Experimenters were blind to the stimulus
in 82% of trials and analyses showed no
difference in subject responses according
to this factor (see the electronic supplementary
material tables SII and SIII).
So, it didn't even matter. Pretty much everyone that would be doing research of this sort is familiar with Clever Hans, so I'm sure they'd be extra careful when designing their experiment.
Indeed. One wonders if the person holding the horse and the person presenting the face photos could see if the photo was of an angry or other face.