ABOUT THE SPEAKER
Leila Takayama - Social scientist
Leila Takayama conducts research on human-robot interaction.

Why you should listen

Leila Takayama is an acting associate professor of Psychology at the University of California, Santa Cruz, where she founded and leads the Re-Embodied Cognition Lab. Her lab examines how people make sense of, interact with, and relate to new technologies. Prior to academia, she was a researcher at GoogleX and Willow Garage, where she developed a taste for working alongside engineers, designers, animators, and more. Her interdisciplinary research continues in her current work on what happens when people interact with robots and through robots.

Takayama is a World Economic Forum Global Futures Council Member and Young Global Leader. In 2015, she was presented the IEEE Robotics & Automation Society Early Career Award. In 2012, she was named a TR35 winner and one of the 100 most creative people in business by Fast Company. She completed her PhD in Communication at Stanford University in 2008, advised by Professor Clifford Nass. She also holds a PhD minor in Psychology from Stanford, a master's degree in Communication from Stanford, and bachelor's of arts degrees in Psychology and Cognitive Science from UC Berkeley (2003). During her graduate studies, she was a research assistant in the User Interface Research (UIR) group at Palo Alto Research Center (PARC).

Photo: Melissa DeWitt

More profile about the speaker
Leila Takayama | Speaker | TED.com
TEDxPaloAlto

Leila Takayama: What's it like to be a robot?

Filmed:
1,183,118 views

We already live among robots: tools and machines like dishwashers and thermostats so integrated into our lives that we'd never think to call them that. What will a future with even more robots look like? Social scientist Leila Takayama shares some unique challenges of designing for human-robot interactions -- and how experimenting with robotic futures actually leads us to a better understanding of ourselves.
- Social scientist
Leila Takayama conducts research on human-robot interaction. Full bio

Double-click the English transcript below to play the video.

00:12
You only get one chance
to make a first impression,
0
760
2656
00:15
and that's true if you're a robot
as well as if you're a person.
1
3440
3176
00:18
The first time that I met
one of these robots
2
6640
3016
00:21
was at a place
called Willow Garage in 2008.
3
9680
2896
00:24
When I went to visit there,
my host walked me into the building
4
12600
3016
00:27
and we met this little guy.
5
15640
1576
00:29
He was rolling into the hallway,
6
17240
1656
00:30
came up to me, sat there,
7
18920
1816
00:32
stared blankly past me,
8
20760
2256
00:35
did nothing for a while,
9
23040
1656
00:36
rapidly spun his head around 180 degrees
10
24720
1936
00:38
and then ran away.
11
26680
1536
00:40
And that was not a great first impression.
12
28240
2176
00:42
The thing that I learned
about robots that day
13
30440
2176
00:44
is that they kind of do their own thing,
14
32640
2176
00:46
and they're not totally aware of us.
15
34840
2136
00:49
And I think as we're experimenting
with these possible robot futures,
16
37000
3239
00:52
we actually end up learning
a lot more about ourselves
17
40263
2673
00:54
as opposed to just these machines.
18
42960
1656
00:56
And what I learned that day
19
44640
1336
00:58
was that I had pretty high
expectations for this little dude.
20
46000
3416
01:01
He was not only supposed to be able
to navigate the physical world,
21
49440
3176
01:04
but also be able
to navigate my social world --
22
52640
2656
01:07
he's in my space; it's a personal robot.
23
55320
2176
01:09
wWhy didn't it understand me?
24
57520
2016
01:11
My host explained to me,
25
59560
1256
01:12
"Well, the robot is trying
to get from point A to point B,
26
60840
3176
01:16
and you were an obstacle in his way,
27
64040
1776
01:17
so he had to replan his path,
28
65840
2016
01:19
figure out where to go,
29
67880
1256
01:21
and then get there some other way,"
30
69160
1696
01:22
which was actually
not a very efficient thing to do.
31
70880
2456
01:25
If that robot had figured out
that I was a person, not a chair,
32
73360
3256
01:28
and that I was willing
to get out of its way
33
76640
2096
01:30
if it was trying to get somewhere,
34
78760
1656
01:32
then it actually
would have been more efficient
35
80440
2216
01:34
at getting its job done
36
82680
1256
01:35
if it had bothered
to notice that I was a human
37
83960
2216
01:38
and that I have different affordances
than things like chairs and walls do.
38
86200
3576
01:41
You know, we tend to think of these robots
as being from outer space
39
89800
3216
01:45
and from the future
and from science fiction,
40
93040
2136
01:47
and while that could be true,
41
95200
1416
01:48
I'd actually like to argue
that robots are here today,
42
96640
2656
01:51
and they live and work
amongst us right now.
43
99320
2776
01:54
These are two robots that live in my home.
44
102120
2856
01:57
They vacuum the floors
and they cut the grass
45
105000
2496
01:59
every single day,
46
107520
1216
02:00
which is more than I would do
if I actually had time to do these tasks,
47
108760
3376
02:04
and they probably
do it better than I would, too.
48
112160
2336
02:06
This one actually takes care of my kitty.
49
114520
2496
02:09
Every single time
he uses the box, it cleans it,
50
117040
2576
02:11
which is not something I'm willing to do,
51
119640
1976
02:13
and it actually makes
his life better as well as mine.
52
121640
2576
02:16
And while we call these robot products --
53
124240
2416
02:18
it's a "robot vacuum cleaner,
it's a robot lawnmower,
54
126680
2696
02:21
it's a robot littler box,"
55
129400
1495
02:22
I think there's actually a bunch
of other robots hiding in plain sight
56
130919
4137
02:27
that have just become so darn useful
57
135080
1856
02:28
and so darn mundane
58
136960
1456
02:30
that we call them things
like, "dishwasher," right?
59
138440
2496
02:32
They get new names.
60
140960
1216
02:34
They don't get called robot anymore
61
142200
1696
02:35
because they actually
serve a purpose in our lives.
62
143920
2416
02:38
Similarly, a thermostat, right?
63
146360
1496
02:39
I know my robotics friends out there
64
147880
1776
02:41
are probably cringing
at me calling this a robot,
65
149680
2336
02:44
but it has a goal.
66
152040
1256
02:45
Its goal is to make my house
66 degrees Fahrenheit,
67
153320
2896
02:48
and it senses the world.
68
156240
1256
02:49
It knows it's a little bit cold,
69
157520
1576
02:51
it makes a plan and then
it acts on the physical world.
70
159120
2616
02:53
It's robotics.
71
161760
1256
02:55
Even if it might not
look like Rosie the Robot,
72
163040
2576
02:57
it's doing something
that's really useful in my life
73
165640
2936
03:00
so I don't have to take care
74
168600
1376
03:02
of turning the temperature
up and down myself.
75
170000
2576
03:04
And I think these systems
live and work amongst us now,
76
172600
3816
03:08
and not only are these systems
living amongst us
77
176440
2336
03:10
but you are probably
a robot operator, too.
78
178800
2656
03:13
When you drive your car,
79
181480
1256
03:14
it feels like you are operating machinery.
80
182760
2216
03:17
You are also going
from point A to point B,
81
185000
2816
03:19
but your car probably has power steering,
82
187840
2216
03:22
it probably has automatic braking systems,
83
190080
2696
03:24
it might have an automatic transmission
and maybe even adaptive cruise control.
84
192800
3736
03:28
And while it might not be
a fully autonomous car,
85
196560
2936
03:31
it has bits of autonomy,
86
199520
1296
03:32
and they're so useful
87
200840
1336
03:34
and they make us drive safer,
88
202200
1816
03:36
and we just sort of feel
like they're invisible-in-use, right?
89
204040
3656
03:39
So when you're driving your car,
90
207720
1576
03:41
you should just feel like
you're going from one place to another.
91
209320
3096
03:44
It doesn't feel like it's this big thing
that you have to deal with and operate
92
212440
3736
03:48
and use these controls
93
216200
1256
03:49
because we spent so long
learning how to drive
94
217480
2176
03:51
that they've become
extensions of ourselves.
95
219680
2696
03:54
When you park that car
in that tight little garage space,
96
222400
2696
03:57
you know where your corners are.
97
225120
1576
03:58
And when you drive a rental car
that maybe you haven't driven before,
98
226720
3256
04:02
it takes some time
to get used to your new robot body.
99
230000
3056
04:05
And this is also true for people
who operate other types of robots,
100
233080
3976
04:09
so I'd like to share with you
a few stories about that.
101
237080
2600
04:12
Dealing with the problem
of remote collaboration.
102
240240
2336
04:14
So, at Willow Garage
I had a coworker named Dallas,
103
242600
2576
04:17
and Dallas looked like this.
104
245200
1576
04:18
He worked from his home in Indiana
in our company in California.
105
246800
4056
04:22
He was a voice in a box
on the table in most of our meetings,
106
250880
2936
04:25
which was kind of OK
except that, you know,
107
253840
2215
04:28
if we had a really heated debate
and we didn't like what he was saying,
108
256079
3377
04:31
we might just hang up on him.
109
259480
1416
04:32
(Laughter)
110
260920
1015
04:33
Then we might have a meeting
after that meeting
111
261959
2217
04:36
and actually make the decisions
in the hallway afterwards
112
264200
2696
04:38
when he wasn't there anymore.
113
266920
1416
04:40
So that wasn't so great for him.
114
268360
1576
04:41
And as a robotics company at Willow,
115
269960
1736
04:43
we had some extra
robot body parts laying around,
116
271720
2336
04:46
so Dallas and his buddy Curt
put together this thing,
117
274080
2496
04:48
which looks kind of
like Skype on a stick on wheels,
118
276600
2936
04:51
which seems like a techy, silly toy,
119
279560
1736
04:53
but really it's probably
one of the most powerful tools
120
281320
2776
04:56
that I've seen ever made
for remote collaboration.
121
284120
2480
04:59
So now, if I didn't answer
Dallas' email question,
122
287160
3496
05:02
he could literally roll into my office,
123
290680
2216
05:04
block my doorway
and ask me the question again --
124
292920
2576
05:07
(Laughter)
125
295520
1016
05:08
until I answered it.
126
296560
1216
05:09
And I'm not going to turn him off, right?
That's kind of rude.
127
297800
2976
05:12
Not only was it good
for these one-on-one communications,
128
300800
2696
05:15
but also for just showing up
at the company all-hands meeting.
129
303520
2936
05:18
Getting your butt in that chair
130
306480
1696
05:20
and showing people that you're present
and committed to your project
131
308200
3216
05:23
is a big deal
132
311440
1256
05:24
and can help remote collaboration a ton.
133
312720
2176
05:26
We saw this over the period
of months and then years,
134
314920
2856
05:29
not only at our company
but at others, too.
135
317800
2160
05:32
The best thing that can happen
with these systems
136
320720
2336
05:35
is that it starts to feel
like you're just there.
137
323080
2336
05:37
It's just you, it's just your body,
138
325440
1696
05:39
and so people actually start
to give these things personal space.
139
327160
3096
05:42
So when you're having a stand-up meeting,
140
330280
1976
05:44
people will stand around the space
141
332280
1656
05:45
just as they would
if you were there in person.
142
333960
2216
05:48
That's great until
there's breakdowns and it's not.
143
336200
2576
05:50
People, when they first see these robots,
144
338800
1976
05:52
are like, "Wow, where's the components?
There must be a camera over there,"
145
340800
3576
05:56
and they start poking your face.
146
344400
1576
05:58
"You're talking too softly,
I'm going to turn up your volume,"
147
346000
2936
06:00
which is like having a coworker
walk up to you and say,
148
348960
2616
06:03
"You're speaking too softly,
I'm going to turn up your face."
149
351600
2896
06:06
That's awkward and not OK,
150
354520
1256
06:07
and so we end up having to build
these new social norms
151
355800
2616
06:10
around using these systems.
152
358440
2216
06:12
Similarly, as you start
feeling like it's your body,
153
360680
3416
06:16
you start noticing things like,
"Oh, my robot is kind of short."
154
364120
3696
06:19
Dallas would say things to me --
he was six-foot tall --
155
367840
2656
06:22
and we would take him via robot
to cocktail parties and things like that,
156
370520
3456
06:26
as you do,
157
374000
1216
06:27
and the robot was about five-foot-tall,
which is close to my height.
158
375240
3336
06:30
And he would tell me,
159
378600
1216
06:31
"You know, people are not
really looking at me.
160
379840
2536
06:34
I feel like I'm just looking
at this sea of shoulders,
161
382400
2696
06:37
and it's just -- we need a taller robot."
162
385120
1976
06:39
And I told him,
163
387120
1256
06:40
"Um, no.
164
388400
1296
06:41
You get to walk in my shoes for today.
165
389720
1936
06:43
You get to see what it's like
to be on the shorter end of the spectrum."
166
391680
3536
06:47
And he actually ended up building
a lot of empathy for that experience,
167
395240
3376
06:50
which was kind of great.
168
398640
1256
06:51
So when he'd come visit in person,
169
399920
1656
06:53
he no longer stood over me
as he was talking to me,
170
401600
2416
06:56
he would sit down
and talk to me eye to eye,
171
404040
2096
06:58
which was kind of a beautiful thing.
172
406160
1736
06:59
So we actually decided
to look at this in the laboratory
173
407920
2656
07:02
and see what others kinds of differences
things like robot height would make.
174
410600
3656
07:06
And so half of the people in our study
used a shorter robot,
175
414280
2856
07:09
half of the people in our study
used a taller robot
176
417160
2416
07:11
and we actually found
that the exact same person
177
419600
2256
07:13
who has the exact same body
and says the exact same things as someone,
178
421880
3336
07:17
is more persuasive
and perceived as being more credible
179
425240
2616
07:19
if they're in a taller robot form.
180
427880
1656
07:21
It makes no rational sense,
181
429560
1816
07:23
but that's why we study psychology.
182
431400
1696
07:25
And really, you know,
the way that Cliff Nass would put this
183
433120
2856
07:28
is that we're having to deal
with these new technologies
184
436000
3016
07:31
despite the fact
that we have very old brains.
185
439040
2736
07:33
Human psychology is not changing
at the same speed that tech is
186
441800
2976
07:36
and so we're always playing catch-up,
187
444800
1816
07:38
trying to make sense of this world
188
446640
1656
07:40
where these autonomous things
are running around.
189
448320
2336
07:42
Usually, things that talk are people,
not machines, right?
190
450680
2736
07:45
And so we breathe a lot of meaning
into things like just height of a machine,
191
453440
4576
07:50
not a person,
192
458040
1256
07:51
and attribute that
to the person using the system.
193
459320
2360
07:55
You know, this, I think,
is really important
194
463120
2216
07:57
when you're thinking about robotics.
195
465360
1736
07:59
It's not so much about reinventing humans,
196
467120
2096
08:01
it's more about figuring out
how we extend ourselves, right?
197
469240
3136
08:04
And we end up using things
in ways that are sort of surprising.
198
472400
2976
08:07
So these guys can't play pool
because the robots don't have arms,
199
475400
4256
08:11
but they can heckle the guys
who are playing pool
200
479680
2336
08:14
and that can be an important thing
for team bonding,
201
482040
3176
08:17
which is kind of neat.
202
485240
1296
08:18
People who get really good
at operating these systems
203
486560
2496
08:21
will even do things
like make up new games,
204
489080
2056
08:23
like robot soccer
in the middle of the night,
205
491160
2136
08:25
pushing the trash cans around.
206
493320
1456
08:26
But not everyone's good.
207
494800
1576
08:28
A lot of people have trouble
operating these systems.
208
496400
2496
08:30
This is actually a guy
who logged into the robot
209
498920
2256
08:33
and his eyeball was turned
90 degrees to the left.
210
501200
2376
08:35
He didn't know that,
211
503600
1256
08:36
so he ended up just bashing
around the office,
212
504880
2176
08:39
running into people's desks,
getting super embarrassed,
213
507080
2616
08:41
laughing about it --
his volume was way too high.
214
509720
2336
08:44
And this guy here
in the image is telling me,
215
512080
2136
08:46
"We need a robot mute button."
216
514240
2096
08:48
And by that what he really meant
was we don't want it to be so disruptive.
217
516360
3496
08:51
So as a robotics company,
218
519880
1616
08:53
we added some obstacle
avoidance to the system.
219
521520
2456
08:56
It got a little laser range finder
that could see the obstacles,
220
524000
3056
08:59
and if I as a robot operator
try to say, run into a chair,
221
527080
3136
09:02
it wouldn't let me,
it would just plan a path around,
222
530240
2496
09:04
which seems like a good idea.
223
532760
1856
09:06
People did hit fewer obstacles
using that system, obviously,
224
534640
3176
09:09
but actually, for some of the people,
225
537840
2096
09:11
it took them a lot longer
to get through our obstacle course,
226
539960
2856
09:14
and we wanted to know why.
227
542840
1560
09:17
It turns out that there's
this important human dimension --
228
545080
3056
09:20
a personality dimension
called locus of control,
229
548160
2296
09:22
and people who have
a strong internal locus of control,
230
550480
3136
09:25
they need to be the masters
of their own destiny --
231
553640
3056
09:28
really don't like giving up control
to an autonomous system --
232
556720
3096
09:31
so much so that they will
fight the autonomy;
233
559840
2136
09:34
"If I want to hit that chair,
I'm going to hit that chair."
234
562000
3096
09:37
And so they would actually suffer
from having that autonomous assistance,
235
565120
3616
09:40
which is an important thing for us to know
236
568760
2576
09:43
as we're building increasingly
autonomous, say, cars, right?
237
571360
3296
09:46
How are different people going
to grapple with that loss of control?
238
574680
3360
09:50
It's going to be different
depending on human dimensions.
239
578880
2696
09:53
We can't treat humans
as if we're just one monolithic thing.
240
581600
3496
09:57
We vary by personality, by culture,
241
585120
2416
09:59
we even vary by emotional state
moment to moment,
242
587560
2456
10:02
and being able to design these systems,
243
590040
1976
10:04
these human-robot interaction systems,
244
592040
2296
10:06
we need to take into account
the human dimensions,
245
594360
2736
10:09
not just the technological ones.
246
597120
1720
10:11
Along with a sense of control
also comes a sense of responsibility.
247
599640
4296
10:15
And if you were a robot operator
using one of these systems,
248
603960
2856
10:18
this is what the interface
would look like.
249
606840
2056
10:20
It looks a little bit like a video game,
250
608920
1936
10:22
which can be good because
that's very familiar to people,
251
610880
2976
10:25
but it can also be bad
252
613880
1216
10:27
because it makes people feel
like it's a video game.
253
615120
2456
10:29
We had a bunch of kids
over at Stanford play with the system
254
617600
2856
10:32
and drive the robot
around our office in Menlo Park,
255
620480
2456
10:34
and the kids started saying things like,
256
622960
1936
10:36
"10 points if you hit that guy over there.
20 points for that one."
257
624920
3176
10:40
And they would
chase them down the hallway.
258
628120
2016
10:42
(Laughter)
259
630160
1016
10:43
I told them, "Um, those are real people.
260
631200
1936
10:45
They're actually going to bleed
and feel pain if you hit them."
261
633160
3296
10:48
And they'd be like, "OK, got it."
262
636480
1616
10:50
But five minutes later,
they would be like,
263
638120
2056
10:52
"20 points for that guy over there,
he just looks like he needs to get hit."
264
640200
3616
10:55
It's a little bit
like "Ender's Game," right?
265
643840
2136
10:58
There is a real world on that other side
266
646000
1936
10:59
and I think it's our responsibility
as people designing these interfaces
267
647960
3416
11:03
to help people remember
268
651400
1256
11:04
that there's real consequences
to their actions
269
652680
2256
11:06
and to feel a sense of responsibility
270
654960
2296
11:09
when they're operating
these increasingly autonomous things.
271
657280
3280
11:13
These are kind of a great example
272
661840
2296
11:16
of experimenting with one
possible robotic future,
273
664160
3256
11:19
and I think it's pretty cool
that we can extend ourselves
274
667440
3856
11:23
and learn about the ways
that we extend ourselves
275
671320
2336
11:25
into these machines
276
673680
1216
11:26
while at the same time
being able to express our humanity
277
674920
2696
11:29
and our personality.
278
677640
1216
11:30
We also build empathy for others
279
678880
1576
11:32
in terms of being
shorter, taller, faster, slower,
280
680480
3216
11:35
and maybe even armless,
281
683720
1416
11:37
which is kind of neat.
282
685160
1336
11:38
We also build empathy
for the robots themselves.
283
686520
2536
11:41
This is one of my favorite robots.
284
689080
1656
11:42
It's called the Tweenbot.
285
690760
1456
11:44
And this guy has a little flag that says,
286
692240
1976
11:46
"I'm trying to get
to this intersection in Manhattan,"
287
694240
2576
11:48
and it's cute and rolls
forward, that's it.
288
696840
2776
11:51
It doesn't know how to build a map,
it doesn't know how to see the world,
289
699640
3456
11:55
it just asks for help.
290
703120
1256
11:56
The nice thing about people
291
704400
1336
11:57
is that it can actually depend
upon the kindness of strangers.
292
705760
3096
12:00
It did make it across the park
to the other side of Manhattan --
293
708880
3896
12:04
which is pretty great --
294
712800
1256
12:06
just because people would pick it up
and point it in the right direction.
295
714080
3456
12:09
(Laughter)
296
717560
936
12:10
And that's great, right?
297
718520
1256
12:11
We're trying to build
this human-robot world
298
719800
2696
12:14
in which we can coexist
and collaborate with one another,
299
722520
3416
12:17
and we don't need to be fully autonomous
and just do things on our own.
300
725960
3376
12:21
We actually do things together.
301
729360
1496
12:22
And to make that happen,
302
730880
1256
12:24
we actually need help from people
like the artists and the designers,
303
732160
3256
12:27
the policy makers, the legal scholars,
304
735440
1856
12:29
psychologists, sociologists,
anthropologists --
305
737320
2216
12:31
we need more perspectives in the room
306
739560
1816
12:33
if we're going to do the thing
that Stu Card says we should do,
307
741400
2976
12:36
which is invent the future
that we actually want to live in.
308
744400
3936
12:40
And I think we can continue to experiment
309
748360
2656
12:43
with these different
robotic futures together,
310
751040
2176
12:45
and in doing so, we will end up
learning a lot more about ourselves.
311
753240
4680
12:50
Thank you.
312
758720
1216
12:51
(Applause)
313
759960
2440

▲Back to top

ABOUT THE SPEAKER
Leila Takayama - Social scientist
Leila Takayama conducts research on human-robot interaction.

Why you should listen

Leila Takayama is an acting associate professor of Psychology at the University of California, Santa Cruz, where she founded and leads the Re-Embodied Cognition Lab. Her lab examines how people make sense of, interact with, and relate to new technologies. Prior to academia, she was a researcher at GoogleX and Willow Garage, where she developed a taste for working alongside engineers, designers, animators, and more. Her interdisciplinary research continues in her current work on what happens when people interact with robots and through robots.

Takayama is a World Economic Forum Global Futures Council Member and Young Global Leader. In 2015, she was presented the IEEE Robotics & Automation Society Early Career Award. In 2012, she was named a TR35 winner and one of the 100 most creative people in business by Fast Company. She completed her PhD in Communication at Stanford University in 2008, advised by Professor Clifford Nass. She also holds a PhD minor in Psychology from Stanford, a master's degree in Communication from Stanford, and bachelor's of arts degrees in Psychology and Cognitive Science from UC Berkeley (2003). During her graduate studies, she was a research assistant in the User Interface Research (UIR) group at Palo Alto Research Center (PARC).

Photo: Melissa DeWitt

More profile about the speaker
Leila Takayama | Speaker | TED.com