Social Presence

1
(No Transcript)
2
Social Presence

• Dr. Frank Biocca
• Chad Harms

3
Social Presence Tutorial Section Outline

• Social nature of networked computing
• Rise of social presence technologies
• Brain basis of social presence
• Model of social presence
• Social responses to computers and agents
• Measurement of social presence

4
Social Computing

• Outline Slide-Hide

5
The internet is a social place
6
People use networks for social communication

• Social interaction leading motivation to use the
  network
• 90 of individuals on-line use e-mail
• 84 of Internet users (90 million) used it to
  contact and get information on groups
• 50 said it helped them get to know people theyd
  otherwise not have met
• 40 said it helped them get more involved with
  groups to which they already belong

7
People use networks for social communication

• Social interaction application are among the
  fastest growing
• E-mailing friends rose from 42(95) to 68(98)
  to 82(00)
• 143 million Americans (54 of the population)
  used the Internet in September 2001 (a 26 jump
  since August 2000)

8
Networking increasing need for higher social
presence

• Future growth in mobile technologies
• Worldwide mobile phone sales between 1996 and
  2000 experienced a compound annual growth rate of
  close to 60 percent (Gartner)
• Mobile Wireless Internet users are expected to
  expand 18-fold, from 39 to 729 million worldwide
  between 2000 and 2005 (Intermarket)

Courtesy Eyematic
9
Networking increasing need for higher social
presence

• Teleconferencing telecollaboration increased
  following terrorist acts.
• American suppliers reported an increase of 140
  in videoconferencing bookings in fall 2001.
  Videoconferencing stocks surged.
• Overseas, British Telecomm reported an increase
  of 85 in videoconferencing and 30 in audio
  conferencing.

10
DEFINITIONS

• Outline Slide-Hide

11
What is social presence?
12
Definitions of social presence

• Co-presence sense of being in the same space
  with another
• Social presence sense of access to another
  intelligence-mind

13
APPLICATIONS

• Outline Slide-Hide

14
Social presence technologies

• Applications where the feeling of social
  presence is a goal.

15
Hey, Im working with you Collaborative
virtual environments
Multiple individuals collaborating via a virtual
environment
Coutesy, Amela Sadagic, National Tele-immersive
Initiative
16
Interactions with masked others.Communication
with Different Avatar Embodiments
17
Interaction with Embodied Agents
Courtesy Eyematic
18
Networked gaming environments
19
QUESTIONSSocial presence

• Outline Slide-Hide

20
Basic problem of social presence research and
design
21
Basic problem of social presence research and
design
22
Key research and design questions

• What properties of the interface increase or
  decrease social presence?
• What are the best ways to measure social
  presence?
• Do users respond in similar social ways to
  computers or agents as they do to physical or
  mediated humans?
• Does increased social presence improve users
  attitude towards the interface, application, web
  site, product?
• Does the level of social presence improve the
  performance of users in information and
  communication tasks?

23
BRAIN BASIS/THEORY OF MIND

• Outline Slide-Hide

24
The brain basis for responding socially to
computers and mediated representations

• A theory of mind mechanism?

25
Evidence for innate mind simulation mechanism

• Baby mimicry of facial expressions.
• Early distinctions between animate/inanimate
• Over reading of intentions from behaviors.
• Evidence of mind reading deficits in autistic
  individuals.

26
Users may simulate the minds of other
characters?The Intentional Stance
Other Mind (Human, Artificial, other)
Self
Simulation of other
27
CAUSES/CORRELATES

• Outline Slide-Hide

28
What causes users to feel social presence and
what are the benefits and effects of social
presence?

• Causes and correlates of social presence

29
A Model of the Social Presence in Networked
Virtual Environments
Causes
Correlates(Effects)
Social Presence
30
A Model of the Social Presence in Networked
Virtual Environments
Network Bandwidth
Interface
Correlates(Effects)
Social Presence
Social Relationship and Task Context
31
A Model of the Social Presence in Networked
Virtual Environments
Network Bandwidth
Interface
I/O devices
Social CuesPresence, absence, distortion
Social Presence
Virtual Body Morphology
Virtual BodyBehavior
Social Relationship and Task Context
32
Technological causes that may affect mediated
interaction and social presence

• Reduced range of sensory social cues
• Reduced sensory information
• E.g., audio only teleconferencing and loss of
  non-verbal turn taking cues
• Amplification of social cues
• Cues not within sensory range in physical
  environment are sensate in virtual environments.
• E.g., sensors make heart rate and
  skin-conductance (i.e. sweating) visible to
  another user.

33
Technological causes that may affect mediated
interaction and social presence

• Distortions of sensory social cues
• Change in body morphology
• E.g., body shape is non-human and non-responsive.
• Distortion of the social context of the
  interaction
• Cultural variation in interactive behavior and
  cue reading
• Inappropriate juxtaposition of information

34
A Model of the Social Presence in Networked
Virtual Environments
Causes
35
A Model of the Social Presence in Networked
Virtual Environments
Cognitive Correlates
Behavioral Correlates
36
Social Presence
Behavioral Correlates
37
Cognitive Correlates
Social Presence
Behavioral Correlates
38
Media EquationHypothesis

• Outline Slide-Hide

39
Do users react socially to computers, agents,
avatars, and real people in the same way?

• Ways in which social interaction in physical and
  virtual environments are the same

40
Example Users being polite to computers.

• Computer presents itself socially as a tutor.
  Praises itself.
• User asked to rate its performance.
• Users rated it higher if questionnaire is on same
  computer rated it lower if rated on another
  computer. (Reeves and Nass)
• As in similar human experiments, participants
  avoid offending the computer!

41
Example Social anxiety and public speaking with
a audience of agents

• Individuals are affected by the behavior of a
  virtual audience
• A negative audience provoked an anxiety response
  in individuals, not found in positive or neutral
  audience
• Perceived confidence in public speaking was lower
  using HMD (greater immersion) than a desktop
  system (Pertaub, slator Barker)

42
Do users feel equally present with an agent or
avatar?

• Worked on a desert survival task in a 3D chat
  room.
• Told their partner was human avatar or
  artificial-agent.
• No difference in co-presence.
• No difference in social presence
• Felt similarity of opinions-emotions whether
  avatar or agent.

(Nowak Biocca, submitted)
43
Other studies showing social response equivalence
of people, avatars, agents and computers.

• Gender stereotypes Users believe female AI
  agents more authoritative in matters of love
  than male agents.
• Personality Tend to choose agents with similar
  personalities.
• Social grouping Rate computers on their team
  better than other computers.
• Specialization Specialist TVs and computers
  rather better than generalist TVs and
  computers.

44
Conclusions from the research

• Brain mechanisms may be wired to automatically
  respond to human forms and intentional agents
  regardless of medium.
• People appear to respond automatically and form
  similar mental models for physical people,
  avatars, agents, and computer interfaces (when
  responding socially).
• Users make social judgments and engage in similar
  social behavior with people, agents, and
  computers.

45
Effects of Interface Form on Social Judgment and
PresenceInterface Formon Social Presence

• Outline Slide-Hide

46
Ways in which interface features affect social
presence and social judgments
47
Ways in which the interface affects social
presence and social judgments.

• The interface is part of the users body or the
  others body.
• Users may infer traits or states of the virtual
  other from unintended technological features.

48
Example Effect of audio-video asynchrony on
social judgments.

• General bias to attribute phenomena to others
  traits than to situational context.
• If audio and video are out of sync speaker may be
  judged as less competent.
• Audio precedence may be more psychologically
  disturbing
• Causes misattributions and evaluations that
  speakers are less interesting, pleasant,
  influential, and more agitated.

Time 2
Time 1
Audio
Video
49
Effects of Interface Form on Social Judgment and
PresenceInterface Formon Social Presence

• Outline Slide-Hide

50
Ways in which virtual body morphology and
behavior affect communication behavior, social
presence, and social judgment.

• Ways in which social interaction in physical and
  virtual environments is the same

51
Ways in which the body of the computer, avatar,
or agent may affect communication behavior.

• Adding emotion
• Managing attention
• Eliciting or suppressing communication behaviors
• Affecting approach-avoidance and boundaries
  between bodies

52
If the environment provides no body or identity
for the other, users may simulate to fill it
in. Evidence that the default other is a
version of ourselves.
Administrator Frank Body image graphic

• Simulation theory of mind predicts that the
  other is always modeled on the self.
• Evidence suggests that in the absence of
  information, social responses to no body 3D
  environments are similar to full anthropomorphic
  body.
• In the absence of disconfirming information,
  others we communicate with in mediated
  environments may be perceived as more similar in
  attitudes than is actually the case.

53
Visual fidelity of an agent may not significantly
affect social responses as long as the key social
cues of expression are present.
Administrator Add detail.

• Selected visual fidelity may be adequate for most
  social responses.
• Visual fidelity (anthropomorphism) had no impact
  on approach or avoidance.

(Nowak Biocca, submitted)
54
E.g., effects of asymmetrical interfaces on
conversation and task control

• Users may often interact in virtual environments
  via interfaces that vary greatly in level of
  immersion and control.
• Users connected to interfaces that convey more
  social cues may have greater social presence.
• Some evidence that users with better immersion
  and control may better influence tasks and be
  perceived as more competent.

55
Generalizations from the research on virtual
bodies

• The interface may be perceived as part of the
  users body or the body of the other.
• Representations of the body have significant
  effects on social judgments.
• Selective fidelity may dominate Fidelity for
  cues that convey expressive states may be more
  important than other aspects of fidelity.
• Social stereotyping based on bodily cues may
  affect social judgments even when the user knows
  the virtual body may not be faithful to the real
  body or when the body is that of an agent.

56
Measurement of Social Presence

• Outline Slide-Hide

57
Ways to measure and assess the social presence of
users
58
Why measure social presence?

• Compare alternative telecommunication channels
  or interfaces
• E.g., audio versus video teleconferencing
• Evaluate an interface feature
• E.g., adding AI sales agents to e-commerce site
• Evaluate appropriateness of communication
  interface for a kind
• E.g., deciding what kind of technology may be
  appropriate for social tasks like instruction,
  negotiation, etc.

59
Types of measures

• Self-report indicators
• Behavioral indicators
• Psychophysiological indicators

60
Self-report scales and measures

• Social presence scale (Short, Williams
  Christie, 1976)
• Oldest and most used.
• Potential limitations from direct evaluation of
  medium.
• Networked Minds Social Presence Battery (M.I.N.D.
  Labs, 2002)
• Set of self-report scales and behavioral
  indicators.
• Designed to allow for cross-media and interface
  component comparisons.
• Being tested for reliability and validity.
• Other measures

61
NetMinds Social Presence Inventory
Sub-dimensions
Orders
3. PerceivedSymmetry
Subjective Symmetry
Inter-subjective Symmetry
Behavioral Interdependence
2. Perceived Psycho-behavioral Accessibility
Perceived Comprehension
Social Presence
Emotional Contagion
Attentional Engagement
1. Co-Presence
62
Behavioral indicators

• Behavioral mimicry
• Facial responses (reciprocity of emotion)
• Kinesic responses (proximity orientation)
• Eye movement (awareness, attention)
• Nonverbal conversational regulative cues

63
Psychophysiological indicators

• Many measures of physiological states related to
  social phenomenon.
• fMRI measures can detect perceptions of
  difference between animate and inanimate.
• There may not be an unambiguous indicator of
  level of social presence because of the variable
  physiological responses to different types of
  social interaction (e.g., affect and arousal)

64
Diagnostic measures of social presence for a
collaborative team or group.

• Getting social with
• Social Presence

65
Subjective Self-other Social Presence Symmetry
My view of Self
My view ofOther
correlation
66
Technique to assess self-other social presence
symmetry and asymmetry
Self
Other
Behavioral Interdependence
Behavioral Interdependence
Correlation
Perceived Comprehension
Perceived Comprehension
2. Psycho-behavioral accessibility
2. Psycho-behavioral accessibility
Emotional Contagion
Emotional Contagion
Correlation
Attentional Engagement
Attentional Engagement
1. Co-Presence
1. Co-Presence
Correlation
67
Other
correlation
correlation
Self
SubjectiveSymmetry
Other
Self
68
correlation
Self
Other
IntersubjectiveSymmetry
Other
Self
69
SUMMARY SLIDE

• Outline Slide-Hide

70
Summary

• Social communication is an increasingly important
  part of networked virtual environments.
• Social presence is a goal of many networked
  technologies.
• Interface shapes interaction with and perception
  of agents and avatars.
• Humans may respond socially in similar ways to
  virtual humans, be they AI agents or real human
  avatars.
• Social presence measures may allow designers to
  assess the social performance of interfaces,
  collaborative environments, and agent-based
  environments.

71
Highly Recommended Readings

• Biocca, F., C. Harms, J. Burgoon (submitted).
  "Criteria for a theory and measure of social
  presence." Presence.
• Biocca, F. C. Harms. Guide to Networked Minds
  Social Presence Inventory. www.mindlab.org.
• Reeves, B. and C. Nass (1996). The media
  equation How people treat computers, televison,
  and new media like real people and places.
  Cambridge, Cambridge University Press.
• Short, J., E. Williams, B. Christie (1976). The
  social psychology of telecommunications. London.,
  John Wiley Sons, Ltd.

72
Recommended Readings

• Biocca, F. (1997). "The cyborg's dilemma
  progressive embodiment in virtual environments."
  Journal of Computer-Mediated Communication 3(2)
• Blascovich, J. (2001). Social influences within
  immersive virtual environments. The social life
  of avatars. R. Schoeder, Springer-Verlag
  127-145.
• de Greef, P. and W. IJsselsteijn (2000). Social
  presence in the photoshare tele-application.
  Presence 2000, Techniek Museum, Delft, The
  Netherlands.
• Nowak, K. and F. Biocca (submitted). "The effect
  of the agency and body anthropomorphism of
  virtual humans on users' sense of presence,
  copresence, and social presence." Presence.
• Pertaub, D. P., M. Slater, et al. (2002). "An
  experiment on public speaking anxiety in response
  to three different types of virtual audiences."
  Presence 11(1) 68-78.