The relationship between infant pointing and language development: A meta-analytic review

TY - JOUR

T1 - The relationship between infant pointing and language development

T2 - A meta-analytic review

AU - Kirk, Elizabeth

AU - Donnelly, Seamus

AU - Furman, Reyhan

AU - Warmington, Meesha

AU - Glanville, Julie

AU - Eggleston, Adam

N1 - © 2022 The Authors Funding Information: This research was supported by a grant from the University of York Pump Priming Fund awarded to Elizabeth Kirk, and by the Australian Research Council (CE140100041). We would like to thank Amy Bidgood, Rechele Brooks, Malinda Carpenter, Cristina Colonnesi, Julie Gros-Louis, Alfonso Igualada, Carina Lüke, Eva Murillo Meredith Rowe, and Catherine Tamis-LeMonda for replying to requests for information and for providing us with additional data, explanation or clarification.

PY - 2022/6

Y1 - 2022/6

N2 - Infant pointing has long been identified as an important precursor and predictor of language development. Infants typically begin to produce index finger pointing around the time of their first birthday and previous research has shown that both the onset and the frequency of pointing can predict aspects of productive and receptive language. The current study used a multivariate meta-analytic approach to estimate the strength of the relationship between infant pointing and language. We identified 30 papers published between 1984 and 2019 that met our stringent inclusion criteria, and 25 studies (comprising 77 effect sizes) with samples ≥10 were analysed. Methodological quality of the studies was assessed to identify potential sources of bias. We found a significant but small overall effect size of r = 0.20. Our findings indicate that the unique contribution of pointing to language development may be less robust than has been previously understood, however our stringent inclusion criteria (as well as our publication bias corrections), means that our data represent a more conservative estimate of the relationship between pointing and language. Moderator analysis showed significant group differences in favour of effect sizes related to language comprehension, non-vocabulary measures of language, pointing assessed after 18 months of age and pointing measured independent of speech. A significant strength of this study is the use of multivariate meta-analysis, which allowed us to utilise all available data to provide a more accurate estimate. We consider the findings in the context of the existing research and discuss the general limitations in this field, including the lack of cultural diversity.

AB - Infant pointing has long been identified as an important precursor and predictor of language development. Infants typically begin to produce index finger pointing around the time of their first birthday and previous research has shown that both the onset and the frequency of pointing can predict aspects of productive and receptive language. The current study used a multivariate meta-analytic approach to estimate the strength of the relationship between infant pointing and language. We identified 30 papers published between 1984 and 2019 that met our stringent inclusion criteria, and 25 studies (comprising 77 effect sizes) with samples ≥10 were analysed. Methodological quality of the studies was assessed to identify potential sources of bias. We found a significant but small overall effect size of r = 0.20. Our findings indicate that the unique contribution of pointing to language development may be less robust than has been previously understood, however our stringent inclusion criteria (as well as our publication bias corrections), means that our data represent a more conservative estimate of the relationship between pointing and language. Moderator analysis showed significant group differences in favour of effect sizes related to language comprehension, non-vocabulary measures of language, pointing assessed after 18 months of age and pointing measured independent of speech. A significant strength of this study is the use of multivariate meta-analysis, which allowed us to utilise all available data to provide a more accurate estimate. We consider the findings in the context of the existing research and discuss the general limitations in this field, including the lack of cultural diversity.

UR - http://www.scopus.com/inward/record.url?scp=85126625022&partnerID=8YFLogxK

U2 - 10.1016/j.dr.2022.101023

DO - 10.1016/j.dr.2022.101023

M3 - Review article

AN - SCOPUS:85126625022

SN - 0273-2297

VL - 64

JO - Developmental Review

JF - Developmental Review

M1 - 101023

ER -