Transcript slides - Courses

Computer-Mediated
Communication
Media Richness and Visual Interfaces
Coye Cheshire & Andrew Fiore
//
20 February 2013
Projects and Assignment #1
 Assignment 1 is a short 2-3 page description of your group
project idea and the division of labor within the group.
 Due Feb. 27 (next week) at beginning of class
(one assignment per group, 2 printed copies)
 Groups will be signing up for a meeting with us to discuss
the project the following Wednesday (Mar. 6).
 http://courses.ischool.berkeley.edu/i216/s13/assignment1.php
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
1
Theories
of mediated
communication
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
2
Cues Filtered Out
 Social presence:
Lower bandwidth  Less warm,
others seem less like people
 Lack of non-verbal cues —
disinhibition and hostility (e.g., flaming)
unsealedprophecy.wordpress.com
 1:1 mapping between cues and social
functions — missing cues, missing functions
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
3
Social Identity/Deindividuation Theory
(Cues About Us, Not You or Me)
Visual anonymity
 “deindividuation”
 salience to group identity
“Overinterpreting” based on
limited info could lead to
greater social attraction based
on in-group status; stereotyping
of out-group.
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
4
Media richness:
“ formation of shared
[C]apacity to facilitate the
meaning within a given time
interval.
2/20/13
”
— Dennis & Kinney
Cheshire & Fiore — Computer-Mediated Communication
5
A plausible ranking?
Richer
Face-to-face
Synchronous video
Synchronous audio / asynch. video
Synchronous text / asynch. audio
Asynchronous text
Leaner
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
6
Rich
2/20/13
Elements of richness
 Multiplicity of cues (bandwidth)
 Immediacy of feedback
 Use of natural language
 Personal focus
Cheshire & Fiore — Computer-Mediated Communication
7
Lean
2/20/13
Elements of richness
 Multiplicity of cues (bandwidth)
 Immediacy of feedback
 Use of natural language
 Personal focus
Cheshire & Fiore — Computer-Mediated Communication
8
Channels, cues, and signals
Channel conduit for a particular type of info,
Channel:
e.g., for voice or text
Cue
Cue: “any feature of the world, animate or
inanimate, that can be used ... as a guide to
future action” (Donath 2007) —
i.e., informative, not necessarily intentional
Signal a cue meant to indicate an otherwise
Signal:
hidden quality
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
9
Some types of social cues
Verbal
Non-verbal
Beyond FTF?
Textual
 Production cost to encode
meaning equivalent to FTF in text
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
10
Feedback
Convey the receiver’s understanding to sender,
who can adjust accordingly
Type of feedback
Acknowledgment — understanding
Repair — correction or clarification
Proxy — completion
Immediacy of feedback
Concurrent: synchronous nods, mm-hmms
 a.k.a. backchannel
Sequential: brief interjection
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
11
Media choice vs. use
(Cues to Choose By)
What medium would you choose for a given task?
vs. What medium “performs” best?
 Media Richness (the theory) originally examined
media choice and use in organizations.
Claim: Managers should choose medium based
on task to be effective. More ambiguous tasks are
more efficient in richer media.
But when might we want a “less rich” medium?
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
12
Media choice vs. media use
Types of tasks
 “Uncertain” — missing information
 “Equivocal” — ambiguous interpretations
“Best” medium for an (un)equivocal task
 What do managers say they would choose?
 What do they actually choose?
 What yields the best performance?
(And what is “best performance”?)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
13
Dennis & Kinney hypotheses
 H1a: Performance improves as
multiplicity of cues increases …
 H1b: … more for more equivocal tasks.
 H2a: Performance improves as
immediacy of feedback increases …
 H2b: … more for more equivocal tasks.
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
14
Dennis & Kinney experiment
Tasks
 Low-equivocality: SAT-type questions
 High-equivocality: College admissions
Media
2/20/13
Cues: Low
Cues: High
Feedback:
Delayed
Text chat
(turn-based)
Video
(half-duplex)
Feedback:
Immediate
Text chat
(live typing)
Video
(full-duplex)
Cheshire & Fiore — Computer-Mediated Communication
15
Mean decision time
High cues (AV)
Task
Low cues (CMC)
Immed.
Delayed
Immed.
Delayed
Low equiv.
12.21
17.00
26.29
31.53
High equiv.
13.14
14.35
18.71
23.71
Mean decision quality
High cues (AV)
Task
Low cues (CMC)
Immed.
Delayed
Immed.
Delayed
Low equiv.
0.11
0.07
– 0.03
– 0.14
High equiv.
0.32
– 0.03
– 0.19
– 0.05
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
16
— Clark & Brennan (1991)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
17
“The extra facial cues [in the video conditions] do not
appear to be that valuable, unless the decision was actually
harder to make. Facial cues might be more important in a
situation where feeling and emotion were bigger factors. In
such a simple study, it is doubtful that anyone feels
particularly strong about any opinion.”
— Taeil
“[The authors assume] that the same factors can be used to
evaluate the performance of all tasks. Media richness can be
effective for certain tasks but not others. There are tasks that
don't require audio & video. A simple one-line text or email may
be sufficient. Some other tasks, audio & video can play a bigger
role because the task itself requires real-time social interactions
and exchange. The authors did touch on these concepts in the
conclusions. I just wish that more of these were front-loaded and
incorporated into the experiments.”
— Joyce
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
18
Social Information Processing
(Cues Filtered In)
Walther (1992) re-examined early CMC research:
“Given sufficient time and message exchanges for
interpersonal impression formation and relationship
development to accrue, and all other things being equal,
relational [quality] in later periods of CMC and F2F
communication will be the same.”
 Users compensate for attributes of CMC
(e.g., emoticons, timing of messages)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
19
Hyperpersonal communication
(Cues Bent and Twisted)
Contributing factors:
 Sender:
Selective self-presentation
 Receiver: Perceived commonality,
anticipated future interaction
 Channel effects — control, time to consider responses
 Feedback effects — confirmation of idealized perceptions
Bottom line: Potential for hyper-positive (or hypernegative) perceptions in CMC as compared to FTF
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
20
“
The sensorial parsimony of plain text
tends to entice users into engaging their
imaginations to fill in missing details while,
comparatively speaking, the richness of
stimuli in fancy [systems] has an opposite
tendency, pushing users’ imaginations
into a more passive role.
— Curtis (1992)
”
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
21
Walther, Slovacek, & Tidwell 2001
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
22
Social affinity
Long-term, no photos
Short-term, photos
Long-term, photos
Short-term, no photos
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
23
Farnham & Riegelsberger 2004
Photo profiles
Count
Text profiles
Gaming partner preference
(1 = Don’t want to play with, 7 = Want to play with)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
24
“
The study of CMC effects is not best served by
blanket statements about technology main
effects on social, psychological, and
interpersonal processes, nor by proclamations
that online relationships are less rewarding
than FTF ones. Rather, qualities of CMC are …
more often the product of interesting and
predictable interactions of several mutual
influences than main effects of media.
— Walther et al. (2001)
”
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
25
“In general, I found Walther & Parks' discussion to be
useful, especially the discussion of the hyperpersonal
effects of online communication. I like that this perspective
put pressure on the assumption that FtF communication is,
a priori, the epitome of media richness.”
— Kyle
“I wish they had talked more about the tools that are
prescribed for expression and interaction on various CMC
platforms. For instance, when there is a toolbar containing
several different emoticons, it naturally makes for more
interesting social interaction between the users than a
purely text-based interaction. I was wondering what other
kinds of tools are available that enhance the ability to
express oneself in online communication, and how much
these tools affect users' willingness to use a site.”
— Kayal
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
26
Let the dance begin!
2/8/12
Cheshire & Fiore — Computer-Mediated Communication
27
Abstract visual
interfaces
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
28
Social translucence
 Visibility: make social information apparent
 Awareness: knowing based on what you see
 Accountability: knowing that I know you know
 Why? To recreate a “social physics.”
 Why not “social transparency”?
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
29
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
30
“[T]ranslucence … stands in for the notion
that, in the physical world, cues are
differentially propagated through space —
something which, as social creatures, we
understand and make use of in governing
our interactions. Thus, we know that those
across the room may see we are talking,
but will be unable to hear what we say; and
we adjust our interactions to take
advantage of this.”
— Erickson et al.
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
31
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
32
Babble social proxy
“provide cues about the presence and activity of those
in the current conversation”
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
33
“Socially useful ambiguity”
 Pretending to pay attention, e.g., clicking
the Babble proxy to feign attention to the
conversation
 Plausible deniability: consider the fallibility
of cell phones, email/spam filtering, etc. —
tech. limitations, not design decisions, but
the social utility of these devices would
change without them.
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
34
Lecture proxy
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
35
Auction proxy
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
36
Waiting-in-line proxy
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
37
“They discuss how social ambiguity is preserved in their
system by way of their visualization design which seems to
suggest that it is possible to build an online system with
absolutely no social ambiguity but I don't think that's a
realistic idea.”
— Laura
“[A]s the article noted, someone could be pretending to
participate. But what about active listeners? This form of
participation is not capable within social mediating tools,
unless video mediated, given the missing cues of watching
a person's body language, eyes, etc. -- a clear socialtechnical gap.”
— Tine
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
38
Chat Circles 2
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
39
DEMO:
http://granite.ischool.berkeley.edu/cc
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
40
The Chat Circles avatar
 Vaguely humanoid form, but
stylized, not realistic — no faces!
 Words centered in/around the form
— ties words to identity, “face”
 2D location allows proximity
 Size tied to length of utterance
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
41
Temporality and spatiality
 Utterances vanish after a few seconds
 Hearing range: can see only nearby utterances
 What is the real-world effect mimicked here?
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
42
Chat Circles Movement
 Rhythm of conversation:
growing and shrinking circles set the pace
 Proximity:
friendliness, intimacy, or aggression
 Expressivity:
fidgeting, dancing, leading, following,
playing
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
43
Chat Circles Traces
 Movement traces
 Speech traces
 Visual indicator of
social history of
the chat space
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
44
Chat Circles History
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
45
Visiphone
A
B
A
B
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
46
Faces
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
47
What are faces good for?
Conveying, among other things:





Social presence
Individual identity
Social identity
Emotion
Gaze
Source: galante.com
By means of:
 Structure
 Dynamics
 Decorations
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
48
Ekman (1999)
Ekman, Friesen, & Ellsworth (1972)
(and many others)
Basic emotions
Anger
Disgust
Fear
Characteristics of basic emotions
Joy
1. Distinctive universal signals
Sadness
2. Distinctive physiology
Surprise
3. Automatic appraisal
Contempt
4. Distinctive universals in antecedent events
5. Distinctive appearance developmentally
6. Presence in other primates
7. Quick onset
8. Brief duration
9. Unbidden occurrence
10. Distinctive thoughts, memories images
11. Distinctive subjective experience
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
49
Facial muscles
2/20/13
Action units
Cheshire & Fiore — Computer-Mediated Communication
50
Representing the face:
“Being close
may be worse.”
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
51
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
52
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
53
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
54
The Uncanny Valley
http://www.youtube.com/watch?v=FKTAJBQSm10
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
55
“being there” vs.
“beyond being there”
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
56
Designing with faces and bodies
We read meaning in lots of things,
but especially human forms!
 There is no such thing as neutral.
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
57
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
58
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
59
Chernoff faces
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
60
The New York Times and Prof. Steve C. Wang
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
61
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
62
Indeed, the 2007 managerial statistics, as presented in an
annual register published by the baseball analyst Bill James,
are a relatively dull grid of digits. But the facial maps make
comparisons much easier to grasp.
The St. Louis Cardinals’ Tony La Russa, known as a
constant tinkerer, had his National League-leading 150
different batting orders (in 162 games) translate into an
elongated head and wider eyes.
By contrast, the Philadelphia Phillies’ Charlie Manuel — who
said this spring that he used far fewer lineups because he
preferred to “get into a routine and stay with it” — had a much
squatter face and dots for eyes.
— The New York Times
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
63
Faces in
interfaces
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
64
Chit Chat Club
(Karahalios and Dobson)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
65
Chit Chat Club
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
66
Second Life facial expressions
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
67
Second Life expression plug-in
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
68
Facial Expression Analysis
(Mateos: http://dis.um.es/~ginesgm/fip/problems.html#expression)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
69
Rather, the key is a balance between the information
provided and the message that is sent. If minimal
information is provided, a minimal message should be
sent. The problem with many face-based interfaces is that
they are sending too complex a message upon the receipt
of too little data. The face is so highly expressive, and we
are so adept at reading (and reading into) it, that any level
of detail in its rendering is likely to provoke the
interpretation of various social messages; if these
messages are unintentional, the face is arguably hindering
communication more than it is helping. — Donath
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
70
Eyes
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
71
2/20/13 & Kohshima
Cheshire
& Fiore — Computer-Mediated Communication
Kobayashi
2001
72
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
Kobayashi & Kohshima 2001
73
Video chat
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
74
The gaze angle problem, or…
Why so glum?
Source: http://staffx.webstore.ntu.edu.sg/personal/astjcham/Web/Research/percepter.htm
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
75
Source: http://www.freepatentsonline.com/7126627.html
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
76
Yang & Zhang 2004
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
77
Source: D. Nguyen
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
78
Cameras
Projectors
MultiView
Display
2/20/13
Source:
D. NguyenCheshire & Fiore — Computer-Mediated Communication
79
Lag, lip synch, social judgments
 When audio precedes video by 5 video fields, viewers
evaluate people on television more negatively (e.g. less
interesting, more unpleasant, less influential, more
agitated, less successful).
 Audio-video asynchrony has no effect on viewer's
memory for audio information.
 Viewers can accurately tell when a television segment is
in perfect synch, and when it is 5 fields out of synch.
Viewers cannot accurately tell the same segments are 2.5
fields out of synch.
 Even though detection is low when asynchrony is
moderate (2.5 fields), viewer evaluations are still affected.
(Reeves and Voelker 1993)
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
80
“With all the modes at our disposal (picture, video, FaceTime),
the emoticon is still the best transmitter of one's 'chosen'
expression over text. Why use the emoticon, when you can send
someone a picture of you smiling, or looking puzzled, or angry?
The answer, it seems, is that we have collectively "understood"
this medium of communication, and matched a :) to it.
[…]As a frequent user of GoogleHangouts and Skype
conferencing, I find them oddly unsatisfying for reasons I can't
fully explain.” — Jen
“[T]he fact that there are attempts to bring gaze into the
mediated world is bizarre and scary at the same time. If we
can create a space on a virtual platform that allows for gaze
to exist then inter-personal relationships will almost die
because all communication can be simulated over a
medium.” — Maurice
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
81
For next Wednesday…
Visualizations and Visual Interfaces
 Monmonier, M. (1996) Chapters 3 and 10. In How to Lie with Maps.
Chicago, Ill.: University Of Chicago Press.
 Erickson, T. (2003) Designing visualizations of social activity: six
claims. In Extended abstracts of ACM Computer-Human Interaction.
 Donath, J. (2011) Visualizing Conversation.
 Narayan, S., Cheshire, C. (2010) Not too long to read: The tldr
Interface for Exploring and Navigating Large-Scale Discussion
Spaces. In Proceedings of the 43rd Annual Hawaii International
Conference on System Sciences. (HICSS). Computer Society Press.
Remember to write your review!
2/20/13
Cheshire & Fiore — Computer-Mediated Communication
82