Language Development: A Literature Review

Arabella Williams

As empirical examinations of language development have increased, a number of developmental trends have emerged. In particular, psycholinguistic research has revealed several salient facets of language development, including insight into the qualities and influences of parental input, linguistic principles (e.g., iconicity, concreteness, etc.), adult-directed speech (ADS), child-directed speech (CDS), children’s media, and picture books. These findings are essential in accurately characterizing how language comprehension and production function in childhood and beyond. Moreover, these phenomena can be best understood when examined under the framework of a perceptual model, namely the Fuzzy Logical Model of Perception (FLMP). 

​

The FLMP models “perceptual cognitive performance” and intends to explain how the human brain recognizes patterns (Massaro, 2017a, p. 1). It includes three sequential and intersecting processing stages: the evaluation stage, the integration stage, and the decision stage (Massaro, 2012). These stages outline the process by which the human brain automatically gathers and combines sensory information in order to come to the conclusion best supported by the evidence (Massaro, 2012). FLMP posits that this process of sensory integration occurs both in recognizing objects and perceiving speech, suggesting that these two phenomena are not actually specialized processes as some theorists have previously claimed (Massaro, 2012). Rather than language being perceived via a specialized process, FLMP contends that language processing is better represented as a form of pattern recognition (Massaro, 2012). It also assumes that new vocabulary is acquired by way of  “contextual or stimulus learning, symbolic or response learning, and the association between the context and symbol” (Massaro, 2017a, p. 1).

​

This model was created with the intention of remedying the faults evident in previous theories of speech perception. In particular, motor theory, a hypothesis based on the idea that motor processes underlie language learning. This theory has been falsified by numerous clinical findings, namely Broca’s aphasia (Massaro, 2015b). Rejection of motor theory is further supported by assertions that speech comprehension generally precedes speech production, as evidenced by observations of considerable complexity in pre-verbal and early-verbal childrens’ understandings of speech (Massaro & Rowe, 2015). If children can understand speech but not produce it, it is thus unlikely that motor abilities are responsible for language development (Massaro, 2015b). Representation theory is also challenged by this assertion, as the theory is based upon the idea that comprehension and production have the same internal representations (Massaro & Rowe, 2015). Theorists have since argued that speech perception is influenced by a plethora of factors, not just motor movements (Massaro, 2015b). Some of these factors include: neurological physiology/function, pattern recognition, psychophysics, as well as top-down and bottom-up information (Massaro, 2015b). The FLMP integrates these various ideas, providing the comprehensive representation of language input lacking in previous theories (Massaro, 2015b). 

​

Of the many findings that have been meaningfully modeled by FLMP, one of the most significant has been parental input. In Massaro’s (2016) study, parental input was examined as a potential contributor to vocabulary acquisition, among factors such as articulation difficulty, iconicity, imageability, and lexical categorization. Previous findings have suggested that “many of the words children know are those that they have heard, especially in child directed speech” (p. 1). Massaro operationalized parental input as parental input frequency, or how often various words are used in CDS. By comparing these components with word frequencies in children's speech, Massaro discovered that parental input had the biggest impact by far. While the influence of the other factors decreased with age, parental input was steadily influential throughout development. In other words, what parents say to their children is immensely significant in how children acquire language, even more so than other factors. 

​

While not as impactful as parental input, the linguistic principles of iconicity, imageability, concreteness, and articulation difficulty appear to be important influences on language perception and acquisition. Iconicity, or “the motivated correspondence between word form and meaning”, can be found readily throughout the English language (Massaro & Perlman, 2017, p. 1). Words like “moo”, “slosh”, “buzz”, and “teeny” are highly iconic, and appear to be learned earlier in development than non-iconic words (Massaro & Perlman, 2017; Perry et al., 2017). Iconic words also tend to be exhibited more frequently in children’s speech and child-directed speech, suggesting that “iconicity scaffolds the production and comprehension of spoken language during early development” (Perry et al., 2017, p. 1). 

​

Closely related to iconicity is the concept of funniness, with many iconic words also being considered funny, like ‘zigzag’ and ‘waddle’ (Dingemanse & Thompson, 2020). Linguistic analysis by Dinemanse and Thompson (2020) has elucidated certain aspects of the relationship between iconicity and funniness, and has distinguished the characteristics of words that are rated as high in iconicity and funniness, high in iconicity and low in funniness, and low in iconicity and high in funniness. High-iconicity high-funniness words in their sample featured structural markedness, particularly “complex onsets, complex codas, and the verbal suffix -le” that convey playfulness and performativity (p. 217). High-iconicity low-funniness words frequently depicted events with negative associations and vivid imagery (e.g., ‘crash’ and ‘slash’), while low-iconicity high-funniness words like ‘buttocks’ exhibited associations with “taboos and socio-semantic categories” (p. 217). Interestingly, Dingemanse and Thompson found that word frequency was negatively correlated with funniness. However, they discovered that log letter frequency was actually a better predictor of both funniness and iconicity than word frequency.

​

Research has also established that words rated higher in imageability (e.g., how well a word can be imagined) and concreteness (e.g., how well a word can be experienced via one’s senses) are easier to learn (Massaro, 2016). Words that are easier to articulate may also be easier to learn, while words that are more difficult to articulate may be harder to learn (Massaro, 2016). It is important to note that iconicity, concreteness, and articulation difficulty are all negatively correlated with parental input, suggesting that parents use primarily non-concrete, non-iconic, and easier to articulate words (Massaro, 2017a). Iconicity, concreteness, and articulation difficulty do appear to facilitate early language acquisition; however, their influence decreases significantly throughout development (Massaro & Perlman, 2017; Massaro, 2016). As children age, the impact of concepts such as articulation difficulty, iconicity, and imageability decreases linearly (Massaro, 2016). For iconicity in particular, Perry et al. (2017) found that in their sample, “93% of all children… decreased their use of words rated high in iconicity over time” (p. 4). 

​

More insight can be gained on these findings with the use of the FLMP. As mentioned previously, vocabulary acquisition under the FLMP is enabled by contextual information, symbolic information, and context-symbol association (Massaro, 2017a). Under this model, articulation difficulty is related to symbol learning, concreteness is linked to context learning, iconicity is associated with context-symbol association learning, and parental input influences all three of these processes (Massaro, 2017a). Essentially, “word input initiates learning” while concreteness, iconicity, and articulation difficulty selectively influence vocabulary acquisition via context, association, and symbol learning (Massaro, 2017a, p. 2). 

​

While children’s word input comes primarily from their parents, children are often exposed to alternative forms of input, such as from picture books and other children’s media. Comprehensive conclusions on how these mediums differ from parental input or CDS are still missing, but researchers have established some important findings regarding children’s media. Massaro (2015a) discovered that the language in children’s books is more complex and featured a more extensive vocabulary than child-directed speech and adult-directed speech. When picture books were compared to a database of CDS utterances, analysis revealed that picture books actually had 3x more rare words than CDS. Massaro (2017b) also discovered differences in the reading level of picture books, CDS, and ADS. According to several measures of reading grade level, Massaro found that picture books had the highest average grade level (4.2) compared to child-directed speech (1.9) and adult-directed speech (3.0). Consistent with Massaro’s (2015a) findings, it appears that picture books feature more challenging language than CDS and ADS, likely due to differences in formality between written and spoken language (Massaro, 2017b). Reading books to children thus provides a “linguistic and cognitive complexity not typically found in speech to children” (Massaro, 2017b, p. 63).

Whitehurst et al. (1998) examined how exposure to picture books can impact children’s language ability. Parents in the experimental group of this study were instructed to “increase their rates of open-ended questions, function/attribute, and expansions; to respond appropriately to children’s attempts to answer these questions; and to decrease their frequency of straight reading and questions that could be answered by pointing” while reading to their child over the course of a month (p. 552). Compared to the control group who were simply instructed to read in their usual way, children in the experimental group scored higher on expressive language ability. Children in the experimental group also used more phrases, spoke fewer single words, and had a higher MLU (mean length of utterance) than the control group. These findings suggest that engaging children in dialogue while reading picture books may encourage significant gains in expressive language acquisition. 

​

Evans and Saint-Aubin (2013) looked deeper into the relationship between vocabulary acquisition and picture books. By tracking and examining eye movements in French preschoolers aged 50 to 62 months as they were read three picture books, they discovered that children’s eye movements were stable and generally aimed toward the illustrations in the book. Through administrations of a scale measure of children’s receptive vocabulary (the ÉVIP) and adaptions of this scale measure,  Evans and Saint-Aubin also found that children’s receptive vocabulary of uncommon words increased over the course of the readings. Furthermore, it was revealed that this improvement was correlated with children’s pretest receptive vocabulary. Based on these findings, the researchers concluded that “print affords children the opportunity to continually roam through the illustrations and search for matches between what is said and what is depicted, and to learn the meaning of new words” (p. 607).

​

Shinksey (2020) examined how different forms of picture books may impact how children learn words. Using either a lift-the-flap book or a normal picture book, two-year-old children were taught a novel term for an unfamiliar food. When tested, children who were read the lift-the-flap book performed significantly worse in an identification task than children who saw the no-flap book. This finding was specific to the new word and did not generalize to children’s recognition of higher frequency words in the books. This finding supports “cognitive load accounts suggesting that tactical features distract from the book’s content” (p. 1), and may suggest that lift-the-flap books hinder word learning. 

​

Under the same topic of book features and learning, Strouse et al. (2018) reviewed how different aspects of books can influence learning, reasoning, and symbolic understanding. They concluded that “certain features (e.g., fantasy) may be more disruptive in some domains (i.e., problem solving and moral lessons) than others (i.e., word and physics learning)”, and that this relationship is mediated by which developmental stage a child is at (p. 12). For instance, realistic pictures and tactile elements appear to hinder word and letter learning in younger children because children rely on surface-level features to facilitate book to context transfer. Even fantasy in books can inhibit learning in young children, particularly those lacking an understanding of what is realistically possible. However, book features seem to be less impactful in domains where basic features are insufficient for learning and deeper comprehension and situational transfer are necessary, like problem solving and morality. Moreover, as children age and are better able to distinguish reality from fantasy, fantastical stories may prove less harmful to learning and transferability. 

​

The influence of other children’s media on children’s vocabulary acquisition has been investigated, in particular, television media aimed toward babies. A study conducted by DeLoache et al. (2010) tested if infants aged 12 to 18 months would learn any words from a DVD targeted toward educating babies. The researchers had parents regularly show their infants a popular DVD specifically advertised for children aged 12 months and older, which depicted a person in a house or yard labelling various everyday household items repeatedly (three times per item with several minutes in between each repetition) while performing a gesture during the first and final repetition. DeLoache and colleagues found that after a month of exposure, the children who watched the DVD did not learn any more words than children who did not watch it. The most vocabulary gains actually occurred in an alternative group, one in which parents did not show any video to their child and instead tried to teach their children the words themselves. 

​

DeLoache et al. (2010) concluded that what infants learn from baby media is minimal, and likely overestimated. This is consistent with previous findings that “very young children often fail to use information communicated to them via symbolic media, including pictures, models, and video” (p. 4).  On the other hand, the child’s linguistic input is easily envisioned as falling on a fuzzy continuum from live face-to-face dialog, remote video conferencing, audio phone calls, television media, to picture books. These interactions might have various degrees of effectiveness, but it is unlikely that one of them would be totally ineffective.

​

Recently, Massaro (2015a, 2017b) established that the language in children’s books is more complex and featured a more extensive vocabulary than child-directed speech and even adult-directed speech. After eliminating the 5,000 most common English words from the Corpus of Contemporary American English (COCA, 1990-2012), a sampled database of children’s picture books had three times as many rare words as a database of CDS and 1.64 times as many rare words as a database of ADS. One implication of this finding is that reading picture books aloud to children may expose them to a more extensive vocabulary at an earlier age than exposure to CDS or ADS alone.

​

It is also reasonable to measure linguistic complexity in terms of grade level readability. Readability can be best understood as the reading grade level required of a reader to effectively read some form of text (Readability Formulas, n.d.). There are several different formulas that have been used in calculating readability, each using factors such as the number of words, the length of sentences, and the average number of syllables to determine the reading grade level of the passage. These readability formulas are generally computed from each sentence as the basic unit of analysis.

​

Massaro (2017b) used some of these formulas, particularly the Flesch-Kincaid grade level formula, Gunning Fog Index, Coleman grade level, SMOG index, and automated readability index (ARI), to evaluate the readability of samples of CDS, ADS, and picture books. According to these measures of reading grade level, Massaro found that picture books had the highest average grade level (4.2) compared to child-directed speech (1.9) and adult-directed speech (3.0). Thus, popular children’s picture books had a higher reading difficulty than CDS and ADS. This strengthens the argument that reading picture books to children is beneficial, as it appears to familiarize them with more challenging language and a more extensive vocabulary than they would otherwise encounter. Reading books to children thus provides a “linguistic and cognitive complexity not typically found in speech to children” (p. 63). Analogous to the importance of the frequency of CDS, it is important to note that the benefits of reading picture books aloud to children are contingent upon how frequently children are being read to. 

​

As the majority of research on sources of language input has centered on the nature and implications of children’s books, CDS, and ADS, it is also important to note that many children also spend a considerable amount of time watching television shows (Vandewater et al., 2006). A sobering observation by Vanderwater et al. (2006) is that children in the U.S. watch between 2 to 5 hours of television a day on average, which accounts for “more time [spent] than in any other single discretionary activity except for sleep” (p. 2). Researchers in this study also found that television viewership was not correlated with time spent reading or being read it, though it was negatively correlated with time spent interacting with parents and siblings. This could point to a complex relationship between viewing television and other sources of language input (i.e., child-directed speech and picture books), and may even suggest the potential of frequent television viewership to limit or disrupt children’s exposure to CDS or other interactive speech. Despite this, little empirical consideration has been placed on the linguistic content of children’s television. 

​

Some observational research has associated television viewership, particularly time spent watching baby DVDs and videos every day, with lower scores on a measure of language development in infants aged 8 to 16 months (Zimmerman et al., 2007). However, this research does not take into account the actual content of the media children consume, but rather speaks to the potential implications of TV viewership in early childhood. Zimmerman et al.’s findings also did not replicate this relationship in toddlers aged 17 to 24 months, and thus may have limited application. 

​

In sum, these findings demonstrate the multifaceted nature of language acquisition. There are many contributors to this process, and the specifics of what each factor contributes remains for the most part unclear. Additional inquiry into this topic is essential for uncovering the complexities of speech input and language learning. With the lack of research conducted on picture books and children’s media in general, I suggest that researchers investigate intersections between language acquisition and these mediums. This could include analyzing if and how reading grade level and vocabulary are differently manifested across sources of speech input. Examination into this area of study is a valuable pursuit as it will allow for a better understanding of how children learn to speak, which will in turn allow parents and professionals to better optimize learning among our youth. 

​

​

References

DeLoache, J. S., Chiong, C., Sherman, K., Islam, N., Vanderborght, M., Troseth, G. L., Strouse, G. A., & O’Doherty, K. (2010). Do babies learn from baby media? Psychological Science, 21(11), 1570–1574. https://doi.org/10.1177/0956797610384145 

​

Dingemanse, M., & Thompson, B. (2020). Playful iconicity: structural markedness underlies the relation between funniness and iconicity. Language and Cognition, 12(1), 203–224. https://doi.org/10.1017/langcog.2019.49 

​

Evans, M. A., & Saint-Aubin, J. (2013). Vocabulary acquisition without adult explanations in repeated shared book reading: An eye movement study. Journal of Educational Psychology, 105(3), 596–608. https://doi.org/10.1037/a0032465 

​

Massaro, D. W. (2012). Acquiring literacy naturally: Behavioral science and technology could empower preschool children to learn to read naturally without instruction. American Scientist, 100, 324–333.

​

Massaro, D. W. (2015a). Two different communication genres and implications for vocabulary development and learning to read. Journal of Literacy Research, 47(4), 505–527. http://dx.doi.org.oca.ucsc.edu/10.1177/1086296X15627528

​

Massaro, D.W. (2015b). Speech perception. In J. D. Wright (Ed.), International Encyclopedia of the Social & Behavioral Sciences (2nd ed., Vol. 23, pp. 235–242). Oxford: Elsevier. ISBN: 9780080970868 

​

Massaro, D. W. (2016). Multiple influences in vocabulary acquisition: Parental input dominates. Proceedings of the 17th Annual Conference of the International Speech Communication Association (Interspeech 2016), 878–882. ISSN 2308-457X. www.isca-speech/archive/interspeech_2016/pdfs/0037.PDF

​

Massaro, D. W. (2017a). Modeling multiple influences on vocabulary acquisition: Context, symbol, and association learning. Unpublished paper. 

​

Massaro, D. W. (2017b). Reading aloud to children: Benefits and implications for acquiring literacy before schooling begins. The American Journal of Psychology, 130(1), 63–72. http://dx.doi.org.oca.ucsc.edu/10.5406/amerjpsyc.130.1.0063

​

Massaro, D. W., & Perlman, M. (2017). Quantifying iconicity’s contribution during language acquisition: Implications for vocabulary learning. Frontiers Communication, 2(4). https://doi.org/10.3389/fcomm.2017.00004 

​

Massaro, D. W., & Rowe, B. (2015). Comprehension outscores production in language acquisition: Implications for theories of vocabulary learning. Journal of Child Language Acquisition and Development – JCLAD, 3(3), 121–152. ISSN: 2148-1997

​

Perry, L. K., Perlman, M., Winter, B., Massaro, D. W., & Lupyan, G. (2017). Iconicity in the speech of children and adults. Developmental Science, 21(3), e12572. https://doi.org/10.1111/desc.12572 

​

Readability Formulas. (n.d.). Automatic readability checker, a free readability formula consensus calculator. https://readabilityformulas.com/free-readability-formula-tests.php

​

Shinskey, J. L. (2021). Lift-the-flap features in “first words” picture books impede word learning in 2-year-olds. Journal of Educational Psychology, 113(4), 641–655. https://doi.org/10.1037/edu0000628

​

Vandewater, E. A., Bickham, D. S., & Lee, J. H. (2006). Time well spent? Relating television use to children’s free-time activities. Pediatrics, 117(2), e181–e191. https://doi.org/10.1542/peds.2005-0812

Whitehurst, G. J., Falco, F. L., Lonigan, C. J., Fischel, J. E., DeBaryshe, B. D.,Valdez-Menchaca, M. C., Caulfield, M. (1988). Accelerating language development through picture book reading. Developmental Psychology, 24(4), 552–559.

​

Zimmerman, F. J., Christakis, D. A., & Meltzoff, A. N. (2007). Associations between media viewing and language development in children under age 2 years. The Journal of Pediatrics, 151(4), 364–368. https://doi.org/10.1016/j.jpeds.2007.04.071