Signs and symptoms among monolingual English-speaking children with LLE are based on parent-report measures. An extensively used set of criteria for LLE is an expressive vocabulary of fewer than 50 words and no two-word combinations by 24 months of age (Paul, 1991; Rescorla, 1989). However, it is essential to review these criteria at regular intervals (e.g., every 6 months) to assess language growth and to determine whether the child is in fact a late talker, whether language skills fall outside of developmental trajectories, or whether signs and symptoms clearly indicate a language disorder.
In making the determination, it is also important to consider other language development factors, including rate of vocabulary growth, speech sound development, emerging grammar, language comprehension, social language skills, use of gestures, and symbolic play behaviors (Olswang, Rodriguez, & Timler, 1998; Wetherby, Allen, Cleary, Kublin, & Goldstein, 2002).
For example, when compared with toddlers of the same age with typical language development, late talkers may demonstrate
- phonological differences once they do produce their first words, including less complex/mature syllable structures, lower percentage of consonants correct, and smaller consonant and vowel inventories (Mirak & Rescorla, 1998; Paul & Jennings, 1992; Rescorla & Ratner, 1996);
- delayed comprehension and communicative use of symbolic gestures (Thal, Marchman, & Tomblin, 2013);
- use of shorter and less grammatically complex utterances—particularly for toddlers with expressive and receptive delays (Thal et al., 2013);
- comprehension of fewer words (Thal et al., 1991; Thal et al., 2013).
Research also suggests that delays and differences in babbling before the age of 2 can predict later delays in expressive vocabulary, limited phonetic repertoire, and use of simpler syllable shapes (Fasolo, Majorano, & D'Odorico, 2008; Oller, Eilers, Neal, & Schwartz, 1999; Stoel-Gammon, 1989).
Approximately 50% to 70% of late talkers are reported to catch up to peers and demonstrate normal language development by late preschool and school age (Dale, Price, Bishop, & Plomin, 2003; Paul, Hernandez, Taylor, & Johnson, 1996). In a study by Rice, Taylor, and Zubrick (2008), the prevalence of language impairment at age 7 was 20% for children with a history of LLE compared with 11% for controls. That is, only one in five late talkers had language impairment at age 7.
Although many late talkers go on to perform within the normal range on expressive and receptive language measures by kindergarten age (Ellis Weismer, 2007; Rescorla 2000, 2002), their scores on such measures continue to be lower than those of children with a history of typical language development, matched for socioeconomic status (SES; Paul, 1996; Rescorla, 2000, 2002).
For example, school-age children who had been identified as late talkers demonstrated
- lower scores at age 5 on language measures that tap complex language skills, like narrating a story;
- poorer performance on measures of general language ability, speech, syntax, and morphosyntax at age 7;
- poorer performance on reading and spelling measures at ages 8 and 9;
- lower scores on aggregate measures of vocabulary, grammar, verbal memory, and reading comprehension at age 13;
- lower scores on vocabulary/grammar and verbal memory factors at age 17.
(Girolametto, Wiigs, Smyth, Weitzman, & Pearce, 2001; Rescorla, 2002, 2005, 2009; Rice et al., 2008)
Delayed language comprehension has been shown to be a significant predictor of language outcomes in late talkers (Ellis Weismer, 2007; Thal et al., 1991). Deficits in comprehension are associated with language deficits at later ages (Thal et al., 1991).
For some children, LLE may be an early indicator of specific language impairment (SLI). Children who continue to have poor language abilities below chronological age expectations (by late preschool or school age) that cannot be explained by other factors (e.g., low nonverbal intelligence, sensory impairments, or autism spectrum disorder) may be identified at that point as having SLI (Archibald & Gathercole, 2007).