Zachary Wojtowicz

Very nice paper on the importance of shared context for communication.

The pairing of insightful theory + clean experiments is phenomenal here.

#eeac Day 5: Instead of @zwojo.bsky.social fascinating JMP "Context and Communication" I wanted to highlight another related provocative paper from him arxiv.org/pdf/2407.14452 on mental proofs and how AI might intervene in social contexts where mental proofs are crucial. #Econsky #econjmp

Part of ongoing work to bridge disciplines through a shared interest in mental resources

#EconSky #PsychSky #CosSky

Curious about attention, but don’t have time to read 300 psych & econ articles?

We’ve got your back.

• Comprehensive overview of both literatures

• Novel resource-based definition of attention

• Implications for risk, time, health, finance, & more

• Promising directions for future research

🔗 👇

This is the issue that the new cognitive turn in behavioral econ has to resolve head on.

Is the dimensionality of factors that the new frameworks are introducing going to be so high to make them vacuous? 1/n

Common knowledge is a valuable social and economic resource!

In the paper, I discuss implications for law, theory of firm, diversity, AI-IO, & more

I demonstrate the implications in a stylized contracting experiment

Pairs with shared context (prior talk about similar items) were better at understanding…

And more likely to play honestly!

In the paper, I formally show how shared context helps people efficiently coordinate on meaning

Common knowledge (in the sense of Aumann) bounds efficiency

Most communication is highly contextual: messages can mean wildly different things in different circumstances (e.g., “I’ll have the usual”)

Coordination failures (misunderstandings) have major consequences

Direct communication is a powerful tool for solving coordination problems

But communication is itself a coordination problem!

Hey #EconSky! I’m a postdoc in psychology and economic theory @ Harvard

My #EconJMP studies how shared context supports efficient communication and cooperation

🧵⬇️

I also demonstrate this in an online experiment where

1. Buyer orders a good
2. Seller delivers a good
3. Buyer accepts / rejects delivery

• Seller wants to misdeliver if possible (think up-sell)
• Buyer wants to forgive genuine misunderstanding, but punish deception

In the paper, I formally prove that the more relevant context a group shares, the easier it is for them to communicate new information

Common knowledge (in the sense of Aumann) places a lower bound on communication costs

Language is highly contextual: messages can mean wildly different things in different circumstances (e.g., “I’ll have the usual”)

Coordination failures matter: E.g. the \$3.5b lawsuit over whether each plane on 9/11 was a separate “occurrence”

Direct communication is a powerful tool for solving coordination problems

But communication is itself a coordination problem!

Empirics: A stylized game where

BUYER orders a good
SELLER delivers a good
BUYER accepts / rejects delivery

SELLER wants to misdeliver if possible (think up-sell)

BUYER wants to forgive genuine misunderstanding, but punish deception

In fact, the more relevant context a group shares, the easier it is for them to communicate new information

Common knowledge (in the sense of Aumann) places a lower bound on communication costs

Theory: Rigid languages are less effective at compressing information

Contextual equilibria Pareto dominate in sender-receiver games with differentially costly (type indep.) messages

Such ambiguity has far-reaching consequences

E.g. the \$3.5b insurance dispute over whether each plane on 9/11 was a separate “occurrence”

Big Q: Given the risks, why not adopt a totally unambiguous system?

We don’t make this easy on ourselves!

Language is highly contextual: messages can mean wildly different things in different circumstances (e.g., “I’ll have the usual”)

The function of zebra stripes - Nature Communications The function of zebra stripes remains unclear as none of the options have been validated ecologically. Here, the authors, matching striping patterns to environmental variables, test the main hypothese...

I’m no biologist, but perhaps convergent evolution re-discovering how annoying biting flies are! (Especially on the rear)

www.nature.com/articles/nco...

Hi Stephanie! There’s a contracting experiment in my JMP and more on the way. Would love to connect with other experimentalists

This is great! Would love to be added

Thanks for putting this together! Would love to be included