Poster #21 - Maternal Sensitivity Predicts Trajectories of Infant Respiratory Sinus Arrhythmia across Maternal Engagement and Disengagement

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Biological Processes: Psychophysiology

Integrative Statement

Patterns of respiratory sinus arrhythmia (RSA), an index of parasympathetic nervous system (PNS) activation, are crucial in socioemotional development (Porges & Furman, 2011). Polyvagal theory (Porges, 2007) posits that adaptive patterns of RSA are characterized by flexibility, such that, during challenge, infant RSA lowers, to actively attend. When challenge is removed (as in normal social engagement with the mother), infant RSA increases, promoting relaxation.
The Still-Face paradigm (SFP, Tronick et al., 1978) is useful for studying RSA during social engagement with the mother (2-min), interruption by challenge (2-min of maternal disengagement), followed by 2-minute reunion (maternal reengagement). On average, most infants show adaptive RSA patterns (high RSA, low RSA, high RSA); group mean data, however, do not fit half of infants (Bazhenova et al., 2001; Moore & Calkins, 2004).
The present study represents methodological advances over previous research, which have only examined group means for each episode of the SFP and episode-to-episode changes. We used growth mixture models to identify trajectories of RSA across the entire SFP and examine links between trajectories and maternal sensitivity.
Method
198 6-month-old infants and their low-income (35%<$10,000, $10,000>18%<$30,000) mothers (38% African American, 16% White, 24% Multiracial, 13% Latina) participated in a larger study of maternal caregiving. RSA was extracted in 30-second epochs during episodes of the SFP (Porges, 1985). Sensitivity was reliably coded from tapes of normal play and reunion episodes of the SFP. Infant level of distress during the reunion episode was coded by reliable coders.
Results
A growth mixture model was tested with mean RSA for each episode of the SFP as data points. Lower values of fit criteria indicated that the growth mixture model (BIC=1536.47, AIC=1493.73, SBIC=1495.29) was a better fit than the baseline model (BIC=1584.38, AIC=1558.07, SBIC=1559.04), and three classes were specified (Figure 1).
Class 1 (“unchanging”) contained infants whose RSA stayed around baseline levels over the entire course of the SFP. Class 2 (“non-augmenters”) experienced withdrawal of RSA in the shift to the still face episode and and failed to augment RSA upon reunion with their mothers. Class 3 contained too few people to analyze.
Global maternal sensitivity during SFP (sensitivity to distress, nondistress, intrusiveness, and positive regard, ICC = .667) and infant distress at reunion predicted class membership (Table 1). Class 1 (“unchanging”) had mothers who were highest in maternal sensitivity and contained infants who were significantly less distressed at reunion than infants in the other classes. Class 2 (“non-augmenters”) had mothers who were the lowest in maternal sensitivity and contained infants who were significantly more distressed at reunion than infants in Class 1.
Conclusion
Utilizing growth mixture modeling, we found groups of infants who do not fit into the adaptive patterns posited by polyvagal theory (high RSA, low RSA, high RSA). Instead, we found one group that experienced high RSA across the SFP and another group that experienced low RSA across the SFP. These trajectories were related to maternal sensitivity, with lower sensitivity associated with less adaptive patterns of infant vagal regulation.

Authors

• Netta Admoni, Lehigh University
  Presenting Author

• Susan S. Woodhouse, Lehigh University
  Non-Presenting Author