Hello, friend!

Thanks for clicking in. We’re back with a veritable buffet of the very best vertebrate ephemera. You get some advice for learning about people and organizations, fun facts about claw machines, recent research, and our robot of the week. It’s a good one! 🐷

Skip your other meetings

There’s no time like next week to get in on Presenting with Confidence, June 4 & 5. Whether you’re interviewing, fundraising, pitching, or just advocating for your work in a chaotic organization, you’ll emerge sounding more like yourself, but better. Tell your friends and co-workers, too.

It’s 6 hours of whatever the opposite of bastards grinding you down is.

If you want to understand why people do things…

I started thinking about this after a participant in yesterday’s research workshop asked how to tell in a job interview whether a particular company has a functional decision-making culture.

My answer: ask the interviewer about a recent team accomplishment and get them to describe how that happened, step by step, and who was involved. The story of the process will reveal the decisions and the decision-makers.

Humans are so interesting, and so tricky. Often in making a good-faith attempt to understand each other and work together, we make everything worse. Sometimes a few words can send a whole conversation off the rails. (If anyone has ever told you to “calm down,” you know.) In conversation the most direct path isn’t always the shortest.

If you want to know the reason a person did a particular thing, try to avoid saying “Why did you do that?”. It’s very easy for that phrase to sound more critical than intended and elicit a defensive reaction. And most of us are faulty interpreters of our own motivations, so you might spark the additional frustration of a question that’s hard to answer.

Ask about the experience instead. Get them to describe the preceding events or the context around the action you’re interested in. You’ll probably be able to pick up on causes and reasons with careful listening. And, you know, be normal and conversational about it. Don’t say “Tell me about the surrounding context” or make it sound like an interrogation. This is true for both interviews and casual conversations.

Anyone who participates in research is familiar with the value of open-ended questions and prompts like this — a practice that goes right out the window as soon as you have an emotional attachment to the answer.

Inquiring about goals is typically a lot less fraught than asking about causes. And it can be a useful way to refocus meetings that start to go off the rails. (“What’s our goal here?”)

Fun fact: The Russian language differentiates between why/what is the cause (Почему) and why/what is the goal or purpose (Зачем), which is handy for clarity.

The claw decides

Research has consistently shown that social media can inspire a fear of missing out on key relationships and experiences, and a claw victory is alluring.

Been thinking about claw machines a lot for some reasons. This 2015 Vox piece (🔒) describes the fascinating way they’re rigged. The secret is variable claw strength, set by the owner. The drop rate is adjustable, too.

Intermittent reinforcement is a hell of a drug.

Two preprints for your consideration

Understanding Artificial Neural Networks: Mysterianism about Known Mechanism is Mysticism

Don’t believe anyone who claims LLMs are a black box. This fresh paper from Olivia Guest et al references the work on understanding of one my favorite professors, Catherine Elgin.

The Largest Review of Synthetic Participants Ever Conducted Found Exactly What You'd Expect. Synthetic Users Don't Work.

Enjoy this thorough write-up of a pretty thorough review of 182 studies on LLM-generated synthetic participants.

🤖 Robot of the Week 🤖

This is a good one. Check out that hopping action. Too bad it’s $7000. Worth it.

Have a delightful weekend!