This piece is the seventh installment of a biweekly series written by David A. Foster (Center bias), based on his new book, Moderates of the World, Unite! Read the first post in the series. 

Immigration is a critical issue in the current election. As a conscientious voter, please review the topic areas below and (mentally) rate your own degree of understanding of each topic on a five-point scale.

Here is a simple question: aside from the last two, can you recall any of these topics being addressed in recent years in the news?

I know I cannot. For the vast majority of us, what we know comes from the daily news, and the news hardly educates us at all.

As discussed in the first article in this series, journalism has been decimated by the internet. And recall the news’ four main information biases laid out by Lance Bennett: personal, dramatic, fragmented, and politics as a game. We can no longer expect the for-profit Fourth Estate to teach democratic citizens about national issues. Widespread deficits in understanding result in low-quality public discourse and high vulnerability to distraction and propaganda.

Adult citizens are asked throughout their lives to vote on the basis of specific national issues, ideologies, programs, and problems. Not as experts, of course, but they do need at least some competence at separating fact from fiction. General education in college is helpful in this regard, though only one-third of adult US citizens over age 25 have a bachelor’s degree.

Solution “A”

Perhaps the solution lies in more publicly-funded media. The US is an outlier in this regard. A 2021 study by Timothy Neff and Victor Pickard evaluated the annual amount spent per capita in countries around the world. Here is a small sample:

In the US, the annual budget for the federally funded Corporation for Public Broadcasting (CPB) is close to $500 million. For perspective, in the aggregate, total federal discretionary spending is $1.6 trillion per year; thus, CPB funding constitutes about 0.031% of federal discretionary spending, that is, less than one-thousandth of it. (Military spending, by contrast, constitutes 46.0% of discretionary spending.)

So, fine—the amount of public spending on media in the US is miniscule. But if more were to be spent, it is not clear why it should be spent on news reporting or journalism, which is not an efficient or effective way to broadly raise public understanding in a way that can produce more competent democratic participants.

How to Produce Competent Democratic Participants

There is practically an infinite amount of information out there on the web that people could use to educate themselves about public matters. In a myriad of websites, on Wikipedia, on YouTube, and so on. But they don’t. They’re busy.

In an ideal world, there would be good courses about public issues that people would take. Where they’d get some background, and key concepts, and facts. But, of course, a course is too large of a commitment, for almost everyone.

Motivation, audience, subject matter… this, in large measure, is an instructional design problem.

The lay view of instructional design is that it is about putting information in front of people, and prettifying and “jazzing it up.” The real, actual core of instructional design, though, is what I refer to as learner empathy. Which includes: thinking deeply about your target audience and their concerns, doubts, and most-likely questions; helping them build a mental picture; providing personalized pathways; inspiring them; and making the best use of their time. The philosophy is called learner-centered design.

Policymakers and other elites seldom appreciate the importance of creating different, specialized materials for the public because they unconsciously project from their own existence. They do not recognize that most people are not aggressive autodidacts like they themselves are. (You, dear reader, as an autodidact may share this complacence! You do not remember what it is like to not know what you now know. I call this expert amnesia, and Steven Pinker refers to it as The Curse of Knowledge.)

General Strategy

The basic proposal is to create a new, national, publicly-funded, independently-governed, mobile-friendly, citizen education website that provides authoritative and engaging short videos, with a heavy focus on issues projected to be debated in future elections. A design goal is to quickly and consistently give each visitor insights they want, through design techniques that I’ve described further in my book. Let us call it the US Voter Education Source, or VES.

We can take inspiration from the runaway success that the conservative PragerU (Right bias) has had with its hundreds of five-minute animated videos. Certainly we should be encouraging young people to read more, but for most adults, consuming videos is simply easier; in cognitive terms, further, synchronizing audio with well-crafted images results in deeper comprehension. Five minutes is ideal—easy to swoop in and consume quickly, but enough to provide some substance.

VES doesn’t provide courses, so it’s tempting to think of it like the content that’s currently available on the web, such as YouTube. There are important differences. One is that free content pieces throughout the web are generally created with razor-thin budgets—seldom with a team of subject-matter experts, instructional designers, and multimedia professionals.

The second is that web content consumption is “unplanned,” in the following sense. Each content piece on the web is intended to be consumable as standalone. What then gets consumed next may be determined by algorithmic suggestions by YouTube, tuned via machine learning to what you will like rather than what is needed to deepen understanding. You may well get something that is too advanced, more entertaining, or only marginally relevant. VES, in contrast, must have a top-down overall design in which content sequencing is carefully orchestrated, with an eye towards progressive deepening of understanding.

Providing information is not the same as providing education. The whole point of education is that learners don’t really know what they don’t know, and thus they need shepherding from experts who “know the territory” for a topic. It’s a huge difference from the “individual knows best” foundation of the web.

But, Why Now?

I am proposing a huge, costly new imposition into the public sphere: interactive educational materials about public matters for average citizens. Well-intentioned, though maybe a little unimaginative? Sure, everyone favors education, and the US could easily afford it, but there never has been much interest in doing it. Why should now be any different?

Two reasons. The first, certainly, is because the societal disruption of social media, partisan media, and disappearing journalism has made it urgent, perhaps even an emergency. The second, though, is that technology is handing us some brand-new possibilities.

The raw pieces have been coming together in recent decades: internet and mobile technologies, interactive multimedia, and now generative AI. Until now, it’s been too expensive to provide really fun, engaging learning episodes that could compete with the totalizing entertainment surrounding each citizen. But the next generation of online learning material may be a tipping point.

Here are a just a couple key elements:

1. Personalization. Interactive personalization in computer-based learning materials has been tried at least since the 1980s, though it has always required many expert labor hours to produce. Generative AI, though, is going to change the game.

My colleagues and I have been experimenting and developing tools and methods for combining multimedia learning content with large language models like GPT and Claude. This will be a burgeoning research front as more people recognize the new architectural possibilities. We’re finding that combining interactive multimedia with a tightly integrated AI “coach” creates higher levels of engagement with a less labor-intensive development effort.

Just as three examples where an AI “coach” can be trained and inserted:

1. Visitors can interrupt a video to ask a question and can expect a reply that is cogent, is knowledgeable about the video, and is even cognizant of the visitor’s knowledge and interest.
2. The coach can generate a small list of the most promising follow-up questions.
3. The coach understands the relationships between the videos and, when called upon, can direct the visitor in directions that will most efficiently increase their understanding.

To be clear, it’s not just turning a visitor loose in a specialized ChatGPT. The more important part is all the multimedia content that is scripted and organized by expert instructional designers. Like PragerU, but nonpartisan, more structured, more interactive, and more personalized.

2. Partisan adaptation. Coaxing partisan citizens into trusting the system will be critical. A scheme that will help in this regard is to speak in the language of partisan values. Three versions of each video can be produced: one with a liberal slant, one with a conservative slant, and a neutral version. We can visualize providing visitors with an option something like this:

After the choice is made, all subsequent navigation will automatically present the slant (A, B, or C) that was selected by the visitor.

There is nothing sneaky or underhanded about this approach. Nothing is hidden, and nothing prevents anyone from seeing all the versions. And learning is made more enjoyable and satisfying. More positive word-of-mouth occurs, and return visits become more likely.

Cost and Branding

The cost to produce and maintain VES would be considerable, but let’s put it into perspective. Just as a back-of-envelope calculation, if individual videos each cost $5,000 to make and 5,000 per year were produced, that would just be $25 million. (Compare, remember, to CPB’s miniscule annual federal budget of $500 million.)

There would, of course, be other large expenses, not least of which is promotion and branding. VES must be professionally positioned as scrupulously fair, and as The Place where you want to go for predictably quick, fun, satisfying briefings on public topics you’re a little fuzzy about.

Let’s start the conversation. I personally need VES, so I can get smart—fast—about immigration!

Read the rest of the series: