Hi Mark The way your question is formulated turns out not to work out for the intention I perceive you have. You said, "...in the area of performance regarding the number of bit being attended to consciously at any one time..." I assume in the above phrase you mean bits (as opposed to bit) or possibly even bytes. The difficulty is the Miller in his original work was artfully vague about the definition of what a chunk was - you have interpreted this to mean bit(s) or bytes which can be rigorously defined (although it is not trivial to do so in actual systems. The issue that is the wild card here is the definition of chunk, bit and byte as each of these concepts has to be relativized to the information available in the system or set under discussion, and most critically, the code (the vocabulary - repesentational system (here I intend the most general sense of representational system not exclusively V A K)) utilized in the experimental situation. As a simple example: 186000 10100101001110100101011001001 (or whatever the binary representation of the first base 10 number happens to be) speed of light in mph Now note that all three codes or representations carry the same information - however the binary representation is well past the 7 + or - 2 limits identified by Miller. Thus until the code itself is fixed, specified, the question is not well formed. BTW, I find this coding issue fascinating. Read Babel 17 by Delaney for an interesting extended example of coding and its consequences in a fictional form. All the best, John