What you say and how you say it: Socioeconomic Differences in Parent Mental-state Talk

• In Event: Poster Session 14
  In Poster Session: PS 14 Topic - Social Cognition

Abstract

Parents’ use of mental-state terms (e.g., want, think, know) is positively related to a variety of developmental outcomes in their children, including mental-state vocabulary, emotion understanding, and understanding others’ beliefs (Devine & Hughes, 2018; Taumoepeau & Ruffman, 2016). Research has also sought to characterize aspects of parent mental-state talk such as how frequently parents use different types of mental-state terms or whose mental state they talk about (Taumoepeau & Ruffman, 2008).

However, the majority of research on parent mental-state talk has focused on families of higher socioeconomic status (SES). Little is known about the nature of parent mental-state talk in lower SES families or whether this talk varies by SES. Given evidence that lower SES parents produce less child-directed speech (Hoff, 2003), SES might impact parent mental-state talk as well.

A second question is whether the standard approach to measuring mental-state talk – counting the number mental-state labels – fully captures the ways that parents convey mental states. Research on co-narration suggests that lower SES families often use other rhetorical devices such as dramatic language and reported speech to convey mental-state information (Miller, Cho, & Bracey, 2005). This suggests that counting mental-state labels might underestimate how much parents talk about mental states.

The present study sought to address these gaps in the literature by examining parent mental-state talk in a socioeconomically diverse sample using a novel coding approach that included both the number of mental-state labels and other rhetorical devices that could convey mental state information. Participants were 48 parent-child dyads; children’s age ranged from 27.0 to 39.1 months (M = 32.1 months). SES was measured via the highest level of education reported by either parent (Table 1). Parent-child dyads viewed a picture-book intended to evoke mental-state talk (Taumoepeau & Ruffman, 2006). Parents were instructed to talk about the pictures like they would at home.

Interactions were transcribed and parent mental-state talk was coded in two ways. First, we coded three types of direct mental-state labels: emotion, desire, and cognition terms (Ensor & Hughes, 2008). Second, we coded indirect ways parents could convey mental states: words that strongly implied a mental state without labeling it directly (i.e. yucky), words that described actions driven by mental states (i.e. “The baby is crying.”), and reported speech (“That boy says, ‘I am so tired.’”).

Controlling for the total number of parent utterances, there was a significant effect of parent education on parents’ use of cognition labels, F(3, 39) = 5.26, p = .004, with higher SES parents using more cognition labels (Table 2). No differences were found for emotion or desire labels. Results also revealed an effect of parent education on the proportion of mental-state talk that was indirect references rather than direct labels, F(3, 44) = 2.93, p = .044. Lower SES parents used a higher proportion of indirect references to mental states.

Together, these findings suggest that mental-state talk might be both quantitatively and qualitatively different in lower SES families. Implications for the relationship between mental-state talk and socio-cognitive development will be discussed.

Authors

• James Z Sullivan, University of California - Merced
  Presenting Author

• Gabriel Nguyentran, Villanova University
  Non-Presenting Author

• Rose M Scott, University of California - Merced
  Non-Presenting Author