bound the communication complexity of distributed problems.

Is there an analogue of information complexity for query complexity? There are many parallels between query complexity and communication complexity; oftentimes (but not always!) a lower bound in one model gets translated to a lower bound in the other model. Sometimes this translation is quite nontrivial.

Is there a notion of information complexity that is useful for lower bounding the query complexity of problems?

A first pass seems to indicate that information complexity is not very useful; for example, the query complexity of computing the OR of $N$ bits is $\Omega(N)$ for randomized algorithms and $\Omega(\sqrt{N})$ for quantum algorithms, whereas the most straightforward adaption of the notion of information complexity indicates that the information learned by any query algorithm is at most $O(\log N)$ (because the algorithm stops when it sees the first $1$ in the input).