– English Lyrics

Lyrics & Translation

[English]

[Music]

[Applause]

What happened on September 23rd, 1999?

This is the day that the Mars Climate

Orbiter went on a mission to Mars and

actually failed considerably. This was

this device that NASA sent over to Mars.

It was designed to measure the weather

on Mars and also to serve as this

communication device for the Mars polar

lander which was supposed to arrive a

couple months later. It just completely

failed. It hit the atmosphere, burst

into a million pieces, leaving the folks

who worked at NASA befuddled, upset,

pretty pissed off. These are not them.

You have to use your imagination a

little bit in this talk. Now, these

things happen. Sending things into space

is tricky business. And quite frankly,

as a psychologist, who am I to judge a

bunch of very smart engineers who work

for NASA for failing and having a bad

day at work? But what makes this story

extra special and super fascinating from

my perspective is that the entire

failure comes down to one thing. failed

communication between these team members

and more specifically

the people working on this project were

not talking to each other about the

right stuff at the right time. It really

is that simple. So I'm going to break

this down for you. So it all started

when they had to calculate the flight

path. So anytime you send something into

space, you have to tell it where to go.

And to do so, you have to calculate the

flight path. Now the folks working at

NASA for the Jet Propulsion Laboratory,

they calculated the flight path using

the metric system and so they had the

Newton as their unit of force. Those

working at Loheed Martin in the UK, they

were using the pound as their unit of

force. When you're using these two

different systems, the whole thing was

off by a factor of about 4.4.

Now, at no point did the folks at NASA

say to those over in the UK, "Hey guys,

you're using the Newton as a unit of

force, right?" No, no, no, guys. No, no.

we were using the pound. That

conversation never actually happened.

And so we have our first big

communication mishap. Two sets of teams

failed to communicate about basic

information that seemed pretty obvious

to everybody. And just to put things

into perspective, imagine instead of

sending something into space, these

folks were making a cake. Okay? If I

thought I was using a pound of butter

and you thought we were using a kilogram

of butter, that's about 2.2 lbs of

butter, your cake would end up tasting

pretty gross like this one looks here.

Now, before we get too judgy about this,

it sounds pretty silly. We do this all

the time. We walk into meetings, you

know, if we get lucky, there's that

annoying, overly conscientious person

who says things like, "Before we get

started, everyone, let's level set and

talk about, you know, whether there's

ink in the printer at the newspaper we

work or, you know, whatever obvious

thing they want to get on the same page

with." And we usually roll our eyes at

this person and we tell them to stop

talking because we want the meeting to

end soon and we want to get on to the

important stuff. And we say things like,

"We've done this a million times. do we

really need to cover that kind of thing?

And the answer is yes. We should have

that 20 second conversation so that our

probe does not explode when it hits the

atmosphere, but we often do not do this.

Now, the good news is this is NASA.

There is no single point of failure. And

so just because you miscalculate your

flight path doesn't mean the whole

mission is going to actually end in just

a complete failure. And so people

started noticing things were wrong. And

the good news is you can actually

recalculate a flight path. So people

started bringing this up in various

meetings and they even had a conference

about it. But then there was a big

mistake that happened. Communication

mishap number two. The people holding

that critical information were ignored

for a very dumb reason that I'm pretty

sure everyone in this room can

recognize. They did not fill out the

right form. Now all of us know that if

you send a important message over Slack

and everybody was on email or over email

and everybody was on Slack, we just miss

that critical information.

But we don't think that a failure to

fill out the right form is going to be

the difference between our mission to

Mars failing and succeeding. We think

that critical information will

eventually make its way to the important

people at the top. But often this is not

the case. We get very married to our

processes and these can actually be our

Achilles heel in really important group

decision-making contexts. Now, things

didn't end here. There were actually

some last stitch efforts to save this

mission that didn't go so well. Someone

got on the phone with another person.

They sounded urgent about fixing it, but

that person didn't actually recognize

urgency. I think the quote was something

around like they didn't sound anxious

enough and so they weren't taken

seriously. So this miscommunication also

worked around non-verbal behaviors, tone

of voice and so on and so forth. So

there's lots of ways in which this

mission went arry. But I've been

studying miscommunication for over 20

years now. And I have to say that what

actually happened at NASA is much more

the norm than the exception that even

when people are making really critical

decisions, they often fall flat on their

faces and often for these very simple

reasons. And this is the case even when

we give people every piece of

information they need to make the right

choice. So now I want to talk about a

very classic experiment done in social

science. So imagine that you're sitting

in a room with these people and your job

is simple. It's to hire the best job

candidate among a list of four. And we

give you all the information you need.

Everyone is handed a piece of paper with

a bunch of information about all of

those job candidates. information like

applicant A is disorganized, applicant B

has strong leadership skills, applicant

C has won many cake baking awards. A lot

of this information is what we're going

to call overlapping information or

shared information. Everybody has it.

But here's the trick. One special member

of this team has what's called unique

information. Special information about

applicant C that only they have. And

here's how this task goes. If this

person does not share that unique

information, applicant C will come

across as the worst job candidate. If

they do share this unique information,

applicant C is going to come across as

the best applicant. So just to be clear,

the only thing that needs to happen for

this team to make the right choice is

that this special person shares the

information about applicant C. The team

hears it, incorporates into their

decision-making, and they indeed pick

the right person. Now, much like the

real world, people don't know exactly

which pieces of information everyone

else has. They just know some is

overlapping and some is not. This is

called the hidden profile task. It is a

very triedand-true task. And researchers

from the University of Southern

California did a huge analysis over 40

years of this and found that most of the

time teams make the wrong choice. By and

large, small teams, big teams, you know,

huge teams, tiny teams, teams online,

teams in person, teams in which the

person who is holding that unique

information is an expert doesn't matter.

And I have actually found in my own

research of about 370 teams, 20%

unanimously pick applic. So the question

is what's going on here? Well, the

obvious explanation and one that we

often see is that teams focus on that

shared information the most. They kind

of throw around the stuff that they all

know. They focus very little on that

critical information about applicancy

that only one person knows. But we also

found something much more basic going on

in the research in my lab. We took all

the transcripts from these conversations

going on teams and we had human coders

read them and code for the presence or

absence of those unique pieces of

information. We also had AI code them in

large language models. And we actually

found that it is very hard to tell if

those special pieces of information

about applicant C are being shared.

Verbatim, not a problem. As long as

someone literally said applicant C is

disorganized, we were fine. But this is

not how people talk in teams. They start

a sentence, they stop a sentence, they

use synonyms all the time. And in well

functioning teams, people will echo each

other's contributions, often restating a

piece of information in a slightly

different way that alters its meaning.

And so what we learned is these critical

pieces of information are incredibly

fragile. They're like little pieces of,

you know, information in the wind that

can kind of blow away. And because of

this, we lose this information, but we

can't actually tell that our

interactions with one another aren't

going as well as we think they are. And

critically, because in these

interactions, everyone is motivated to

make the right decision. No one person

is trying to bulldoze or push their

person through. These team interactions

actually feel good. And so we can be

communicating terribly and not know it

because the red flags that we usually

look for, those interruptions and so on

and so forth, simply aren't there,

making this type of poor communication

just really clever and underneath the

surface of what's going on in these team

interactions.

Now in this study, people are all

speaking the same language quite

literally, but also social scientists

are very good about holding things

constant that could potentially explain

this effect. Use of jargon, use of, you

know, different types of cultural

languages and so on and so forth. But in

the real world, that is not how we talk.

We show up to these interactions using

all kinds of different languages. And I

don't mean that literally. I mean the

local languages that we often develop in

our communities, in our friend groups,

in our workplaces, acronyms, synonyms,

terms of phrases that we use all the

time and we don't even realize it. And

we often call these things hidden

languages. And they are everywhere.

There, you probably have already used 50

of them today without even realizing it.

They're all over our resumes, often in

the forms of random letters strung

together that very few people recognize.

And these are great. They actually make

our lives more efficient and they build

a sense of community. They build a sense

of identity. But if you don't know them,

and most of us know what it feels like

to sit in a meeting where people are

using them and we don't know them, you

feel really stupid and you feel really

left out. And it's awkward to actually

ask people what they mean. Things like,

"Let's get this done ASAP. I hold a BA,

MBA, and I'm the CEO. This person is

very proud." And they're showing it with

all of these letters. Yeet, which I've

learned is a thing. It is both a verb

and a noun. Um, I yeated the ball. I am

yeet. I think that's true. I probably

got that one wrong. That idea is cringe.

NASA. And sometimes these things

actually disappear just as quickly as

they show up. I'm not allowed to say

cringe anymore, according to my

12-year-old. Now, if anyone's ever gone

to another country, you realize some of

these things don't track. I have a lot

of German colleagues who've told me that

there's a phrase that says something

like it's all train station to me. That

is a terrible English translation of a

German phrase that means something like

it's all Greek, which also doesn't make

a ton of sense. So, we use these phrases

all of the time and they can infiltrate

the ways in which we speak. But one

thing you probably don't realize is just

how quickly they actually develop. And

so, some researchers from Keltech showed

this through this very clever

experiment. They handed people photos of

offices that looked nearly identical to

each other. And in pairs, they had them

describe these photos to one another.

And they looked at how quickly they

actually came up with their own hidden

languages. So, at first they started off

pretty slowly. This one has a computer,

three ferns, a cup of coffee, so on and

so forth. This one has a computer,

headphones, a cup of coffee, a phone.

But over time, and and by over time, I

mean a matter of minutes, they got very

efficient. Team A would call this wall

fern. Team B call this one tidy vibe.

Why? Nobody knows. We just know that

they do it. Team A calls this one lots

of stuff and team B calls this one

wannabe writer. Now imagine that we now

have someone from another team come join

yours. Team A is working well together

and someone from team B joins that team.

What happens? Do they learn their hidden

languages? Do they start over? They

actually get really irritated with each

other pretty quickly for not

understanding one another. Team A says,

"It's wernern. You know the wall fern."

And team B says something like, "Stop

saying wallern. Half of them have wall

ferns. Tell me if there's a tidy vibe."

So, we do not actually realize that

these hidden languages are dominating

our conversations. And they ended up

taking so long to do this. Most of them

actually didn't finish the task at all.

Now, there's this huge theme here in

this talk that we don't know what others

don't know. We don't know if they're on

the same page with ostensibly obvious

pieces of information like whether we're

using the Newton or the pound as our

unit of force. We don't know what the

hidden languages are that they're using.

And we don't know if they are sharing

critical information in the ways that we

think they are. So, the question is,

what are we supposed to do with all of

this mess if we want to make smarter

team decisions? So, I think the first

thing that we should do is be that

annoying person in the room who says

things like, "Let's level set." Which,

by the way, is also a hidden language.

State the obvious. It is a good idea to

start those meetings, to start those

conversations, even if it makes people

roll their eyes. That 20 second

conversation about the obvious thing

that should be going on in that meeting

should happen. And it's okay to be the

one to do it. It's okay to be the one to

make that the norm in the meeting.

realize that not all critical

information appears as such or is

obvious to everyone. So in the research

I talked about, they quite literally

handed people all the information they

needed to make the right hiring

decision. But that is not the real

world. We walk into rooms, we might not

know if our information is critical. We

might think it is, but there's a norm

against sharing it. So imagine for

instance that you are making some really

important decision at work and your boss

is in the room and that boss is arguing

to give her direct report Tom a raise

but you just saw Tom come out of her

hotel room three times that weekend

retreat you guys got back from. Should

you share that information? It feels

critical to you but it could also just

be seen as a nasty little nugget of

gossip if it's shared to the wrong

person. So we often don't know. There

are norms that we could be violating by

sharing critical information. We might

be sitting on something important and we

have no idea because we are new in the

workplace. Don't assume people are

always sharing it. People are actually

more likely to withhold something if

they're afraid that they're violating a

norm or if they're afraid it's not going

to go down as well as they think it

might. It's okay to restate information.

In fact, you should a few times. We

learned that people often assume that

critical information is shared and

that's a false assumption.

Make sure that information sticks. Say

it in the way you wanted it to say, in

the way you wanted it to be heard.

Restate it and do it a couple times and

do it at the end of that interaction so

that you can make sure that your message

actually goes out as intended. It

doesn't get restated even by a

well-meaning member of your team.

And make it comfortable for people to

ask, "What did you mean by that?"

I think most of us know what it feels

like to be sitting on the other end of a

conversation in which a lot of jargon is

thrown around, a lot of acronyms, and we

feel silly asking. But make it a norm to

just say, "What did you mean by that?

What were those letters? What did that

phrase mean?" We will be much less

likely to be annoyed with one another

when those kinds of conversations are

happening. No matter what you do, don't

say Walern. Thank you.

Heat. Heat.