Caregiver-infant interaction and the still-face

1
Caregiver-infant interaction and the still-face

• Messinger

2
Questions

• What does it mean that interaction is
  bidirectional? How do baby and parent influence
  each other?
• Describe infant-mother dyads A B in detail.
• Do you see mutual influence in their
  interactions?
• How can maternal smiles prompt both infant gazes
  away from mother and infant smiles
• How does infant behavior in face-to-face
  interaction change during the first six months?
• Does the still-face procedure show evidence that
  infants are intentional
• what does the developmental evidence show?
• evidence from modified still-faces?
• What does face-to-face synchrony and still-face
  behavior predict?

3
Two to six months

• Waking time increases
• Crying decreases
• Motor movement become more purposeful
• Infants maintain eye contact
• They increasingly initiate as well as respond to
  caregiver
• A good time to play

4
Social play

• Enjoyable interaction engaged in for its own sake
  
• "the mutual maintenance of attention and arousal
  within a range that facilitates positive
  expressions like smiles and coos
• (Stern, 1974a, p. 404).
• Often occurs during face-to-face interactions in
  the middle-class West
• Prevalence in other groups?

5
Here now

• For the baby, no evidence of reference
• To objects
• To past events
• Emotional expression of what is immediately
  occurring
• Though infants show periods when tend to look
  consecutively at mother for longer (or shorter)
  periods

6
Caregivers role

• Exaggerated vocalizations, facial expressions,
  and movements
• Slowing down and simplification
• Rhythm and repetition
• Matching and attunement
• Variability in this repertoire
• Under- and over-stimulation
• clinical implications

7
Open system

• Caregiver Energy influx
• a continually changing array of sounds, motions,
  facial expressions, tactile and kinesthetic
  events modulating the level nature, timing, and
  patterning of stimulation Stern (1974, p. 407)

8
Infant Actions

• Infant smiles, vocalizations, etc. activities
• Occur for time and more frequently when
• Infants are gazing at their mothers
• And when their mothers are expressive
• Majority of time infants not expressive
• Simply gazing at or away from caregiver

9
Infants role

• Functional control of gazing, smiling, and
  vocalizing are integrated into play
• Infant seeks stimulation that is optimally
  arousing
• Different from earlier homeostatic perspectives

10
A guide to infant actionIf you acted last act
again!
Neural Networks Messinger, Ruvolo, et al., 2010
11
Infants exhibit stable patterns of attention
during interaction
12
Periods of sustained interest

• In both gazing at (and away) from mothers face,
  infants
• tend to follow a longer gaze with a longer gaze
• and a shorter gaze with a shorter gaze
• for up to two consecutive gazes
• Infants have periods of sustained visual interest
  that span various gazes
• So gazes away are not micro-rejections
• Theyre interest in something else

13
Developmental results

• With age
• The duration of infants' gazes at mother's face
  became briefer
• The duration of gazes away from mothers face
  became more lengthy
• Reflects increasing familiarity with mother's
  face and increasing interest in other features of
  the environment

14
Necessity and sufficiency

• Mother smiles are not sufficient to elicit an
  infant smile before six months, but they are
  necessary
• Parents, then, may feel responsible for whether
  or not their infants smile. Symons, 1994 503
• Infant smiles are sufficient to elicit a mother
  smile, but are not necessary.
• Mothers typically smile in response to infant
  smiles and do so within a two second time
  interval
• Malatesta, 1982 143 Van Egeren, 2001 857.
• However, mothers often smile in the absence of an
  infant smile.

15
Moms and dads

• Mothers displayed more positive affect than
  fathers fathers used more physical play.
• Infants were equally positive and negative with
  both parents. Cohn et al.

16
Moms dads patterns with baby

• Infant positive displays with mother build more
  gradually
• Infant affective contour with mother was
  rhythmic
• 1 episode of positive arousal framed by social
  gaze.
• Positive displays appear more suddenly with
  father
• Affective contour with father contained several
  peaks of positive arousal of shorter duration.
• Feldman, 2003 1129.
• Prediction
• Symbolic complexity was comparable and preserved
  the parent-specific contours, with quicker
  latencies, higher frequencies, and shorter
  durations of complex symbolic episodes with
  father.
• Feldman, 2007

17
Mutual regulation (bi-directional)

• both partners reciprocally modify their actions
  based on feedback they receive from their
  partners (Tronick et al., 1978, p. 2).
• This is how infants learn the meaning of their
  own expressive behavior (Tronick et al., 1978,
  p. 1).
• video

18
Early interaction can be bidirectional

• Infant gaze away Mother smile
• A two-way street
• But mothers and infant actions have different
  consequences.
• Mothers respond to infant attention with positive
  displays.
• Infants respond to positive displays by reducing
  attention.

19
In general

• Mothers affectively intensify interaction
• But infants regulate exposure to affective
  intensity.
• Maternal positive expressions increase odds both
  of infant reciprocation and infant turning away
• Developmentally, infants become more likely to
  reciprocate and to elicit

20
Pattern in middle-class western dyads

• Infant gazes at the mother who smiles, makes an
  exaggerated display, and/or vocalizes.
• The infant smiles. The mother smiles or is
  already smiling. The infant gazes away.
• The mother tones things down and waits or tries
  to get the infant's attention.
• Clinical implications

21
Early interaction

• Mother tends to gaze at infant almost
  continuously
• Through two months, infants are transfixed by
  caregivers face
• Infant and caregiver interactions resemble those
  of romantic lovers in the beginning of the
  relationship

22
With increasing age

• The duration of infants' gazes at mother's face
  became briefer
• The duration of gazes away from mothers face
  became more lengthy
• Reflects increasing familiarity with mother's
  face and increasing interest in other features of
  the environment

23
But infants become more discriminating

• And proportion of time when mother is smiling
  that infant gazes at mother does not decline
• Time gazing away from mother replaces time gazing
  at mother when she was not expressive

24
Kaye and Fogel, 1980

• 37 Infants and their mothers
• Face to face play was observed at 6 weeks, 13
  weeks, and 26 weeks
• Variables of interest
• Infant attention
• Mothers facial attention
• Greetings from infant and mother

25
(No Transcript)
26
Infant gaze at mother declines except when
mother is smiling
Percent of session
Inexact depiction of Kaye Fogel, 1978
27
(No Transcript)
28
In Conclusion

• Proportion of time gazing at their mother
  decreases with time as quality of interaction
  increases.
• Mothers behavior to get infants
  attentionPosture-changing Smiles,
  Exaggeration Less facial and isolated
  Bounces expressivenessand Touches
• Greetings mother mother initiated with
  infant reciprocal initiated
  spontaneous greetings communication

29
Infant greetings

• Attends to mother and then opens mouth, smiles,
  vocalizes, or laughs
• Rare at 6 weeks and dependent on mother attending
• More common at 3 months
• By 6 and especially 9 - months, no longer
  dependent on mother attention
• Infant initiates

30
Infant gazing at mothers face and mother smiling
tend to co-occur
31
How does this occur?

• What accounts for large proportion of overlap
  between infant gazing at mother and mother
  expressivity, especially at later ages?
• Who creates this pattern?
• Mother, infant, or both?

32
(No Transcript)
33
(No Transcript)
34
Tronick Individual?Dyad

• Infant Engagement
• Negative
• Passive/Withdrawn
• Protest
• Neutral
• Object
• Social Monitor
•  
• Social Positive
• Social Positive

• Mother Engagement
• Negative
• Disengaged
• --
• Neutral
• Social Monitor
• (No Vocalizing)
•  
• Social Positive
• Social Monitor (Positive Vocalizing)
•  Social Positive

35
Tronick Goal

• Capture cyclic changing engagement patterns of
  dyad
• Assumptions about what phases are more engaged
  than others
• Contrast with behavioral codes with fewer
  assumptions
• That may also be less immediately meaningful

36
Tronicks Dyadic phases

• Initiation
• Mutual orientation
• Greetings
• Play dialogues
• Disengagement

37
Tronick (1989) Emotions and Emotional
Communication in Infants

• Central Argument
• The Affective Communication system
• Is formed first between mother and infant based
  around infant goals
• Which in turn shapes later development
• Therefore its proper functioning will lead to
  better outcomes later in life
• What about?
• Temperament, Enviromental Factors, Health
• Tronick asserts that anything that will
  consistently change infant affect will have an
  effect later on

38
Infant Caretaker Emotions

• Infants and adults take part in an affective
  communication system
• The affect of each infant and mother changes the
  emotional experience and behavior of the other
• Infants have goals, evaluate these goals, and
  change their emotions and behavior based on these
  evaluations
• Evaluation that goal is being accomplished Joy
• Evaluation of failure Anger or Sadness
• The caretaker aids and supplements the
  goal-directed behavior of the infant

39
Infant Caretaker Emotions

• Infant uses regulatory behaviors to actively
  change their affective states
• Other-directed regulatory behaviors
• Self-directed regulatory behaviors
• Infants also respond to others affective states,
  which in turn affects the infants affective
  state
• Therefore, infant emotions are specific and
  meaningful reactions

40
Infant Caretaker Emotions

• Normal affective communication system
• Frequently moves from affectively positive,
  mutually coordinated states to affectively
  negative, miscoordinated states
• BUT these miscoordinated states are repaired
  and returned to coordinated states
• Infant feels effective and caretaker is reliable
• Abnormal affective communication system
• Mostly negative, miscoordinated states with no
  repair to positive, coordinated states
• More likely in depressed mothers
• Infant feels ineffective and caretaker is
  unreliable
• Pathway to psychopathology

41
Tronicks model

• Results of interaction leads to pattern of
  response

42
Knowing what we know now?

• Is it really so simple, even in just the effect
  of a system like ACS?

43
Face-to-face interaction can be bi-directional

• Infants tend to influence parents who sometimes
  influence infants
• Mothers are not simply inserting their actions in
  the pauses between infant actions
• Influencing the probability the partner will
  engage in a particular action
• Not determining that the action will occur.

44
Prenatal cocaine exposure

• Subtle decrements in Social Monitoring
• Not smiling
• Greater exposure more passive withdrawl
• More time when infant is neutral and mother
  negative
• Mothers more negative

45
Dynamic systemsPredicting interactions in time

• Motivated by social robotics

• Applying machine learning to face-to-face
  interaction

46
Social-emotional development

• Smile turn-taking increased with age,
• mean r 0.43, p lt 0.001

Turn-taking
No Turn-taking
Mother Smile
Mother Smile
Infant Smile
47
Bi-directionality in how infants and moms smile
together?

• Joy may be created during mutual smiling
• Through mutual responses/communication during the
  smiling
• Does mom smile harder when baby does?
• Does baby smile harder when mom does?
• Same for mouth opening?
• Does this change with age?

Mom
Baby
48
Affective Valence Non-expert raters

• Person on the street
• 16-20 non-expert (undergrads)
• Separate ratings of infants parents
• Affective valence
• Ratings above the tic mark indicate positive
  emotion (joy, happiness, pleasure) below the
  mark indicate negative emotion (distress,
  sadness, anger).

49
Parent Rating - FACS Concordance
Baker, Haltigan, Messinger, in
press, International J. of Behavioral Development
50
Expert coding non-expert ratingConvergent
validity
Reunion
Face-to-Face
Still-Face
51
Infant Rating FACS Concordance
r .83
r .59
r .84
r .99
r .86
r .80
52
Still-Face and Risk Effects
Face-to-Face
Still-Face
Reunion
Face-to-Face
Still-Face
Reunion
53
What about interactive process?

• Who leads whom?
• No group models directly address this question
• Can we capture these processes with ratings?

54
Infant leads parent
Chow, Haltigan, Messinger, 2010, Emotion
55
Does emotional influence vary in time?
56
Emotional influence itself varies with time
during interactive episodes
Interactive Influence
Face-to-Face
Still-Face
Reunion
57
Variability in time

• Can also be seen in computer vision measurements
  of two dyads face-to-face interactions
• Messinger, et al., 2009

58
Variability at Every Level
59
Whos leading whom?
Windowed Correlations
60
Infant Smile, Mother Smile
61
(No Transcript)
62
Variability in Cross-Correlation
63
(No Transcript)
64
Infant Smile, Mother Smile
65
Tickle/Looming and Gaze
Tickle
Gaze
66
Variability at Every Level
67
Infant Non-Smiling Actions
68
Back and forth

• At 6 and 9 months but not at 3 months, periods of
  engagement originate with the mother's using
  positive affective expressions to try to elicit
  her disengaged infant.
• (2) maternal positive expression precedes the
  onset of infant's positive expression and (3)
  when the infant becomes positive, the mother will
  remain positive until the infant again becomes
  disengaged.
• However, at 9 months, there was a significant
  probability of the infant's becoming positive
  before the mother.
• Cohn, J. F., Tronick, E. Z. (1987). Mother
  infant face-to-face interaction The sequence of
  dyadic states at 3, 6, and 9 months.
  Developmental Psychology, 23(1), 68-77.

69
What does face-to-face teach?

• My partner is responsive to me
• I feel things with my partner
• I take turns with my partner
• Depends on action modality
• Smiles become shared
• Vocalizations not

70
Prediction
71
Through mutual influence, infants come to
understand themselves as social beings who affect
and are affected by others
Mutual Influence

• Infant??Mother
• Infant?Mother and Mother?Infant
• Relation to bi-directional influence, synchrony

72
Participating in a synchronous exchange may
sensitize infants to the emotional resonance and
empathy underlying human relationships across the
life span. Feldman, 2007
73
(No Transcript)
74
Evidence of infant expectations/intentionality

• How do we know if infant really is attempting to
  get interaction back on track?
• Cease interaction
• See if infant attempts to re-commence interaction

75
Face-to-face-still-face

• Face-to-face interaction
• play with Johnny the way you do at home
• Still-face
• Hold a poker face. No interaction of any kind.
• Reunion
• Ok, play with Johnny again.

76
Idealized still-face description

• Infant orients and greets mother
• Gazes at mother and smiles
• But then looks away
• Gazes at mother, even smiles briefly, yet warily
  in less and less convinced attempts to get the
  interaction back on track (p. 8),
• Eventually withdrawing, with body and gaze
  oriented away, giving up.
• Reality Individual differences in expression of
  positive and negative affect

77
Typical results

• Compared to face-to-face, gazing at mother
  declines in still-face and rises to intermediate
  levels in the reunion
• Compared to face-to-face, smiling declines in
  still-face and rises to intermediate levels in
  the reunion
• Compared to face-to-face, negative facial
  expressions rise in the still-face and decline to
  intermediate levels in the reunion

78
Typical results

• Smiles decline w some rebound

• Cry-faces go up, then down a bit

79
The many faces of the Still-Face Paradigm A
review and meta-analysisby Mesman, van
IJzendoorn, Bakermans-Kranenburg

• The Still-Face Paradigm (SFP) first introduced
  by Tronick, Als, Adamson, Wise, and Brazelton
  (1978) to test the hypothesis that infants are
  active contributors to social interaction
• In the original Still-Face Procedure (Tronick et
  al., 1978)
• (1) normal interaction episode
• an interval of 30 s
• (2) the still-face episode
• The general Still-Face procedure
• (1) a baseline episode, the adult performs normal
  interaction
• (2) the still-face episode, the adult becomes
  unresponsive and maintains a neutral face
  expression
• (3) a reunion episode, the adult resumes normal
  interaction
• The Still-Face Effect marked changes in infant
  behavior
• Increased gaze aversion
• Less smiling
• More negative effect

80
The SFP A review and meta-analysisby Mesman,
van IJzendoorn, Bakermans-Kranenburg

• Theoretical explanations for the classic
  Still-Face Effect (Tronick et al., 1978 Gianino
  Tronick, 1988 Tronick Weinberg, 1997
  Tronick, 2005 Tronick etal., 1998 Field, 1994
  Stoller Field, 1982 Fogel, 1982)
• The importance of an adult as a regulator of
  infant arousal who shares meaning and intent with
  the infant such that the infant may use the adult
  as an aid to increasing his self-regulation
  skills.
• The infant is an active contributor to the
  interaction as evidenced by clear signaling to
  elicit optimal responses from the caregiver.
• When failing to do so during the still-face
  episode, the infant is left to regulate its own
  emotions, which is reflected in increases in
  negative affect and gaze aversion as the infant
  has only a limited array of regulatory
  capacities.

81
The SFP A review and meta-analysisby Mesman,
van IJzendoorn, Bakermans-Kranenburg

• The paper includes a narrative review of the
  results of all empirical studies using the SFP
  and a series of meta-analyses
• The narrative review (85 studies)
• clarifying the elements of the still-face effect
• infant physiological responses to the still-face
• stability of infant responses to SFP
• infant characteristics in relation to the
  still-face effect
• maternal behavior in relation to infant responses
  in the SFP
• maternal psychopathology in relation to infant
  responses to SFP
• infant behavior in the SFP in relation to
  attachment
• infant responses to the SFP in relation to other
  behaviors
• adaptations of the SFP to special populations of
  children
• the SFP as a method for other research questions

82
The SFP A review and meta-analysisby Mesman,
van IJzendoorn, Bakermans-Kranenburg

• Meta-analyses
• Hypotheses
• Infant gaze and positive affect decrease from
  baseline to the still-face episode, while
  negative affect increases
• Infant gaze and positive affect increase from the
  still-face episode to the reunion, while negative
  affect decreases
• Infant gaze and positive affect are higher and
  negative affect lower in the baseline episode
  than during the reunion
• Results
• Infant gaze and positive affect decrease from
  baseline to the still-face episode, while
  negative affect increases
• ? in gaze and positive affect
• ? in negative affect and neutral affect
• Decrease in gaze -gt different for age groups
  (stronger for 6 mo. or older)
• Drop in gaze from baseline to still-face -gt less
  strong when touch is allowed
• Decrease in pos. affect -gt stronger when
  interacted with a parent and the baseline was 120
  s or longer

83
The SFP A review and meta-analysisby Mesman,
van IJzendoorn, Bakermans-Kranenburg

• Meta-analyses
• Results (contd)
• Infant gaze and positive affect increase from
  the still-face episode to the reunion, while
  negative affect decreases
• ? in gaze and positive affect
• ? in neutral affect
• No difference in negative affect (absence of
  recovery)
• Increase in gaze -gt different for age groups
  (stronger for 6 mo. or older)
• Infant gaze and positive affect are higher and
  negative affect lower in the baseline episode
  than during the reunion
• ? in positive affect
• ? in negative affect
• No difference in gaze and neutral affect

84
Continuous Ratings Replicate Traditional Effects
and
Note. Mean ratings were divided by 100 for
comparability with correlations.
85
Continuous ratings replicated traditional
still-face effects for both parents and infants
86
Autonomic results

• Heart rate (HR) increases during still-face
• Variability in HR declines
• HR declines slightly from SF to Re-engagement
• Variability goes up
• Autonomic measures returned to baseline in
  re-engagement after still-face
• Unlike behavior
• Where reunion was high in positive affect and
  negative affect

87

• http//www.youtube.com/watch?vapzXGEbZht0
  Tronick FFSF at 15 s

88
Does still-face

• Violates infant expectations of interaction and
  desire for social interaction
• En-face posture says, Hello lets play! while
  still-face says, I will not play!
• So that infants fuss and occasionally smile as
  intentional bids to engage non-responsive parent?

89
Or, does still-face

• Simply withdraw parental scaffolding
• vocal, facial, and gestural support of infant
  behavior
• So infant fusses and withdraws?
• Developmental and experimental evidence

90
Developmental test of still-face

• Expectations of social interaction increase in
  the first 6 months of life
• An increase in intentional bids (e.g., smiling to
  elicit mother) should increase during still-face

91
Individual Differences

• Relatively stability of infant positive and
  negative emotion between the face-to-face and
  reunion
• Interest and gaze scanning stable between the
  still-face and other episodes
• Carter and Weinberg Tronick

92
Stability
Weinberg, Tronick, Cohn, Olson
93
In sum

• Behavioral consistency between face-to-face play
  and still-face is less than consistency between
  face-to-face and re-engagement
• Unclear whether happiness in interaction predicts
  bids to continue during still-face
• This stability may be absent for infant siblings
  of children with autism

94
But

• Infant intentionality image of a future desired
  state and repeated behaviors to attain it may
  be developing, rather than present or absent
• May also be individual differences in
  intentionality and affective reactions to the
  still-face

95
Intentionality

• Image of goal - behavior - reaction
• Infants engage in non-elicited communicative
  behaviors
• They expect reactions
• But at 6 months, there is little evidence that
  they combine all of these features

96
Does still-face

• Violates infant expectations of interaction and
  desire for social interaction
• En-face posture says, Hello lets play! while
  still-face says, I will not play!
• So that infants fuss and occasionally smile as
  intentional bids to engage non-responsive parent?

97
Or, does still-face

• Simply withdraw parental scaffolding
• vocal, facial, and gestural support of infant
  behavior
• So infant fusses and withdraws?
• Developmental and experimental evidence

98
Developmental test of still-face

• Expectations of social interaction increase in
  the first 6 months of life
• An increase in intentional bids (e.g., smiling to
  elicit mother) should increase during still-face
• But only looking away from mom increases during
  still-face, not bidding

99
Longitudinal still-face studies

• Do not show increases in smiling at mother
• Only self-regulatory behaviors increase
• e.g., gazing away, hand-sucking
• Suggests infant is responding to parental
  non-responsivity, lack of interactive support

100
ExperimentalModified still-face test

• Parent gazes over infants shoulder
• Direction of gaze shows parent is not available
• Less contradiction
• No differences in infant behavior between
  traditional and modified still-face
• Also suggests infant is responding to parental
  non-responsivity, lack of interactive support
• Delgado, Messinger, Yale 2002
• But see Markova

101
(No Transcript)
102
What other factors influence when infants gaze
at and away from mother?

• Clinical dimension for caregiver
• Control and micro-rejections
• Nature of infant attention
• Only responsive to here and now stimuli?
• Ahistoric
• Or are there periods of sustained interest?

103
Different agenda?

• Infants spend more time gazing at mother when
  mother is smiling.
• This appears to occur because mothers smile more
  frequently when their infants are gazing at them.
• But infants gaze away from mother more frequently
  when mothers are smiling.
• No difference in infant gazing at mother.

104
Additional readings

• Kaye Fogel
• Weinberg Tronick, 1994 1996 Tronick, et al.
  1978
• Lamb, et al., 1987
• Field
• Bruner, J. S. Sherwood, V. Early rule
  structure The case of peekaboo

105
Interaction during the still-face?

• Rating data
• Correlations with gaze.
• CRS Figures

106
How long do infants gaze at mother?

• Duration of an infant's gaze at mother's face
  positively associated with the durations of
  previous 2 gaze at her mother's face and the
  duration of the gaze at mothers face previous to
  that
• ps lt .001 and .01, respectively

107
When will an infant stop gazing away from mother
gaze at her?

• The duration of infant's gazes away from mother's
  face were positively associated with the duration
  of both the infant's previous gaze away from
  mother's face and the duration of the gaze away
  from mothers face previous to that
• ps lt .001 and .025, respectively.