TL;DR: My professional background makes me especially sensitive to how RPGs model the discovery of clues. If you want to skip the philosophy, wait for Part II.

---

There are two things I’m a world-class expert on: the reconstruction of mathematical logic from natural language argumentation and the logic of discovery. Admittedly, these field are so “niche” that (almost) every expert is a world-class expert. Still, I’ve published research on both topics and proved a couple of novel results, so my claim to expertise is not only that I know the textbook backward: it’s that, if there were a textbook, I’d be writing it.

Given that background, I’ve always been quite sensitive to how RPGs handle the discovery of clues. Now, before going any further, here’s a massive caveat: I’m not a game designer, and I don’t necessarily have the same priorities as one. If you want tips on designing adventures centered around discoveries, you may be better off checking “classic” RPG references about Node-Based Scenarios or the Three-Clue Rule.

My goals are much more modest and, so to speak, “local”: I want to make room for an in-game equivalent to the discovery of unknown unknowns. If you’re not curious about what “unknown unknown” means, or don’t care about the my motivation, you can stop reading right now and wait for Part II and the mechanical bits. This post is all about philosophy (and a bit of cognitive neuroscience at the end).

The Philosophy Bit: Unknowns, Knowns and Otherwise

The distinction most relevant to the topic of discovery is between known unknowns and unknown unknowns. There’s a whole body of philosophical literature about that distinction and another of logical literature too. I prefer the latter (it’s more precise), but the math can become hairy pretty fast, so I’ll stick to philosophy which, honestly, is good enough for this discussion.

Philosophy works with words and definitions, so here’s one:  a known unknown is something you don’t know but are prepared to ask a question about. The operative word here is “prepared,” as we will soon see. Also, not every question will do: it must be a genuine question, not a rhetorical one. When you ask such a question, I immediately learn that there is something…

• …you don’t know: otherwise, you would not have to ask about it;
• …you want to know: otherwise, you wouldn’t care to ask about it;
• …you know that you don’t know: otherwise, you wouldn’t think to ask about it.

Now, for what “being prepared to” means. First, you must know the words (or have the concepts) to ask the question. Second, you must be aware that there is a question to ask with those words (or about those concepts) in the current state of the world. For instance, you may know the words “mathematical,” “model,” “Sherlock Holmes,” and “reasoning,” but unless you belong to a particular subset of my readers, you would not be prepared to ask the question:

What does a mathematical model of Sherlock Holmes’ reasoning look like?

Indeed—unless you belong to a particular subset of my readers—before reading the question, you did not even know that “a mathematical model of Sherlock Holmes’ reasoning” could be a thing. It was an unknown unknown. So, an unknown unknown is something you don’t know and are not even prepared to ask a question about. In the discovery model I work with, the set of questions someone is prepared to ask in a given situation is called their range of attention. People don’t have infinite attention spans, so questions come in that range and out of it depending on circumstances. 

Enter Sherlock Holmes

A typical plot device of Sherlock Holmes’ stories is a question in Holmes’ range of attention, but not in Watson’s, Lestrade’s, or some others, whose answer unlocks a deductive path to a solution. One of the best-known examples is in Silver Blaze, where the question of whether a watchdog barked at a thief makes it possible to deduce the thief’s identity. Ironically, the person who answers Holmes’ question is a police officer (Inspector Gregory) who did not consider that question relevant—and thus, missed the clue and arrested the wrong suspect.

Now, here’s the crux of the issue: genuine discovery only occurs if an inquirer increases her range of attention by herself. In Silver Blaze, Holmes picks up on a detail—a trained watchdog—and matches it with a generality—trained watchdogs bark at strangers. Then, Holmes asks the question about the dog because he anticipates what he could deduce from each possible answer. Learning that the dog did not bark, Holmes concludes that the dog knew the thief. Gregory, who did not pick on the detail, nor matched it with the generality, couldn’t reach the same conclusion—and arrested a suspect whom the dog does not know. 

And now, for RPGs. Suppose you’re GM-ing an investigative scenario and, you’d like your players to pick on relevant details by themselves. That is, you want them to think about your description, broaden their range of attention, and ask questions. The minute you give them any indication of which detail to pick up on, you become Holmes pointing Gregory to the dog. It’s not a genuine discovery anymore. It’s “guided discovery.”

Wrapping Up: Why Bother with Discovery—or: the Cognitive Science Bit

Guided discovery is cool. It’s related to “collaborative discovery,” which is much cooler. Research shows that collaborative discovery requires creativity. That was a shameless self-promotion plug, but I’m honest, so here are two caveats. First, the research is theoretical cognitive science: it’s based on deductions from available data, not on ad hoc data collection. Still, the data is solid. Second, the similarity is relatively abstract. In short, the same math describes what happens in a network of neurons and a network of people when discovering an unknown unknown. But the long and short of it is this: genuine discovery is close kin to creativity—guided discovery, less so. 

And that’s why I bother about making room for genuine discovery in my games. First, greater creativity at the game table makes the game more unpredictable and more enjoyable for everyone. Second, it makes it possible to solve more complicated problems, so as a GM, I can present my players with more challenging situations. They have to think harder to solve them, which increases their pay-off when they finally do. Solving more challenging problems reinforces their reliance on creative thinking, which makes it more unpredictable, etc. That’s a positive feedback loop of creativity. Now, if they’d expect me to guide their discovery whenever they hit a snag, I’d have the opposite: a negative feedback loop, with less and less creativity at the table.

If you think it’s a valid priority, stay tuned for Part II. And that’s all for today, folks.