About this srcd poster session

---

Panel information
Panel 20. Social Cognition

Abstract

---

Background: The language a person speaks and their accent can reveal aspects of their history (Kinzler, 2020). For example, it may indicate where the person grew up and the larger social groups they belong to. Strikingly, even infants seem to view language as a social cue. Infants as young as 10 months prefer to interact with people who speak the same language or accent as their caregivers (Kinzler et al., 2007). Furthermore, when infants observe others speaking, they expect those who speak the same language to socially engage rather than disengage (Liberman et al., 2018). But what exactly do infants infer about people who speak in a similar language or accent? Some argue that babies prefer those who speak the same language as their caregivers due to an expectation of receiving more information from them (Begus et al., 2016). While some studies challenge this explanation, it remains an open question why toddlers expect those who speak the same language to engage. Building on this prior work, we examine whether toddlers, ages 13 to 18 months, expect those who speak with the same accent to respond to one another’s distress. We used this measure because previous research suggests that infants and toddlers use social intimacy cues, such as saliva sharing (Thomas et al., 2022) and imitation (Kudrnova et al., 2024), to make predictions. We reasoned that if toddlers use accent to infer close relationships, they should make similar predictions when presented with distress. Moreover, we were interested in whether it matters if the matching accents correspond to toddlers' own accents or if they can reason about any two people who share accents.

Methods: Participants were 13-to-18-month-old toddlers (N=33) from monolingual English-speaking homes. We showed toddlers an animated video of three characters conversing. The central character shared an accent with one side character but not the other. In one condition, the shared accent matched that of the toddlers' caregivers (e.g., American-Accented English). In the other condition, the shared accent did not match that of the toddler’s caregivers. Then, the central character was shown in distress (it made a crying noise and frowned). We hypothesized that toddlers would expect the character with the shared accent to respond: to look first and longer at the character who shared an accent with the central speaker.

Results: Contrary to our predictions, we found that toddlers looked first at the character whose accent differed in both conditions (BF10= 90.7), regardless of whether the shared accent was similar to the one the toddler was exposed to in their home environment. While this rules out the possibility that toddlers always look at the character whose accent matches their caregivers, it presents a mystery given the pattern of results from previous studies. Future studies will test whether it matters what the central character does—if so, it will rule out the possibility that toddlers are more interested in the ‘left out’ character. Moreover, it is possible that toddlers in this experiment looked at the source of the distress.

Author information

Author	Role
Denisse Lopez Flores, Harvard	Presenting author
Christina M. Steele, Harvard University	Non-presenting author
Ashley J. Thomas, Harvard University	Non-presenting author

⇦ Back to session