Infants preferences for infantdirected speech IDS over adultdirected speech ADS may be due to emotio

1
Influences Of Emotional Valence And Speaking Rate
On Preferences For Infant-directed Speech Across
Age Robin P. Coopera, Christine Kitamurab, Karen
Mattockb, Denis Burnhamb
a Virginia Tech, Blacksburg, VA, USA b MARCS
Auditory Laboratories, University of Western
Sydney, Sydney, AU
Introduction and Purpose
Emotional Categorization and Acoustic Features
8 IDS utterances (4 High
Affect 4 Low Affect) were digitally slowed to 2x
their normal tempo. Average acoustics and
category ratings are presented below

• Infants preferences for infant-directed speech
  (IDS) over adult-directed speech (ADS) may be due
  to emotional valence, with IDS typically being
  more positive in emotional tone than ADS. For
  example
• 6-month-olds showed preferences for high affect
  IDS over low affect IDS even when pitch was
  equated1
• 6-month-old infants showed no preference for IDS
  over ADS when they were matched for affect2
• The emotional valence of speech (and music) is an
  effective modulator of infant attention5. What
  is it about speech that makes it sound more
  emotional?
• Absolunte pitch and pitch variance
• Timbre
• Tempo (or relative duration of syllable lengths)
• With affect controlled, pitch characteristics
  alone can not account for infants preferences
• Tempo also appears to affect infant attention
  young infants preferred slow vs. normal IDS when
  pitch was equivalent, although this tempo
  preference seemed to diminish with age3,4
• It may be that tempo is another factor that
  enhances infants perception of emotion in
  speech, but may only do so in younger infants,
  before infants become more phonemically aware of
  their languages
• We examined 16- and 30-week-olds preferences for
  IDS that varied in tempo (normal vs. slow) and
  affect (low vs. high affect).

Normal
Results Experiments 1, 2, 3
Discussion
Experiment 1 High Affect/Slow vs. High
Affect/Normal (n 18) Age x Tempo interaction
(F(1,34) 7.44, p lt .05)

• For the younger infants
• Preferred High Affect Slow gt High Affect Normal,
  suggesting that tempo adds unique information to
  IDS
• Slower speech may increase younger infants
  perception of affect (even though utterances were
  equated for affect via adult ratings)
• This was particularly clear in Experiment 3 in
  which High Affect/Normal Low Affect/Slow in
  this age group this suggests that other features
  in addition to tempo act in concert to create
  younger infants percept of emotional tone
• In a previous study3, young infants did not
  prefer IDS-Fast over IDS-Normal, even though
  pitch characteristics were the same.
• For the older infants
• Generally attended more to High Affect IDS that
  was normal in tempo this was particularly clear
  in Experiment 3 in which High Affect Normal gt Low
  Affect Slow in this age group.
• As infants begin to process speech phonemically,
  prosodic exaggerations (like tempo) carry less
  perceptual weight, although not entirely
• Older infants continue to prefer IDS gt ADS but
  are becoming more attuned to language-specific
  features of IDS

Methodology
Experiment 2 High Affect/Slow vs. Low
Affect/Slow (n 18) No significant 2-way effects

• Eighteen 16-week-olds and Eighteen 30-week-olds
  in each experiment
• Infants tested in a two-choice preference
  procedure
• 2 10-sec familiarization trials
• 6 20-sec preference trials speech played only
  when infant fixated a visual target on either the
  left or right
• Side x speech counterbalanced across infants

Experiment 3 High Affect/Normal vs. Low
Affect/Slow (n 18) Age x Affect/Tempo
interaction (F(1,34) 5.92, p lt .05)

• Each infant had to accumulate 10 sec (50) visual
  attention overall to remain in analysis
• Primary measure was average look duration to a
  particular side

References
1Kitamura, C., Burnham, D. K. (1998). The
infants response to maternal vocal affect. In
C. Rovee-Collier, L. Lipsitt, H. Hayne (Eds.),
Advances in infancy research, vol. 12, pp.
221-236. Stamford, CT Ablex. 2Singh, Morgan,
Best (2002). Infants listening preferences
Baby talk or happy talk? Infancy, 3,
365-394. 3Cooper, J. S., Cooper, R. P., Ostroff,
W. L., Aslin, R. N. (2004). The temporal
characteristics of infant-directed speech affect
attention in younger but not older infants.
Unpublished manuscript. 4McIlreavy, M., Bhullar,
N., Cooper, R. P. (2004, May). Behavioral and
psychophysiological responses of 1- and
4-month-olds to differing rates of
infant-directed speech. Poster at ICIS, Chicago,
IL. 5Trainor, L., Austin, C., Desjardins, R.
(2000). Is infant-directed speech prosody a
result of the vocal expression of emotion?
Psychological Science, 11, 188-195.
Presented at the 14th Biennial International
Conference on Infant Studies, Chicago, Illinois,
May, 2004