Note: This is somewhat of a digression on what I was doing but I thought I'd post it and see if anyone had any feedback.
I read a research study a few months ago that presented people with some known falsehoods as fact. Then these same people were then informed that the facts they had been told were actually false and then they were provided with the actual facts. Then each person was tested on the information that they had been given and scored on correctness. The researchers found that a surprising amount of people maintained the initial information as fact even after being corrected. Some of the people even dismissed the facts as the incorrect information.
The study seemed to conclude (damn where is it...anyone familiar with this study?) that the reason for this was that people had an easier time learning information than they had re-learning information or changing their mind. I think the initial information was also enforced by a confirmation bias.
With the results of this study in mind I began to wonder how this impacted a persons education and if or to what extent is behaves in propaganda. Since even the ordering of how information is presented is statistically significant then the influence of bias appears to be virtually inescapable. (Conditioning could be a viable alternative perhaps...)
So how would this would play out in a simple thought experiment using "Heads or Tails?" as the initial information?
If we are to follow the predictions of the experiment and assume only two mutually exclusive possible outcomes exist then we are to assume that a majority of people will call heads even though the probability of each side being correct is 50/50. So conversely the opposite would seem to be true if we phrased the question "Tails or Heads?", however in this example context seems to matter. Since most people would seem to recognize that tails is traditionally mentioned second then I predict that tails will not only be called by a majority but it will be called by a greater majority than heads was in the first example. The reason for this is that not only does tails enjoy an advantage by being listed first but it also benefits from an inaccurately perceived underdog bias. So it receives the actual advantage and an underdog / contrarian bias.
However, if we are to continue with the "Heads or Tails" example then some people may have made their decision prior to the question, something of a team favorite. I predict this is less of a factor than an advantageous position or an underdog bias because a perceived shift in power will cause heads to lose a greater majority of fans than tails. Perhaps this could be a driving force behind bandwagon trends.
In the end I predict a perceived underdog with the actual advantage could be a statistically important passive bias. The interesting part of this bias is that it is passive and is not necessarily observable through syntactic analysis. I think that it is a bias that many people have passively deduced and even act upon but some may not. So my suggestion is to not only be aware of the order that topics are presented but be especially perceptive of an underdog that is presented first in an argument.
I can't say that I remember studies taking this bias into account, but to do so should be fairly simple. Just randomize the order of questions and answers even on something as basic as forms needs to be randomized or evenly distributed. Half of the people should get "A or B", and the other half should get "B or A".