1
00:00:00,299 --> 00:00:05,610
Herald: Hello, and welcome back to the FM-
channel. Our next speaker is Kai Konsa.
2
00:00:05,610 --> 00:00:11,340
He's a professor at the Graduate School of
Media Design at Cairo University Japan.
3
00:00:11,340 --> 00:00:15,870
His talk will be about focus on Wirfs.
From what I understood is that it's about
4
00:00:15,870 --> 00:00:20,930
shivers and goosebumps in media
performances that are usually spontaneous.
5
00:00:20,930 --> 00:00:27,140
But somehow he and his team managed to
induce them artificially. But I'm very
6
00:00:27,140 --> 00:00:33,560
forward. I'm looking very forward to this
talk as I don't really understand it. But
7
00:00:33,560 --> 00:00:40,520
now, if you want to ask questions in the
end, RC at the Channel rc3-fm in the
8
00:00:40,520 --> 00:00:46,129
rocket chat at the Channel FM and on
Twitter, and the videos and the hashtag
9
00:00:46,129 --> 00:00:51,210
3FM without the Dash, you can ask
questions that will be answered in the Q&A
10
00:00:51,210 --> 00:00:53,790
session afterwards.
repated but on english
11
00:00:53,790 --> 00:00:58,590
Allardyce parking Seotioa your current
idea that some 200 images of a player for
12
00:00:58,590 --> 00:01:04,830
native of translated and that and now I'm
looking forward to a hopefully very
13
00:01:04,830 --> 00:01:09,770
interesting talk.
Konsa: Hello and welcome to my talk
14
00:01:09,770 --> 00:01:16,630
"Frisson Waves", augmenting esthetic
chills in classical music performances.
15
00:01:16,630 --> 00:01:25,830
This is conceptual, early research work
from collaboration of a lot of artists,
16
00:01:25,830 --> 00:01:33,799
designers, researchers, and I'm just
speaker to introduce it to you a little
17
00:01:33,799 --> 00:01:39,729
bit. My name is Kai. And in the next 20
minutes, I will talk to you a little bit
18
00:01:39,729 --> 00:01:46,580
about what is Frisson. Give you a bit of
motivation and background information, why
19
00:01:46,580 --> 00:01:55,320
we are interested in this feeling. And
then I will talk about how can we
20
00:01:55,320 --> 00:02:03,380
recognize into youth and also share
Frisson. And then at the end, I'll talk
21
00:02:03,380 --> 00:02:11,620
about some conclusion and a little bit of
outlook. So the question is, what is
22
00:02:11,620 --> 00:02:18,480
Frisson? You might not have heard the
term, I actually haven't heard Frisson
23
00:02:18,480 --> 00:02:25,840
before we started the research. Two and a
half years, three years ago. But I
24
00:02:25,840 --> 00:02:31,849
definitely knew the feeling. So if you're
listening attentively to a musical piece,
25
00:02:31,849 --> 00:02:39,480
sometimes you might get goose bumps or
some shiver down your spine. And that is
26
00:02:39,480 --> 00:02:46,650
usually triggered from the music. Frisson
is from the French shivers and sorry, my
27
00:02:46,650 --> 00:02:55,800
pronunciation with the German accent. So
you applied for that, I hope. And it's
28
00:02:55,800 --> 00:03:01,849
this psycho physiological phenomenon, that
we feel when we get these goose bumps or
29
00:03:01,849 --> 00:03:11,519
shivers that are triggered from music, but
also other. If you insist and you might
30
00:03:11,519 --> 00:03:20,610
wonder why goosebumps, how can goosebumps
be related to a positive feeling? There is
31
00:03:20,610 --> 00:03:32,819
actually no need to answer theories. I'm
one that I particularly like, is that
32
00:03:32,819 --> 00:03:37,319
Frisson is often induced over music or
over some kind of stimulus that is
33
00:03:37,319 --> 00:03:42,361
repetitive, that has a certain pattern,
and that at one point the pattern breaks
34
00:03:42,361 --> 00:03:50,629
and that surprises you. So this triggers
your autonomous nervous system. So the
35
00:03:50,629 --> 00:03:56,099
fight or flight response you get to
surprise you wonder alertness goes up and
36
00:03:56,099 --> 00:04:03,370
you realize that there's no danger and you
will relax. And feel this esthetic chills.
37
00:04:03,370 --> 00:04:13,349
So the talk, I will give today is an
exploration of the feeling of Frisson with
38
00:04:13,349 --> 00:04:21,500
technology. So how could we detect, induce
or transmit it using especially variable
39
00:04:21,500 --> 00:04:26,419
sensors and actuators and to be a little
bit aclimactic? I can already tell you
40
00:04:26,419 --> 00:04:33,340
that this is still in process. So this is
really exploratory work. However, you
41
00:04:33,340 --> 00:04:42,540
might also wonder why do you care about
this? Why do you want to do this? And you
42
00:04:42,540 --> 00:04:48,880
know, one reason, of course, is. Because
we can and because it's fun, and I think
43
00:04:48,880 --> 00:04:55,510
that's definitely, you know, kind of one
aspect of of the research, however. Also
44
00:04:55,510 --> 00:05:02,509
another reason is so our lab in Yokohama
works in human factors, research, so HCI
45
00:05:02,509 --> 00:05:09,130
and human computer interaction. And we
lately revisited a lot of work also from
46
00:05:09,130 --> 00:05:16,550
cybernetics, also nonlinear dynamics in
terms of research and also in terms of art
47
00:05:16,550 --> 00:05:22,550
and performance. We are very much inspired
by Stilnox work on extending and
48
00:05:22,550 --> 00:05:31,430
augmenting our body. And there's this
realization, if you work on research that,
49
00:05:31,430 --> 00:05:35,560
you know, knowledge is not merely
functional, there's always some kind of
50
00:05:35,560 --> 00:05:42,650
enjoyment in understanding a concept. And
I think also this community will really
51
00:05:42,650 --> 00:05:47,880
understand that type of feeling and this
sense of wonder and this feeling we also
52
00:05:47,880 --> 00:05:53,760
want to explore. We want to understand
ourselves better in terms of cognition
53
00:05:53,760 --> 00:06:00,660
perception, but also in terms of our
feeling. And actually, last year, I gave
54
00:06:00,660 --> 00:06:06,960
also a talk on on boiling mind on an
Frisson loop that we played with and
55
00:06:06,960 --> 00:06:12,460
started researching on. And to some
extent, this expression. If this all beef
56
00:06:12,460 --> 00:06:17,400
talk is just the continuation of this, and
overall, we are also looking for more
57
00:06:17,400 --> 00:06:23,880
creative ways to use physiological data or
other wearable computing sensing that is
58
00:06:23,880 --> 00:06:33,000
not related to surveillance. So extended
to this, we also wonder what does it mean
59
00:06:33,000 --> 00:06:40,449
to be life? It's easy if you think about
transmitting audio video, easy in
60
00:06:40,449 --> 00:06:45,490
quotation marks because yeah, there are
some experts that know a lot about that.
61
00:06:45,490 --> 00:06:53,220
And I see also the effort that goes into
the remote experience and not the
62
00:06:53,220 --> 00:06:59,430
congresses or conferences. However, we
still don't know how to transmit an
63
00:06:59,430 --> 00:07:05,250
atmosphere or a feeling that's much more
difficult. I think the Congress is a very
64
00:07:05,250 --> 00:07:11,430
nice example for that because it moved
from Berlin to Hamburg to Leipzig. But
65
00:07:11,430 --> 00:07:17,139
every time I visited, I kind of felt at
home. I felt, Oh yeah, these are, you
66
00:07:17,139 --> 00:07:23,319
know, kind of the people I like. These are
the culture, the community I belong to,
67
00:07:23,319 --> 00:07:29,050
even though it's at different places. And
we wonder, you know, kind of how can we
68
00:07:29,050 --> 00:07:36,569
transmit at this type of feeling and to
efforts that we get inspired from from
69
00:07:36,569 --> 00:07:42,340
this work is one is neuro life. That's a
project, an EU project with co-
70
00:07:42,340 --> 00:07:50,330
investigator Jamie Ward and also
cybernetic being project here in Japan,
71
00:07:50,330 --> 00:07:56,560
headed by Quarterman Minami. So that deals
with things like parallel agency and
72
00:07:56,560 --> 00:08:01,259
similar. And both of them are actually
also collaborators in the work that I will
73
00:08:01,259 --> 00:08:07,630
present today. So this is the high level
overview why we are interested in this
74
00:08:07,630 --> 00:08:14,710
song, but now getting back to the esthetic
chills. And first, the question is how
75
00:08:14,710 --> 00:08:22,620
could we go about and try to detect or
recognize them? Looking into related work,
76
00:08:22,620 --> 00:08:29,800
of course, we'll see Frisson is that
chills, of course, affect our physiology.
77
00:08:29,800 --> 00:08:35,090
And the first thing that you notice is, of
course, the payload erection. So the goose
78
00:08:35,090 --> 00:08:40,710
bumps that you can get on your arm so the
hairs go up. So we could try to detect
79
00:08:40,710 --> 00:08:44,990
that. However, that might be a little bit
difficult because some people might not
80
00:08:44,990 --> 00:08:53,690
have so much hair on on them and so on. So
then looking into other physiological
81
00:08:53,690 --> 00:09:00,940
changes, respiratory rate is going up for
the sweat glands, electro dermal activity.
82
00:09:00,940 --> 00:09:05,690
You will see more peaks. That's a stress
and excitement indicator, and heart rate
83
00:09:05,690 --> 00:09:10,890
goes up, blood pressure goes up and
usually heart rate variability related
84
00:09:10,890 --> 00:09:17,209
features go down. Because also, if you saw
last year's talk, we already built a
85
00:09:17,209 --> 00:09:25,329
system to record electro dermal activity.
So the sweating on the hand as well as
86
00:09:25,329 --> 00:09:34,050
heart rate we just thought will move along
and use that. Luckily, we also did a
87
00:09:34,050 --> 00:09:38,760
redesign of the wristband bands in the
meantime, so they look a little bit nicer
88
00:09:38,760 --> 00:09:45,260
now and you see also a life demo on my
background right now. So you see EDA and
89
00:09:45,260 --> 00:09:54,300
heart rate behind and if I press here. You
should also see some noise on the sensor.
90
00:09:54,300 --> 00:10:03,149
The visualization, by the way, is done by
Kirill Ragodzin. So thanks for the work!
91
00:10:03,149 --> 00:10:08,990
And then moving forward, so we use these
wristbands to set up a controlled
92
00:10:08,990 --> 00:10:14,810
experiment to detect esthetic chill
events. We just added a trigger, so to add
93
00:10:14,810 --> 00:10:19,620
some self-reporting to it. So in this
case, we really use the user as a self-
94
00:10:19,620 --> 00:10:27,470
report to classify or to label the Frisson
events that has, of course, you know, also
95
00:10:27,470 --> 00:10:33,520
some limitations. So you hope that that's
good enough to capture it. And we used
96
00:10:33,520 --> 00:10:39,740
some music pieces also from related work
and did some counterbalancing and run this
97
00:10:39,740 --> 00:10:48,700
lab study just in know kind of controlled
space or with headphones and so on. We
98
00:10:48,700 --> 00:10:54,640
finished this, but then we also wondered,
you know, how does it look like in real
99
00:10:54,640 --> 00:11:02,560
life in the wild experiments? So we also
organized a concert. With 18 audience
100
00:11:02,560 --> 00:11:08,800
members for one album musical program, and
the set up was the same, so everybody got
101
00:11:08,800 --> 00:11:15,620
a wristband and a trigger. We also added a
third for the pianist, so using EDA from
102
00:11:15,620 --> 00:11:23,470
the foot actually works also relatively
well and then recorded here the data and
103
00:11:23,470 --> 00:11:35,829
hope that people would report their focus
on their esthetic chills. Here's now one
104
00:11:35,829 --> 00:12:32,310
video. A short minute video that shows you
the recording. piano music How about the
105
00:12:32,310 --> 00:12:39,400
analysis? I have to see? I'm sorry the
this is still ongoing, so we don't really
106
00:12:39,400 --> 00:12:43,710
have a lot of results yet. And of course,
there were a lot of issues with the life
107
00:12:43,710 --> 00:12:48,810
recording. If you're interested in doing
something similar, contact some of the
108
00:12:48,810 --> 00:12:54,810
technical stuff or also me. We can give
you hints and doing this now over 15 or 20
109
00:12:54,810 --> 00:13:00,300
years and always do something is going
wrong, depending on the setting and so on.
110
00:13:00,300 --> 00:13:05,490
Now I also know more about the classical
music concerts. However, we got some
111
00:13:05,490 --> 00:13:11,000
useful data. The problem there was we
could also train a machine learning model
112
00:13:11,000 --> 00:13:16,010
because we really wanted to dect it real
time. And it seemed to work really well.
113
00:13:16,010 --> 00:13:22,520
We are just still not sure if it really
works or not, so we want to be very
114
00:13:22,520 --> 00:13:26,410
careful about that. So we get higher
accuracy spec. But given the limited
115
00:13:26,410 --> 00:13:31,800
amount of uses we had also and we want to
look into that a little bit more. However,
116
00:13:31,800 --> 00:13:36,220
the analysis, as well as the data sets,
will be publicly available. And if you
117
00:13:36,220 --> 00:13:41,279
want to get them a little bit earlier to
also contact me. So then moving on, this
118
00:13:41,279 --> 00:13:46,940
is the progress on detection. How does it
look like for triggering or inducing
119
00:13:46,940 --> 00:13:55,899
Frisson? So there's also a lot of cool
related work. I just show or highlight two
120
00:13:55,899 --> 00:14:03,350
of them. One is work by Shoko Fukushima at
all. And they're using the electrostatic
121
00:14:03,350 --> 00:14:12,250
effect on the arm to control payload
erection. And they use it to increase the
122
00:14:12,250 --> 00:14:17,670
surprise feeling of somebody, so you put
your arm inside and they can control the
123
00:14:17,670 --> 00:14:24,060
payload erection. Other work is from Ha at
all, whether using three Peltier elements
124
00:14:24,060 --> 00:14:31,550
on your back, on your spine and deactivate
them upwards to also induce Frisson or
125
00:14:31,550 --> 00:14:39,610
static chills. The problem with those two
setups, it's quite hard to get them into a
126
00:14:39,610 --> 00:14:44,190
into a concert hall. And, you know, some
people might not really have much hair on
127
00:14:44,190 --> 00:14:49,699
their arms or so on, so there might be
limitations for it. So then, you know, for
128
00:14:49,699 --> 00:14:56,009
first iteration, we decided to go for a
neck prototype, because kind of the neck
129
00:14:56,009 --> 00:15:02,620
is also a part of some of the Frisson
responses. So you get either chills down
130
00:15:02,620 --> 00:15:08,509
the spine or up the neck or also your hair
might stand up. So we thought it's a good
131
00:15:08,509 --> 00:15:12,630
start and we used first healthy elements
or thermal modules and also vibrant
132
00:15:12,630 --> 00:15:18,350
tactile feedback. In later iterations we
moved just to a thermal feedback to
133
00:15:18,350 --> 00:15:25,009
activate this on the back of the neck
around on the upper side of the trapezius
134
00:15:25,009 --> 00:15:32,260
muscle, and they would activate with
slight cold feedback. So for an initial
135
00:15:32,260 --> 00:15:37,940
tests, it seemed to work or just this,
just with 10 participants, around 30
136
00:15:37,940 --> 00:15:44,380
minutes per participant, we had two music
pieces that are based on related works or
137
00:15:44,380 --> 00:15:51,080
Chopin and Gustav Holst. We counterbalance
the conditions or music pieces with
138
00:15:51,080 --> 00:15:56,880
neckband without neck bent with neck bent,
with activation and without activation.
139
00:15:56,880 --> 00:16:03,570
And from an initial test, we can see that
it seems that slight cold feedback really
140
00:16:03,570 --> 00:16:09,010
provides more instances of reported
Frisson. So there is a slight positive
141
00:16:09,010 --> 00:16:12,139
feedback, but you know, it was still quite
little participants and we'll have to
142
00:16:12,139 --> 00:16:17,750
continue and see also with a little bit of
redesign. So we want to change the order
143
00:16:17,750 --> 00:16:25,130
and placement of the pelty elements for
the continuation work as well. Now moving
144
00:16:25,130 --> 00:16:30,650
to the last part, so we talked about
detection, induction, and now let's talk
145
00:16:30,650 --> 00:16:37,880
about sharing or transmitting Frisson. He
had the idea would be, you know, you are
146
00:16:37,880 --> 00:16:45,199
listening to a musical piece, a classical
piece and one person gets Frisson, does
147
00:16:45,199 --> 00:16:52,230
this detect it over the wristband and then
it's distributed ripples through the
148
00:16:52,230 --> 00:17:00,670
neighbors? They get activated over the
Nick Bend and hopefully also free Frisson
149
00:17:00,670 --> 00:17:08,309
again around the same time it just after
the red circled person felt that the
150
00:17:08,309 --> 00:17:13,199
esthetic chills. So in this case, then,
you know, we would have all of the
151
00:17:13,199 --> 00:17:18,309
audience members need to wear sensors and
actuators, and we would need to freshen
152
00:17:18,309 --> 00:17:27,530
detection and also then the activation
based on that. And for that, we also
153
00:17:27,530 --> 00:17:33,700
organized a not a concert in this case, 50
audience members. The program was around
154
00:17:33,700 --> 00:17:43,270
1,5 hours. And the set up was, as you see
here. So performers on the top, and then
155
00:17:43,270 --> 00:17:51,340
we had two sections. One saw 25 users
would wear just the wristband as a kind of
156
00:17:51,340 --> 00:17:58,360
control group. And the second group, 25
users would wear wristband and neck bent,
157
00:17:58,360 --> 00:18:02,929
so it would get to actually the detection
and also the activation. That's all, you
158
00:18:02,929 --> 00:18:10,180
know, 50 plus wristbands needed charging
and 25 neck bands were manufactured, and
159
00:18:10,180 --> 00:18:18,000
this is the picture from the actual
concert with an nick bent section. And
160
00:18:18,000 --> 00:18:22,640
here is how this should work, so, you
know, you have first one person, you
161
00:18:22,640 --> 00:18:29,960
detect the Frisson and then you ripple it
out to the neighbors, then the next person
162
00:18:29,960 --> 00:18:34,710
might feel Frisson we detected over the
wristband and then replay it out to the
163
00:18:34,710 --> 00:18:40,680
other people that haven't gotten
activation yet and so on. So you have then
164
00:18:40,680 --> 00:18:50,559
a wave of Frisson hopefully moving through
the audience members. This is not a setup
165
00:18:50,559 --> 00:18:54,720
up, yab he (NAME!) who also did a lot of
the organization parts or so on at the
166
00:18:54,720 --> 00:19:02,630
piano, and he is then a small video that
summarizes the work. And at the end you
167
00:19:02,630 --> 00:19:08,090
see also the servers, the recording
server, the activation server and the I
168
00:19:08,090 --> 00:19:38,100
know the detection server and also the
activation server. piano music playes
169
00:19:38,100 --> 00:20:25,070
one person plays piano, another plays
chello So the question you might have
170
00:20:25,070 --> 00:20:35,140
now, did it work? Hmm. Not not completely
sure. Again, here, work in progress or
171
00:20:35,140 --> 00:20:39,870
analysis is ongoing, and we can't also
really see because, yeah, we had this
172
00:20:39,870 --> 00:20:44,890
control group and we could see more
Frisson events in the sharing group. But
173
00:20:44,890 --> 00:20:49,460
how to interpret that, that's really,
really difficult. We are also working on
174
00:20:49,460 --> 00:20:54,049
the design of the wristbands as well as
the neckband and especially for the
175
00:20:54,049 --> 00:20:59,570
neckband. We got a couple of uses, I think
five or six or seven that really didn't
176
00:20:59,570 --> 00:21:05,909
like the neckband, not the activations or
the slight called activation was OK, but
177
00:21:05,909 --> 00:21:10,870
just because it was a little bit too tight
and a little bit too uncomfortable. So
178
00:21:10,870 --> 00:21:16,049
we're working on a redesign we have for
the next concert in april. All of the data
179
00:21:16,049 --> 00:21:23,130
make it also publicly available. Soon
enough, look also a little bit more what
180
00:21:23,130 --> 00:21:27,880
we can find out about what happened. This
brings me to the end of the presentation.
181
00:21:27,880 --> 00:21:34,590
I hope you enjoyed it. Yeah, I just wanted
to thank a couple of people first and
182
00:21:34,590 --> 00:21:44,370
foremost Yann He, who organized this
Frisson , who introduced us. And also the
183
00:21:44,370 --> 00:21:49,180
teams are dismissed for their second
concert. The extended team thanks a lot
184
00:21:49,180 --> 00:21:54,720
for everybody who was involved here, then
also all of the names. So these are the
185
00:21:54,720 --> 00:21:58,750
people that did the actual work did not
just doing the presenting like I do right
186
00:21:58,750 --> 00:22:04,980
now. I hope I haven't missed anybody. So
also, thanks to George, Dingding, Denny
187
00:22:04,980 --> 00:22:10,470
and so on and all of the other people
involved in group planning the studio
188
00:22:10,470 --> 00:22:19,529
Apollo and also the piano NIST's and
interactive performers. So thanks a lot.
189
00:22:19,529 --> 00:22:27,270
Yeah, that brings me to the end of the
presentation. As I said, we have a third
190
00:22:27,270 --> 00:22:31,320
concert, probably in April next year in
Yokohama, Tokyo area. So if you're
191
00:22:31,320 --> 00:22:36,670
interested, let me know also if you have a
general interest in effect or similar
192
00:22:36,670 --> 00:22:43,400
phenomenon. Also, just write me an email.
It would be good if you mentioned Frisson
193
00:22:43,400 --> 00:22:49,750
remote experience in the subject, so I can
just filter that out and something
194
00:22:49,750 --> 00:22:54,820
completely different. We also have a
conference next year. March submission
195
00:22:54,820 --> 00:23:01,110
deadline is January 7th, augmented humans
in Japan and Germany and cyberspace that
196
00:23:01,110 --> 00:23:11,419
deals with maybe similar work. So thanks a
lot for listening and I hope. Yeah, I told
197
00:23:11,419 --> 00:23:16,850
you something interested in the last 20
minutes. By.
198
00:23:16,850 --> 00:23:21,620
Herald: Hello, and welcome back to the FM
Channel, thank you Kay for the very
199
00:23:21,620 --> 00:23:25,400
interesting talk, and Kay should actually
be with us to answer a few questions.
200
00:23:25,400 --> 00:23:27,400
Hello, Kay.
Kay: Hello.
201
00:23:27,400 --> 00:23:33,210
Herald: And we actually do have a few
questions already. And the first one
202
00:23:33,210 --> 00:23:39,279
sounds a bit more like a comment, but I
tell you anyway. So one one viewer noted
203
00:23:39,279 --> 00:23:42,960
that this a technical cybernetics and
systems course here in Illmenau, the two
204
00:23:42,960 --> 00:23:46,980
in the know and wanted to know if you were
aware of this already.
205
00:23:46,980 --> 00:23:53,881
Kay: Actually, it wasn't, but it sounds
that sounds quite fun and getting more and
206
00:23:53,881 --> 00:23:57,762
more interested in cybernetics as well.
And I think it's useful to revisit some of
207
00:23:57,762 --> 00:24:03,730
the ideas around feedback loops, as I said
at the beginning. So that's cool if you if
208
00:24:03,730 --> 00:24:08,000
you're already looking into that. And I
think especially if people go into HCI
209
00:24:08,000 --> 00:24:12,240
fields, I think it's quite useful to get a
little bit of that background.
210
00:24:12,240 --> 00:24:17,809
Herald: Yes, nice. OK, let's go to the
next question. It's about neural networks,
211
00:24:17,809 --> 00:24:22,620
which have the neural network was us. Have
you considered neural net different doing
212
00:24:22,620 --> 00:24:27,240
differential equations such as echo-state
network or reservoir computing, which are
213
00:24:27,240 --> 00:24:30,490
good when modeling stiff time
consciousness processes?
214
00:24:30,490 --> 00:24:37,039
Kay: That's actually a really good
question and actually also good hint for
215
00:24:37,039 --> 00:24:42,460
what to do next. I just tried to look up.
I saw the question already. Also in the
216
00:24:42,460 --> 00:24:47,200
chat, I try to look up what we used, and I
think at the beginning we just used
217
00:24:47,200 --> 00:24:52,149
support vector machines, so not neural
networks. And no, no, we are using some
218
00:24:52,149 --> 00:24:57,940
neural network, but I don't know the
configuration and I couldn't check. I'll
219
00:24:57,940 --> 00:25:03,480
get back to the person who asked the
question. What was interesting for me was
220
00:25:03,480 --> 00:25:08,559
that the the data looks already quite
good. The sensor data looks actually quite
221
00:25:08,559 --> 00:25:15,610
good. And I would assume that in most
cases, any classifier will do a decent job
222
00:25:15,610 --> 00:25:21,760
for the lab experiments for the other
works. I think, yeah, that sounds quite
223
00:25:21,760 --> 00:25:29,000
interesting. I also want to go more
towards, um yeah, nonlinear dynamics work
224
00:25:29,000 --> 00:25:35,700
as well in terms of of of estimating
Frisson or different feelings. But that's
225
00:25:35,700 --> 00:25:41,140
that's a really good hint, that it would
be more question also for Jo Ann Hunt,
226
00:25:41,140 --> 00:25:47,340
coauthor of the paper that is also linked.
She is our data analyst and so on, and
227
00:25:47,340 --> 00:25:51,909
notes what what she used to the first
classifier was a support vector machine
228
00:25:51,909 --> 00:25:56,130
fairly basic and I think recently we use a
neural network, but I thought it's just
229
00:25:56,130 --> 00:26:01,810
very straightforward. PyTorch long
training, but nothing, nothing special and
230
00:26:01,810 --> 00:26:06,020
nothing fancy so far.
Herald: OK, nice, I guess I guess she will
231
00:26:06,020 --> 00:26:10,330
probably also hear from the questions
then. And then let's go to the next
232
00:26:10,330 --> 00:26:15,620
question. Have you considered or tested
the effects of editorial stimulus, such as
233
00:26:15,620 --> 00:26:20,890
attempting to cross Frisson Waves and in
boring situations instead of like
234
00:26:20,890 --> 00:26:29,980
interesting ones?
Kay: That's also quite a good or
235
00:26:29,980 --> 00:26:35,000
interesting question. I mean, there were
some some audience members also that
236
00:26:35,000 --> 00:26:41,059
mentioned that Nick Bend was actually a
little bit uncomfortable. So I'm not
237
00:26:41,059 --> 00:26:47,120
really sure if we cost Frisson with them.
And I don't know what would happen if you,
238
00:26:47,120 --> 00:26:53,320
I think you probably would just make the
situation uncomfortable anyways. I'm not
239
00:26:53,320 --> 00:27:00,710
sure what would happen then, if you
stimulating called feedback. Actually, you
240
00:27:00,710 --> 00:27:05,710
might get the fear response in these cases
if you're in a boring situation, I'm not
241
00:27:05,710 --> 00:27:12,909
sure if you if you. Hmm. Yeah, I actually
I don't know. It's definitely an
242
00:27:12,909 --> 00:27:21,240
interesting idea to to use it in boring
situations. Can you get somebody to change
243
00:27:21,240 --> 00:27:28,740
their? The feeling and get to the more
excited state? We are playing often, we
244
00:27:28,740 --> 00:27:32,630
played a little bit with the thermal
feedback and it was always interesting if
245
00:27:32,630 --> 00:27:37,450
you change the thermal feedback. So
instead of if you see something hot in VR,
246
00:27:37,450 --> 00:27:42,230
you give cold stimulus or so on. It really
is a little bit confusing and interesting.
247
00:27:42,230 --> 00:27:49,279
I haven't thought about that in Frisson
situation. And if it works for boring
248
00:27:49,279 --> 00:27:53,480
work, but it's definitely cool or
interesting. So if somebody wants to play
249
00:27:53,480 --> 00:27:58,850
with that, I would be up for also giving a
little bit of help or ideas in that
250
00:27:58,850 --> 00:28:04,130
direction.
Herald: OK, thank you for. For answering
251
00:28:04,130 --> 00:28:11,809
these questions, unfortunately, that don't
seem to be any more of them. So that's it.
252
00:28:11,809 --> 00:28:16,260
Thank you very much for the very
interesting talk. That's a topic I haven't
253
00:28:16,260 --> 00:28:22,040
really thinking that much about, but thank
you very much. It was very interesting.
254
00:28:22,040 --> 00:28:26,340
Kay: Thanks a lot also for having me, and
it's always fun and I always enjoyed the
255
00:28:26,340 --> 00:28:30,940
feedback. Yet there's also a candlelight
life demonstration behind me. So you saw
256
00:28:30,940 --> 00:28:34,539
my excitement level kind of increasing or
decreasing with the questions.
257
00:28:34,539 --> 00:28:38,680
Herald: Oh wow, that's pretty cool.
Kay: OK. Yeah, thanks a lot.
258
00:28:38,680 --> 00:28:46,540
Herald: Then here and here on the FM
channel, the next thing happening will at
259
00:28:46,540 --> 00:28:51,010
11:00 p.m. the lightening think talk or
not on the film channel. But one of the
260
00:28:51,010 --> 00:28:56,059
next things happening at three will be at
11:00 p.m. The lightning talks at remote
261
00:28:56,059 --> 00:29:02,210
range. And here on on our channel,
actually at 12:00 a.m. or midnight, there
262
00:29:02,210 --> 00:29:07,040
will be the next terror news show. And
that until then by.
263
00:29:07,040 --> 00:29:15,000
Subtitles created by many many volunteers and
the c3subtitles.de team. Join us, and help us!