I do live sound as a hobby, so I really hate getting feedback! ;)

When I hear reinforcement I think of classical conditioning, where reinforcement is a specific term for something you do to increase the frequency of the desired response. It’s the opposite of punishment and can be negative or positive, which aren’t exactly what you’d think they are (and can get confusing.)

I think feedback more as something that can be desirable (monitor speakers that allow the band to hear itself) or undesirable (a signal being reflected back into a radio and causing interference.) Both of these examples are a response to something that affects the following output.

I don’t have a problem with the unqualified use of “feedback” in the colloquial sense, since information given to the player is also feedback in the technical sense (although its effect can be a change in player behavior rather than something intrinsic to the system and much thus less predictable). There can also be some overlap in usage. A reward can be construed as positive feedback if it allows the player to more easily get future rewards. A punishment may also be positive feedback (in the negative direction) if one ignores any impact on player behavior.

I remember teaching out of a game design textbook that actually had these definitions backwards in the text. That made it extra confusing when I tried to disambiguate to the students that “positive feedback” in the engineering sense doesn’t mean always “a good thing.”

Perhaps the discussion could be served by talking about the ends rather than the means. Negative feedback serves to stabilize or converge systems that have been perturbed, or to maintain them in a steady-state or equilibrium condition. Positive feedback encourages divergence, amplification of differences, and system instability. Using stability or equilibrium to talk about this will run us up against colloquialism again, because players perceive individual histories rather than averages. “Convergence” and “divergence” mechanics, or “amplifiers” and “dampers” might be useful ways to talk about mechanics or sets of mechanics that act in the way you describe.

Within that simulation step, however, is where the current game state is checked against the rules (is the temperature too high, too low?), applies balance mechanics (blue shell now available to player 7, who is in last place), all of which leaves ‘feedback’ available for the intent you describe above (and fits, I think the first two definitions in the link that Ralph Koster provided on twitter, more so than the last 3 he highlighted).

I’d note that while the blue shell seems a great example of negative feedback as you describe, it isn’t because someone is in first place, but rather it becomes *available* because someone is a certain distance behind. It’s not a direct punishment of the leading player, but rather a balancing negative feedback provided to the losing player, which seems somewhat counterintuitive. Using your definition, it’s technically appropriately ‘negative’ feedback because the losing racer is cold, and the system kicks the heat on. But colloquially it’s not negative because it potentially helps the losing player (which is a reward). As a counterexample: automatically throttling the speed of the leader player would be a negative feedback (and a penalty) for the lead player. In a sense, the blue shell is a great example of the linguistic tension you’re encountering, and why it’s tricky: implementing negative feedback in this case also serves as a reward for the (currently losing) player, whereas throttling is an example of negative feedback that serves as a penalty/punishment.

The *use* of the blue shell, on the other hand, is a traditional ousting/paralyzing move for many kinds of race games (cf Parlett) which represents interaction between players and arguably isn’t ‘feedback’ at all, but simply a ‘move.’

What about “Response”?