Accent Matters: Children are less likely to investigate surprising claims provided by non-native speakers

• In Event: Belief revision in a social world: How other people shape children’s ability to revise beliefs

Abstract

By attending to what other people tell them and asking questions, children can learn about non-observable entities such as the existence of germs or God (Evans, 2001; Harris & Koenig, 2006; Harris et al., 2006; Ronfard, Zambrana, Hermansen, & Kelemen, 2018; Shtulman, 2013). Testimony from other people can also influence children’s learning about entities they could learn about on their own. When such testimony from others is surprising and counterintuitive, children may be especially motivated to seek first-hand information to decide whether to revise their beliefs or not, just as they do upon encountering surprising phenomena (e.g., Bonawitz, van Schijndel, Friel, & Schulz, 2012). Indeed, 6- to 8-year-old children are significantly more likely to explore objects when their intuitions about those objects have been contradicted rather than confirmed (Ronfard, Chen, & Harris, 2018).

What factors determine whether children seek to test a surprising claim? Here, we examine whether the way an adult speaks –whether they speak with a native accent or not – affects how likely a child is to test a claim. We manipulated the accent of the informant because prior research has demonstrated that children, like adults, value information provided by native speakers more than information provided by non-native speakers (Kinzler, Corriveau, & Harris, 2011; McDonald & Ma, 2016). We presented children (N = 144, M = 6.6 years) with five different-sized Russian dolls (that were painted and therefore unrecognizable, Figure 1) and asked them to indicate the heaviest one. All children stated that the biggest doll was the heaviest. We provided all children with counter-intuitive testimony, stating that the smallest doll was the heaviest and that the biggest was the lightest (a false claim). Half of the children heard this claim from a native speaker while half heard this claim from a non-native speaker. Next, all children were given an opportunity to seek empirical evidence (e.g., by picking up the dolls) when the experimenter left the room.

To investigate children’s exploration of the dolls, we coded the number of times children picked up each doll by condition (see Figure 2). Inspection of Figure 2 reveals that children who received testimony from the native-accented speaker explored more than children who received testimony from the foreign-accented speaker (p <.001). Critically these effects of accent proved robust. They were observed across two different accents (Chinese-accented and Ukrainian-accented English) and, in line with Paquette-Smith et al. (2019)’s study on children’s friendship choices, the effects of accent did not vary based on children’s particular accent exposure.

In sum, children’s inferences about the experimenter’s credibility influenced their belief in her claim and whether they chose to test it. This is noteworthy because it provides additional evidence that children do not believe everything they are told (Harris et al., 2018). It also suggests that children’s testing of surprising claims may be driven by a desire to confirm rather than falsify the claim they have been given, as they are more motivated to test a claim delivered by a more credible informant.

Authors

• Samuel Ronfard, University of Toronto
  Presenting Author

• Thomas St. Pierre, University of Toronto
  Non-Presenting Author

• Katherine White, University of Waterloo
  Non-Presenting Author

• Elizabeth K. Johnson, University of Toronto
  Non-Presenting Author