1
00:00:05,960 --> 00:00:08,800
My research focuses on cognitive abilities

2
00:00:08,800 --> 00:00:10,960
and how people differ in those abilities

3
00:00:10,960 --> 00:00:13,880
and what are factors that can affect

4
00:00:13,880 --> 00:00:16,200
these cognitive abilities
and what we can do about it.

5
00:00:16,200 --> 00:00:21,160
So for example, I'm interested in
how everyday life experiences

6
00:00:21,160 --> 00:00:24,320
shape the way we think and in how we can improve

7
00:00:24,320 --> 00:00:28,200
those cognitive abilities, for example,
through training interventions.

8
00:00:28,200 --> 00:00:32,280
So in this particular project,
we were interested in whether we can train

9
00:00:32,280 --> 00:00:36,320
working memory in people
through computer based interventions.

10
00:00:36,320 --> 00:00:37,760
So what working memory is,

11
00:00:37,760 --> 00:00:41,720
that is simultaneous maintenance
and processing of information,

12
00:00:41,720 --> 00:00:42,640
in the current moment.

13
00:00:42,640 --> 00:00:45,280
What we do for these studies
is that people come in

14
00:00:45,280 --> 00:00:49,040
for testing their cognitive abilities
before the training intervention

15
00:00:49,080 --> 00:00:53,200
and then they come back in the lab
after the training intervention.

16
00:00:53,200 --> 00:00:58,280
So in this project, we used open software
to code the training intervention.

17
00:00:58,280 --> 00:01:01,440
The reason for that was that
we wanted this intervention,

18
00:01:01,440 --> 00:01:04,640
if successful, to be free
for anyone to use later

19
00:01:04,640 --> 00:01:09,680
so that we wouldn't have any issues
with proprietary software and so on.

20
00:01:09,680 --> 00:01:11,280
We also shared all of our data

21
00:01:11,280 --> 00:01:16,200
because we had quite a big dataset
from almost 200 participants

22
00:01:16,200 --> 00:01:20,320
who did a very extensive cognitive testing
and these training interventions.

23
00:01:20,320 --> 00:01:21,400
And so we thought, well,

24
00:01:21,400 --> 00:01:24,240
we are interested in very specific
research questions here,

25
00:01:24,240 --> 00:01:28,040
but other people might be much
more interested in other questions

26
00:01:28,040 --> 00:01:32,160
that we didn't even think of, but
that can be addressed with our dataset.

27
00:01:32,160 --> 00:01:35,200
And so we just wanted to make sure
that all the effort

28
00:01:35,200 --> 00:01:39,400
that we put into collecting
these data is really worth it.

29
00:01:39,400 --> 00:01:44,040
One big issue really was that we didn't
plan to share the data from the start.

30
00:01:44,040 --> 00:01:48,640
We kind of started this project back
when this wasn't a big thing yet,

31
00:01:48,640 --> 00:01:51,280
so we didn't really think about

32
00:01:51,280 --> 00:01:56,080
how would we really output the data,
how would we write a code book and so on.

33
00:01:56,080 --> 00:01:58,800
Only after we had all the
data collected already,

34
00:01:58,800 --> 00:02:00,600
then we kind of came up
with the idea that,

35
00:02:00,600 --> 00:02:02,920
‘Oh yeah, we should actually
share this data.’

36
00:02:02,920 --> 00:02:05,040
Just creating that code book,

37
00:02:05,040 --> 00:02:09,360
so that means writing up
all the different variables in the dataset

38
00:02:09,360 --> 00:02:13,720
and describing them to a level of detail
that other people can understand

39
00:02:13,720 --> 00:02:14,880
what you actually did,

40
00:02:14,880 --> 00:02:18,120
that was quite an effort
and that is like just extra time

41
00:02:18,120 --> 00:02:20,840
that is not planned in
for a particular project.

42
00:02:20,840 --> 00:02:25,200
And in this particular case, for example,
the PhD student who did the project,

43
00:02:25,200 --> 00:02:27,840
she was already out of
academia at that point.

44
00:02:27,840 --> 00:02:29,120
She got a great job

45
00:02:29,120 --> 00:02:33,040
as a data scientist and so on,
but there was no time for her to do that.

46
00:02:33,040 --> 00:02:36,920
So this was just like something
that we didn't really think of before.

47
00:02:36,920 --> 00:02:39,880
And so it was all a little bit hectic
and then actually doing it.

48
00:02:39,880 --> 00:02:41,880
In retrospect, I would definitely say

49
00:02:41,880 --> 00:02:44,920
I would have thought
about how to share the data before.

50
00:02:44,920 --> 00:02:48,160
I would have made a proper data
management plan, for example.

51
00:02:48,160 --> 00:02:52,240
I would have also thought more about

52
00:02:52,240 --> 00:02:57,960
the variables that we should maybe also assess

53
00:02:57,960 --> 00:02:59,840
or what kind of different levels
of those variables

54
00:02:59,840 --> 00:03:02,760
we should assess to make it
maybe even more useful to share.

55
00:03:02,760 --> 00:03:06,560
There have definitely been
a range of benefits in sharing these data.

56
00:03:06,600 --> 00:03:09,680
For example, in a teaching context,
I can use these data

57
00:03:09,680 --> 00:03:12,480
for teaching statistics for example. 

58
00:03:12,480 --> 00:03:16,520
There are some researchers
who reused it to look into the reliability

59
00:03:16,560 --> 00:03:19,920
and validity of these particular measures
that we used.

60
00:03:19,920 --> 00:03:24,800
So in my research going forward,
I try to plan in

61
00:03:24,800 --> 00:03:28,320
using or following
FAIR principles from the start.

62
00:03:28,320 --> 00:03:31,120
So now, for example,
a data management plan

63
00:03:31,120 --> 00:03:35,480
is always a starting point
with my PhD students, for example.

64
00:03:35,480 --> 00:03:39,000
Also, I'm trying to factor
in the time that it takes.

65
00:03:39,000 --> 00:03:43,480
And already, while coding my tasks,
for example, that I use

66
00:03:43,480 --> 00:03:46,960
for testing cognitive abilities,
I already think about how to output

67
00:03:46,960 --> 00:03:49,360
those variables so that it
will be understandable

68
00:03:49,360 --> 00:03:51,480
without me having to recode something

69
00:03:51,480 --> 00:03:53,840
after having collected the data.

70
00:03:53,840 --> 00:03:58,240
I would definitely just go for it 
and try sharing your data. 

71
00:03:58,240 --> 00:04:00,640
Start with maybe with a small project

72
00:04:00,640 --> 00:04:04,920
where you know that not much can happen
if you share these data.

73
00:04:04,920 --> 00:04:08,120
So for example, use a data
set that you have already published,

74
00:04:08,120 --> 00:04:12,720
give it a try to create a code book,
upload it to ORDA

75
00:04:12,720 --> 00:04:15,760
or to the Open Science Framework
and see what happens with it,

76
00:04:15,760 --> 00:04:19,600
see whether other people are reusing it
and whether some collaborations

77
00:04:19,600 --> 00:04:21,200
can come out of it as well.

78
00:04:21,200 --> 00:04:24,800
I think one aspect that is
sometimes a little bit overlooked

79
00:04:24,800 --> 00:04:28,960
is that we start sharing data now,
but what is also really important

80
00:04:28,960 --> 00:04:32,720
is to share the methods and materials
that we use.

81
00:04:32,720 --> 00:04:35,920
So for example, the specific software
that we use for

82
00:04:35,920 --> 00:04:39,760
creating the task
that we use in cognitive assessment,

83
00:04:39,760 --> 00:04:43,800
the paradigms that we used, instructions
and the actual stimuli.

84
00:04:43,800 --> 00:04:46,720
So for example, what kind of pictures
are we showing and so on.

85
00:04:46,720 --> 00:04:49,640
Sharing those materials,
that could be really useful

86
00:04:49,640 --> 00:04:54,000
or is very useful, in order to make
our research more reproducible as well.

87
00:04:54,000 --> 00:04:57,080
It can also lead to others recreating

88
00:04:57,080 --> 00:05:01,440
or reusing your tasks and paradigms,
which is good for your research

89
00:05:01,440 --> 00:05:04,240
as it makes it more prolific as well.