1
00:00:03,080 --> 00:00:05,920
Speaker 1: Welcome to Stuff to Blow your Mind from how Stuff

2
00:00:05,960 --> 00:00:14,560
Speaker 1: Works dot com. Hey, welcome to Stuff to Blow your Mind.

3
00:00:14,600 --> 00:00:17,800
Speaker 1: My name is Robert Lamb and I'm Joe McCormick and Robert.

4
00:00:17,960 --> 00:00:20,639
Speaker 1: I want to take you back to a conversation we had.

5
00:00:21,040 --> 00:00:23,360
Speaker 1: I think it was last December. He was right after

6
00:00:23,520 --> 00:00:26,800
Speaker 1: I went to see the new, the most recent Star

7
00:00:26,800 --> 00:00:29,680
Speaker 1: Wars movie, Rogue one. Oh yes, uh, And I am

8
00:00:29,720 --> 00:00:32,040
Speaker 1: on the cusp, the very cusp of seeing it myself

9
00:00:32,040 --> 00:00:35,000
Speaker 1: and waiting for it to become rent rental options. So,

10
00:00:35,200 --> 00:00:37,480
Speaker 1: oh it's not yet. No, still got still a week

11
00:00:37,520 --> 00:00:40,320
Speaker 1: or two out. So has everything been spoiled for you

12
00:00:40,400 --> 00:00:43,640
Speaker 1: so far? No? People have been a little um cooler

13
00:00:43,680 --> 00:00:45,800
Speaker 1: on this one. Um. I think some things are probably

14
00:00:45,800 --> 00:00:47,960
Speaker 1: spoiled for me, but not like that last one where

15
00:00:47,960 --> 00:00:51,000
Speaker 1: everybody just really felt the need to, you know, just

16
00:00:51,120 --> 00:00:53,239
Speaker 1: lay it all out on social media. Let me let

17
00:00:53,280 --> 00:00:55,200
Speaker 1: me spoil one thing for you here and they go

18
00:00:55,280 --> 00:00:59,280
Speaker 1: to space and there's a war there, ye, stars among

19
00:00:59,360 --> 00:01:01,720
Speaker 1: the stars. But there is one thing in the movie. Okay,

20
00:01:01,720 --> 00:01:04,080
Speaker 1: so mild spoiler for Rogue one coming out. It's something

21
00:01:04,080 --> 00:01:06,720
Speaker 1: that probably everybody already knows. It's also not really about

22
00:01:06,720 --> 00:01:09,240
Speaker 1: the content of the movie, but just about which characters

23
00:01:09,240 --> 00:01:12,400
Speaker 1: you see. But if you are ready, are you ready

24
00:01:12,400 --> 00:01:14,920
Speaker 1: for the smild spoiler? Okay, we get to see if

25
00:01:14,920 --> 00:01:17,679
Speaker 1: you remember back to the original Star Wars back go

26
00:01:17,800 --> 00:01:21,319
Speaker 1: back to the seventies, Peter Cushing as Grand mof Tarkin,

27
00:01:21,440 --> 00:01:24,440
Speaker 1: the guy who was in fact Darth Vader's boss on

28
00:01:24,480 --> 00:01:29,520
Speaker 1: the Star Him. Yeah, he's enough of this, Vada release him.

29
00:01:29,560 --> 00:01:32,080
Speaker 1: And we love Peter Cushing because he was in all

30
00:01:32,080 --> 00:01:34,200
Speaker 1: these old monster movies. He he goes back to the

31
00:01:34,200 --> 00:01:37,319
Speaker 1: Hammer movies. He was. He was Dr Frankenstein, like the

32
00:01:37,440 --> 00:01:40,120
Speaker 1: villainous Dr Frankenstein of the Hammer films, and he I

33
00:01:40,160 --> 00:01:42,759
Speaker 1: think was the hero of the version of the Mummy

34
00:01:42,880 --> 00:01:46,399
Speaker 1: that has Christopher Lee as as the Mummy. Uh, the one.

35
00:01:46,440 --> 00:01:48,360
Speaker 1: I've got the poster for it in the house, except

36
00:01:48,360 --> 00:01:51,840
Speaker 1: it's the Belgian posters, so it's La Malediction de Feron.

37
00:01:53,160 --> 00:01:56,720
Speaker 1: But yeah, so Peter Cushing was the original Grand mof Tarkin,

38
00:01:56,880 --> 00:02:00,200
Speaker 1: this bad Empire guy who was Darth Vader's ball us.

39
00:02:00,800 --> 00:02:03,240
Speaker 1: And the thing they do in the New Rogue One

40
00:02:03,600 --> 00:02:07,080
Speaker 1: is they bring him back. He's dead, he has passed away.

41
00:02:07,600 --> 00:02:10,560
Speaker 1: This movie takes place a little bit before the original

42
00:02:10,600 --> 00:02:13,600
Speaker 1: Star Wars is supposed to have taken place, but they

43
00:02:14,120 --> 00:02:17,160
Speaker 1: bring this character back and they have an actor stand

44
00:02:17,160 --> 00:02:19,680
Speaker 1: in as him, but it's not just a recast role.

45
00:02:19,960 --> 00:02:22,280
Speaker 1: They try to make it look as if this is

46
00:02:22,400 --> 00:02:26,200
Speaker 1: Peter Cushing standing here delivering the lines with c g I.

47
00:02:26,520 --> 00:02:28,480
Speaker 1: And this is an odd choice because, all right, so

48
00:02:29,600 --> 00:02:32,160
Speaker 1: you're gonna have Darth Vader in there, that's easy to do.

49
00:02:32,280 --> 00:02:34,680
Speaker 1: Darth Vader is a dude in a suit, voiced by

50
00:02:34,760 --> 00:02:37,880
Speaker 1: James Earl Jones. James Earl Jones is still alive, so

51
00:02:37,960 --> 00:02:40,720
Speaker 1: you can check that one off the list. But Grandma Tarkin,

52
00:02:40,919 --> 00:02:43,720
Speaker 1: like you said, uh that the actor is dead. So

53
00:02:43,800 --> 00:02:45,640
Speaker 1: it seems to me like the first easiest thing to

54
00:02:45,680 --> 00:02:47,440
Speaker 1: do is just don't have those scenes. If you know

55
00:02:47,480 --> 00:02:51,120
Speaker 1: it's gonna be problematic, don't even mess with it, or um,

56
00:02:51,639 --> 00:02:54,919
Speaker 1: just use an actual living actor such as Wayne pie

57
00:02:55,000 --> 00:02:58,320
Speaker 1: Graham who played him in Revenge of the Sith. Or

58
00:02:58,480 --> 00:03:00,720
Speaker 1: go with Ben Cross, who's another actor that I've seen

59
00:03:00,720 --> 00:03:03,680
Speaker 1: over the years brought up as potential Tarkan casting, or

60
00:03:03,880 --> 00:03:06,280
Speaker 1: head go with Ralph Finds, Like, clearly you have the

61
00:03:06,360 --> 00:03:09,480
Speaker 1: money to throw down the well of expensive c g

62
00:03:09,600 --> 00:03:11,920
Speaker 1: I equipment. Just go ahead and hire Ralph Finds. I

63
00:03:11,960 --> 00:03:15,600
Speaker 1: know he's pricy, but he's great and the consummate evil Brett. Yeah,

64
00:03:15,600 --> 00:03:17,920
Speaker 1: he even kind of looks like a younger Peter Cushing.

65
00:03:18,280 --> 00:03:20,680
Speaker 1: He's got that same kind of angular face, like the

66
00:03:20,720 --> 00:03:24,000
Speaker 1: thin long face with the jaw and the scowl. It's

67
00:03:24,040 --> 00:03:27,800
Speaker 1: all there. And it's not like fans of various franchise

68
00:03:27,960 --> 00:03:30,760
Speaker 1: are not clearly cool with recasting. It's not like we're

69
00:03:30,800 --> 00:03:34,280
Speaker 1: gonna be thrown into, you know, a traumatic spin, because

70
00:03:34,320 --> 00:03:37,520
Speaker 1: you can look to a Game of Thrones, James Bond, Twilight,

71
00:03:37,600 --> 00:03:40,680
Speaker 1: Harry Potter, etcetera, like we we get it. We can

72
00:03:40,800 --> 00:03:43,640
Speaker 1: roll with a recast. Now. I want to go into

73
00:03:43,680 --> 00:03:47,680
Speaker 1: completely different directions thinking about this c g I grandmof Tarken.

74
00:03:48,480 --> 00:03:52,560
Speaker 1: One is that I didn't like it in the movie. Okay,

75
00:03:52,600 --> 00:03:55,320
Speaker 1: I saw it and I was just like, don't want this.

76
00:03:55,640 --> 00:03:57,920
Speaker 1: It pulled me out of the movie. It made me

77
00:03:58,000 --> 00:04:01,240
Speaker 1: stop being in the story and just thinking about how

78
00:04:01,240 --> 00:04:03,800
Speaker 1: did they do that? I don't On one hand, it

79
00:04:03,840 --> 00:04:07,240
Speaker 1: looked great, Like when you see the movie, I think

80
00:04:07,240 --> 00:04:10,280
Speaker 1: you will kind of have to agree. It's unless I'm

81
00:04:10,280 --> 00:04:14,160
Speaker 1: missing something. It's the best c g I simulation of

82
00:04:14,200 --> 00:04:17,760
Speaker 1: a real person that I've ever seen, Like, it looks amazing,

83
00:04:18,240 --> 00:04:21,720
Speaker 1: but it still looks not quite good enough that I

84
00:04:21,760 --> 00:04:24,159
Speaker 1: can just accept it and go with it. I kept

85
00:04:24,200 --> 00:04:28,240
Speaker 1: continually thinking, like, what am I looking at? It's almost

86
00:04:28,320 --> 00:04:31,320
Speaker 1: really him, but it's not quite really him, and it

87
00:04:31,400 --> 00:04:35,039
Speaker 1: made me feel icky. So in this it made you

88
00:04:35,080 --> 00:04:37,880
Speaker 1: descend into what we've come to know as a as

89
00:04:37,920 --> 00:04:41,120
Speaker 1: a as a species as the Uncanny Valley. Right, So

90
00:04:41,240 --> 00:04:43,240
Speaker 1: today is going to be the first of two episodes

91
00:04:43,279 --> 00:04:45,359
Speaker 1: we want to do about the Uncanny Valley, And this

92
00:04:45,400 --> 00:04:47,919
Speaker 1: first one we wanted descend into the Uncanny Valley, but

93
00:04:47,920 --> 00:04:50,520
Speaker 1: not just talk about it in terms of the standard

94
00:04:50,520 --> 00:04:53,039
Speaker 1: pop culture phenomenon, because this is one of those side

95
00:04:53,040 --> 00:04:56,200
Speaker 1: tech concepts that is totally filtered down into the mainstream.

96
00:04:56,279 --> 00:05:00,280
Speaker 1: Everybody talks about the Uncanny Valley. It's a totally norm mole,

97
00:05:00,720 --> 00:05:04,400
Speaker 1: ground level pop culture phenomenon now, especially with as much

98
00:05:04,480 --> 00:05:06,360
Speaker 1: bad C g I as we encounter in the movie.

99
00:05:06,839 --> 00:05:09,479
Speaker 1: But there it's also a scientific field of study. It's

100
00:05:09,480 --> 00:05:12,279
Speaker 1: something that people are looking into with empirical research to

101
00:05:12,320 --> 00:05:14,960
Speaker 1: try to figure out does it really exist, If it

102
00:05:15,000 --> 00:05:17,800
Speaker 1: does really exist, what causes it, what can be done

103
00:05:17,839 --> 00:05:19,680
Speaker 1: about it? So we want to look at it from

104
00:05:19,680 --> 00:05:22,719
Speaker 1: both of these angles today, right, So we should probably

105
00:05:22,800 --> 00:05:25,679
Speaker 1: roll through some just fun examples of this. We're gonna

106
00:05:25,680 --> 00:05:27,600
Speaker 1: try and not to go too long on this. If

107
00:05:27,640 --> 00:05:30,000
Speaker 1: we do, we'll cut it and save it for trailer talk.

108
00:05:30,040 --> 00:05:32,960
Speaker 1: Either way, we'll probably do a Facebook live trailer talk

109
00:05:33,440 --> 00:05:37,040
Speaker 1: on an upcoming Friday about some of these movies. Okay,

110
00:05:37,080 --> 00:05:38,880
Speaker 1: So I want to go back to a much earlier

111
00:05:38,960 --> 00:05:42,480
Speaker 1: experience for me, Robert, did you see The Mummy Returns

112
00:05:42,560 --> 00:05:45,120
Speaker 1: in two thousand one? Remember this one? I don't think

113
00:05:45,160 --> 00:05:47,480
Speaker 1: I saw The Mummy Returns. I saw the Money and

114
00:05:47,560 --> 00:05:50,640
Speaker 1: I remember digging it at the time, but not not

115
00:05:50,760 --> 00:05:52,880
Speaker 1: the old hammer one or the Universal one. No, no,

116
00:05:53,040 --> 00:05:55,479
Speaker 1: the yeah, the the the re the reboot of the Money.

117
00:05:55,560 --> 00:05:59,440
Speaker 1: Is his name, Brendan Frasier. Yeah, and what Arnold Voslo?

118
00:05:59,600 --> 00:06:02,120
Speaker 1: Yeah he was he. I I enjoyed him as they

119
00:06:02,320 --> 00:06:04,120
Speaker 1: kind of brought in some of these aspects of the

120
00:06:04,160 --> 00:06:07,839
Speaker 1: tragic Mummy figure, which I liked. Yeah, yeah, yeah, okay.

121
00:06:07,839 --> 00:06:10,480
Speaker 1: But in two thousand one we got the Scorpion King.

122
00:06:10,600 --> 00:06:12,960
Speaker 1: This is a character that appears in the Mummy Returns

123
00:06:13,160 --> 00:06:14,640
Speaker 1: and he's pretty much if you want to picture it,

124
00:06:14,640 --> 00:06:16,520
Speaker 1: if you haven't seen the movie. Actually you should look

125
00:06:16,600 --> 00:06:18,640
Speaker 1: up video of this. We're gonna tell you to go

126
00:06:18,720 --> 00:06:20,760
Speaker 1: look up images and video quite a few times in

127
00:06:20,800 --> 00:06:24,040
Speaker 1: these episodes because some visual aids will help. But if

128
00:06:24,040 --> 00:06:26,840
Speaker 1: you want to picture it, picture the concept of a centaur,

129
00:06:27,400 --> 00:06:31,000
Speaker 1: except replace the horse parts with scorpion parts and some

130
00:06:31,080 --> 00:06:34,600
Speaker 1: other random arthropod bits, and the man part on top

131
00:06:34,680 --> 00:06:38,919
Speaker 1: is Dwayne the Rock Johnson. Except it's not Dwayne the

132
00:06:39,000 --> 00:06:41,400
Speaker 1: Rock Johnson. There's a there's a bit of a problem

133
00:06:41,440 --> 00:06:44,480
Speaker 1: with the rocks. So the corpion Corpion, the Scorpion King

134
00:06:44,839 --> 00:06:49,360
Speaker 1: scuttles into action in the film, and you can tell

135
00:06:49,440 --> 00:06:52,520
Speaker 1: immediately something is wrong because it's not just the rock.

136
00:06:52,600 --> 00:06:56,080
Speaker 1: It's this c g I upper body designed to look

137
00:06:56,160 --> 00:06:58,640
Speaker 1: like the rock. It's supposed to be him, but it

138
00:06:58,680 --> 00:07:01,679
Speaker 1: doesn't look right. It looks like somebody took the rock,

139
00:07:02,480 --> 00:07:05,680
Speaker 1: skinned him, and then took the skin suit and then

140
00:07:05,800 --> 00:07:09,039
Speaker 1: boiled it and then maybe ironed it and rubbed it

141
00:07:09,080 --> 00:07:12,200
Speaker 1: down with wax, and then stretched it over somebody else,

142
00:07:12,240 --> 00:07:16,800
Speaker 1: like a bald Crispin Glover wearing a waxed up the

143
00:07:16,920 --> 00:07:20,520
Speaker 1: rock suit. And and that would be fine if that's

144
00:07:20,560 --> 00:07:22,840
Speaker 1: what they were going for. But I guess the disconnect

145
00:07:22,880 --> 00:07:25,480
Speaker 1: here is that clearly they wanted this to be like

146
00:07:25,560 --> 00:07:29,400
Speaker 1: the Rock as a scorpion center and not this um,

147
00:07:29,400 --> 00:07:32,640
Speaker 1: creepy in human above the nation that you've described, right,

148
00:07:32,720 --> 00:07:35,000
Speaker 1: And it's sort of I guess works because it's okay

149
00:07:35,000 --> 00:07:37,280
Speaker 1: if he's creepy, because he's a monster. But he was

150
00:07:37,320 --> 00:07:39,920
Speaker 1: creepy in a way that he clearly wasn't supposed to be.

151
00:07:40,440 --> 00:07:42,800
Speaker 1: It wasn't just that, oh he's a monster. He looks weird.

152
00:07:42,960 --> 00:07:45,400
Speaker 1: Something looked wrong with him. And this was at a

153
00:07:45,440 --> 00:07:49,160
Speaker 1: time when computer generated animations were hot, right two thousand one.

154
00:07:49,400 --> 00:07:51,840
Speaker 1: They seemed to be getting better all the time, and

155
00:07:51,920 --> 00:07:55,040
Speaker 1: yet they were terrible, producing these characters that were not

156
00:07:55,080 --> 00:07:59,560
Speaker 1: only not convincingly human, they were literally physically unpleasant to

157
00:07:59,600 --> 00:08:02,320
Speaker 1: look at. They were repulsive. Yeah, it was a period

158
00:08:02,320 --> 00:08:05,360
Speaker 1: when everyone was just foolishly optimistic about what we could

159
00:08:05,360 --> 00:08:07,880
Speaker 1: achieve with c g I, and you know, into in

160
00:08:08,040 --> 00:08:11,080
Speaker 1: a sense, maybe that hasn't gone away. We're still very

161
00:08:11,400 --> 00:08:14,840
Speaker 1: the Rogue one example, Like, clearly everyone is very optimistic

162
00:08:14,880 --> 00:08:18,120
Speaker 1: about how great this looked, and even though to your point,

163
00:08:18,160 --> 00:08:21,600
Speaker 1: it does look great, but within the context of the film,

164
00:08:21,880 --> 00:08:24,720
Speaker 1: something doesn't quite work. Yeah, I would say, Now for

165
00:08:24,800 --> 00:08:26,720
Speaker 1: some people, we can get into this more. I think,

166
00:08:26,800 --> 00:08:29,680
Speaker 1: especially in the next episode when we then the next one,

167
00:08:29,720 --> 00:08:32,560
Speaker 1: we want to try to go beyond the valley, the

168
00:08:32,640 --> 00:08:36,320
Speaker 1: Uncanny Valley. But I will say at this point, for

169
00:08:36,440 --> 00:08:40,080
Speaker 1: some people, Tarkan was not over the line or under

170
00:08:40,120 --> 00:08:41,720
Speaker 1: the line. I don't know where you put the line.

171
00:08:41,840 --> 00:08:44,440
Speaker 1: But for for some people it worked, and I do

172
00:08:44,520 --> 00:08:46,920
Speaker 1: think that's an interesting thing to acknowledge that while for

173
00:08:47,000 --> 00:08:50,880
Speaker 1: me I experienced this, uh, not to the same extent

174
00:08:50,960 --> 00:08:53,880
Speaker 1: as the Scorpion King, but a kind of Scorpion King revulsion,

175
00:08:54,000 --> 00:08:56,640
Speaker 1: not everybody did. Now, one thing about Tarkan is that

176
00:08:56,840 --> 00:09:00,520
Speaker 1: the Tarkan c Jack character is correct me if I'm wrong,

177
00:09:00,520 --> 00:09:02,640
Speaker 1: because I have not seen it myself yet. But he

178
00:09:02,720 --> 00:09:06,960
Speaker 1: is interacting with human actors in this in his scenes, yeah,

179
00:09:07,000 --> 00:09:09,400
Speaker 1: I think so, or at least he's in a film

180
00:09:09,440 --> 00:09:11,960
Speaker 1: with other human actors, even if he's not sharing the

181
00:09:11,960 --> 00:09:15,560
Speaker 1: exact same scene with them. So you might think, well,

182
00:09:15,600 --> 00:09:17,880
Speaker 1: if you just had a movie just full of brilliant

183
00:09:17,920 --> 00:09:20,680
Speaker 1: looking Tarkans, maybe it would be okay. And maybe it would.

184
00:09:20,960 --> 00:09:25,560
Speaker 1: But some of the classic examples of Uncanny Valley happened

185
00:09:25,600 --> 00:09:27,880
Speaker 1: to be films that are filled with nothing but c

186
00:09:28,000 --> 00:09:30,000
Speaker 1: g I characters. Yeah, how about one from the same

187
00:09:30,080 --> 00:09:32,720
Speaker 1: year as The mom of Your Turns two thousand one,

188
00:09:32,800 --> 00:09:35,640
Speaker 1: If you go back to Oh Yes, Final Fantasy, The

189
00:09:35,679 --> 00:09:41,520
Speaker 1: Spirits Within, I remember kind of liking it. I do too.

190
00:09:41,600 --> 00:09:43,240
Speaker 1: It was a film that I think I kind of

191
00:09:43,840 --> 00:09:47,959
Speaker 1: half watched, half worked on, like some just college coursework

192
00:09:48,040 --> 00:09:50,760
Speaker 1: or something just on in the background. And and maybe

193
00:09:50,760 --> 00:09:52,520
Speaker 1: that was the right level of immersion in it. But

194
00:09:52,559 --> 00:09:55,080
Speaker 1: I remember digging it. But at the same time, there

195
00:09:55,120 --> 00:09:57,720
Speaker 1: are a lot of dead puppet eyes in this movie.

196
00:09:57,840 --> 00:10:00,719
Speaker 1: Oh yeah, and it's so I saw it at the time.

197
00:10:00,720 --> 00:10:03,520
Speaker 1: I remember having mixed feelings about the animation, like in

198
00:10:03,600 --> 00:10:06,079
Speaker 1: some senses, I remember thinking, Wow, that looks so cool.

199
00:10:06,880 --> 00:10:09,160
Speaker 1: That again may have been a product of its time.

200
00:10:09,240 --> 00:10:12,040
Speaker 1: We can talk about that more, how our expectations change

201
00:10:12,120 --> 00:10:16,240
Speaker 1: as things go on. But also I don't know. There

202
00:10:16,240 --> 00:10:18,320
Speaker 1: were multiple things wrong with that movie, one of which

203
00:10:18,360 --> 00:10:21,720
Speaker 1: being that the last line of spoken dialogue in the movie,

204
00:10:21,760 --> 00:10:23,600
Speaker 1: as a friend of mine pointed out at the time,

205
00:10:23,760 --> 00:10:29,960
Speaker 1: was oh it's warm. Well I could I don't remember

206
00:10:29,960 --> 00:10:31,960
Speaker 1: the line, so I can't speak to it how well

207
00:10:31,960 --> 00:10:36,320
Speaker 1: it landed, But I can see that being a problem credits. Now,

208
00:10:36,440 --> 00:10:39,720
Speaker 1: another big one, this came out just three years later, is,

209
00:10:39,840 --> 00:10:42,839
Speaker 1: of course, The Polar Express. Now, when people talk about

210
00:10:42,840 --> 00:10:45,040
Speaker 1: the Uncanny Valley these days, I'd say this is a

211
00:10:45,120 --> 00:10:47,640
Speaker 1: top three mentioned. Yeah, this is one of the defining

212
00:10:47,760 --> 00:10:51,120
Speaker 1: nightmares of our time now based on a wonderful children's

213
00:10:51,120 --> 00:10:53,800
Speaker 1: book about the magic of Christmas time. Yeah, the book

214
00:10:53,840 --> 00:10:56,520
Speaker 1: is wonderful, but it's certainly one of these examples. If

215
00:10:56,559 --> 00:10:58,800
Speaker 1: you take a very brief children's book and you try

216
00:10:58,840 --> 00:11:01,760
Speaker 1: and adapt it into a a feature length motion picture,

217
00:11:02,440 --> 00:11:05,880
Speaker 1: that's very difficult to do. In fact, I'm really grasping

218
00:11:05,920 --> 00:11:08,760
Speaker 1: for an example where anybody actually pulled it off. Like,

219
00:11:08,800 --> 00:11:11,640
Speaker 1: the best adaptations of children's books that come to mind

220
00:11:11,920 --> 00:11:16,320
Speaker 1: are all very short, uh, very short films generally. I'm

221
00:11:16,400 --> 00:11:20,559
Speaker 1: thinking of Dr Seus's adaptations from the seventies and eighties,

222
00:11:21,360 --> 00:11:25,120
Speaker 1: not The Polar Express, which is just an exercise in

223
00:11:25,160 --> 00:11:31,400
Speaker 1: psychic trauma brought on by just seemingly intentionally weaponized Uncanny Valley. Um,

224
00:11:31,520 --> 00:11:35,160
Speaker 1: you know, the soulless puppet people. I've never seen this movie,

225
00:11:35,160 --> 00:11:37,079
Speaker 1: but I looked up clips to see what people were

226
00:11:37,080 --> 00:11:41,360
Speaker 1: talking about, and oh man, they they are not kidding it.

227
00:11:41,960 --> 00:11:44,320
Speaker 1: I don't know how children made it through this movie.

228
00:11:44,559 --> 00:11:48,000
Speaker 1: It has these it has these creepy elves, It's got

229
00:11:48,000 --> 00:11:52,360
Speaker 1: a creepy Tom Hanks as a train conductor. Nothing seems right,

230
00:11:52,440 --> 00:11:56,280
Speaker 1: Everything seems like it's just about to everybody's about to

231
00:11:56,280 --> 00:11:59,280
Speaker 1: start melting and screaming. Yeah. I think this is one

232
00:11:59,280 --> 00:12:02,480
Speaker 1: where it was a poor idea in my opinion, and

233
00:12:02,640 --> 00:12:06,080
Speaker 1: uh in the technology was not there to to rescue

234
00:12:06,080 --> 00:12:08,800
Speaker 1: the idea. Now the next one we're going to discuss though.

235
00:12:09,000 --> 00:12:12,040
Speaker 1: I think it was a great idea on paper, but

236
00:12:12,120 --> 00:12:15,319
Speaker 1: it just didn't work out on the screen. And that's

237
00:12:15,320 --> 00:12:18,280
Speaker 1: of course two thousand sevens Baywolf. Now as this Robert

238
00:12:18,360 --> 00:12:21,480
Speaker 1: Zemeckis who did this, Yeah, Robert Zemeckis helmed it. And

239
00:12:21,520 --> 00:12:24,760
Speaker 1: then the writing it was Neil Gaiman and Roger Avery.

240
00:12:24,840 --> 00:12:27,920
Speaker 1: So some you know, some some some big names just

241
00:12:27,960 --> 00:12:31,160
Speaker 1: attached and to the the ideas behind this, uh this movie,

242
00:12:31,160 --> 00:12:34,240
Speaker 1: and of course based on the story of Grindel and Beowulf,

243
00:12:34,280 --> 00:12:37,560
Speaker 1: which is a classic you would think, you know, hard

244
00:12:37,600 --> 00:12:41,760
Speaker 1: to miss action narrative. I think the Beowolf could make

245
00:12:41,760 --> 00:12:44,160
Speaker 1: a really great movie if somebody did it right. Yeah,

246
00:12:44,280 --> 00:12:46,960
Speaker 1: I think so too. I have yet to see that movie.

247
00:12:47,240 --> 00:12:51,319
Speaker 1: But but but certainly has all the the potential in

248
00:12:51,360 --> 00:12:54,040
Speaker 1: the world. And they had a pretty cool vocal cast

249
00:12:54,080 --> 00:12:56,240
Speaker 1: as well, I think. And Angelina Jolie is in it

250
00:12:56,320 --> 00:12:59,439
Speaker 1: as the monster's mother. They have ray Winstone as Bayolwolf. Yeah,

251
00:12:59,480 --> 00:13:02,560
Speaker 1: he does the voice of Beowulf, and uh, and who

252
00:13:02,640 --> 00:13:05,199
Speaker 1: was it that plays the monster? We were just crisp

253
00:13:06,960 --> 00:13:09,520
Speaker 1: now I've seen it all back home. Not one of

254
00:13:09,559 --> 00:13:12,600
Speaker 1: my favorite monster depictions of Grenville, by the way, But

255
00:13:13,200 --> 00:13:15,400
Speaker 1: he's a monster. We can get past that. But everybody

256
00:13:15,400 --> 00:13:18,480
Speaker 1: else in the film really has the uncanny valley that

257
00:13:18,640 --> 00:13:20,640
Speaker 1: going on to to a high degree. I think I

258
00:13:20,640 --> 00:13:24,199
Speaker 1: read a quote somewhere where film critic was talking about

259
00:13:24,240 --> 00:13:27,319
Speaker 1: how the monsters in the movie were only slightly less

260
00:13:27,320 --> 00:13:31,319
Speaker 1: frightening than the humans. Yeah, yeah, the humans. It just

261
00:13:31,440 --> 00:13:35,040
Speaker 1: it just didn't land. Now at this point you're probably thinking, well,

262
00:13:35,400 --> 00:13:38,720
Speaker 1: how about video games, because they're certainly when you're thinking

263
00:13:38,720 --> 00:13:42,560
Speaker 1: about computer animated human beings interacting with each other, staring

264
00:13:42,720 --> 00:13:46,000
Speaker 1: right into the camera, you think of video games. Yeah,

265
00:13:46,040 --> 00:13:49,480
Speaker 1: And and I think, and you know, here's here's the

266
00:13:49,480 --> 00:13:51,360
Speaker 1: thing here. I have to say that I haven't noticed

267
00:13:51,400 --> 00:13:54,040
Speaker 1: it as often these days I think a lot of

268
00:13:54,080 --> 00:13:57,800
Speaker 1: game animators have found ways to get around the uncanny Valley. Yeah.

269
00:13:58,160 --> 00:13:59,800
Speaker 1: I don't want to get to ahead of our flow here,

270
00:13:59,800 --> 00:14:02,920
Speaker 1: but I think one thing that I've noticed they sometimes

271
00:14:02,920 --> 00:14:06,840
Speaker 1: do is that they don't actually go for photo realism,

272
00:14:06,880 --> 00:14:09,120
Speaker 1: and they go for a kind of more real than

273
00:14:09,200 --> 00:14:13,920
Speaker 1: real combination of like a comic book style type character illustration,

274
00:14:14,520 --> 00:14:18,440
Speaker 1: and then these other realistic aspects that when when you

275
00:14:18,480 --> 00:14:21,040
Speaker 1: look at a video game character, you would never mistake

276
00:14:21,040 --> 00:14:24,000
Speaker 1: it for a photograph of a person, even even one

277
00:14:24,080 --> 00:14:27,680
Speaker 1: that's got really good graphics. But much like the way

278
00:14:27,920 --> 00:14:30,480
Speaker 1: dialogue is written in films, you know, you don't want

279
00:14:30,480 --> 00:14:34,080
Speaker 1: to make dialogue sound like real people talk, because that

280
00:14:34,120 --> 00:14:36,680
Speaker 1: would be horrible to listen to, but you do want

281
00:14:36,680 --> 00:14:40,240
Speaker 1: to make it sound quote realistic. You don't want to

282
00:14:40,280 --> 00:14:43,640
Speaker 1: make your characters look too realistic in animation, but you

283
00:14:43,680 --> 00:14:46,640
Speaker 1: do want to make them look quote realistic. In other words,

284
00:14:46,680 --> 00:14:50,160
Speaker 1: they feel real. Yeah, this reminds me of a game

285
00:14:50,200 --> 00:14:52,200
Speaker 1: franchise that I haven't I don't think I've ever played

286
00:14:52,200 --> 00:14:54,520
Speaker 1: more than a demo of this, but the Gears of

287
00:14:54,520 --> 00:14:57,480
Speaker 1: War series. So all the people in this kind of

288
00:14:57,480 --> 00:14:59,600
Speaker 1: look like like if you're gonna be critical, but you

289
00:14:59,640 --> 00:15:02,040
Speaker 1: might say everyone looks kind of like they're weird guerrilla people.

290
00:15:02,280 --> 00:15:04,440
Speaker 1: Like it was a like we're in an alternate world

291
00:15:04,520 --> 00:15:08,880
Speaker 1: where unrealistically huge upper bodies. Yeah, as if evolution took

292
00:15:08,920 --> 00:15:13,520
Speaker 1: a slightly different turn into an intelligent primates. Uh. And

293
00:15:13,640 --> 00:15:16,480
Speaker 1: yet they look real. They don't look like they don't

294
00:15:16,480 --> 00:15:19,000
Speaker 1: get an uncanny effect rolling off them, Like you know,

295
00:15:19,040 --> 00:15:20,760
Speaker 1: you look at them, you can see pores, you can

296
00:15:20,800 --> 00:15:25,680
Speaker 1: see hair follicills. They look real, but they are but

297
00:15:25,720 --> 00:15:29,920
Speaker 1: that they are certainly not going for authentic human being

298
00:15:29,920 --> 00:15:32,200
Speaker 1: there all right, Now, I want to put out one

299
00:15:32,200 --> 00:15:34,480
Speaker 1: more example here before we move on, and it's a

300
00:15:34,600 --> 00:15:39,360
Speaker 1: rare example of uncanny valley avoidance, a very specific type

301
00:15:39,360 --> 00:15:42,840
Speaker 1: of uncanny valley avoidance, and that is au from a

302
00:15:42,880 --> 00:15:46,120
Speaker 1: fantastic stop motion short that was produced by the National

303
00:15:46,160 --> 00:15:48,560
Speaker 1: Film Board of Canada. And you can find this online

304
00:15:48,560 --> 00:15:50,560
Speaker 1: if you just do a search for it. It's Madam

305
00:15:50,640 --> 00:15:54,880
Speaker 1: Tutley Putly and it's a wonderful little little film, very

306
00:15:54,960 --> 00:15:59,320
Speaker 1: very French feel to it. Characters on a train, weirds,

307
00:15:59,320 --> 00:16:02,800
Speaker 1: frightening things occurring. Uh, definitely check it out. But the

308
00:16:03,120 --> 00:16:06,520
Speaker 1: trick to it there, these are stop motion animated characters,

309
00:16:06,880 --> 00:16:10,040
Speaker 1: and their eyes just feel so alive. They stare right

310
00:16:10,080 --> 00:16:12,640
Speaker 1: into you, and you don't you don't question for a

311
00:16:12,680 --> 00:16:15,760
Speaker 1: second that these are that these are people. And the

312
00:16:16,000 --> 00:16:18,600
Speaker 1: trick that they employed is that they used real human eyes,

313
00:16:19,200 --> 00:16:22,280
Speaker 1: not in a you know, depraved, evil puppet master kind

314
00:16:22,320 --> 00:16:25,400
Speaker 1: of way, either. They videotape the eyes of human actors

315
00:16:25,440 --> 00:16:27,800
Speaker 1: and then blended the footage with that of the puppets.

316
00:16:28,240 --> 00:16:31,480
Speaker 1: That sounds like an incredible gambit, because that sounds like

317
00:16:31,520 --> 00:16:34,360
Speaker 1: that could have produced some of the worst Uncanny Valley

318
00:16:34,400 --> 00:16:37,000
Speaker 1: feelings ever if it went wrong. Yeah, and and I

319
00:16:37,040 --> 00:16:38,720
Speaker 1: don't know, there may be some people who watch this

320
00:16:38,840 --> 00:16:41,880
Speaker 1: short and and have the opposite effect and and think

321
00:16:41,880 --> 00:16:44,560
Speaker 1: that it's super creepy. I found it to be like this,

322
00:16:44,560 --> 00:16:48,960
Speaker 1: this interesting example of circumventing the Uncanny Valley. But I'll

323
00:16:49,000 --> 00:16:50,680
Speaker 1: leave it for you guys to decide. I'll include a

324
00:16:50,720 --> 00:16:52,360
Speaker 1: link to this one as well as some of the

325
00:16:52,360 --> 00:16:54,480
Speaker 1: other sources we're talking about on the landing page for

326
00:16:54,520 --> 00:16:56,960
Speaker 1: this episode. It's stuff to blow your mind dot com.

327
00:16:57,000 --> 00:16:58,760
Speaker 1: All right, Well, we are going to take a quick

328
00:16:58,800 --> 00:17:01,040
Speaker 1: break and when we come back will get into the

329
00:17:01,080 --> 00:17:04,120
Speaker 1: origin of the scientific idea of the Uncanny Valley and

330
00:17:04,200 --> 00:17:11,399
Speaker 1: its history and research. All right, we're back. So the

331
00:17:11,480 --> 00:17:13,919
Speaker 1: Uncanny Valley. Where does this even come from? Right? So

332
00:17:13,920 --> 00:17:16,600
Speaker 1: we've already been talking about it because most people have

333
00:17:16,720 --> 00:17:19,639
Speaker 1: heard of this, they're somewhat familiar with it. I was

334
00:17:19,680 --> 00:17:22,679
Speaker 1: talking to Rachel about it though. She was saying, you know,

335
00:17:23,640 --> 00:17:26,600
Speaker 1: at least to her, it had this connotation of just

336
00:17:26,760 --> 00:17:32,040
Speaker 1: generally synthetically generated images being creepy in one way or another.

337
00:17:32,359 --> 00:17:34,480
Speaker 1: So maybe we should get into the specifics of the

338
00:17:34,520 --> 00:17:37,200
Speaker 1: origin of the idea. So let's go back to the

339
00:17:37,280 --> 00:17:42,840
Speaker 1: year nineteen seventy. Everything's great, Wait is it? I don't know,

340
00:17:43,200 --> 00:17:46,639
Speaker 1: but but everybody, everybody's looking forward to the future in

341
00:17:46,760 --> 00:17:49,440
Speaker 1: terms of creating humanoid robots. What are we going to

342
00:17:49,520 --> 00:17:52,879
Speaker 1: be able to do well? The Japanese roboticist massa Hiro

343
00:17:52,960 --> 00:17:57,560
Speaker 1: Mori of the Tokyo Institute of Technology. He wrote a

344
00:17:57,600 --> 00:18:00,760
Speaker 1: paper that was published in this Japanese journal Energy that

345
00:18:00,920 --> 00:18:05,840
Speaker 1: coined the term uncanny Valley to describe a problem that

346
00:18:05,880 --> 00:18:09,359
Speaker 1: he was predicting with increasingly humanoid robots. And this was

347
00:18:09,400 --> 00:18:15,240
Speaker 1: based on just some observations he'd had of of different events.

348
00:18:15,280 --> 00:18:18,000
Speaker 1: So you might say incidents in the progress of designing

349
00:18:18,040 --> 00:18:21,520
Speaker 1: humanoid robots, such as consumer electronics shows in Japan and

350
00:18:21,560 --> 00:18:25,120
Speaker 1: stuff like that. So what he predicted was that as

351
00:18:25,520 --> 00:18:27,920
Speaker 1: you had a humanoid robot, robot that looks like a

352
00:18:28,000 --> 00:18:32,680
Speaker 1: human and its likeness to a human increased, our attitude

353
00:18:32,960 --> 00:18:36,560
Speaker 1: towards them would improve. Our affinity would go up as

354
00:18:36,600 --> 00:18:40,920
Speaker 1: they became more human, until they reached a certain tipping

355
00:18:41,000 --> 00:18:44,879
Speaker 1: point of similarity to humans, where suddenly our affinity, our

356
00:18:44,960 --> 00:18:50,600
Speaker 1: friendly attitude, almost immediately shifts and plunges down into strong revulsion.

357
00:18:51,880 --> 00:18:55,600
Speaker 1: Being human is likable, being sort of human is likable,

358
00:18:55,720 --> 00:19:00,320
Speaker 1: but being almost human is horrible and repulsive. And then

359
00:19:00,320 --> 00:19:03,119
Speaker 1: of course at the final end, uh you you would

360
00:19:03,200 --> 00:19:05,040
Speaker 1: have a real human. So you can think of the

361
00:19:05,119 --> 00:19:08,000
Speaker 1: uncanny valley as a phase in a graph, an X

362
00:19:08,160 --> 00:19:11,480
Speaker 1: Y graph, and along the horizontal axis on the bottom,

363
00:19:11,640 --> 00:19:14,439
Speaker 1: you've got the degree of similarity to a human, and

364
00:19:14,480 --> 00:19:17,000
Speaker 1: then on the vertical axis you've got the degree of

365
00:19:17,040 --> 00:19:20,800
Speaker 1: our affinity for the object. And more hypothesized, this graph

366
00:19:20,800 --> 00:19:23,240
Speaker 1: would have these two peaks. You'd start with zero on

367
00:19:23,320 --> 00:19:26,040
Speaker 1: both axes, because a thing that has no human like

368
00:19:26,200 --> 00:19:29,920
Speaker 1: traits basically gets no human affinity response one way or another.

369
00:19:29,960 --> 00:19:32,119
Speaker 1: And we just don't you know, how much do you

370
00:19:32,200 --> 00:19:35,720
Speaker 1: really like an industrial conveyor belt. You're just sort of

371
00:19:35,760 --> 00:19:39,960
Speaker 1: neutral on it. But as you increase the humanity, you

372
00:19:40,080 --> 00:19:44,720
Speaker 1: give a robot arms or something that looks like a face, eyes, limbs,

373
00:19:45,119 --> 00:19:48,639
Speaker 1: you climb this gentle, gradual slope to the first peak

374
00:19:48,680 --> 00:19:53,240
Speaker 1: and affinity. Um, you know, and he didn't name the peak,

375
00:19:53,240 --> 00:19:55,000
Speaker 1: but I think we should name the peak. I think

376
00:19:55,000 --> 00:19:58,000
Speaker 1: this first peak should be called something like the cuteness peak.

377
00:19:58,119 --> 00:20:01,399
Speaker 1: That's not exactly right because it's not exactly cuteness, but

378
00:20:01,480 --> 00:20:05,920
Speaker 1: it's recognizing something kind of human about what you're looking at. Yeah, Like,

379
00:20:06,680 --> 00:20:08,880
Speaker 1: I mean, we don't have to describe cute to everyone here,

380
00:20:08,920 --> 00:20:11,960
Speaker 1: but certainly this is hello kitty territory, this is the

381
00:20:12,240 --> 00:20:16,120
Speaker 1: this is the domain of large eyed It's vaguely infant

382
00:20:16,200 --> 00:20:19,560
Speaker 1: or kitten like creatures that would never be mistaken for

383
00:20:19,640 --> 00:20:22,200
Speaker 1: human or real, but they resonate with us for a

384
00:20:22,280 --> 00:20:24,359
Speaker 1: number of reasons. We could do a whole podcast infect

385
00:20:24,400 --> 00:20:27,320
Speaker 1: we have an old podcast episode about the science of cute.

386
00:20:27,359 --> 00:20:31,160
Speaker 1: Why that connects with us? Yeah, so they would include

387
00:20:31,200 --> 00:20:33,720
Speaker 1: that would include all kinds of robots that just kind

388
00:20:33,720 --> 00:20:37,520
Speaker 1: of have general, very basic faces that don't try to

389
00:20:37,560 --> 00:20:40,520
Speaker 1: have human skin or anything like that. That just might

390
00:20:40,600 --> 00:20:45,440
Speaker 1: have like kind of a mouth and some cartoonish eyes. Yeah, sure,

391
00:20:45,520 --> 00:20:48,159
Speaker 1: there you go. That the c three po boldly on

392
00:20:48,200 --> 00:20:52,520
Speaker 1: the cuteness. But at a certain point after this first speak,

393
00:20:52,600 --> 00:20:55,480
Speaker 1: this graph drops off steeply. So you keep going along

394
00:20:55,480 --> 00:20:58,680
Speaker 1: the x axis, but then the y axis drops off,

395
00:20:58,760 --> 00:21:01,880
Speaker 1: not just down to zero, but far below zero, into

396
00:21:01,880 --> 00:21:05,199
Speaker 1: the negative affinity range. And this part of the graph

397
00:21:05,280 --> 00:21:08,760
Speaker 1: is the uncanny valley. As the similarity to a real

398
00:21:08,880 --> 00:21:12,520
Speaker 1: human continues to increase near a dent. In other words,

399
00:21:12,720 --> 00:21:16,600
Speaker 1: as it becomes indistinguishable from a real human, our affinity

400
00:21:16,680 --> 00:21:20,280
Speaker 1: sharply shoots back up the second peak toward reality. So

401
00:21:20,320 --> 00:21:23,320
Speaker 1: I'd call this second peak the reality peak. It's when

402
00:21:23,359 --> 00:21:28,480
Speaker 1: you become, for all intents and purposes, a real human being. Yeah.

403
00:21:28,800 --> 00:21:31,640
Speaker 1: I would also say that if if robots were candy,

404
00:21:31,920 --> 00:21:35,240
Speaker 1: the bottom of the uncanny valley would be banana flavored

405
00:21:35,280 --> 00:21:38,320
Speaker 1: candy like that for me has always been a flavor

406
00:21:38,400 --> 00:21:41,040
Speaker 1: where it's like clearly like not only like runts that

407
00:21:41,119 --> 00:21:44,360
Speaker 1: have the bananas. I think, so like like great candy,

408
00:21:44,560 --> 00:21:47,600
Speaker 1: like grape candy doesn't really taste like grapes, but it's

409
00:21:47,800 --> 00:21:50,119
Speaker 1: it's far enough from the Uncanny Valley of candy that

410
00:21:50,160 --> 00:21:53,760
Speaker 1: you're you're okay, whoa, You're right, banana candy actually does

411
00:21:53,880 --> 00:21:57,080
Speaker 1: taste like bananas in a way that makes it not

412
00:21:57,320 --> 00:22:01,720
Speaker 1: really good. Yeah, Like I've candy fans, I don't eat

413
00:22:01,720 --> 00:22:04,639
Speaker 1: that much candy anymore. So maybe the technology has advanced,

414
00:22:04,640 --> 00:22:08,399
Speaker 1: but uh, my memory of the banana candy is is

415
00:22:08,840 --> 00:22:12,359
Speaker 1: that of an uncanny experience. Now, one thing we should

416
00:22:12,359 --> 00:22:14,679
Speaker 1: note is that so this original paper was published in

417
00:22:14,800 --> 00:22:19,439
Speaker 1: nineteen seventy twelve. English translation was published in uh the

418
00:22:19,520 --> 00:22:22,280
Speaker 1: I Triple A Robotics and Automation magazine. And that's what

419
00:22:22,320 --> 00:22:24,760
Speaker 1: I was using is my reference, that English translation from

420
00:22:26,200 --> 00:22:29,120
Speaker 1: uh and so it has some graphs here. It has

421
00:22:29,160 --> 00:22:32,679
Speaker 1: Morey's original graphs or interpretations of them, and we can

422
00:22:32,720 --> 00:22:35,280
Speaker 1: get into a little more detail on the nuances of

423
00:22:35,320 --> 00:22:37,640
Speaker 1: Morey's theory. But one thing I did read was that

424
00:22:38,480 --> 00:22:43,000
Speaker 1: many years later, somebody contacted more and he and they

425
00:22:43,040 --> 00:22:45,240
Speaker 1: were talking to him about this idea he had of

426
00:22:45,280 --> 00:22:47,719
Speaker 1: the uncanny Valley, and they were like, well, does does

427
00:22:47,760 --> 00:22:51,040
Speaker 1: anything lie beyond the peak of reality? And he said, hey,

428
00:22:51,040 --> 00:22:53,919
Speaker 1: oh yeah, actually there is such a thing. And he

429
00:22:54,000 --> 00:22:56,400
Speaker 1: said you know, beyond the real human, you'd have sort

430
00:22:56,440 --> 00:23:03,000
Speaker 1: of like artistic ideals, like the realm of forms even right. Yeah,

431
00:23:03,040 --> 00:23:05,199
Speaker 1: So well, I think he used an example of like

432
00:23:05,240 --> 00:23:08,400
Speaker 1: a statue of Buddha, you know, a beautiful, perfect statue

433
00:23:08,400 --> 00:23:11,359
Speaker 1: of Buddha. It's almost like it we have greater affinity

434
00:23:11,400 --> 00:23:15,280
Speaker 1: for it than we have for a realistic human because

435
00:23:15,280 --> 00:23:18,440
Speaker 1: we've been, well, to a large point, we've been conditioned

436
00:23:18,480 --> 00:23:20,960
Speaker 1: to right. Yeah, that that kind of gets into this,

437
00:23:20,960 --> 00:23:24,159
Speaker 1: this idea of conditioned familiarity that we not only have

438
00:23:24,240 --> 00:23:27,960
Speaker 1: with my religious icons, but also with pop culture icons,

439
00:23:28,320 --> 00:23:31,200
Speaker 1: so not only the Buddha, but also Robbie the robot,

440
00:23:31,800 --> 00:23:36,040
Speaker 1: or or even the Terminator or well, yeah, that does

441
00:23:36,160 --> 00:23:38,960
Speaker 1: make me think that in some ways, if if aesthetic

442
00:23:39,000 --> 00:23:41,800
Speaker 1: ideals and things that were familiar with through our culture

443
00:23:42,240 --> 00:23:45,080
Speaker 1: might be even beyond humans. I mean, again, this is

444
00:23:45,119 --> 00:23:47,840
Speaker 1: not like rigorous research, This is just what more he says,

445
00:23:47,880 --> 00:23:51,119
Speaker 1: he thinks, who predicts, Uh, could could there be like

446
00:23:51,160 --> 00:23:54,480
Speaker 1: a robot that we really love that's actually better than

447
00:23:54,520 --> 00:23:56,960
Speaker 1: a than a normal human? Well, you know, there's a

448
00:23:57,040 --> 00:23:59,920
Speaker 1: study that came out last year, I believe from Penn

449
00:24:00,000 --> 00:24:02,520
Speaker 1: State University that was kind of interesting alone. These lines.

450
00:24:02,840 --> 00:24:06,679
Speaker 1: So the researchers survey three seventy nine adults ages ages

451
00:24:06,760 --> 00:24:09,680
Speaker 1: sixty to eighty six, and they asked them for specific

452
00:24:09,720 --> 00:24:13,240
Speaker 1: memories of robot films they'd seen in their general attitudes

453
00:24:13,320 --> 00:24:16,480
Speaker 1: towards robots and and the you know, with the age here,

454
00:24:16,560 --> 00:24:20,720
Speaker 1: as you might imagine, they're really looking at at potential

455
00:24:20,760 --> 00:24:22,600
Speaker 1: care robots, like the idea of like what kind of

456
00:24:22,680 --> 00:24:24,680
Speaker 1: robots should help you use the bathroom? Do you want

457
00:24:24,720 --> 00:24:26,960
Speaker 1: something that looks like kind of like a person, or

458
00:24:27,000 --> 00:24:29,840
Speaker 1: do you want something that looks like a forklift with

459
00:24:30,480 --> 00:24:32,840
Speaker 1: a forklift mated with an easy chair. But if you

460
00:24:32,880 --> 00:24:36,000
Speaker 1: look at the ages used here and twenty sixteen, when

461
00:24:36,040 --> 00:24:38,520
Speaker 1: the study took place, you can say that these people

462
00:24:38,640 --> 00:24:41,080
Speaker 1: grew up with science fiction. Oh yeah, I mean they

463
00:24:41,160 --> 00:24:44,159
Speaker 1: might not have personally consumed a lot of it, but

464
00:24:44,200 --> 00:24:46,720
Speaker 1: it's in the culture, right, Yeah. They they definitely had

465
00:24:46,760 --> 00:24:49,760
Speaker 1: access to it. And the researchers found that individuals who

466
00:24:49,760 --> 00:24:53,919
Speaker 1: could recall more cinematic robot portrayals were increasingly likely to

467
00:24:53,960 --> 00:24:56,560
Speaker 1: hold positive attitudes towards robots in general. So it didn't

468
00:24:56,600 --> 00:25:01,320
Speaker 1: matter if they remembered murderous killbots or well meaning hell probots. Uh,

469
00:25:01,400 --> 00:25:05,880
Speaker 1: they need the mere memory of multiple robotic portrayals correlated

470
00:25:05,880 --> 00:25:09,920
Speaker 1: to pro robot vibes, so to study findings. They also

471
00:25:09,960 --> 00:25:14,520
Speaker 1: backed up the importance of human looking human asque robots

472
00:25:14,760 --> 00:25:18,159
Speaker 1: to invoke a sympathetic user response. But the researcher stress

473
00:25:18,200 --> 00:25:21,720
Speaker 1: that robot designers might want to incorporate robotic features that

474
00:25:21,800 --> 00:25:25,760
Speaker 1: older adults will remember from their cinematic past. So it's

475
00:25:25,800 --> 00:25:28,560
Speaker 1: saying that, like, don't just try to make it like

476
00:25:28,600 --> 00:25:30,800
Speaker 1: a human. Try to make it like the robots we

477
00:25:30,840 --> 00:25:33,639
Speaker 1: have known and loved. Yeah, like make it fun. You know,

478
00:25:34,680 --> 00:25:36,320
Speaker 1: if I'm if I need a robot to help me

479
00:25:36,359 --> 00:25:38,840
Speaker 1: go to the bathroom, may make it. Make make them

480
00:25:38,840 --> 00:25:41,119
Speaker 1: the robots from Silent Running, you know, Huie, Dowie and

481
00:25:41,119 --> 00:25:43,760
Speaker 1: Louie a little little guys. Then at least I can

482
00:25:43,800 --> 00:25:46,840
Speaker 1: engage my nostalgia a little bit totally. So I want

483
00:25:46,840 --> 00:25:48,959
Speaker 1: to look at a few more nuances from Maury's original

484
00:25:49,000 --> 00:25:51,440
Speaker 1: paper in nineteen seventy. So one thing I do think

485
00:25:51,440 --> 00:25:53,239
Speaker 1: it's very interesting and I want to come back to

486
00:25:53,400 --> 00:25:57,359
Speaker 1: as we explore this topic more. More hypothesizes in the

487
00:25:57,359 --> 00:26:00,600
Speaker 1: original paper that our perception of an un handy valley

488
00:26:00,680 --> 00:26:04,119
Speaker 1: might depend on the context in which we're we're viewing

489
00:26:04,160 --> 00:26:06,800
Speaker 1: the being. And the example he gives here is he's

490
00:26:06,800 --> 00:26:10,720
Speaker 1: talking about buon Rocko puppets, and so he says, quote,

491
00:26:10,760 --> 00:26:13,480
Speaker 1: I don't think that on close inspection of bun Rocku

492
00:26:13,520 --> 00:26:17,600
Speaker 1: puppet appears similar to a human being. But when we

493
00:26:17,720 --> 00:26:20,480
Speaker 1: enjoy a puppet show in the theater, were seated at

494
00:26:20,520 --> 00:26:24,080
Speaker 1: a certain distance from the stage, the puppets absolute size

495
00:26:24,200 --> 00:26:28,280
Speaker 1: is ignored. Its total appearance, including hand and eye movements,

496
00:26:28,640 --> 00:26:31,040
Speaker 1: is close to that of a human being. So, given

497
00:26:31,080 --> 00:26:34,360
Speaker 1: our tendency as an audience to become absorbed in this

498
00:26:34,440 --> 00:26:37,240
Speaker 1: form of art, we might feel a high level of

499
00:26:37,280 --> 00:26:40,960
Speaker 1: affinity for the puppet. I think that's interesting. So it's

500
00:26:41,040 --> 00:26:44,119
Speaker 1: it's not just the object, but it's also the context

501
00:26:44,160 --> 00:26:46,840
Speaker 1: in which we experience the object. You might have very

502
00:26:46,840 --> 00:26:50,159
Speaker 1: different feelings about a bun Rocku puppet lying on the

503
00:26:50,160 --> 00:26:54,160
Speaker 1: floor versus one that you go to see in the

504
00:26:54,200 --> 00:26:57,040
Speaker 1: context of staging a play. Yeah. I think that the

505
00:26:57,080 --> 00:27:00,679
Speaker 1: puppet argument is something to keep in mind throughout considerations

506
00:27:00,680 --> 00:27:02,359
Speaker 1: of the Uncanny Valley because there are a lot of

507
00:27:02,359 --> 00:27:03,720
Speaker 1: people that there are a lot of people who have

508
00:27:03,800 --> 00:27:06,840
Speaker 1: kind of um an irrational version to puppets in general,

509
00:27:06,880 --> 00:27:09,960
Speaker 1: and certainly if you take just a still puppet and

510
00:27:09,960 --> 00:27:12,399
Speaker 1: you hold it up, there are various puppets that one

511
00:27:12,480 --> 00:27:15,600
Speaker 1: might find a little bit uncanny or creepy, etcetera. But

512
00:27:15,640 --> 00:27:19,159
Speaker 1: in the process of performing with a talented performer is

513
00:27:19,240 --> 00:27:21,720
Speaker 1: going to bring that to life. Like that's the art form.

514
00:27:22,040 --> 00:27:25,280
Speaker 1: And and there's so many different varieties of puppetry. Certainly

515
00:27:25,280 --> 00:27:30,560
Speaker 1: they're they're too broad categories are it's situations where the

516
00:27:30,600 --> 00:27:33,919
Speaker 1: puppeteer is visible and puppet situation where puppeteer is not.

517
00:27:34,040 --> 00:27:36,200
Speaker 1: You know, so you have your basic the Muppets situation

518
00:27:36,320 --> 00:27:38,520
Speaker 1: where you don't see the puppeteers, but there are plenty

519
00:27:38,560 --> 00:27:42,480
Speaker 1: of art forms of puppetry performance styles in which the

520
00:27:42,520 --> 00:27:46,280
Speaker 1: puppeteer is very visible, either completely or just their face.

521
00:27:46,440 --> 00:27:48,480
Speaker 1: You see their eyes, you see that there's a person

522
00:27:48,520 --> 00:27:51,920
Speaker 1: involved here, and uh, and there's not this this mystery

523
00:27:52,000 --> 00:27:55,399
Speaker 1: or this sense of deception, right, yeah, I think conceptual

524
00:27:55,760 --> 00:27:59,159
Speaker 1: clues like that are very important. Also is when you

525
00:27:59,200 --> 00:28:01,880
Speaker 1: consider the the the idea of going to a puppet theater,

526
00:28:02,359 --> 00:28:07,440
Speaker 1: it also includes a certain attitude charging effect in the audience,

527
00:28:07,880 --> 00:28:10,720
Speaker 1: Like an an audience member goes to a puppet theater

528
00:28:11,160 --> 00:28:14,840
Speaker 1: prepared to suspend their disbelief, like you know what I mean,

529
00:28:14,960 --> 00:28:17,920
Speaker 1: Like you put yourself in an intentional state of open

530
00:28:17,960 --> 00:28:21,760
Speaker 1: mindedness about what you're viewing, and you give yourself an

531
00:28:21,760 --> 00:28:24,959
Speaker 1: interpretive framework through which to Like if you were not

532
00:28:25,119 --> 00:28:28,600
Speaker 1: prepared to watch a puppet theater story and suddenly a

533
00:28:28,640 --> 00:28:31,840
Speaker 1: puppet was just moving around, that might be a lot creepier.

534
00:28:33,240 --> 00:28:35,920
Speaker 1: So part of the Uncanny Valley effect is probably also

535
00:28:36,080 --> 00:28:39,080
Speaker 1: in the viewer themselves and in the so the context

536
00:28:39,160 --> 00:28:41,560
Speaker 1: is not just where you are, what's going on with

537
00:28:41,560 --> 00:28:45,400
Speaker 1: what you're looking at, but what you're expecting to see. Now.

538
00:28:45,480 --> 00:28:48,880
Speaker 1: One more thing that Morey points out is he thinks

539
00:28:48,920 --> 00:28:51,480
Speaker 1: that they're going to be very different rules governing the

540
00:28:51,600 --> 00:28:57,000
Speaker 1: Uncanny Valley for still objects versus moving objects. And essentially

541
00:28:57,000 --> 00:29:00,960
Speaker 1: his hypothesis is that movement is going to amplify both

542
00:29:01,000 --> 00:29:04,239
Speaker 1: the peaks and the valleys of the graph. So if

543
00:29:04,240 --> 00:29:07,080
Speaker 1: you imagine the graph we said earlier, gentle slope up

544
00:29:07,080 --> 00:29:09,960
Speaker 1: to first peak, you know, kind of cute whatever has

545
00:29:10,040 --> 00:29:13,440
Speaker 1: some human characteristics, then a dip down into too close

546
00:29:13,480 --> 00:29:16,280
Speaker 1: to human but not there, and then a final rise

547
00:29:16,360 --> 00:29:20,080
Speaker 1: up to actually human. He he would say if it's moving,

548
00:29:20,200 --> 00:29:22,360
Speaker 1: the peaks are going to be higher and the valleys

549
00:29:22,400 --> 00:29:25,520
Speaker 1: are going to be lower. Okay, so a thing that

550
00:29:25,720 --> 00:29:28,800
Speaker 1: is moving gets greater affinity if it's good if it's

551
00:29:28,840 --> 00:29:31,560
Speaker 1: at one of these two peaks, but it's even more

552
00:29:31,600 --> 00:29:35,160
Speaker 1: revolting and unpleasant if it's at the valley. Uh So

553
00:29:35,320 --> 00:29:38,280
Speaker 1: this this makes me think of Samara in The Ring,

554
00:29:38,920 --> 00:29:41,320
Speaker 1: those scenes where Samara is emerging from the TV or

555
00:29:41,320 --> 00:29:45,240
Speaker 1: the well, her movement is is jerky and and I

556
00:29:45,320 --> 00:29:48,400
Speaker 1: understand that they created that effect by having the actor

557
00:29:48,600 --> 00:29:52,080
Speaker 1: or actress walk backwards and then reversing the footage. So

558
00:29:52,160 --> 00:29:55,120
Speaker 1: you have this, you have this this movement that is

559
00:29:55,640 --> 00:29:58,600
Speaker 1: you know, natural, but being reversed it it feels very

560
00:29:58,680 --> 00:30:01,960
Speaker 1: unnatural and and it's hard to really pinpoint what's not

561
00:30:02,120 --> 00:30:05,800
Speaker 1: working for you about it. Right. So More in the

562
00:30:05,920 --> 00:30:09,520
Speaker 1: end concludes he gives this recommendation based on his hypothesis.

563
00:30:09,520 --> 00:30:12,760
Speaker 1: He says, don't go for realism, right, It's gonna be

564
00:30:12,880 --> 00:30:16,239
Speaker 1: so hard if you're designing a humanoid robot. Now, a

565
00:30:16,240 --> 00:30:18,840
Speaker 1: lot of what we're we're talking about in these episodes

566
00:30:18,960 --> 00:30:22,520
Speaker 1: is animation. He's talking primarily about humanoid robots, but typically

567
00:30:22,560 --> 00:30:26,600
Speaker 1: these uh two fields get somewhat conflated in discussion of

568
00:30:26,640 --> 00:30:28,760
Speaker 1: the Uncanny Valley, because in both cases you're trying to

569
00:30:28,760 --> 00:30:32,720
Speaker 1: create something that looks pleasingly human. Um. He wrote, it's

570
00:30:32,720 --> 00:30:35,240
Speaker 1: gonna be so hard to get out of the valley

571
00:30:35,320 --> 00:30:38,280
Speaker 1: up the second peak, that that's the reality peak is

572
00:30:38,840 --> 00:30:42,800
Speaker 1: so steep. Instead, roboticists should not try, and instead they

573
00:30:42,800 --> 00:30:45,600
Speaker 1: should aim for the very tip of the first peak.

574
00:30:46,160 --> 00:30:48,440
Speaker 1: Stick on the cuteness peak because we know we can

575
00:30:48,440 --> 00:30:52,880
Speaker 1: get there. Think think Wally or other cute humanoid robots.

576
00:30:52,920 --> 00:30:55,520
Speaker 1: The first peak is not really that hard to attain.

577
00:30:55,920 --> 00:30:58,200
Speaker 1: People respond well to it, So why do you need

578
00:30:58,240 --> 00:31:01,240
Speaker 1: to try to go past it? Um? You know. As

579
00:31:01,280 --> 00:31:04,760
Speaker 1: for animated humans, I think a good analogy might be here.

580
00:31:04,840 --> 00:31:09,040
Speaker 1: Here's one Pixars The Incredibles versus Final Fantasy The Spirits,

581
00:31:09,080 --> 00:31:12,160
Speaker 1: within which we already mentioned. The former they don't look

582
00:31:12,200 --> 00:31:15,400
Speaker 1: like real humans at all, right, they're cute, cartoonish, non

583
00:31:15,520 --> 00:31:19,080
Speaker 1: realistic humans, but they're quite pleasant. The latter goes for

584
00:31:19,400 --> 00:31:23,840
Speaker 1: and fails at photo realism and creates these characters that

585
00:31:23,880 --> 00:31:26,880
Speaker 1: are stiff and unsettling. In other words, he says, don't

586
00:31:26,880 --> 00:31:28,880
Speaker 1: try to climb out of the valley. Just don't go

587
00:31:28,920 --> 00:31:32,040
Speaker 1: into the valley to begin with. Yeah, this this really

588
00:31:32,040 --> 00:31:34,520
Speaker 1: brings to mind just the idea of like filmmakers and

589
00:31:34,600 --> 00:31:38,320
Speaker 1: creators standing on the the the edge of this physical

590
00:31:38,440 --> 00:31:40,880
Speaker 1: valley and there's a local guide. They're saying, don't do it.

591
00:31:41,160 --> 00:31:43,520
Speaker 1: Don't do it, the value will consume you. And they're like, no,

592
00:31:43,600 --> 00:31:45,880
Speaker 1: we're Lucasfilm. We can do it. Yeah, we got all

593
00:31:45,880 --> 00:31:49,120
Speaker 1: this high text gear. There's no way that anything's gonna

594
00:31:49,160 --> 00:31:50,840
Speaker 1: take us down there. And then they go down there

595
00:31:50,840 --> 00:31:54,040
Speaker 1: and it's just Jurassic Parker congo with They just get

596
00:31:54,080 --> 00:31:57,720
Speaker 1: torn apart. You know, I do. I do think the

597
00:31:58,120 --> 00:32:01,360
Speaker 1: dinosaurs in Jurassic Park come out of the uncanny valley

598
00:32:01,360 --> 00:32:04,760
Speaker 1: for dinosaurs. They do. Yeah, And that introduces an interesting

599
00:32:04,760 --> 00:32:08,720
Speaker 1: wrinkle in that, like Maury is talking about human wide qualities,

600
00:32:09,280 --> 00:32:13,080
Speaker 1: it would probably be a related but different thing to

601
00:32:13,200 --> 00:32:19,880
Speaker 1: just say animal reality versus specifically human reality. Yeah, because

602
00:32:20,000 --> 00:32:23,000
Speaker 1: I mean when for non human creatures, certainly we've been

603
00:32:23,000 --> 00:32:26,000
Speaker 1: able to nail that. For ages to stop motion creatures,

604
00:32:26,040 --> 00:32:28,560
Speaker 1: often even if their movements are kind of herky jerky,

605
00:32:28,600 --> 00:32:33,080
Speaker 1: they feel great. Um that heard of stop motion robots

606
00:32:33,520 --> 00:32:36,400
Speaker 1: in older films, So that I've never had a problem

607
00:32:36,440 --> 00:32:39,200
Speaker 1: buying into them, and yeah, you're look into the eye

608
00:32:39,240 --> 00:32:41,840
Speaker 1: of the track or the t rex or the velociraptor

609
00:32:42,000 --> 00:32:44,800
Speaker 1: and you never doubt for a second. Yeah. But so

610
00:32:44,920 --> 00:32:46,760
Speaker 1: one thing I think we should point out is that

611
00:32:46,960 --> 00:32:50,800
Speaker 1: as prescient as Mori was of what would become this

612
00:32:51,080 --> 00:32:55,120
Speaker 1: widely recognized pop culture phenomenon his paper, it's not it's

613
00:32:55,120 --> 00:32:59,760
Speaker 1: not research, really, it's just sort of observation and interesting speculation.

614
00:33:00,400 --> 00:33:02,840
Speaker 1: So what we should shift now to do, I think,

615
00:33:02,960 --> 00:33:05,760
Speaker 1: is talk about whether there's really any evidence that the

616
00:33:05,840 --> 00:33:09,160
Speaker 1: Uncanny Valley number one exists at all? Is it really

617
00:33:09,160 --> 00:33:12,400
Speaker 1: a thing number two? Is it a is it a

618
00:33:12,480 --> 00:33:15,440
Speaker 1: unified phenomenon, or is it there there's some separate things

619
00:33:15,480 --> 00:33:18,680
Speaker 1: getting pulled into the net together here. And then finally

620
00:33:18,720 --> 00:33:21,560
Speaker 1: maybe we should look at if it's real, what causes it?

621
00:33:22,680 --> 00:33:25,920
Speaker 1: Why do human brains tend to react this way? So

622
00:33:25,960 --> 00:33:27,920
Speaker 1: maybe we should take a quick break, and then when

623
00:33:27,960 --> 00:33:35,800
Speaker 1: we come back we will get into more recent research. Thank. Okay,

624
00:33:35,840 --> 00:33:39,160
Speaker 1: So there's there's really no denying that there is some

625
00:33:39,280 --> 00:33:43,720
Speaker 1: kind of creepy humanoid synthetic figure effect. Right. We we've

626
00:33:43,760 --> 00:33:45,880
Speaker 1: all seen these c g I movies, We've all seen

627
00:33:45,880 --> 00:33:49,640
Speaker 1: these creepy robots and had that feeling of don't like it.

628
00:33:50,320 --> 00:33:55,000
Speaker 1: But that doesn't necessarily mean that the uncanny Valley, as

629
00:33:55,320 --> 00:33:59,640
Speaker 1: described by Mori or as popularly conceived in culture, is

630
00:33:59,720 --> 00:34:03,600
Speaker 1: in fact a correct description of what's happening there. Right

631
00:34:03,680 --> 00:34:06,720
Speaker 1: just because it it feels truthy, just because it lines

632
00:34:06,840 --> 00:34:09,560
Speaker 1: up with to a certain degree with how we feel

633
00:34:09,560 --> 00:34:12,239
Speaker 1: about the world, doesn't mean that it is. You know,

634
00:34:12,280 --> 00:34:15,960
Speaker 1: that it is an actual effect that's taking place, and

635
00:34:16,280 --> 00:34:18,640
Speaker 1: or that it's even a fixed effect, etcetera. There are

636
00:34:18,560 --> 00:34:21,560
Speaker 1: a lot of factors to contemplate here, Like for my

637
00:34:21,600 --> 00:34:24,800
Speaker 1: own part, I've always found it interesting and I definitely

638
00:34:25,120 --> 00:34:29,120
Speaker 1: think there's something to it. However, you line it up

639
00:34:29,160 --> 00:34:33,719
Speaker 1: with similar cases in life, such as say, individuals that

640
00:34:33,760 --> 00:34:37,759
Speaker 1: you may encounter who have some degree of facial disfigurement,

641
00:34:37,800 --> 00:34:40,160
Speaker 1: and it might be extremely mild. It might be it

642
00:34:40,239 --> 00:34:42,960
Speaker 1: might be nothing more than a uh in then you

643
00:34:43,000 --> 00:34:45,759
Speaker 1: know a lazy eye, or or you know some sort

644
00:34:45,760 --> 00:34:49,600
Speaker 1: of cleft lip or cleft palate to scenario it, or

645
00:34:49,640 --> 00:34:52,080
Speaker 1: it just might be like their faces maybe not all

646
00:34:52,120 --> 00:34:56,680
Speaker 1: that symmetrical, and you know nobody whose face is perfectly symmetrical.

647
00:34:57,800 --> 00:35:01,399
Speaker 1: But with all of these individual rules, you interact with them,

648
00:35:01,400 --> 00:35:04,960
Speaker 1: you get to know them maybe, and whatever kind of

649
00:35:05,000 --> 00:35:09,640
Speaker 1: like initial um reaction is present, be it just kind

650
00:35:09,640 --> 00:35:13,200
Speaker 1: of a huh or that goes away, and you can

651
00:35:13,280 --> 00:35:15,640
Speaker 1: unless you're a total jerk, unless you're a total jerk,

652
00:35:15,800 --> 00:35:18,440
Speaker 1: or you're gonna be able to relate to that person.

653
00:35:18,480 --> 00:35:20,600
Speaker 1: You're gonna be able to communicate with that person, and

654
00:35:20,640 --> 00:35:22,640
Speaker 1: you're not going to be thrown for a curve every

655
00:35:22,680 --> 00:35:25,080
Speaker 1: time they make eye contact with you. Yeah, I would

656
00:35:25,120 --> 00:35:27,920
Speaker 1: agree with that. So there is certainly, like in Maury's

657
00:35:27,920 --> 00:35:32,680
Speaker 1: original formulation, he he would, I think, put different kinds

658
00:35:32,719 --> 00:35:37,520
Speaker 1: of um physical abnormality somewhere on the ascending slope, on

659
00:35:37,600 --> 00:35:41,279
Speaker 1: the on the on the uncanny valley slope. So you

660
00:35:41,360 --> 00:35:43,759
Speaker 1: have a normal, healthy human up at the peak, I

661
00:35:43,800 --> 00:35:46,560
Speaker 1: guess somewhere below the artistic ideal of the great Buddhist

662
00:35:46,560 --> 00:35:49,759
Speaker 1: statue or something. But you'd have normal, healthy human. Then

663
00:35:49,800 --> 00:35:52,520
Speaker 1: somewhere below them would be people who have who look

664
00:35:52,640 --> 00:35:56,359
Speaker 1: like there is something wrong with them in terms of

665
00:35:56,400 --> 00:35:59,759
Speaker 1: having uh, you know, perfect health and symmetricality. I mean

666
00:35:59,760 --> 00:36:02,480
Speaker 1: so because just an ill person, you encounter someone who

667
00:36:02,719 --> 00:36:05,080
Speaker 1: is clearly a little bit sick or a little bit

668
00:36:05,120 --> 00:36:08,320
Speaker 1: hungover or whatever. You can tell and it it causes

669
00:36:08,360 --> 00:36:10,640
Speaker 1: a light to go off in your head. Yeah. Uh.

670
00:36:10,680 --> 00:36:14,360
Speaker 1: And and yet we can quite easily adapt to people

671
00:36:14,680 --> 00:36:17,080
Speaker 1: like you know, you see somebody like that, it is

672
00:36:17,200 --> 00:36:20,319
Speaker 1: you just know, it is not proper to react to

673
00:36:20,360 --> 00:36:23,440
Speaker 1: somebody with revulsion. Oh yeah, like right now in Atlanta

674
00:36:23,719 --> 00:36:27,080
Speaker 1: as we're recording, this pollen is everywhere. So there are

675
00:36:27,120 --> 00:36:29,560
Speaker 1: several people in my life. I'm not really affected by

676
00:36:29,600 --> 00:36:33,319
Speaker 1: the pollen so much, but it totally debilitates some of

677
00:36:33,360 --> 00:36:37,759
Speaker 1: my coworkers, some of my friends and red face puffy eyes. Yeah.

678
00:36:37,840 --> 00:36:40,960
Speaker 1: And and sometimes they're like walked out on allergy medication

679
00:36:41,000 --> 00:36:44,360
Speaker 1: to boot and you just get you just you know,

680
00:36:44,400 --> 00:36:47,040
Speaker 1: you accept it. You realize, oh, well, you know, my

681
00:36:47,040 --> 00:36:48,880
Speaker 1: my friend here is going to be kind of a

682
00:36:48,880 --> 00:36:51,840
Speaker 1: pollen zombie for a couple of weeks. But that doesn't

683
00:36:51,840 --> 00:36:54,560
Speaker 1: mean we can't hang out. It doesn't mean we can't

684
00:36:54,560 --> 00:36:58,600
Speaker 1: work on this or that. Yeah. So definitely, The Uncanny

685
00:36:58,680 --> 00:37:02,480
Speaker 1: Valley has plenty of critics, and plenty I think a

686
00:37:02,600 --> 00:37:05,480
Speaker 1: very fair criticisms leveled at it. I just want to

687
00:37:05,520 --> 00:37:08,440
Speaker 1: go back to one popular article I came across a

688
00:37:08,480 --> 00:37:12,160
Speaker 1: two thousand ten article in Popular Mechanics by Eric Softge,

689
00:37:13,040 --> 00:37:15,280
Speaker 1: where he sort of points out that at the time,

690
00:37:15,640 --> 00:37:18,120
Speaker 1: people were, as I think they are still now treating

691
00:37:18,160 --> 00:37:21,640
Speaker 1: the Uncanny Valley as a proven fact, but in fact,

692
00:37:21,800 --> 00:37:24,360
Speaker 1: at the time, he says, you know, there's really almost

693
00:37:24,440 --> 00:37:28,000
Speaker 1: no convincing evidence that such a thing even exists, and

694
00:37:28,120 --> 00:37:31,640
Speaker 1: he speaks to an expert named Carl McDorman, director of

695
00:37:31,680 --> 00:37:35,560
Speaker 1: the Android Science Center at Indiana University, and McDorman, who

696
00:37:35,719 --> 00:37:38,799
Speaker 1: has conducted research on the valley, offered his opinion in

697
00:37:38,840 --> 00:37:41,560
Speaker 1: the article, saying quote, it turns out that there may

698
00:37:41,560 --> 00:37:44,760
Speaker 1: be more than one Uncanny Valley. It's not the overall

699
00:37:44,800 --> 00:37:47,920
Speaker 1: degree of human likeness that makes a robot or animated

700
00:37:48,000 --> 00:37:51,520
Speaker 1: character uncanny. It's more a matter of mismatch. If you

701
00:37:51,600 --> 00:37:55,000
Speaker 1: have an extremely realistic skin texture, but at the same

702
00:37:55,040 --> 00:37:59,800
Speaker 1: time cartoonish eyes or realistic eyes and an unrealistic skin texture,

703
00:38:00,280 --> 00:38:04,160
Speaker 1: that's very uncanny, uh and the art. So that's an

704
00:38:04,200 --> 00:38:06,840
Speaker 1: idea about the perceptual mismatch that I do want to

705
00:38:06,880 --> 00:38:10,520
Speaker 1: revisit later in this episode. But the article also speaks

706
00:38:10,520 --> 00:38:14,000
Speaker 1: to a guy named David Hanson who's a roboticist who

707
00:38:14,040 --> 00:38:18,160
Speaker 1: specifically specializes in creating very realistic humanoid robots. I think

708
00:38:18,160 --> 00:38:22,560
Speaker 1: he did that that Einstein head thing. Oh yeah, nothing. Uh. So,

709
00:38:22,640 --> 00:38:26,280
Speaker 1: Hansen claims that even if people find overly realistic robots

710
00:38:26,320 --> 00:38:29,280
Speaker 1: creepy at first, they get used to them within minutes.

711
00:38:29,360 --> 00:38:31,080
Speaker 1: This is sort of what you were just talking about.

712
00:38:31,120 --> 00:38:34,760
Speaker 1: I think, you know, you become acclimated even to something

713
00:38:34,840 --> 00:38:37,720
Speaker 1: that you might uh at some kind of base level,

714
00:38:37,840 --> 00:38:41,960
Speaker 1: have a negative reaction to. Yeah. I keep thinking of

715
00:38:41,960 --> 00:38:44,120
Speaker 1: aving in isolation in this because it's the game I'm

716
00:38:44,160 --> 00:38:46,160
Speaker 1: currently playing, and uh, and I feel like that the

717
00:38:46,200 --> 00:38:48,759
Speaker 1: c g I characters are are are pretty well done

718
00:38:48,760 --> 00:38:51,680
Speaker 1: in there. I haven't felt that the tinge of of

719
00:38:52,080 --> 00:38:55,000
Speaker 1: Uncanny Valley so washing over me. Some of the voice

720
00:38:55,000 --> 00:38:58,440
Speaker 1: actings a little weak. But but but speaking of the voice,

721
00:38:58,560 --> 00:39:01,640
Speaker 1: like the that the androids you encounter though with the

722
00:39:01,680 --> 00:39:06,319
Speaker 1: sexs and uh androids that key. Yeah, and when I

723
00:39:06,320 --> 00:39:08,680
Speaker 1: first on Canny Valley, well yeah, but when I first

724
00:39:08,800 --> 00:39:11,400
Speaker 1: encountered them, yeah, they had the uncanny intentionally kind of

725
00:39:11,440 --> 00:39:15,359
Speaker 1: creepy appearance and the very creepy robot voice. But yet

726
00:39:15,640 --> 00:39:19,040
Speaker 1: when they were not actively attacking me, I kind of was.

727
00:39:19,160 --> 00:39:20,800
Speaker 1: I was kind of cool with it. It wasn't until

728
00:39:20,840 --> 00:39:24,719
Speaker 1: heded becoming violent. That that that the mere sound of

729
00:39:24,760 --> 00:39:27,400
Speaker 1: their voice or the appearance of one UH down the

730
00:39:27,560 --> 00:39:30,040
Speaker 1: you know, in the distance down the hallway, would would

731
00:39:30,080 --> 00:39:32,680
Speaker 1: cause my nerves to react. I mean, those things are

732
00:39:32,680 --> 00:39:35,680
Speaker 1: funny there, h They're a good part of that game.

733
00:39:36,640 --> 00:39:40,040
Speaker 1: But anyway, so in this UH article, the author also

734
00:39:40,120 --> 00:39:43,560
Speaker 1: cites some other unnamed robot roboticists, as well as his

735
00:39:43,600 --> 00:39:47,520
Speaker 1: own experience when he's talking about meeting robots that he

736
00:39:47,560 --> 00:39:50,560
Speaker 1: had previously seen on video, and one thing he says is,

737
00:39:50,680 --> 00:39:53,880
Speaker 1: you know, an Uncanny Valley effect that was present when

738
00:39:53,920 --> 00:39:55,880
Speaker 1: I saw a video of this robot went away when

739
00:39:55,920 --> 00:39:58,560
Speaker 1: I saw it in person. I don't know if that's

740
00:39:58,600 --> 00:40:01,919
Speaker 1: generally true of people, he claims it's true. But even

741
00:40:01,960 --> 00:40:04,200
Speaker 1: if this is truly the case for robots, I'm not

742
00:40:04,200 --> 00:40:07,160
Speaker 1: sure how it would apply to animations. Probably wouldn't apply

743
00:40:07,239 --> 00:40:11,520
Speaker 1: to animations. UM. But I think that there are some

744
00:40:11,560 --> 00:40:15,799
Speaker 1: good threads to start tugging at here, because it's probably

745
00:40:15,880 --> 00:40:18,960
Speaker 1: the case that there are more dimensions to the Uncanny

746
00:40:19,000 --> 00:40:22,560
Speaker 1: Valley than more he imagined in nineteen seventy, meaning more

747
00:40:22,600 --> 00:40:27,480
Speaker 1: than just that X axis of um closeness to realistic

748
00:40:27,520 --> 00:40:31,799
Speaker 1: human appearance versus distance from realistic human appearance. Yeah, I

749
00:40:31,800 --> 00:40:34,040
Speaker 1: mean just that what makes a person human, what makes

750
00:40:34,040 --> 00:40:37,240
Speaker 1: a lightness human. There's arguably a whole chorus of things

751
00:40:37,280 --> 00:40:40,279
Speaker 1: going on there. Yeah, so it would make sense that

752
00:40:40,280 --> 00:40:42,839
Speaker 1: that that chorus would play into the Uncanny Valley. Yeah.

753
00:40:42,840 --> 00:40:45,920
Speaker 1: So I do think that there are multiple other dimensions

754
00:40:45,960 --> 00:40:48,719
Speaker 1: to be explored. But I also don't think that means

755
00:40:48,719 --> 00:40:51,720
Speaker 1: we can conclude that there's nothing to the Uncanny Valley.

756
00:40:51,960 --> 00:40:54,440
Speaker 1: And in the past decade there's actually been an explosion

757
00:40:54,480 --> 00:40:56,680
Speaker 1: of research on the Uncanny Valley. So I think we

758
00:40:56,680 --> 00:40:59,280
Speaker 1: should look at a few interesting studies on the effect.

759
00:40:59,560 --> 00:41:02,960
Speaker 1: All right, well, uh, first one here that I came

760
00:41:03,000 --> 00:41:06,759
Speaker 1: across was a two thousand nine Princeton University study and

761
00:41:06,760 --> 00:41:09,560
Speaker 1: they looked into the effects of uncanny avowe of the

762
00:41:09,640 --> 00:41:15,040
Speaker 1: Uncanny Valley on macaque monkeys, so so non human subjects. Yeah,

763
00:41:15,080 --> 00:41:17,120
Speaker 1: because that that makes sense, right if you if you

764
00:41:17,120 --> 00:41:19,959
Speaker 1: want to see if this is an evolved response, let's

765
00:41:20,000 --> 00:41:23,480
Speaker 1: look beyond the complications of human intelligence and human culture

766
00:41:23,840 --> 00:41:26,919
Speaker 1: and looked something closely related to us. Is it biological

767
00:41:27,080 --> 00:41:30,319
Speaker 1: rather than say cultural? Right, And so they showed a

768
00:41:30,360 --> 00:41:34,640
Speaker 1: selection of the primates close to real quote unquote computer

769
00:41:34,760 --> 00:41:38,320
Speaker 1: visuals of macaques to see if they responded with coups

770
00:41:38,360 --> 00:41:41,800
Speaker 1: and lip smacking as they do with their fellow monkeys. Uh.

771
00:41:41,840 --> 00:41:46,120
Speaker 1: And these these close to real computer visuals were essentially

772
00:41:46,640 --> 00:41:50,520
Speaker 1: lawnmower man monkeys. If you see they kind of asking

773
00:41:50,640 --> 00:41:55,120
Speaker 1: if I've seen lawnmower man. You know I've seen lawn man. Yes,

774
00:41:55,400 --> 00:41:57,800
Speaker 1: so yeah, think a lot more man. Uh and you

775
00:41:58,160 --> 00:42:00,759
Speaker 1: kind of have an idea that that level of compute animation,

776
00:42:01,120 --> 00:42:03,480
Speaker 1: and the monkeys did not want any part of it,

777
00:42:03,520 --> 00:42:06,719
Speaker 1: and they averted their eyes, they acted frightened when confronted

778
00:42:06,719 --> 00:42:10,960
Speaker 1: with lawnmower man monkey. So it's not much, i admit,

779
00:42:11,400 --> 00:42:14,759
Speaker 1: but it's a little experimental evidence for the argument that

780
00:42:15,000 --> 00:42:18,279
Speaker 1: uncanny valley is an evolutionary response. Right, so if you

781
00:42:18,280 --> 00:42:21,640
Speaker 1: can observe it in monkeys, there's probably some element of

782
00:42:21,680 --> 00:42:25,600
Speaker 1: it that that is biological in the brain. It's instinctual

783
00:42:25,760 --> 00:42:28,839
Speaker 1: and not just something we've all learned to say about

784
00:42:28,920 --> 00:42:33,399
Speaker 1: weirdly looking animated characters and robots. And that would be

785
00:42:33,480 --> 00:42:36,640
Speaker 1: maybe a weak piece of evidence, but still a piece

786
00:42:36,680 --> 00:42:38,520
Speaker 1: of evidence you could put in the column of saying

787
00:42:38,600 --> 00:42:41,200
Speaker 1: there is something there. The valley does to some extent

788
00:42:41,280 --> 00:42:45,600
Speaker 1: exist now the next study that I ran across. This

789
00:42:45,640 --> 00:42:48,960
Speaker 1: comes back, this to one of the graphics that you

790
00:42:49,080 --> 00:42:52,680
Speaker 1: pulled out of the believe the original uh study correct? Yeah, yeah,

791
00:42:52,760 --> 00:42:56,560
Speaker 1: the original Morey's original graphs. So in this graph and

792
00:42:56,680 --> 00:42:58,920
Speaker 1: we talked about diving down into the valley and then

793
00:42:59,320 --> 00:43:01,960
Speaker 1: steadily try to claw yourself out on the other side,

794
00:43:02,480 --> 00:43:05,200
Speaker 1: very very steep ascent. Yeah, so you hit bottom and

795
00:43:05,239 --> 00:43:08,880
Speaker 1: that's where you have a zombie and as you begin

796
00:43:08,960 --> 00:43:12,160
Speaker 1: to scale out of the uncanny valley, he has um uh,

797
00:43:12,520 --> 00:43:16,400
Speaker 1: myoelectric hand and prosthetic hand down there. As you climb

798
00:43:16,400 --> 00:43:20,759
Speaker 1: back up, eventually hitting ordinary doll and and puppets and

799
00:43:20,880 --> 00:43:24,680
Speaker 1: ill person and maybe hitting healthy person at the very top. Again.

800
00:43:24,760 --> 00:43:27,200
Speaker 1: But it's interesting you have prosthetic hand down there, because

801
00:43:27,200 --> 00:43:31,640
Speaker 1: this next study looks at prosthetic and robotic and human hands. Yeah,

802
00:43:31,680 --> 00:43:33,799
Speaker 1: this is in the original study. More He talks about

803
00:43:33,920 --> 00:43:38,080
Speaker 1: the variable creepiness of prosthetic hands. And I found I

804
00:43:38,120 --> 00:43:39,799
Speaker 1: found this interesting because I don't know about you, but

805
00:43:39,800 --> 00:43:43,960
Speaker 1: but growing up I felt like crazy robot hands. Especially

806
00:43:44,000 --> 00:43:48,000
Speaker 1: we're everywhere like every G. I. Joe show or Heman

807
00:43:48,160 --> 00:43:52,439
Speaker 1: type franchise, there's always somebody it could be a villain,

808
00:43:52,480 --> 00:43:55,200
Speaker 1: it could be a hero. But there were crazy robot

809
00:43:55,239 --> 00:43:58,400
Speaker 1: hands galore, UH, and I always found them cool, and

810
00:43:58,920 --> 00:44:01,520
Speaker 1: I feel like a lot of us probably even fetishize

811
00:44:01,520 --> 00:44:03,799
Speaker 1: them to a certain point, Like we we didn't understand

812
00:44:04,040 --> 00:44:06,319
Speaker 1: what it would necessarily be like to lose and lose

813
00:44:06,360 --> 00:44:10,880
Speaker 1: a hand and the shortfall and the ability of technology

814
00:44:10,920 --> 00:44:14,720
Speaker 1: at the time and even today to replace that missing limb.

815
00:44:15,120 --> 00:44:18,000
Speaker 1: But we thought, well, that looks cool. Superpowered robot hands

816
00:44:18,040 --> 00:44:21,040
Speaker 1: signed me up, right. But back to the study two

817
00:44:21,040 --> 00:44:23,960
Speaker 1: thousand thirteen University of Manchester study, and they looked at

818
00:44:24,000 --> 00:44:29,120
Speaker 1: prosthetic hands. UH. They used of forty three right handed participants,

819
00:44:29,200 --> 00:44:31,920
Speaker 1: thirty six female and seven male, and they were all

820
00:44:31,960 --> 00:44:35,120
Speaker 1: looking at photos, and the photos were divided into three categories.

821
00:44:35,320 --> 00:44:38,719
Speaker 1: Human hands, robotic hands like, no question about it, that's

822
00:44:38,719 --> 00:44:41,600
Speaker 1: a robot hand I'm looking at like straight up terminator

823
00:44:41,600 --> 00:44:46,160
Speaker 1: exoskeleton or or or even less human, and then prosthetic hands.

824
00:44:46,880 --> 00:44:50,080
Speaker 1: The results, I have to say, reading through some some

825
00:44:50,160 --> 00:44:52,640
Speaker 1: of the writing about this UH and the original press release,

826
00:44:52,760 --> 00:44:57,080
Speaker 1: the results were kind of confusing sounding. They the subjects

827
00:44:57,080 --> 00:45:01,160
Speaker 1: here preferred human hands and robot hands, but but rated

828
00:45:01,400 --> 00:45:05,120
Speaker 1: and certainly rated prosthetic hands is more uncanny, but prosthetics

829
00:45:05,160 --> 00:45:10,520
Speaker 1: that looked more human were less eerie. Okay, So so

830
00:45:10,640 --> 00:45:13,600
Speaker 1: something that's clearly a robot that's not too creepy. Something's

831
00:45:13,600 --> 00:45:16,520
Speaker 1: clearly a human that's not too creepy. If something is

832
00:45:16,520 --> 00:45:19,760
Speaker 1: a robot trying to be human, that might be more creepy,

833
00:45:19,920 --> 00:45:23,839
Speaker 1: but as it gets better at being human, it's less creepy,

834
00:45:24,040 --> 00:45:25,719
Speaker 1: I think. So, I think that's my take. I mean,

835
00:45:25,800 --> 00:45:29,200
Speaker 1: it also makes me wonder if if the hand alone

836
00:45:29,920 --> 00:45:33,200
Speaker 1: is an is like a subset of the uncanny valet,

837
00:45:33,200 --> 00:45:36,080
Speaker 1: because certainly if you're if you're just working with a

838
00:45:36,160 --> 00:45:38,920
Speaker 1: hand and trying to replicate the movements, the look, the

839
00:45:39,000 --> 00:45:42,080
Speaker 1: feel of a human limb for an observer, not we're

840
00:45:42,080 --> 00:45:44,000
Speaker 1: not going to even get into the the you know,

841
00:45:44,040 --> 00:45:47,600
Speaker 1: the problems of creating something that the user can experience

842
00:45:47,880 --> 00:45:50,920
Speaker 1: as a lifelike limb. But if you're just looking at it,

843
00:45:50,920 --> 00:45:52,760
Speaker 1: if you don't have to worry about its eye contact,

844
00:45:52,800 --> 00:45:55,480
Speaker 1: you don't have to worry about micro expressions. Uh, it

845
00:45:55,840 --> 00:46:00,399
Speaker 1: seems like it would be an easier peak to surmount. Yeah,

846
00:46:00,400 --> 00:46:02,960
Speaker 1: so the if that is in fact the correct interpretation,

847
00:46:02,960 --> 00:46:06,320
Speaker 1: that would seem to undercut the steepness in Maury's original

848
00:46:06,400 --> 00:46:09,480
Speaker 1: graph right on the on the final peak. Yeah, that's

849
00:46:09,480 --> 00:46:13,080
Speaker 1: I mean, that's what I'm wondering. Because the hand had

850
00:46:13,080 --> 00:46:17,719
Speaker 1: taken in isolation, is thinking would be easier to replicate. Yeah.

851
00:46:18,719 --> 00:46:21,400
Speaker 1: Uh and Uncanny Valley. Let's face it, when we talk

852
00:46:21,480 --> 00:46:23,960
Speaker 1: about it, most of the time we're talking about faces.

853
00:46:24,080 --> 00:46:29,480
Speaker 1: Right now, Speaking of faces, there's another study. UM. This

854
00:46:29,840 --> 00:46:32,319
Speaker 1: is a two thousand and eleven University of California, San

855
00:46:32,360 --> 00:46:36,160
Speaker 1: Diego study. UM. This was published in the Social Cognitive

856
00:46:36,160 --> 00:46:39,200
Speaker 1: and Effective Neuroscience, and they did exactly what you'd expect

857
00:46:39,280 --> 00:46:42,600
Speaker 1: researchers to do when confronted with the Uncanny Valley. Grab

858
00:46:42,640 --> 00:46:45,000
Speaker 1: the f m r I and see what our brains

859
00:46:45,000 --> 00:46:46,960
Speaker 1: are doing when we're looking at all these images. So

860
00:46:47,040 --> 00:46:50,480
Speaker 1: all these fMRI I studies, all right, well, what what

861
00:46:50,560 --> 00:46:52,960
Speaker 1: did they find? All right, I'll roll through the basics

862
00:46:52,960 --> 00:46:55,360
Speaker 1: of the study here? So twenty subjects, not a not

863
00:46:55,400 --> 00:46:59,000
Speaker 1: a huge study here, aged thirty six. And here were

864
00:46:59,040 --> 00:47:02,759
Speaker 1: some of the the caveats they had in selecting these individuals.

865
00:47:02,960 --> 00:47:07,120
Speaker 1: No experience working with robots, no time spent in Japan,

866
00:47:07,520 --> 00:47:10,279
Speaker 1: no friends or family from Japan because they wanted to

867
00:47:10,320 --> 00:47:14,959
Speaker 1: avoid uh any you know, potential cultural exposure that would

868
00:47:14,960 --> 00:47:18,480
Speaker 1: have made them, would make them more accepting of androids. Okay,

869
00:47:18,480 --> 00:47:21,000
Speaker 1: So the idea is that maybe in Japan people just

870
00:47:21,520 --> 00:47:25,680
Speaker 1: experience humanoid robots way too much already there too, they're

871
00:47:26,280 --> 00:47:29,600
Speaker 1: acclimatized to them. Yeah, that that's the the argument they made,

872
00:47:29,600 --> 00:47:32,600
Speaker 1: and laying out the study, let's let's not even go there,

873
00:47:32,760 --> 00:47:35,360
Speaker 1: let's just deal with people who have less exposure to robots.

874
00:47:36,239 --> 00:47:39,200
Speaker 1: And they were shown twelve videos of a humanoid robot

875
00:47:39,480 --> 00:47:42,480
Speaker 1: named repley Q two. Oh man, I'm looking it up

876
00:47:42,560 --> 00:47:45,839
Speaker 1: right now. It's it's it's rough, but well, they watched

877
00:47:45,920 --> 00:47:49,200
Speaker 1: video twelve videos of this robot doing various things, and

878
00:47:49,239 --> 00:47:52,120
Speaker 1: they were shown videos of humans doing the same things.

879
00:47:52,320 --> 00:47:55,920
Speaker 1: And in fact, the robots movements and mannerisms were patterned

880
00:47:55,960 --> 00:47:59,359
Speaker 1: directly after the humans. So you had a you had

881
00:47:59,480 --> 00:48:02,480
Speaker 1: a human version of the actions, you had an android

882
00:48:02,600 --> 00:48:05,480
Speaker 1: version of the actions, uh, you know, a lifelike robot,

883
00:48:05,640 --> 00:48:08,080
Speaker 1: and then you had a a stripped down version of

884
00:48:08,120 --> 00:48:10,439
Speaker 1: the androids. So basically the android with all its skin

885
00:48:10,520 --> 00:48:13,280
Speaker 1: ripped off, so it looks more like a robot, clearly

886
00:48:13,320 --> 00:48:16,320
Speaker 1: a robot, and it's doing the same motions as well.

887
00:48:16,880 --> 00:48:19,560
Speaker 1: So this broke it all down to a human with

888
00:48:19,560 --> 00:48:22,840
Speaker 1: biological appearance in movement, a robot with mechanical appearance and

889
00:48:22,880 --> 00:48:26,720
Speaker 1: mechanical motion, and a human seeming agent with the exact

890
00:48:26,800 --> 00:48:32,200
Speaker 1: same mechanical movements as the robot. Then in came the

891
00:48:32,280 --> 00:48:36,040
Speaker 1: f M R I scans. So the main brain area

892
00:48:36,080 --> 00:48:38,520
Speaker 1: of note here, the the area that that that that

893
00:48:38,680 --> 00:48:42,480
Speaker 1: lit up where we saw the most activity, the parietal

894
00:48:42,520 --> 00:48:46,799
Speaker 1: cortex on both sides of the brain, specifically in the

895
00:48:46,840 --> 00:48:49,960
Speaker 1: areas that connect the part of the brain's visual cortex

896
00:48:50,000 --> 00:48:53,360
Speaker 1: that process bodily movements with the section of the motor

897
00:48:53,440 --> 00:48:57,719
Speaker 1: cortex thought to contain mirror neurons. So those would be

898
00:48:57,719 --> 00:49:01,360
Speaker 1: like the the empathy parts of the brain where you know,

899
00:49:01,440 --> 00:49:04,440
Speaker 1: we we see something going on in some other creature

900
00:49:04,520 --> 00:49:07,359
Speaker 1: like us, and we empathize with it. Exactly. Yeah, So

901
00:49:07,480 --> 00:49:10,759
Speaker 1: when viewing the human looking android, the brain lit up

902
00:49:10,800 --> 00:49:15,000
Speaker 1: at the recognition of a human form, but registered essentially

903
00:49:15,280 --> 00:49:18,359
Speaker 1: a computing error over the movement. Something didn't match up.

904
00:49:18,960 --> 00:49:21,320
Speaker 1: Uh so it's it's not. According to this study, it

905
00:49:21,360 --> 00:49:23,480
Speaker 1: would seem that it's not the biological movement or the

906
00:49:23,480 --> 00:49:28,080
Speaker 1: biological appearance, it's the congruents or lack of congruents between

907
00:49:28,120 --> 00:49:30,560
Speaker 1: the two. You look alive but you're dead, you look

908
00:49:30,640 --> 00:49:32,880
Speaker 1: dead but you move you or you speak as if

909
00:49:32,880 --> 00:49:36,239
Speaker 1: you're alive. Uh. So the researchers noted that this is

910
00:49:36,280 --> 00:49:41,200
Speaker 1: something that could be retuned through exposure, but it could

911
00:49:41,239 --> 00:49:43,920
Speaker 1: be at the heart of what's going on with the

912
00:49:44,000 --> 00:49:46,960
Speaker 1: Uncanny Valley. Interesting. Well, I think we should look at

913
00:49:47,000 --> 00:49:50,799
Speaker 1: one more study, uh, potentially providing recent support for the

914
00:49:50,800 --> 00:49:53,520
Speaker 1: existence of the Uncanny Valley, and then maybe after that

915
00:49:53,560 --> 00:49:55,640
Speaker 1: we should break and then come back next time to

916
00:49:55,880 --> 00:49:58,600
Speaker 1: get into the causes, what what would be causing this

917
00:49:58,680 --> 00:50:02,239
Speaker 1: effect and uh and future. So I want to look

918
00:50:02,280 --> 00:50:05,160
Speaker 1: at a study that came out in two six in

919
00:50:05,200 --> 00:50:09,080
Speaker 1: the journal Cognition by Mather and Rifling called Navigating a

920
00:50:09,120 --> 00:50:13,239
Speaker 1: Social World with Robot Partners. A Quantitative cartography of the

921
00:50:13,320 --> 00:50:17,880
Speaker 1: Uncanny Valley. Cute invocation of map making there because it

922
00:50:17,920 --> 00:50:19,640
Speaker 1: does kind of make sense. I like the idea of

923
00:50:19,680 --> 00:50:23,400
Speaker 1: mapping the valley because that indicates that it may expand

924
00:50:23,480 --> 00:50:27,040
Speaker 1: beyond just the one dimensional dip and is in fact

925
00:50:27,080 --> 00:50:30,480
Speaker 1: more of a topographical space, you know, like we can

926
00:50:30,520 --> 00:50:34,120
Speaker 1: extend into three dimensions. But anyway, So to get into

927
00:50:34,120 --> 00:50:36,880
Speaker 1: the study, the author's note that while the Uncanny Valley

928
00:50:36,920 --> 00:50:39,640
Speaker 1: has very strong intuitive support, people tend to take it

929
00:50:39,680 --> 00:50:43,840
Speaker 1: as fact, experimental evidence for it has been limited and inconsistent.

930
00:50:43,920 --> 00:50:47,120
Speaker 1: As as we mentioned earlier, some studies seem to find

931
00:50:47,160 --> 00:50:50,560
Speaker 1: evidence for the valley, others don't you know, they say this,

932
00:50:50,560 --> 00:50:55,680
Speaker 1: This isn't necessarily a thing. So there are multiple experiments here. First,

933
00:50:55,760 --> 00:50:57,560
Speaker 1: they did a thing that I think was pretty smart.

934
00:50:57,640 --> 00:51:01,319
Speaker 1: If they were trying to chart a linear progression of

935
00:51:01,400 --> 00:51:04,000
Speaker 1: the up and down peaks and valleys, they tried to

936
00:51:04,080 --> 00:51:09,000
Speaker 1: generate an objectively determined gradient of more and less human

937
00:51:09,080 --> 00:51:11,719
Speaker 1: looking robots. So what a lot of these studies do

938
00:51:11,880 --> 00:51:15,640
Speaker 1: is maybe along the macaques study ideas, they show you

939
00:51:15,680 --> 00:51:18,520
Speaker 1: a lawnmower man, they show you a real person, they

940
00:51:18,520 --> 00:51:21,279
Speaker 1: show you a robot, uh, and they ask you to

941
00:51:21,440 --> 00:51:23,880
Speaker 1: characterize you know, how do you feel about these? What

942
00:51:24,000 --> 00:51:26,960
Speaker 1: they did here is that they gathered a very large

943
00:51:27,000 --> 00:51:30,680
Speaker 1: sample or relatively large sample of eighty images quote from

944
00:51:30,760 --> 00:51:34,720
Speaker 1: the wild meaning from the Internet. So these wild type

945
00:51:34,840 --> 00:51:37,279
Speaker 1: robots samples, and they had a bunch of inclusion and

946
00:51:37,360 --> 00:51:39,799
Speaker 1: exclusion criteria. I don't want to get into all of them,

947
00:51:39,800 --> 00:51:42,680
Speaker 1: but they tried to limit it to where it would

948
00:51:42,760 --> 00:51:45,600
Speaker 1: it would kind of throw out all these variables they

949
00:51:45,600 --> 00:51:48,080
Speaker 1: could complicate things like they tried to keep just certain

950
00:51:48,160 --> 00:51:52,120
Speaker 1: types of pictures of faces of real robots that are

951
00:51:52,160 --> 00:51:56,440
Speaker 1: built and uh and they had some exclusion criteria like

952
00:51:56,520 --> 00:51:58,920
Speaker 1: it couldn't be a well known character of a famous person.

953
00:51:59,760 --> 00:52:02,759
Speaker 1: Uh um, it couldn't have objects overlapping the face, It

954
00:52:02,800 --> 00:52:05,600
Speaker 1: couldn't be a toy, it had to be a real

955
00:52:05,719 --> 00:52:08,960
Speaker 1: humanoid robot. And then they had subjects rate these images

956
00:52:09,480 --> 00:52:12,920
Speaker 1: on what they call the mechano Humanoid scale, basically to

957
00:52:13,000 --> 00:52:16,880
Speaker 1: come up with an objectively derived score for each image

958
00:52:16,920 --> 00:52:19,719
Speaker 1: by using this this empirical research, by going to a

959
00:52:19,719 --> 00:52:22,200
Speaker 1: bunch of people and saying, hey, how mechanical is this?

960
00:52:22,520 --> 00:52:26,239
Speaker 1: How human is this? And then after they had a

961
00:52:26,360 --> 00:52:28,680
Speaker 1: rating for each of these eight images, and Robert have

962
00:52:28,800 --> 00:52:32,239
Speaker 1: included an image, uh, I think down here to show you,

963
00:52:32,280 --> 00:52:35,000
Speaker 1: like what all these robots where you can kind of see.

964
00:52:35,040 --> 00:52:38,239
Speaker 1: It starts with things that look not human at all,

965
00:52:38,440 --> 00:52:41,520
Speaker 1: just like a lump of wires and junk, and then

966
00:52:41,560 --> 00:52:44,160
Speaker 1: it proceeds up to something that looks like a picture

967
00:52:44,160 --> 00:52:47,439
Speaker 1: of a guy. Yes, yeah, very much. Though you start

968
00:52:47,480 --> 00:52:52,680
Speaker 1: off with very kind of wallyesque heads. Then you move

969
00:52:52,719 --> 00:52:57,480
Speaker 1: in through like like skinless gremlins, and then through the

970
00:52:57,520 --> 00:53:02,240
Speaker 1: sort of the the expected hierarchy of umanoid robots. Okay,

971
00:53:02,280 --> 00:53:05,080
Speaker 1: so they've got this thing, and then they rate all

972
00:53:05,120 --> 00:53:09,080
Speaker 1: these images and sort them into an ascending scale of humanness.

973
00:53:09,800 --> 00:53:12,279
Speaker 1: And then they took ratings in multiple different ways of

974
00:53:12,360 --> 00:53:16,640
Speaker 1: likability and trustworthiness. Now, in likability, they claimed to find

975
00:53:16,680 --> 00:53:21,520
Speaker 1: a robust uncanny Valley effect, where likability increased linearly with

976
00:53:21,760 --> 00:53:24,799
Speaker 1: humanoid qualities up to a certain point, and then it

977
00:53:24,840 --> 00:53:28,440
Speaker 1: took a negative dip as the humanoid qualities continued to

978
00:53:28,560 --> 00:53:31,600
Speaker 1: increase past that point, and then once again began to

979
00:53:31,760 --> 00:53:34,560
Speaker 1: rise at the far end of the scale. Now, one

980
00:53:34,560 --> 00:53:36,320
Speaker 1: thing I want to say, just looking at the results

981
00:53:36,400 --> 00:53:39,240
Speaker 1: is it does not appear that people were the most

982
00:53:39,440 --> 00:53:42,400
Speaker 1: bothered by the things that were the most human looking.

983
00:53:43,000 --> 00:53:45,799
Speaker 1: Like given my understanding of the uncanny valley, I would

984
00:53:45,800 --> 00:53:48,120
Speaker 1: have expected the stuff at the very top end of

985
00:53:48,120 --> 00:53:51,440
Speaker 1: the scale to be the most disturbing. But they actually

986
00:53:51,560 --> 00:53:53,799
Speaker 1: kind of liked the stuff at the very top end

987
00:53:53,800 --> 00:53:56,760
Speaker 1: of the scale. It was somewhere closer to the upper

988
00:53:56,920 --> 00:53:59,840
Speaker 1: half middle of the scale that they really didn't like.

989
00:54:00,800 --> 00:54:03,680
Speaker 1: Um So, to whatever extent, there is a real uncanny Valley,

990
00:54:03,719 --> 00:54:06,720
Speaker 1: it might not lie so close to the quote realism

991
00:54:06,760 --> 00:54:10,000
Speaker 1: into the spectrum as we think. They also performed some

992
00:54:10,120 --> 00:54:13,320
Speaker 1: trust experiments by creating a scenario where subjects would be

993
00:54:13,360 --> 00:54:16,319
Speaker 1: asked to trust these robots to invest money for them,

994
00:54:17,080 --> 00:54:19,920
Speaker 1: and the results there were basically they claimed that the

995
00:54:20,000 --> 00:54:24,600
Speaker 1: trust uh experiments did show some Uncanny Valley effects, but

996
00:54:24,640 --> 00:54:27,080
Speaker 1: the results were a little more complicated than on the

997
00:54:27,120 --> 00:54:32,200
Speaker 1: straightforward superficial likability scale, the likability really did look like

998
00:54:32,440 --> 00:54:37,120
Speaker 1: Uncanny Valley was being displayed. They also performed experiments with

999
00:54:37,520 --> 00:54:41,600
Speaker 1: a more traditional quote controlled series of composed face images,

1000
00:54:41,680 --> 00:54:44,720
Speaker 1: so would just be a series of basically the same

1001
00:54:44,760 --> 00:54:47,799
Speaker 1: face as a robot than a little bit more human,

1002
00:54:47,840 --> 00:54:50,239
Speaker 1: a little bit more human, little bit more human on

1003
00:54:50,280 --> 00:54:53,680
Speaker 1: this gradient of human nous. And they generally claimed to

1004
00:54:53,719 --> 00:54:56,360
Speaker 1: find that there was evidence for the Uncanny Valley effect

1005
00:54:56,360 --> 00:55:00,520
Speaker 1: in both likability and trust with both the why caught

1006
00:55:00,680 --> 00:55:04,799
Speaker 1: robot image samples and with these composed face images that

1007
00:55:04,840 --> 00:55:08,080
Speaker 1: they came up with. But as always, more studies are needed.

1008
00:55:08,080 --> 00:55:12,240
Speaker 1: But that looks like there is one study showing pretty

1009
00:55:12,280 --> 00:55:16,839
Speaker 1: solid evidence that there is something like an uncanny Valley effect. Yeah,

1010
00:55:16,840 --> 00:55:19,399
Speaker 1: and I like the idea that that that that it's

1011
00:55:19,600 --> 00:55:22,319
Speaker 1: it's an uncanny valley, but maybe it's just a more

1012
00:55:22,520 --> 00:55:27,200
Speaker 1: more nuanced from a a topographical standpoint. You know, they're

1013
00:55:27,200 --> 00:55:30,399
Speaker 1: they're more a little little bumps and little valleys within

1014
00:55:30,920 --> 00:55:34,239
Speaker 1: the overall valley, little caves you can crawl into and

1015
00:55:34,640 --> 00:55:37,839
Speaker 1: just yourself inside, and maybe even caves that turn into

1016
00:55:37,880 --> 00:55:40,480
Speaker 1: tunnels that emerge on the other side. Yeah, that that's

1017
00:55:40,520 --> 00:55:42,400
Speaker 1: an interesting thing. I mean, like they point out that

1018
00:55:42,400 --> 00:55:44,799
Speaker 1: there's a lot of variability in their data. Actually, like

1019
00:55:44,840 --> 00:55:48,440
Speaker 1: it wasn't um If you look at their their plot

1020
00:55:48,520 --> 00:55:51,239
Speaker 1: chart of where all the data points fall and then

1021
00:55:51,280 --> 00:55:53,280
Speaker 1: they plot a line going through it. If you plot

1022
00:55:53,320 --> 00:55:55,399
Speaker 1: a line going through all their data, it does show

1023
00:55:55,400 --> 00:55:58,319
Speaker 1: the uncanny valley effect. But you know, there there are

1024
00:55:58,360 --> 00:56:01,960
Speaker 1: outliers all over the place, like there is some there

1025
00:56:02,239 --> 00:56:05,839
Speaker 1: are some robots that are just consistently more like more

1026
00:56:05,880 --> 00:56:08,279
Speaker 1: than the other ones. I find it interestingly that the

1027
00:56:08,560 --> 00:56:10,759
Speaker 1: some of the higher rated ones, or at least uh

1028
00:56:10,840 --> 00:56:14,319
Speaker 1: I think what number seventy nine in particular, kind of

1029
00:56:14,360 --> 00:56:19,160
Speaker 1: looks like a generic human as opposed to say, go

1030
00:56:19,239 --> 00:56:22,120
Speaker 1: down to seventy four that looks like a very specific

1031
00:56:22,200 --> 00:56:24,799
Speaker 1: human like if I had to pick him or pick

1032
00:56:24,840 --> 00:56:29,920
Speaker 1: the human he's patterned after, assumingly out of a police lineup,

1033
00:56:30,000 --> 00:56:31,600
Speaker 1: I feel like I'd be able to do it. But

1034
00:56:31,680 --> 00:56:34,480
Speaker 1: also seventy four looks angry. I'm sorry, folks, you can't

1035
00:56:34,480 --> 00:56:37,319
Speaker 1: see what we're talking about, but it's frowning at you,

1036
00:56:37,719 --> 00:56:40,680
Speaker 1: kind of like should I kill all humans or just

1037
00:56:41,120 --> 00:56:43,960
Speaker 1: shrug it off? And maybe two day's the day that

1038
00:56:44,000 --> 00:56:47,239
Speaker 1: does introduce There are a lot of complicating factors here,

1039
00:56:47,239 --> 00:56:50,040
Speaker 1: and the authors acknowledged this, like, these images don't all

1040
00:56:50,080 --> 00:56:53,279
Speaker 1: have necessarily the same emotional affect, like some of them

1041
00:56:53,320 --> 00:56:57,160
Speaker 1: seem happy, some seem unhappy. There's enough variability across the

1042
00:56:57,200 --> 00:57:01,120
Speaker 1: board that you can think you're getting a reasonably decent

1043
00:57:01,280 --> 00:57:05,800
Speaker 1: answer when you plot reactions across all samples. But yeah,

1044
00:57:05,840 --> 00:57:08,799
Speaker 1: there's definitely a lot of different stuff going on here

1045
00:57:09,120 --> 00:57:12,880
Speaker 1: in addition to just being more or less human. I

1046
00:57:12,960 --> 00:57:15,759
Speaker 1: like how thirty four on our on our chart here

1047
00:57:16,040 --> 00:57:20,000
Speaker 1: it seems to rely heavily on animated mustache and eyebrows.

1048
00:57:20,120 --> 00:57:22,200
Speaker 1: Oh yeah, what is that? It looks like a It

1049
00:57:22,240 --> 00:57:25,320
Speaker 1: looks like a very mustache. I can't add to what

1050
00:57:25,360 --> 00:57:28,280
Speaker 1: you've just said. It's got a white mustache and brow

1051
00:57:28,400 --> 00:57:31,520
Speaker 1: and beard, and it's saying by it looks like a

1052
00:57:31,560 --> 00:57:34,320
Speaker 1: lot of these incomplete puppets are stripped away puppets, you

1053
00:57:34,320 --> 00:57:36,120
Speaker 1: see where they're like, all right, we got a lot

1054
00:57:36,120 --> 00:57:37,680
Speaker 1: of work to do on this thing, but at least

1055
00:57:37,680 --> 00:57:39,880
Speaker 1: we got the eyebrows in a mustache in place. But see,

1056
00:57:39,920 --> 00:57:42,200
Speaker 1: I find that one very likable. It doesn't look very

1057
00:57:42,240 --> 00:57:44,840
Speaker 1: human at all, but it's very I want to play

1058
00:57:44,880 --> 00:57:47,080
Speaker 1: with it. Yeah, okay, Robert, Well, we've got a bunch

1059
00:57:47,080 --> 00:57:48,400
Speaker 1: of more stuff to talk about, but I think we

1060
00:57:48,440 --> 00:57:50,760
Speaker 1: should call it there and come back and finish our

1061
00:57:50,760 --> 00:57:53,760
Speaker 1: discussion of the Uncanny Valley next time. Yeah, we'll get

1062
00:57:53,760 --> 00:57:56,400
Speaker 1: into we'll go beyond the Uncanny Valley. Yeah, so we'll

1063
00:57:56,480 --> 00:57:59,640
Speaker 1: we'll talk about what might cause the Uncanny Valley effect

1064
00:57:59,640 --> 00:58:02,080
Speaker 1: to what it were extent it does exist, and we

1065
00:58:02,120 --> 00:58:05,480
Speaker 1: can talk about you know, what happens when you ascend

1066
00:58:05,600 --> 00:58:09,200
Speaker 1: that that far slow? All right? Well, hey, in the meantime,

1067
00:58:09,400 --> 00:58:11,480
Speaker 1: head on over to stuff to Blow your Mind dot com.

1068
00:58:11,560 --> 00:58:14,000
Speaker 1: That is where you will find all the podcast episodes.

1069
00:58:14,040 --> 00:58:17,360
Speaker 1: You'll find videos, blog posts, as well as links out

1070
00:58:17,400 --> 00:58:19,880
Speaker 1: to our various social media accounts, and the landing page

1071
00:58:19,880 --> 00:58:22,600
Speaker 1: for this episode should include some links to some of

1072
00:58:22,600 --> 00:58:25,440
Speaker 1: the resources we're talking about here today. And if you

1073
00:58:25,440 --> 00:58:27,520
Speaker 1: want to get in touch with us, as always with

1074
00:58:27,600 --> 00:58:30,120
Speaker 1: feedback on this episode or any other, or you just

1075
00:58:30,120 --> 00:58:31,880
Speaker 1: want to say hi, or you want to let us

1076
00:58:31,920 --> 00:58:34,040
Speaker 1: know an episode topic you'd like us to cover in

1077
00:58:34,040 --> 00:58:36,720
Speaker 1: the future, you can email us at blow the Mind

1078
00:58:36,760 --> 00:58:48,360
Speaker 1: at how stuff works dot com for more on this

1079
00:58:48,560 --> 00:58:51,080
Speaker 1: and thousands of other topics. Is it how stuff works

1080
00:58:51,080 --> 00:59:12,040
Speaker 1: dot Commentary five five five point part proper First Far

1081
00:59:11,240 --> 00:59:11,400
Speaker 1: Far