Her research, as it was presented yesterday, centered on the study of human information exchanges, and the behaviors that are associated with those exchanges. As of late, she has partially based her research on the expressions that people make through language through other means such as deception, ambiguity, and identifying the effects of credibility and emotions on the believability of information. The emphasis this year is on identifying misleading news (yes, "Fake News") with text analytics systems.
-Dr. Rubin and her team have been working on trying to create an algorithm that is more capable of detecting deception or not truths in textual statements processed by that algorithm. Over lunch, she explained to me, SU Professor John Glass and a few other students that the term "lie" has a pejorative tone, that it is an act of malice, and thus her team prefers deception and not-truths as research terms. However, she recognized that in the real world, truths and not-truths do not exist in a binary state, where it is one or the other. Whether we may want to believe it or not, there is a graduation that exists, between true and false and, similarly, the way we view or receive these falsehoods also varies. For example, she mentioned satire as a field during her presentation, and specifically mentioned no longer on-air The Colbert Report, as being an instance in which we were somewhat accepting of lies as they were told, as we approved of the result of the lie. Using The Colbert Report as an example, she described how Conservative viewers would likely be angry at the show, as they would lock onto the lies that were told and would identify them as malicious, while Liberal viewers would likely view them as satire, as something to be laughed at, a tongue-in-cheek statement.
-This concept, of course, becomes a lot less fun in the current media climate. According to a 2016 Gallup Poll, "Americans' trust and confidence in the mass media 'to report the news fully, accurately and fairly' has dropped to its lowest level in Gallup polling history, with 32% saying they have a great deal or fair amount of trust in the media. This is down eight percentage points from last year." Reporters, journalists, photographers and all sorts of other media workers are verbally berated and sometimes physically attacked while doing their jobs. This idea of fake-news and malicious half-truths and non-truths has become so ubiquitous that we all accept it as true, despite the desire to roll our eyes and shake our heads whenever we hear it brought up at the dinner table. That's why it is critical that research such as that being done by Dr. Rubin et. al. is so important. She says that humans are notoriously bad at detecting deception, with the success rate hovering around 54% of the time. In other words, in optimal conditions, its basically a flip of a coin that determines whether a human will detect a deceptive statement. By contrast, in similarly controlled environments, their algorithm has been able to reach a rate of success in the 65-75% range, when presented with choices between true and deceptive texts. If Dr. Rubin and her team are successful in developing an algorithm that can determine the veracity of online statements, the potential benefits to society are innumerable. This could change political dialogue, alter the discussion of medical practices like abortion and vaccination, and innumerable social fields. It could alter any number of other fields.
-So what does Dr. Rubin hope can be done? Not every website can be The Onion. Very few websites are as overt with their satire, and fewer have established themselves as the smart-alecs of the internet like they have. These humor/satirical sites are treated and regarded accordingly. But what happens when a site that is less-well known publishes something that seems like satire? Without research, it is hard to determine the veracity of that text. Dr. Rubin briefly explained the principle of truth bias, which is an innate drive in humans that predisposes us to believing what we hear, and is a really interesting field of psychology. This same wiring means that we will believe what we read online. We can credit this idea with the fact that our society functions as well as it does, on the micro level. If we can add some element of truth-detection to the internet, that ever-growing, ever-advancing, and ever more-important pillar of our society, then we will be able to detect deception and illusion before it ever has a chance to spread online, and into our discourses. It is a fantastic field of study, and I wish Dr. Rubin and her team all the possible successes in their work.
You can read Dr. Rubin's blog here.
This blog was originally created for coursework while at the Newhouse School. But this blog quickly became something more, as it gave me a platform, however small, on which to share my thoughts. So feel free to read through these posts, and if you like what you read, please share. Enjoy.