Positive and Negative Affective Vocalizations in 2-Year-Olds with ASD

Background:  Research on affect in young children with ASD suggests that they are less socially reactive, smile less, and show more distress reactions than typical peers (Cassel et al., 2011; Zwaigenbaum et al., 2005). Child temperament is particularly important to study because children with difficult temperaments are at increased risk of child abuse and neglect (Fisher, Hodapp, and Dykens, 2008). Plumb and Wetherby (2012) found that infants later diagnosed with ASD produced more non-speech distress vocalizations than developmentally delayed children and typically developing controls. However, the field has yet to conclusively determine whether diagnostic group differences exist in delight and distress sounds (Sheinkopf et al., 2000; Paul et al., 2011; Schoen, Paul, & Chawarska, 2011).

Objectives: Assess whether patterns of non-speech affective vocalization distinguish children in three groups: children who have an older sibling with ASD and are also diagnosed with ASD (ASD); children at high-familial risk for ASD by virtue of having an older sibling with ASD, but not currently diagnosed with ASD themselves (HR-), and children at low-familial risk for ASD with no ASD diagnosis themselves (TDC).

Methods: The Communication and Symbolic Behavior Scales (CSBS; Shumway & Wetherby, 2009) was administered to 33 2-year-olds (ASD: 9, TDC: 12, HR-: 12) participating in the Infant Brain Imaging Study (IBIS; Estes et al., 2015). Two individuals coded child non-speech vocalizations into four categories: (1) delight (laugh), (2) distress (cry, whine, fuss), (3) atypical (squeal, growl, yell, grunt), or (4) other (sound effect, uncodable); (“vegetative” vocalizations like coughs were excluded from the current analyses). For this study, we calculated the percentage of time occupied by distress and delight vocalizations relative to total amount of non-speech vocalization produced by each child.

Results:  A repeated measures ANOVA with group (ASD/TDC/HR-) as a factor and affective vocalization type (delight/distress) as a repeated measure revealed a trend toward an interaction between group and type, F(2,29)=3.01, p=.065, η_p²= .17 (see Figure 1). Planned paired contrasts revealed that the ASD group produced significantly more distress than delight vocalizations (p<.05), the TDC group produced marginally more delight vocalizations than distress vocalizations (p<.08), and the HR- group produced equal amounts of each type (p=n.s.). Independent samples t-tests showed the ASD group produced a significantly higher percentage of distress vocalizations (39%) than the TDC group (14%, p<.05). In contrast, the TDC group produced a significantly higher percentage of delight vocalizations (43%) as compared to the ASD group (12%, p<.05). The HR- group was not significantly different from the other two groups on either variable.

Conclusions:  Emotion regulation difficulties in ASD are common and impairing (Mazefsky et al., 2013) and dysregulation in toddlers is associated with increased parenting stress (Davis & Carter, 2008). Fine-grained identification of distress vs. delight vocalizations in children’s natural communication, demonstrated here, could serve as a way to monitor progress as clinicians and parents help children learn to regulate their emotions. We plan to code distress/delight vocalizations at 6 and 12 months as well, to understand developmental continuity and deviation in this domain (anticipated complete, May 2017).