Whenever we talk about fake news, what most people have in mind is information that is undeniably false. Unfortunately, the reality isn’t quite that simple. News that are outright fake are easy to detect and debunk, but most misinformation that are difficult to dispel often contain some truth in them, making it hard for artificial intelligence to pick them out, or for legislation to definitively call its inaccuracy.
In this first part of a mini-series, I will be sharing about the types of misinformation that people are susceptible to, some examples of such misinformation, and the cognitive biases that are in play. In the rest of the mini-series, I will talk about the circumstances that give rise to susceptibility, and the reasons that make people susceptible.
Types of Misinformation
As mentioned, determining what is misinformation is not always black-and-white – a lot of it lies in the grey area. First Draft, a non-profit organisation that is attempting to tackle misinformation by supporting journalists, academics and technologists, has categorised misinformation into 7 different types.
As can be seen from the infographic, the types of misinformation lie on a scale indicating the intent to deceive, from satire or parody that is meant for entertainment, to fabricated content that has been intentionally designed to deceive, also known as disinformation. The other types of misinformation contain or originate from some genuine content, but have been modified or misused either deliberately or unintentionally.
This brings us to the issue of falsehoods being technically true but factually false, as well as technically false but factually true. An example of something that is technically true but factually false is one where a statement claims the government is losing $5 billion a year in investments, when the government is actually losing say $4.95 billion a year. The statement is technically right to say the government is losing money, but factually, the amount is wrong. On the other hand, an example of something is technically false but factually true is one where the leader of a state has been praised to have the heart of a lion, but alternative media twists the words and say the leader has been compared to an animal. Such a statement would be factually right since a lion is an animal, but technically wrong because that wasn’t the original meaning of the praise.
The above examples may be oversimplified and a little extreme, but they illustrate the tricky problem of defining what is a falsehood. It also helps to explain why falsehoods can be so convincing, because some truth is often muddled in the information.
Resurfacing of Old Articles
A classic example of how genuine content can be used to mislead people is through the resurfacing of old articles. Old articles are actual news content, but when they are irrelevantly resurfaced in response to a more recent incident, interpretation of the old article becomes out of context, and emotions can be unnecessarily flared.
One such case happened after the Boston Bombing in 2013, when a Daily Mail article about an Afghan bombing by the US military was widely shared over social media. Many readers who shared the article probably only read the headline, and believed that the Afghan wedding bombing happened around the same time but was less publicised compared to the Boston Bombing. However, the truth is the Daily Mail article was actually more than a decade old.
Just to clarify, this isn’t to say that the Afghan wedding bombing that happened in 2002 was any less important. But the intention of resurfacing a dated and not quite related incident is rather questionable. It certainly resulted in quite a bit of anger and confusion, which is not the most helpful during a period of chaos. Then again, to say that the Daily Mail article is false isn’t quite right either, since it was an incident that actually happened. One can only imagine how difficult it would be for artificial intelligence to pick this up, if it was just purely looking for fake news.
Thankfully, some news outlets are starting to recognise this problem, and have implemented measures to counter it. The Guardian is one such agency that has done something to make the age of their articles more salient. Since April 2019, The Guardian has started adding an eye-catching age stamp right at the top of an old article in a bright yellow box; and on social media, the year in which an article was published is clearly printed in the image link, in a bright yellow box once again.
This is an extremely crucial change, because The Guardian realised that readers often glance over these important details, or even share article links without reading the content. While it may be difficult to force people to read the whole article before sharing, it is easier to design the content in a way that shoves the necessary information into their faces. Luckily, The Guardian does it the right way, and to read more about why they decided to make this change, check out this article.
Despite this simple and elegant solution to a long-standing problem, one question still remains: why aren’t other news outlets or social media following suit?
One of the reasons why such misinformation works is because almost everyone exhibits some form of confirmation bias.
Confirmation bias is the tendency to seek information that confirms our hypotheses or beliefs. Many articles that write about confirmation bias often describe it as some sort of sin or defect of the human mind, but objectively, it is actually just a type of search strategy, albeit an imperfect one. To search for information, we all have to start from somewhere, and it usually makes sense to start with what we know. But to only consider some parts of information before making a judgment, that’s where the problem comes in.
Misinformation that contains some aspect of truth works because when readers consume such information, they tend to focus on the parts that appeal to them the most. This is especially the case for people who have strongly-held beliefs, and will most likely put little weight on the less truthful parts of the information. Take the resurfacing of old articles for example, a person who is unsympathetic to the events of the Boston Bombing would likely argue that the Daily Mail article is still relevant because it is true, even if the incident happened more than a decade ago.
Again, I would like to reiterate that the point isn’t about whether the Boston Bombing or the Aghan wedding bombing is of a greater concern, but rather, to illustrate how misinformation preys on our biases even when the logic is weak.
The unfortunate reality is that most of the time, misinformation that people are most susceptible to are either technically true or factually true. Considering that everyone may have a different opinion of what the truth is, it will almost certainly backfire if we tried to force antagonistic information down the throats of deniers.
Hence, to convince a misinformation believer, it is important to first step into their shoes to understand what makes them believe in what they believe. It is then necessary to acknowledge the parts of what they believe that are true, but at the same time gently point out the logical inconsistencies in the connection of information, which will ultimately invalidate the conclusion despite the pockets of truth. This will hopefully make the information more palatable, and help them to see from a different perspective.
Of course this is easier said than done, but it is nonetheless a start. And thanks to news outlets like The Guardian, the so-called truth in the misinformation will have less of an effect, when the age of the articles are more explicitly displayed. Hopefully, this will inspire more tech and social media companies to incorporate simple nudges in their solutions for the fight against misinformation.