WEBVTT

00:00:00.000 --> 00:00:02.610 align:middle line:90%


00:00:02.610 --> 00:00:07.080 align:middle line:84%
Laura, do you think it's
possible to do pupillometry

00:00:07.080 --> 00:00:08.400 align:middle line:90%
in the real world?

00:00:08.400 --> 00:00:10.380 align:middle line:90%
I mean, outside the laboratory?

00:00:10.380 --> 00:00:11.130 align:middle line:90%
Yes.

00:00:11.130 --> 00:00:13.360 align:middle line:90%
And, in fact, we have done this.

00:00:13.360 --> 00:00:16.020 align:middle line:84%
We have done pupillometry
with performing musicians

00:00:16.020 --> 00:00:18.930 align:middle line:84%
in actual concert
halls on the stage.

00:00:18.930 --> 00:00:19.430 align:middle line:90%
All.

00:00:19.430 --> 00:00:21.960 align:middle line:84%
Right I thought that
was not possible.

00:00:21.960 --> 00:00:26.610 align:middle line:84%
So, in an ecological
situation, how did you do that?

00:00:26.610 --> 00:00:27.655 align:middle line:90%
Well, it's not that easy.

00:00:27.655 --> 00:00:29.780 align:middle line:84%
There are many things that
you have to think about.

00:00:29.780 --> 00:00:32.750 align:middle line:84%
Yeah, because there are
all these nuisance factors

00:00:32.750 --> 00:00:35.310 align:middle line:90%
like changes in luminance.

00:00:35.310 --> 00:00:35.810 align:middle line:90%
Yeah.

00:00:35.810 --> 00:00:38.150 align:middle line:90%
They eyes move around.

00:00:38.150 --> 00:00:40.260 align:middle line:90%
Yes, so you have to be careful.

00:00:40.260 --> 00:00:44.360 align:middle line:84%
For example, you have to try to
keep the lighting very stable.

00:00:44.360 --> 00:00:46.100 align:middle line:84%
In our case, looking
at musicians,

00:00:46.100 --> 00:00:49.100 align:middle line:84%
you want to avoid using
stage lights that are shining

00:00:49.100 --> 00:00:50.810 align:middle line:90%
into the eyes of the musicians.

00:00:50.810 --> 00:00:54.410 align:middle line:84%
And also, for us,
it was a good idea

00:00:54.410 --> 00:00:57.980 align:middle line:84%
to give the musicians something
to look at so they are actually

00:00:57.980 --> 00:01:01.070 align:middle line:84%
playing from their scores,
which helps give some more

00:01:01.070 --> 00:01:03.540 align:middle line:84%
control into the data
that we're getting.

00:01:03.540 --> 00:01:04.040 align:middle line:90%
OK.

00:01:04.040 --> 00:01:09.380 align:middle line:84%
So, what kind of musicians
have you been studying?

00:01:09.380 --> 00:01:12.830 align:middle line:84%
We've done some studies so
far with string quartets.

00:01:12.830 --> 00:01:14.533 align:middle line:90%
We've had two string quartets.

00:01:14.533 --> 00:01:15.200 align:middle line:90%
Classical music.

00:01:15.200 --> 00:01:16.220 align:middle line:90%
Yes.

00:01:16.220 --> 00:01:16.850 align:middle line:90%
Yes.

00:01:16.850 --> 00:01:19.460 align:middle line:84%
And we've had a
student classical group

00:01:19.460 --> 00:01:22.420 align:middle line:90%
and an expert classical group.

00:01:22.420 --> 00:01:25.660 align:middle line:90%
So tell me a bit about it.

00:01:25.660 --> 00:01:29.050 align:middle line:84%
Well, the hypothesis
that we've been testing

00:01:29.050 --> 00:01:32.650 align:middle line:84%
is that these musicians
are performing

00:01:32.650 --> 00:01:35.740 align:middle line:84%
a very difficult task when
they're on stage performing

00:01:35.740 --> 00:01:36.820 align:middle line:90%
complex music.

00:01:36.820 --> 00:01:40.310 align:middle line:84%
And this engages their
attention in different ways.

00:01:40.310 --> 00:01:43.930 align:middle line:84%
And, so, on one hand, we
would expect their attention

00:01:43.930 --> 00:01:48.100 align:middle line:84%
to fluctuate in relation to
different demands of the music.

00:01:48.100 --> 00:01:51.430 align:middle line:84%
And, on the other hand, we would
also expect their attention

00:01:51.430 --> 00:01:55.990 align:middle line:84%
to kind of reflect the social
demands of the situation

00:01:55.990 --> 00:01:58.070 align:middle line:90%
that they are playing in.

00:01:58.070 --> 00:01:58.570 align:middle line:90%
All right.

00:01:58.570 --> 00:02:01.510 align:middle line:84%
So, you are interested in
the changes of the pupil

00:02:01.510 --> 00:02:06.100 align:middle line:84%
but also in the movement of
the eyes at the same time?

00:02:06.100 --> 00:02:08.139 align:middle line:84%
Yes, both of these
things at the same time.

00:02:08.139 --> 00:02:11.860 align:middle line:84%
The pupil is useful to tell
us about how much attention

00:02:11.860 --> 00:02:15.730 align:middle line:84%
is being employed
at any given moment.

00:02:15.730 --> 00:02:18.670 align:middle line:84%
And the gaze is
useful because we then

00:02:18.670 --> 00:02:21.490 align:middle line:84%
know about when the musicians
are looking at each other

00:02:21.490 --> 00:02:23.180 align:middle line:84%
and where they're
looking at the score.

00:02:23.180 --> 00:02:27.000 align:middle line:84%
So, how did you actually
measure the pupil?

00:02:27.000 --> 00:02:28.900 align:middle line:84%
What kind of
equipment did you use?

00:02:28.900 --> 00:02:30.220 align:middle line:90%
We used mobile eye tracking.

00:02:30.220 --> 00:02:31.690 align:middle line:90%
OK.

00:02:31.690 --> 00:02:34.630 align:middle line:84%
This is better if you're
working with musicians

00:02:34.630 --> 00:02:37.240 align:middle line:84%
because they are not fixed
in front of a computer.

00:02:37.240 --> 00:02:38.950 align:middle line:84%
They can be actually
out in the world,

00:02:38.950 --> 00:02:42.190 align:middle line:84%
and there's some compensation
that happens for the movements

00:02:42.190 --> 00:02:43.900 align:middle line:90%
that they make.

00:02:43.900 --> 00:02:48.940 align:middle line:84%
So, you mean you use something
like the eye tracking glasses?

00:02:48.940 --> 00:02:49.450 align:middle line:90%
Yes.

00:02:49.450 --> 00:02:51.770 align:middle line:84%
We used glasses
that look like this.

00:02:51.770 --> 00:02:54.940 align:middle line:84%
This is one of our cellists
wearing the glasses.

00:02:54.940 --> 00:02:59.470 align:middle line:84%
So, in this case, they were also
fitted with reflective markers

00:02:59.470 --> 00:03:02.510 align:middle line:84%
because we are doing motion
capture at the same time.

00:03:02.510 --> 00:03:05.180 align:middle line:84%
And so we were able to
combine these two measures.

00:03:05.180 --> 00:03:06.490 align:middle line:90%
Right.

00:03:06.490 --> 00:03:10.330 align:middle line:84%
So, all four musicians
in the string quartet

00:03:10.330 --> 00:03:11.680 align:middle line:90%
had a pair of glasses on.

00:03:11.680 --> 00:03:12.790 align:middle line:90%
That's right.

00:03:12.790 --> 00:03:15.700 align:middle line:90%
And that was non-problematic?

00:03:15.700 --> 00:03:18.460 align:middle line:90%
They could do it comfortably?

00:03:18.460 --> 00:03:20.900 align:middle line:84%
Yes, the glasses are
quite comfortable.

00:03:20.900 --> 00:03:24.950 align:middle line:84%
There's a little
bit of difficulty

00:03:24.950 --> 00:03:28.250 align:middle line:84%
for people who are not
used to wearing glasses.

00:03:28.250 --> 00:03:31.790 align:middle line:84%
And the glasses
don't necessarily

00:03:31.790 --> 00:03:35.340 align:middle line:84%
capture the extremes
of the visual field.

00:03:35.340 --> 00:03:37.880 align:middle line:84%
So if people are looking
at a very sharp angle down

00:03:37.880 --> 00:03:41.690 align:middle line:84%
at their instrument then you
don't capture that so well.

00:03:41.690 --> 00:03:43.400 align:middle line:90%
Did you get results?

00:03:43.400 --> 00:03:47.030 align:middle line:84%
We did get very
interesting results here.

00:03:47.030 --> 00:03:48.365 align:middle line:90%
I can show you some graphs.

00:03:48.365 --> 00:03:52.720 align:middle line:90%


00:03:52.720 --> 00:03:56.890 align:middle line:84%
These represents the average
pupil size of the musicians

00:03:56.890 --> 00:03:59.950 align:middle line:84%
during different
social conditions

00:03:59.950 --> 00:04:02.470 align:middle line:90%
that we had them performing in.

00:04:02.470 --> 00:04:04.390 align:middle line:84%
All of these
conditions were done

00:04:04.390 --> 00:04:07.420 align:middle line:84%
in a kind of rehearsal-like
situation with no audience.

00:04:07.420 --> 00:04:09.190 align:middle line:84%
And then the last
condition was done

00:04:09.190 --> 00:04:11.420 align:middle line:90%
in a concert with an audience.

00:04:11.420 --> 00:04:16.329 align:middle line:84%
And so you would expect that
the musicians are perhaps

00:04:16.329 --> 00:04:21.490 align:middle line:84%
having more intense attention
during the concert situation.

00:04:21.490 --> 00:04:24.430 align:middle line:84%
And this is, in fact, what
we see for the student group.

00:04:24.430 --> 00:04:28.300 align:middle line:84%
They had elevated pupil
response during the concert

00:04:28.300 --> 00:04:32.320 align:middle line:84%
compared to the other
conditions, whereas the expert

00:04:32.320 --> 00:04:36.340 align:middle line:84%
group did not show
the same result.

00:04:36.340 --> 00:04:38.320 align:middle line:84%
We did see some
differences across

00:04:38.320 --> 00:04:42.100 align:middle line:84%
the different rehearsal
conditions, which

00:04:42.100 --> 00:04:46.373 align:middle line:84%
we kind of arranged in ways that
would disrupt the musicians'

00:04:46.373 --> 00:04:47.290 align:middle line:90%
normal way of playing.

00:04:47.290 --> 00:04:50.200 align:middle line:84%
And so the experts did
respond with their attention

00:04:50.200 --> 00:04:51.830 align:middle line:90%
to these different conditions.

00:04:51.830 --> 00:04:57.010 align:middle line:84%
So you think these manipulations
that you did challenged the two

00:04:57.010 --> 00:04:59.150 align:middle line:90%
quartets very differently?

00:04:59.150 --> 00:04:59.900 align:middle line:90%
I think so.

00:04:59.900 --> 00:05:01.930 align:middle line:84%
I think they responded
in different ways.

00:05:01.930 --> 00:05:05.740 align:middle line:84%
And you would kind
of expect the experts

00:05:05.740 --> 00:05:09.640 align:middle line:84%
to be able to compensate
for these changes

00:05:09.640 --> 00:05:12.550 align:middle line:84%
in a different way
than the student group.

00:05:12.550 --> 00:05:19.810 align:middle line:84%
For example, there might be less
of an obvious overt response

00:05:19.810 --> 00:05:21.820 align:middle line:84%
from the experts,
whereas you still

00:05:21.820 --> 00:05:24.940 align:middle line:84%
see these pupil
changes which are

00:05:24.940 --> 00:05:27.210 align:middle line:90%
reflecting a covert response.

00:05:27.210 --> 00:05:28.600 align:middle line:90%
Right.

00:05:28.600 --> 00:05:31.000 align:middle line:84%
And you think that what
the pupil is telling

00:05:31.000 --> 00:05:36.070 align:middle line:84%
us is the engagement that the
musicians have in the music

00:05:36.070 --> 00:05:39.790 align:middle line:84%
or reflects the effort
they are putting into it?

00:05:39.790 --> 00:05:43.870 align:middle line:84%
Or they are stressed in
a particular situation?

00:05:43.870 --> 00:05:46.930 align:middle line:84%
Well, I kind of expect that it
reflects all of these things

00:05:46.930 --> 00:05:50.680 align:middle line:84%
because all of these
factors require more

00:05:50.680 --> 00:05:54.560 align:middle line:90%
intense focusing of attention.

00:05:54.560 --> 00:05:55.060 align:middle line:90%
Right.

00:05:55.060 --> 00:05:58.920 align:middle line:84%
Place more strong
demands on attention.

00:05:58.920 --> 00:06:01.080 align:middle line:90%
That's wonderful.

00:06:01.080 --> 00:06:02.900 align:middle line:90%
Thank you.

00:06:02.900 --> 00:06:13.000 align:middle line:90%