So some of the smallest
most insignificant work
is everyday cos reflect
a lot about who we are.
And I say this not
as a linguist,
or a computer scientist
but as a social psychologist.
And today I'd like
to tell you a story
that summarizes
a lot of the research
that my colleagues,
my students and I have done,
that have helped me
to come to this realization.
Now several years ago I was studying
the nature of traumatic experience
and how it is related
to physical health
and kept finding
that just completely perplexed me.
Basically when people have
a major traumatic experience in their life,
they are much more likely
to get sick after that event,
if they keep the events secret,
than if they actually
talk to other people.
So, this really bugged me,
so keeping a secret it seems
is somehow toxic.
####this let me too
brunson experiment we brought people in the laboratory
and we asked them to write about
the most dramatic experiences they've had
especially if they kept a secret
and these are big trauma cedras things like right
they were like uh...
major public humiliations or failure
and the results that we got from this this study were stunning
we discovered that
harried
people write as little as fifteen minutes a day for three or four
consecutive days brought about
meaningful change in people's physical health
and even their immune function
translating up saying experiences into words
makes a difference
book why
since then there been hundreds of studies done by lab so all over the
world trying to answer this and they haven't come up with the single life as
a single explanation
by own approach was to actually
look at the essays that these people wrote
and try to figure out
was there something about the essays i can't predict who would benefit from
writing verses he wouldn't
i tried and i couldn't figure it out
so i gotta number psychologists and other experts to read and write hundreds
of these essays
and they couldn't fixya pattern either
units to try some other strategies cell
with the help of one of my graduate students partha francis we're on a
computer program
and the idea of this computer program was to go into any given text
and calculate the percentage of words in the attacks that were positive emotion
words
negative emotion words
or words related to topics such as death or sex or violence or
religion or family
and as long as we are writing the computer program
i thought well let's go ahead and throw in some parts of speech pronouns
prepositions
because it was easy who cares
i go back
start to analyze these dramatic essays
squiggly discover that
the content of what people were writing about
hetnet didn't matter in terms of if they improved in their help they're not
instead
ulysses with these job boards pronouns and articles of prepositions and so
forth
that didn't matter
not think about this
here people are writing about
deeply disturbing issues
the actual topics that dealt with
tragedies devastation horrible things
the topics themselves in the words associate with those topics made no
difference
instead these little words like aryan vi and and didn't matter
i get looking for the obvious
but in fact
and i don't pay attention to what people were saying
but not how they were saying it
so how do i go about
analyzing watch versus how
well turns out that they're difficult classes of words that makes these makes
that look at this distinction whatever it is
if you're looking at what people are writing about you look at when park all
content word caesar now into the regular verbs it as yet since i'm a adverbs
you know that's stuff of thought this is the stuff of communication we were
trying to talk to somebody gogo
and search terms are all based on these content words
the other group of words are art class of words urgently cult
function works
at funship words
are made up of the most boring
words that you can imagine
they're made up of
pronouns army he she
prepositions to of four
auxiliary verbs and is have
out wake you up by-catch they'll have to wake up i keep talking about these
function worth dot
it turns out these function words
are really interested because first of all
there's only about five hundred
function words in english
so they account for far less than one percent of all the words we know we hear
weary
nevertheless they reflect fifty-five dist
to sixty percent of all the words that we
that we are surrounded with their everywhere
but we don't pay attention to them
in english and in other languages
there the shortest words there are a hand
when they're spoken or with your reading
pcf into your brain at the speed of less than two tenths of a second
meaning that they're processed essentially not consciously
but there's something even more interesting about that
they are social they are
profoundly social
let me give me an example let's say you're walking along
you see in note on the ground
you pick it up and says
i am placing it on the table
well that kinda makes an skied it doesn't
by places on the table this to content words placing in table
all the rest of function worked
i am yet
on p
now the reason this doesn't make sense to most of us
is it was i
no idea amma implies
present tense when was it written
no idea what it is
on v table v table
needs its a table that
dot author knew about
and they intended recipient of this note new about
but nobody else did and in fact
there's no only has meaning
to the uh... authored
and the recipient of the note
at a particular time in a particular location in fact if i took that no to
this all turn out six months later and say
what's this all about there's a good chance that the the off was safe
no idea
function of words are social
they tell us about the author they tell us about the relationship between the
author of the recipient
and relationship between the author
and the topic itself
is take heart of what i want to talk to you about today
by analyzing function words
we start to get a sense
of who people are quick to relationships are
how they think about theirselves
and how they
connect with others
whether a lot of function words
honestly i could talk for several hours about function words but i'm going to
spare you that it just focus on it
couple today to just give you a flavor of her
why they're so interesting
let's start off with pro downs of the storm up with third-person pronouns like
t-shirt ye bang
it turns out some people out there in the world used pronouns these
third-person pronouns at high rates
and other people at lower rates
what kind of person would use
well you have to think about pronouns as and all function words
in terms of
where our people paying attention
if you are using these third-person pronounced by different definition
you're paying attention to other people
you care about other people you're thinking about other people
at people who use these at high rates
are much more socially engaged
we can analyze emails tweets and so forth and get a sense someone's social
engagement just by looking at this
how about first-person singular part as a highly me in mind
okay
you see tential argument somewhat is attending to their thoughts feelings
behaviors
to themselves and subway would use these words more
what kind of person you think uses aren't words the most
you're sitting there you're speaking for them
well somebody who's
self-centered self-important narcissistic
uh...
hungry for power and highest status
you would be completely wrong
in fact the person who is
highest instead issues is on words the least
the new rephrase that
the hire anybody isn't status the less they use a high were ds
the lower so what is in status
the higher they use a spy works
i discovered this by analyzing emails
isdn messages natural conversations business groups and so forth
in the affects were huge
i look at these results and i thought
this must be true for other people but he can't possibly be true for me
you know i'd love everybody equally
isolated analyze my own emails
on the same as everybody else
alexey melt that i get from the undergraduate student
dear doctor pennebaker i would like to know if i could possibly meet with you
because i think i need to change my grades are right back
dear student thank you so much for your email
unfortunately uh... later which prices don't work well
i look at my email to the dean
dare denied jamie pennebaker and i would like to ask if i could do this and i
could get that like to do this in the team rice packed
dear jamie thank you so much for your email so forth
not everybody is being completely polite
nobody's putting anybody down
this is the language of power in status
it tells us where people are paying attention high status person
is looking out at the world the los esperamos to be looking more inwardly
what about others states
lets move beyond status
let's look at emotional states
you would think that someone would be paying more attention to themselves
if they're in pain
it could be physical pain or emotional pain
in fact you alive we look at people who are depressed we've done any studies on
this we know the people who are depressed
pay attention to themselves more and they used the word i'm more
that one of our very first time he's looked at the poetry of suicidal and non
suicidal poets that we get this research where we
with through analyzer poetry and initially i thought
whether big big differences in the degree that they use negative emotion
words
not true
suicidal and non suicidal poets all use negative motion words at high rates
i think it's part of it is part of the job description
the big difference was their use of the word on
suicidal poets used to work hard work
consider this poll on this is by sylvia plath you later committed suicide
despotic listen to the way that she uses the word i
and first-person singular
and this is the optus
and taking some lines from her poland not bad girls love song
eyes shut my eyes and all the world drops dead
i've missed my live and always born again
i think a major up inside my head
by fitting in return the way you said but i grow old and i forget renee
i think i made you up inside my head
you can almost see plan
embracing her sorrow her misery and so forth
and you can compare her writing with uh... other poets who write about in but
not suicidal poets who right about lost love
when they gave him almost see them holding at all
from a distance of their looking at it from aboard distance
third-person perspective
now there's a really interesting things head out uh...
important theory within psychology about depression
and people who are depressed arf
thought to be people who are very high in self-awareness urself focus
and part of this is a also tend to be
extremely uh... honest
there often
today are it
they're they're many study showing that they
have this deficit in they're not able to have
a positive illusions about ourselves
those of us who are depressed
get by every day but holding these insane illusions about the life
but these people
are brutally honest
now this made me wonder
throwaway depression for just a second could return this entire thing upside
down to find out if depressed people
or if we could use the computer program
linguistic lie-detector let me for anybody so
in fact we did some studies for we bra people no lab we do some to lie or tell
the truth
we analyze court transcripts of people who are all found guilty half of whom
were exhausted later exonerated
and the affects rectly quite impressive
we did a pretty good job at telling to someone who is telling the truth verses
lying in one of the best words
was he uses the word i'm people who tell the truth
used the word i border pony what they're saying liars
pertaining to hold off distancing themselves
lie detection depression
status are all something so we can look at but one of the things that i'm most
interested now is looking at groups looking at the relationship between two
people can you tell how to people
forget who won't bite
by analyzing the weight they're using function worked with each other
and the answer is yes we look at the percentage of each class of words and we
come up with the metrics mccall language style matchy
and the more that two people are matching their function word use
the more there on the same page
the more they're talking about something in the same way
now what i said we started to look at this was with speed dating
not going to tell you
love speed dating i've really it just
i would never do it in a million years
but anchorage all of you to go do speed dating it when you do
invite a researcher along because there is no carradine
it is better
we've been involved in st speed dating projects where
people coming in
and in there for a minute date we tape-recorded they know we are heavily
transcribe the not the way they talk
the more they match in their language
the more likely they are to go out on a date we can predict who will go on a
date at rates slightly better than the people themselves here
we've done studies with young dating couples where we uh...
they had to be in our study they have to give this ten days of the instant
messages our audience
and what we do is we analyze their audience with its this style matching
and we do much much better than they do have prediction if they'll be together
three months later
the fact is his these words are telling us how individuals and
pairs of people are connecting
what about groups now this is an area that were network you know we're looking
at working groups summary groups that we've uh... work with people in the
business school would look at uh... people you get to know you groups
we do at occasional groups and what we're finding his bike looking at group
of say five a six people we cannot get a sense of how productive the group will
be and also how cohesive the group
simply by looking at the style matching
but here's where things are starting to get you christine
bite tracking a groups this interacting and say they're all interacting online
we can't have a computer of monitoring how the group is behaving imagine for
example
you are up
in this group
and a computer approach
comes to your group every now and then
anna message comes and says
you guys are not paying attention to one another
you need to be you'd be more attentive to what the other people are saying or
you guys for the last few minutes have strayed off topic
try to get back on topic or that loud mouth in the group the computer comes in
missus
joan for the last five minutes you said fifty percent of the words when she
stand back
any courage others to talk
we have now created a program that does the s
and we've now tested out with hundreds of groups and that we are getting very
very promising results
now you can start to see
why a m so excited about this world the function works
that we're not taking this in all these in
directions that uh... i'd never would've thought about
we've been looking at it in terms of looking at historical records can you
tell if a particular uh... explorer committed suicide tourist murdered we've
got a project on that
can you look at a company and get a sense of how their internal
communications are working how well they are connected with the people in their
company or with their clients we look at that to corporate uh... earnings reports
or the quarterly phone calls to get a sense of the internal groups dynamics of
the company we've worked with the government to try to get a sense of
terrorist groups and how if they are likely to behaved badly
we help people sort out their love lives it
you can start to say that
bike harnessing these
east the power of these function words we can get a sense of individuals and
groups and help people are connected
now but i wouldn't hurt you to do
i'd like you to go home
tonight
and i want you to start looking at your emails here tweets year high amps or
whatever
added doing that
what i hope you start to see is
first of all you learn a little bit more about your relationship with others but
more than anything
i hope you learned a little bit of about yourself