The costs of emotion perception in social interaction Konstantinos Kafetsios Psychology Dept', Univ'

1
The costs of emotion perception in social
interactionKonstantinos KafetsiosPsychology
Dept., Univ. of CretePaulo N. Lopes, Dept. of
Psychology, University of Surrey
2
Emotion in social interaction

• Social interaction is the ultimate theatre
  where interpersonal phenomena can be checked and
  validated
• (Goffman, 1967 Interaction Ritual Essays on
  face-to-face behaviour )

3
Emotion in interpersonal interaction

• Involved in dynamic interpersonal processes
  (Gottman, 1979 Parkinson, Fischer, Manstead,
  2005)
• Emotion contagion (Hatfield, Cacioppo, Rapson,
  1994),
• Social appraisal processes (Manstead Fischer,
  2001)
• Socio-cognitive processes - congruency effects
  (e.g., Forgas, 1994)

4
Emotion Perception (EP)

• Ability and tendency to attend to others
  emotions
• Emotion perception represents a fragmented domain
  of ability that cannot be assessed fully with a
  single test (Hall Bernieri, 2001).
• No single performance test can reliably measure
  emotion perception ability as a whole since the
  process of perceiving another person involves
  also motivational elements (Ickes et al.,1990
  Klein Hodges, 2001).

5
EP Possible benefits

• Those sensitive in others non-verbal messages
• Report higher satisfaction in their relationships
  (Carton, Kessler, Pape, 1999 Noller, 1980
  Rosenthal, et al., 1979)
• Are evaluated as more socially competent
  (Costanzo Archer, 1989) and emotionally warm
  (Funder Harris, 1986).
• However
• Research has examined those social and relational
  outcomes at the intrapersonal level using global
  self-report outcome measures (e.g., relat.
  Satisfaction, personality reports etc.)

6
EP Possible costs

• High skill at decoding nonverbal cues involving
  channels that are difficult to control, such as
  voice and body, experience more difficulties in
  social relationships than those who are less
  skilled (Blanck, et al. 1981 Rosenthal
  DePaulo, 1979 Rosenthal, et al., 1979).
• In a laboratory study, individuals who were the
  target of eavesdropping by their partners felt
  less rapport with them (Puccineli
  Tickle-Degnen, 2004).
• Research on empathic accuracy also suggests that
  the accuracy with which one can infer one
  partners thoughts and feelings in social
  interaction could undermine close relationships
  if it reveals threatening information (Simpson,
  Ickes et al., 1996).

7
Reasons why emotion perception may undermine the
quality of social interactions

• Eavesdropping on subtle, negative emotional cues,
  or perceiving emotional information that others
  do not wish to disclose, can disrupt social
  interaction (Puccineli Tickle-Degnen, 2004).
• Vigilant traits. excessive preoccupation with
  others and a hyper-vigilance to interpersonal
  cues.
• Self-presentational concerns. Ickes and
  colleagues argued that high self-monitors are
  likely to modulate their affect in response to
  others social cues in an effortful way and
  suffer personally as a consequence (Ickes,
  Holloway, Stinson, Hoodenpyle, 2006).

8
Limitations of global self-reports in assessing
the emotional and social outcomes of EP

• Lack of awareness of how one behaves in social
  interaction (Nisbett Wilson, 1977).
• Relationships between personality characteristics
  and behaviour in close relationships examined
  with global, retrospective self-reports are
  disproved when examined with event sampling
  methods at the level of social interaction
• Self-monitoring) Ickes et al., 2006
• Fearful attachment Kafetsios Nezlek, 2002
  Pietromonaco Feldman-Barrett, 1997).
• Gender stereotypes Barrett, et al.1998 gender
  differences in emotion in global
  self-descriptions are eliminated when emotion
  outcomes are examined in the context of social
  interaction.

9
Overall aims of the studies

• 1. Provide more reliable and valid information on
  the possible costs of EP using an event sampling
  methodology of naturalistic social interactions.
• 2. Examine interactions with the social context
• Halberstadt et al. (2001) question the validity
  of findings on emotion perception that are
  divorced from the social context and there is
  evidence suggesting that interpersonal outcomes
  of EP may depend on contextual influences (see
  Elfenbein Ambady, 2002).
• 3. Identify possible mediators

10
Study 1 Objectives

• Look for evidence for the costs and benefits of
  emotion perception in social interaction
• Test whether these relationships could be
  explained by anxious attachment, a dimension of
  adult attachment orientation that is associated
  with hyper-vigilance to other peoples emotions
  and more negative emotion in social interactions
  (Mikulincer Shaver, 2003 Kafetsios Lydaki,
  2004).

11
Study 1 Procedure

• 62 Participants (20 men), av. age 23.4
• Event sampling methodology (pp RIR) seven days.
• 5 Positive affective states (happy, enthusiastic,
  relaxed, attentive, active)
• 5 Negative affective states (anxious, distressed,
  sad, angry, and rejected)
• Participants described 1314 interactions (M
  20.5, SD 9.7) family members (14.8),
  acquaintances (16.9), friends (14.7), good
  friends (15), best friends (24.4), and partners
  (14.2).
• As a measure of self-perceived closeness/intimacy
  we constructed a 5-point scale (1- acquaintance
  to 5- partner) excluding interactions with family
  members.

12
Study 1 Person Level Measures

• Emotion perception (Appraisal of Others Emotions
  AOE WLEIS Wong Law, 2002 a .84),
• e.g. I always know my friends emotions from
  their behaviour
• I am a good observer of others emotions
• Adult attachment (ECR-R Fraley et al., 2000
  Tsagkarakis, Kafetsios, Stalikas, 2007)
• Avoidance M.?. 2.82, SD .96, a .93
• Preoccupation M.?. 3.12, SD .97, a .93.

13
Multilevel modelling (Bryk Raudenbush 1992)
Level 1 Everyday social interaction yij ?0j
rij
Level 2 Individual ?1j ?00 u1j
14
An unconditional model
yij b0j rij The intercept was then modelled
as varying randomly across individuals at level
2. b0j g00 u0j.

• There was considerable variability at both levels
• positive affect (M 4.12, Level 1 variance
  .69, Level 2 variance .33)
• negative affect (M 1.99, Level 1 variance
  .58, Level 2 variance .37).

15
Level 1 yij b0j b1j (Intimacy) rij
Level 2 b0j g00 g01(AOE) g02(SEA) g03(UOE)
g04(ROE) u0j b1j g10 g11(AOE) g12(SEA)
g13(UOE) g14(ROE) u1j
16
Study 1 Multilevel Relationships between
Perceived Ability to Read Others Emotions and
Affect in Social Interaction
17
(No Transcript)
18
Additional analyses

• The relationship between EP and lower positive
  affect in social encounters remained significant
  controlling for anxious attachment
• There were no gender differences in reports of
  affect in social interactions and in
  relationships between emotion perception and
  affect
• The valence of social encounters did not moderate
  the relationship between AOE and positive or
  negative affect

19
Study 1 main finds

• Individuals reporting high ability or tendency to
  decode others emotions report less positive
  affect in everyday interpersonal interactions
• Preoccupation with others (anxious attachment)
  was not found to be a viable mediator of this
  relationship
• Emotion perception and lower positive affect was
  significant independently of the level of
  negative affect in social encounters
  (Eavesdropping hypothesis?)

20
Aims of study 2

• Replicate the findings of study 1
• Assure that it is interaction level processes
  that are at play and not factors to do with
  individual daily variation in emotionality.
• Check for self-deceptive enhancement
• Examine the possible effects of self-monitoring,
  and attachment anxiety
• Test whether the relational context would
  moderate emotion perception effects on positive
  affect in social interaction. If it did, this
  would provide further evidence for the
  motivational elements involved in the ability and
  tendency to perceive other persons emotions in
  social interactions.

21
Study 2 Procedure

• 72 Participants (38 men), av. age 24.51
• Social interaction level affectivity Event
  sampling methodology (pp RIR) seven days.
• 5 Positive affective states (happy, enthusiastic,
  relaxed, attentive, active)
• 5 Negative affective states (anxious, distressed,
  sad, angry, and rejected)
• Person level end-of day affectivity
• Participants were also instructed to complete the
  Positive and Negative Affect Scale (Watson,
  Clark, Tellegen, 1988) at the end of four
  different days. Alphas (mean of four days) were a
  .87 for positive affect and a .90 for
  negative affect.
• Participants recorded a total of 1772
  interactions (M 22.71, SD 11.86). The
  distribution of types of interactions as reported
  by the participant was similar to that found in
  study 1 family members (13), acquaintances
  (17.7), friends (15.2), good friends (16.6),
  best friends (10.5), and partners (16.9).
• The same procedure for constructing the intimacy
  variable was used as in study 1.

22
Study 2 Person Level Measures

• Emotion perception (WLEIS Wong Law, 2002 a
  .84),
• Adult attachment. (ECR-R Tsagkarakis, Kafetsios,
  Stalikas, 2007)
• Avoidance M.?. 2.82, SD .96, a .92
• Preoccupation M.?. 3.12, SD .97, a .91.
• Self-Monitoring. The Revised Self-Monitoring
  Scale (RSMS Lennox Wolfe, 1984). Ability to
  change self (a .74) Sensitivity to the
  expressive behaviour of others (a .76).
• Self-deceptive enhancement Balanced Inventory of
  Desirable Responding (version 6 Paulhus, 1991
  a? .67).

23
Study 2 Person level affectivity in end of day
reports MRCM models

• AOE as a Level 2 predictor of Level 1 positive
  and negative affect and state anxiety.
• AOE was NOT related with either
• positive (g01 -.05, SE .08, p .52) or
• negative (g01 .04, SE .04, p .46) daily
  affectivity
• or state anxiety (g01 .04, SE .05, p .40)
  at daily level.

24
Study 2 Multilevel Relationships between
Perceived Ability to Read Others Emotions and
Affect in Social Interaction
25
?e??t? 2 ? ??t????? t?? s??a?s??µat?? (?S)
26
Finds of study 2

• We replicated findings of study 1. EP was also
  associated with more negative affect
• We found assurances that it is interaction level
  processes that are at play and NOT factors to do
  with
• individual daily variation in emotionality or
• self-deceptive enhancement
• Anxious attachment preoccupation with others
• Self-monitoring sensitivity was related to
  emotion perception but did not mediate the
  effect.
• The relational context was found to moderate
  emotion perception effects on positive affect in
  social interaction.

27

• EP in self and others
• EP can have personal costs and this is certainly
  at the level of social interaction (effortful
  attending to others reduces positive affect).
• However, EP is generally beneficial for others
• Leaders EP was positively related to several
  work well-being outcomes Kafetsios, Nezlek,
  Vassiou, in press).
• EP and the Social context.
• Evidence of the influence of the social contexts
  in person level variables that have to do with
  attending to other persons emotional displays.
• Recently, there is a resurgence of interest on
  the interaction of social situation with
  personality factors (see Reis, 2008) and our
  results are in keeping with this

28
Limitations

• Extend to less positive interactions
• Assess both interactants
• Combine performance/ability with self-report
  measures of emotion perception

29
Thank you

• Health Psych. MSc Students
• Paraskeui Lydaki, Antwnia Chalambalaki
• UoC Small Research Grant that made the completion
  of the second study possible

30
??? µp??e? ?a ?fe??eta? ?

• ?p???e?st??e ? pe??pt?s? t?? est?as?? st?
  a???t??? s??a?s??µa (Black et al., 1979
  Puccineli , 2004)
• ?p???e?st??e ? pe??pt?s? t?? a???t???? at?µ????
  p??d???es?? e?te se ep?ped? ?a??µe?????
  s??a?s??µ?t?? e?te se ep?ped? ?µµ???? desµ??

31
??? µp??e? ?a ?fe??eta? ??

• ?t?µa p?? ????? t?? t?s? ?a est?????ta? sta
  s??a?s??µata t?? ????? st?? d?ap??s?p???? epaf??
  se s??se?? µe µe?a??te?? s??a?s??µat??? ???µa ??a
  t??? ?d???? ?ataß????? µ????? µe?a??te??
  p??sp??e?a p?? ??e? ??a p??s?p??? ??st?? ??a t???
  ?d???? st? ep?ped? t?? ?et???? s??a?s??µat??
• ??t?pa???s?as? ?es??aße? ? d??stas? t??
  a?t?-pa???s?as??, µ????? de? p???e?ta? ??a µ?a
  e?d?at?µ??? d?ad??as?a st? µ??µ????? ep?ped? a???
  ??a µ?a d?ap??s?p??? d?ad??as?a ?p?? eµp???eta?
  t? ????????? p?a?s?? ?p?? ?a? ? t?s? e??? at?µ??
  ?a ????et?se? est?as? sta s??a?s??µata t?? ?????

32
Self-reported emotion ability (WLEIS)

• Self-report measures of emotion perception have
  the advantage of providing a global judgment that
  combines the different channels of nonverbal
  communication.