That’s what I was going to say

 A thought for today…

Yesterday I watched a short video about how implied communication happens. There’s a set of mutual assumptions: when you communicate with me, I assume that you are sharing meaning. And, furthermore, that you understand that what you are sharing will be interpretable by me. 

So, you say something vague and I search for a clue in our relationship or topically. And it works some of the time.

I would add that one of the reasons why we do this is intimacy - it says “I trust you to get me.”

Pick Up Artists also use this format to fake intimacy, test boundaries, and pick up clues about the other person:

“We finish each other’s…”

“Sandwiches”

“That’s what I was going to say!”

AI does it too. And, when AI does it, we make the same assumptions that we do with people - that it “gets us” or it’s digging for information. 

But it’s not. AI has less intention than the dumbest orange cat. It’s just being unclear. 

This is one reason why we unintentionally accept poor quality AI output - because it feels flattering or deep. We have to learn to hold output that feels human to machine standards to use AI well for knowledge work. 

Learning is a Change in Performance

 If you’ve heard me tell this story before, forgive me. It’s a touchstone. 

When I was getting my instructional design degree, my beloved professor said, “learning is a change in behavior.” If there is no change in performance, people haven’t learned. While this may be functionally true in a corporate setting, it’s profoundly untrue in real life.

As an undergraduate, I learned of an experiment where two sets of rats were given access to a maze. The first set of rats found a tasty snack at the end point of the maze. The second set of rats got no treat and were simply removed when they reached the end.

The first set of rats quickly learned how to reach the end of the maze. The second set of rats wandered around infinitely.

Then, the second set of rats received the snack at the end of the maze. During the second trial, every rat in Group 2 overwhelmingly beat the fastest rat in Group 1. 

Did Group 2 fail to learn until there was a change in behavior or was learning quietly happening in the background until there was a reason to apply it?

I’d argue that Group 1 learned how to find the snack while Group 2 learned the maze. 

Also, as an addendum, this is why not all learning should be ISD.