In short, they test a bunch of samples by cold drawing them in one direction, in alternate/reversed directions, and in two other patterns of drawing one way and the other. They concluded that cold-drawing copper wire creates a preferred orientation of unit cells in the core. However, the conductivity decreased with more frequent reversals of the drawing direction. Actual variations appeared to be on the order of about 0.20 percent and the differences in drawing were found not to affect the conductivity of annealed wire. Annealing the copper wire after cold-drawing increases the conductivity.
My take on the bulletin is that the authors determined there are indeed surface and cell differences that can occur based on how copper wire is processed. These differences can be observed by x-ray spectroscopic examination (and possibly electron microscopy?) and can cause very small variations in conductivity that could conceivably be construed as directional differences. However, the differences are erased when the wire is annealed, which is a process applied to all (almost all?) the wire we use. In any event, even for non-annealed wire, the differences measured were so small that it would be hard to imagine that somebody could discern a sonic difference in music played through a home audio system due to the wire direction or drawing process.
Does anyone here know of other scientific papers about wire directionality/conductivity relative to processing, from authors without a dog in the fight (i.e., from anyone other than a cable manufacturer or their surrogates)?
http://www.michigan.gov/documents/deq/Bulletin_MCMT_CopperWire_opt_301717_7.pdf