Food quality

During a sensory comparison, assessors see package labels naming the experimental treatment on each sample. What change would best reduce expectation bias from those labels?