Gordana Keresteš, Mirjana Tonković, Anita Peti-Stantić

(Psycho)linguistic predictors of subjective age of word acquisition: Are there differences for words acquired in different developmental periods?

Keresteš, G. (1), Tonković, M. (1), Peti-Stantić, A. (2)


1 Department of Psychology, Faculty of Humanities and Social Sciences, University of Zagreb, Croatia

2 Department of South Slavic Languages and Literatures, Faculty of Humanities and Social Sciences, University of Zagreb, Croatia


Learning new words—the basic building blocks of language—is a lifelong process, which is the most rapid in childhood and adolescence. There is abundant research evidence showing that the age at which a certain word is acquired (age of acquisition, AoA) influences its processing in both healthy individuals and people with neuropsychological impairments. In general, words acquired early in life, in comparison to those acquired later, are processed faster, and their processing is less affected by various neuropsychological disorders. In contrast to a large body of research demonstrating an AoA effect in different kinds of cognitive tasks (for a review of studies see Juhasz, 2005), empirical evidence on factors determining the age at which people acquire certain words is scarce, although many theories have been proposed to explain language development, including the process of word learning. Most of the theories on language acquisition have been developed within either psychology or linguistics, but there are also hybrid theories attempting to integrate multiple linguistic and non-linguistic psychological factors. A notable hybrid theory is the Emergentist Coalition Model (ECM, Golinkoff et al., 1994; Hirsh-Pasek et al., 2004). One of the main assumptions of ECM is that factors influencing the acquisition of words change as a child develops. Although all cues for learning a certain word are available at all ages, children use different cues in different developmental periods. This is a result of their relying on different perceptual, motor, cognitive, emotional, and social capacities at different ages. According to the ECM, children themselves construct principles of word learning, and these principles emerge and change with development. Supporting the ECM, Ramey et al. (2013) found, in a corpus of over 1000 English nouns, that different combinations of predictors explained the variance of AoA in three different AoA ranges: early, middle, and late. Following Ramey et al.’s approach, the aim of this study was to examine whether factors predicting AoA differ in three AoA ranges within 6000 Croatian nouns, verbs, adjectives, and adverbs. 

The study is a part of the MEGACRO project (Peti-Stantić et al., 2020), where subjective AoA ratings were collected by using a continuous measure, asking Croatian university students to report the exact age (in years) at which they think they knew the meaning of a given word, even if they did not use it by themselves. Participants completed all of the questionnaires in paper-and-pencil form. The administration of questionnaires took approximately 20 minutes.      

We performed multiple regression analyses with AoA as a criterion variable, and the following five variables available in the Croatian Psycholinguistic Database (CPD) as predictor variables: objective word frequency in the Croatian web corpus hrWaC (logarithmically transformed), word length (number of letters), and subjective ratings of word frequency, concreteness, and imageability. All the predictor variables have been shown to influence word processing in earlier research. We ran separate multiple regression analyses for each of the three AoA ranges: the AoA range containing words acquired in early childhood (up to 6.99 years), the AoA range containing words acquired in middle childhood (7-9.99 years), and the AoA range containing words acquired in late childhood and adolescence (ten years and later). There were 1918, 2534, and 1548 words acquired in the early, middle, and late childhood/adolescence range, respectively. 

As expected, predictors of AoA differed across the three AoA ranges. More specifically, whereas the contribution of word length and concreteness to variance in AoA decreased with development, the contribution of imageability increased. Subjective frequency was the most powerful AoA predictor in all developmental periods, although its predictive power was somewhat stronger in early than in middle and late childhood/adolescence. Objective word frequency did not contribute to explaining AoA variance in any of the developmental periods. The whole regression model explained the largest proportion of AoA variance for words acquired in early childhood (34%); less variance was explained for words acquired in middle (21%) and late childhood/adolescence (25%). Overall, the results support the assumption of the ECM that in the process of learning new words children attend to different cues at different ages, which stem from the specificities of each developmental period. The decreasing importance of concreteness, along with the increasing importance of imageability in learning words from early childhood to adolescence may reflect an overall developmental process of children relying progressively less on the concrete reality and more on mental representations in cognitive processing. The continuous and highest importance of subjective frequency in predicting AoA throughout childhood and adolescence implies that the frequency with which children encounter a given word (or, more precisely, their perception of this frequency) is a key factor in learning its meaning. This finding has important practical implications, suggesting that vocabulary interventions should include multiple encounters with target words.


Golinkoff, R. M., Mervis, C. B., & Hirsh-Pasek, K. (1994). Early object labels: the case for a developmental lexical principles framework. Journal of Child Language, 21(1), 125–155. https://doi.org/10.1017/s0305000900008692

Hirsh-Pasek, K., Golinkoff, R. M., Hennon, E. A., & Maguire, M. J. (2004). Hybrid theories at the frontier of developmental psychology: The Emergentist Coalition Model of word learning as a case in point. In D. G. Hall & S. R. Waxman (Eds.), Weaving a lexicon (pp. 173–204). MIT Press.

Juhasz, B. J. (2005). Age-of-Acquisition effects in word and picture identification. Psychological Bulletin, 131(5), 684–712. https://doi.org/10.1037/0033-2909.131.5.684

Peti-Stantić, A., Anđel, M., Gnjidić, V., Keresteš, G., Ljubešić, N., Masnikosa, I., Tonković, M., Tušek, J., Willer-Gold, J., & Stanojević, M.-M. (2021). The Croatian psycholinguistic database: Estimates for 6000 nouns, verbs, adjectives and adverbs. Behavior Research Methods. https://doi.org/10.3758/s13428-020-01533-x

Ramey, C. H., Chrysikou, E. G., & Reilly, J. (2013). Snapshots of children’s changing biases during language development: Differential weighting of perceptual and linguistic factors predicts noun age of acquisition. Journal of Cognition and Development, 14(4), 573–592. https://doi.org/10.1080/15248372.2012.689386

 16 total views,  2 views today

This post is also available in: hrHrvatski (Croatian)