May 14 2019

Truth Decay

What is the greatest threat facing human civilization? This question is obviously meant to be provocative, and is probably inherently unanswerable. But I think there is a reasonable argument to be made that perhaps the greatest threat is the deterioration of fact-based political and social discussion. The argument is that this is a meta-problem that keeps us from effectively addressing all other problems.

There are, of course, potential threats that could override everything else, such as an asteroid barreling down on the Earth or a super pandemic that could wipe out humanity. Most problems we face or are likely to face, however, can be potentially effectively managed, or at least mitigated, if we optimally marshal our resources and planning. The real problem we are facing is that we appear to be increasingly unable to do so.

Some observers identify a large part of the problem as “Truth Decay.” The RAND corporation (a think tank of scientists and researchers), which has been researching the issue, defines Truth Decay as:

1- increasing disagreement about facts and analytical interpretations of facts and data

2- a blurring of the line between opinion and fact

3- the increasing relative volume and resulting influence of opinion and personal experience over fact

4- declining trust in formerly respected sources of facts.

I think that list seems reasonable. Clearly this is a multifaceted problem, and other researchers have identified these various factors before. In The Death of Expertise, for example, Tom Nichols focuses on item 4, the declining trust in experts and the very notion of expertise itself.

But of course we don’t want to assume anything, which would ironically be part of the very problem itself. We first need to ask – are these trends actually happening or are they just illusion and confirmation bias? Also, can we put these trends into historical context? RAND recently conducted a study looking at item #3 – the relative volume of opinion vs fact-based reporting in the media over the last 28 years. This is what they did:

The RAND-Lex tool scanned millions of lines of text in print, broadcast and online journalism from 1989 (the first year such data was available via Lexis Nexis) to 2017 to identify usage patterns in words and phrases. Researchers were then able to measure these differences not only within one outlet or type of media (e.g. print) but also comparatively with other forms of journalism (e.g. print vs. digital).

They identified several trends, which may contribute to Truth Decay. The first is that prior to 2000 broadcast news tended to be more academic and fact-based. After 2000 the news became more narrative based – presented more as simplistic stories, with less complexity and nuance.

Over this same time there was a shift in viewership from broadcast to cable networks. The cable networks contained much more opinion-based reporting, and far less fact-based reporting. They were more likely to have people discussing the news rather than giving a prepared factual report of the news. So essentially we went from watching Walter Cronkite to The View.

In print they saw a similar pattern. Print newspapers have changed the least, but also have shifted toward a more narrative style (just not as much). Meanwhile there was a shift to digital print news, which is more personal and anecdote-based.

All of these trends verify the concern that the overall volume of information being consumed by Americans has shifted from fact-based reporting to personal stories, narratives, discussions and opinions. We are no longer content to have a talking head give us a prepared digested form of “Just the facts, Ma’am” (which is, ironically, itself a bit of false reporting).  We want to be entertained with a story, we want our emotional buttons pressed. We want to be outraged by those we perceive as our ideological foes. We want to listen to people expressing our opinions, and amplifying them.

The research shows that people do also want information, and we do want to be truthful and accurate in our belief. These are competing imperatives, and we arrive at different balances in different contexts. Unfortunately, when the context is emotionally-held beliefs, or notions that comprise our world-view or identity, we favor confirmation bias, motivated reasoning, and partisan ideological thinking. Facts become increasingly subjective.

The study suggests we cannot blame the trend away from objective facts entirely on social media. Across all media the trend is away from facts and toward narrative and opinion.  Social media is a huge part of this trend, however. This study focuses only on content, but does not look at how people spread and consume that content. That is where the social media effect may be greatest.

In a way social media is a giant social experiment. Information is free to flow without filters. We are essentially crowd-sourcing the aggregations and spread of information. This is not entirely true, because there are top-down organizations, corporations, and groups who are gaming the social media system to promote their brand, their product, or their ideology. Even governments, like Russia’s hacking of the 2016 election, are involved. There is even a term for this – astroturfing: pretending to be a spontaneous grassroots movement when instead there is a deliberate force at work.

But astroturfing aside, social psychologists are now starting to gather information about how and what kinds of information spread online through social media. The indications so far are not good. False information seems to have the edge over facts in social media. This makes sense. If a message is optimized for social engagement it will have an advantage over other messages optimized for accuracy. Unconstrained by truth or reality, information can be maximally marketed, even weaponized.

It’s all a complex social phenomenon that we are just starting to grapple with. We clearly need much more research into the phenomena of Truth Decay, and also into potential mitigating solutions. It is the one problem that underlies all other problems.


No responses yet