We consider a framework for  information collecting and
processing more general than that examined by Belnap, which gave 
rise to the famous  4-valued Dunn-Belnap logic. The  framework 
consists of a set of information sources S and  a processor P. 
The sources provide information about formulas of 
propositional classical logic L_C, and we assume that
for each such formula F, a source s in S can say that F is true, 
that F is false, or that it has no knowledge about F.  In turn,
the processor collects information from the 
sources, combines it according to some strategy, and defines the 
resulting combined valuation  of formulas in L_C. 

This general framework has many specific variants, corresponding 
to the assumptions made about the logical properties of the sources, 
the properties of the whole set of sources and the strategy used 
by the processor in combining information. In particular, Belnap 
assumed that the sources provide information about atomic formula 
only, which  the processor combines to yield information 
about complex formulas. In our approach, the sources can also provide 
information  about complex formulas. This generalization is made 
possible by the use of non-deterministic logical matrices (Nmatrices) 
instead of the ordinary ones.