WEBVTT

1
00:00:00.500 --> 00:00:03.500
Digital public displays have become commonplace.

2
00:00:03.500 --> 00:00:08.200
Commercial and academic designers and operators of public screens have sought answers

3
00:00:08.200 --> 00:00:13.500
to the question of how these displays should work in order to drive user engagement.

4
00:00:13.500 --> 00:00:18.500
Research articles in this area predominantly talk about aspects of visual and interaction design,

5
00:00:18.500 --> 00:00:24.000
often disregarding the influence of what information (or content) is shown on the display.

6
00:00:24.000 --> 00:00:26.000
On some level this makes sense:

7
00:00:26.000 --> 00:00:33.500
Researchers aspire to find general design guidelines that can help engage users regardless of the specific purpose of the device.

8
00:00:33.500 --> 00:00:39.000
But this can lead to a "blind spot" if we ignore the content's influence on user engagement altogether.

9
00:00:39.500 --> 00:00:45.500
That is why we decided to survey the research landscape for answers on the inflence of content on public displays.

10
00:00:45.500 --> 00:00:51.500
What can we conclude from existing studies? Are there specific kinds of content that work better than others?

11
00:00:52.000 --> 00:00:57.500
We started with a keyword search in relevant literature databases and aggregating search engines,

12
00:00:57.500 --> 00:01:00.000
going through about 200 potentially relevant results,

13
00:01:00.000 --> 00:01:05.000
and followed promising citations and references to articles that our keyword search had missed.

14
00:01:05.000 --> 00:01:08.500
Many of these results were empirical studies of public displays,

15
00:01:08.500 --> 00:01:11.800
and most of them did not discuss aspects of content at all.

16
00:01:11.800 --> 00:01:16.300
We filtered down the list to 21 articles that made some sort of claim or observation

17
00:01:16.300 --> 00:01:19.300
regarding the impact of public display content.

18
00:01:19.800 --> 00:01:23.300
The main results, each corroborated by more than one study, are:

19
00:01:23.300 --> 00:01:29.700
One, local or locally relevant content can be perceived as more interesting compared to general content.

20
00:01:29.700 --> 00:01:34.000
Two, the same goes for timely relevant and up-to-date content.

21
00:01:34.000 --> 00:01:39.700
And three, user-created content, albeit often overlapping with the previous criteria,

22
00:01:39.700 --> 00:01:44.000
has also been shown in at least two studies to drive increased user engagement.

23
00:01:44.500 --> 00:01:49.000
These results come with some caveats, and there are also studies showing contradictory results.

24
00:01:49.000 --> 00:01:52.000
More details on this can be found in our paper.

25
00:01:52.500 --> 00:01:59.000
Our main conclusion, however, is that aspects of content are underexamined and underspecified in existing research.

26
00:01:59.000 --> 00:02:05.000
We hope to inspire researchers in this area to consider aspects of content more explicitly in future studies,

27
00:02:05.000 --> 00:02:09.000
and to document them so this research gap can be filled.

28
00:02:09.500 --> 00:02:12.000
Because our literature survey was not fully systematic,

29
00:02:12.000 --> 00:02:16.000
but more of a semi-formalized literature probe showing preliminary results,

30
00:02:16.000 --> 00:02:20.000
we have received some push-back on the idea of presenting this work as-is.

31
00:02:20.000 --> 00:02:25.000
Instead, a deeper systematic dive into the literature may uncover more results.

32
00:02:25.000 --> 00:02:27.500
The reason why that has not been done for this study

33
00:02:27.500 --> 00:02:31.000
is that our work was conducted by students in a university course,

34
00:02:31.000 --> 00:02:34.000
where a systematic literature review would have been out of scope.

35
00:02:34.000 --> 00:02:37.500
Because this work is unlikely to be continued by the same team,

36
00:02:37.500 --> 00:02:43.000
we present it for community discussion in its current form and hope that it can nonetheless inspire discussion.

37
00:02:43.500 --> 00:02:46.500
This video has been narrated by Julian Fietkau.

38
00:02:46.500 --> 00:02:52.000
On behalf of my co-authors, Benedikt Sprecher and Thomas Havekost, I thank you for your attention.