Skip to main content
NCBI home page
Journal of Speech, Language, and Hearing Research : JSLHR logoLink to Journal of Speech, Language, and Hearing Research : JSLHR
. 2023 Aug 17;66(10):3925–3939. doi: 10.1044/2023_JSLHR-23-00096

Lexical–Semantic Organization as Measured by Repeated Word Association in Children Who Are Deaf and Hard of Hearing Who Use Spoken Language

Olivia Rush a, Krystal L Werfel b, Emily Lund a,✉
PMCID: PMC10713021  PMID: 37591230

Abstract

Purpose:

This study compares responses of children who are deaf and hard of hearing (DHH) who use spoken language with responses of children who have typical hearing on a repeated word association task to evaluate lexical–semantic organization.

Method:

This study included 109 participants in early kindergarten or who had completed first grade. The younger group included 30 children with typical hearing, 22 with hearing aids, and 21 with cochlear implants. The older group included 16 children with typical hearing, nine with hearing aids, and 11 with cochlear implants. Children were asked to give a word associated with 24 stimuli words. Responses were coded according to their relation to the target.

Results:

An analysis of variance revealed that older children, regardless of hearing status, produced more semantically related responses to prompts than younger children. Children in the younger DHH group differed from children with typical hearing in their production of non–semantically related responses: They produced errored responses at higher rates.

Conclusion:

This preliminary data may indicate an early deficit in recognition of semantic relations between words for children who are DHH and provides a basis for continued longitudinal study of changes in lexical–semantic organization.

Children who are deaf and hard of hearing (DHH) who use spoken language, on average, have a documented deficit in vocabulary knowledge as compared to their same-age peers with typical hearing (Lund, 2016; Lund et al., 2022; McCreery & Walker, 2022; Persson et al., 2022). To date, however, most studies of vocabulary knowledge in children who are DHH have only measured quantity of words known, typically through single-word vocabulary assessments. It is equally important to consider other facets of a child's vocabulary knowledge, such as how the words that they know are stored and retrieved. One construct that likely influences a child's ability to flexibly use the words stored in the lexicon is lexical–semantic organization, or the way a child links and stores words for retrieval (Sheng & McGregor, 2010). The purpose of this study is to evaluate lexical–semantic organization, as measured by repeated word association, in children who are DHH as compared to children of the same age with typical hearing.

Vocabulary Outcomes for Children Who Are DHH

Distinct differences are present between vocabulary outcomes in children who are DHH and their peers with typical hearing: In general, children who are DHH and who use spoken language know fewer words overall as compared to children with typical hearing. McCreery and Walker (2022) assessed single-word, receptive vocabulary of 5- to 10-year-old children who wore hearing aids (n = 177) and who had typical hearing (n = 86) and found that children in the DHH group had a significantly lower vocabulary size indexed to age (d = −0.61). Similarly, Lund et al. (2022) found that 4-year-old children with cochlear implants (n = 47) and children with hearing aids (n = 45) scored lower on a composite measure of expressive and receptive single-word vocabulary than their age-matched peers with typical hearing (n = 66; d = 1.60; 1.00 respectively).

A gap in quantity of vocabulary knowledge is, perhaps, predictable. Children whose parents choose spoken language as their primary communication modality learn to access vocabulary through a device, such as a hearing aid or cochlear implant. Due to United States Food and Drug Administration (FDA) regulations, the minimum age a child may receive cochlear implants in the United States is 12 months of age (FDA, 2022). Average age of hearing aid fitting in the work of McCreery and Walker (2022) was 20.6 months and in the work of Lund et al. (2022) was between 15 and 23 months. Therefore, a child who is DHH begins using auditory input to learn new words later than an age-matched peer with typical hearing. If a child is expected then to “catch up” to the vocabulary knowledge of same-age peers (e.g., Fulcher et al., 2012; Nicholas & Geers, 2007), a child who is DHH would have to learn vocabulary at a faster rate than a child with typical hearing (Lund, 2016).

Children who are DHH likely do not only struggle with vocabulary as a result of delayed input; children may experience interruptions in any of the three subprocesses that children with typical hearing experience during vocabulary acquisition: triggering, configuration, and engagement (Lund, 2020). Triggering refers to the process of recognizing that a word is new and must be learned (Hoover et al., 2010). Children with cochlear implants struggle to pair novel words to nameless pictures in disambiguation tasks, likely reflecting difficulty with triggering (Lederberg et al., 2000; Lund, 2018). The process of configuration involves recognizing and mapping a phonological and semantic representation of a new word to its referent (Leach & Samuel, 2007). Based on currently published literature, it is unclear if children with hearing loss demonstrate deficits in the configuration subprocess (Lund, 2020). It is important to consider the typical environments for word learning. Classrooms are a common setting for implicit and explicit vocabulary learning for preschool children. Listening fatigue and listening effort, proven difficulties that are exhibited in individuals with hearing loss, tend to increase in severity in noisy environments (Hornsby et al., 2016). Because these difficulties increase in noisy settings, such as classrooms, they may contribute to potential difficulty in the configuration stage of word learning. Finally, engagement refers to the necessary interaction that must occur between the new word and the child's existing lexicon, facilitated by the organization of the lexicon; that is, lexical engagement refers to how a new word evolves in its relation to the child's other lexical entries. An individual can learn the form of a word (configuration) without it activating a similar word in its semantic network; activation would be an indicator of engagement (e.g., Leach & Samuel, 2007). A child's ability to recognize, even subconsciously, semantic and taxonomic relations, phonological relations, and syntactic relations between words stored in the mental lexical and the novel word reflects engagement (Luce & Pisoni, 1998; Sheng & McGregor, 2010; Waxman & Gelman, 1986). Thus, the process of engagement is likely influenced by a child's ability to build lexical–semantic organization.

A few studies have considered lexical organization of children with hearing loss in different ways. Children with cochlear implants have demonstrated reduced ability to access superordinate taxonomic organization (i.e., to identify that a tree is also a plant, even if they know the word “plants”) as compared to their same-age peers (Lund & Dinsmoor, 2016). Children who are DHH appear less likely to produce phonologically related responses to the prompted word in rhyming and/or in verbal fluency tasks (Ground et al., 2014; Wechsler-Kashi et al., 2014). Based on performances on analogy tasks, single word association tasks, and verbal fluency tasks, researchers have proposed that adults and children who are DHH may use different lexical–semantic organization strategies than their peers with typical hearing (Kenett et al., 2013; Marschark et al., 2004). Further research is needed to assess the specific organizational differences appearing for children who are DHH. Additionally, changes in lexical–semantic organization for children with hearing loss must be measured with a task that captures change in lexical organization strategies.

Lexical–Semantic Organization

To study the lexical–semantic organizational differences between children who are DHH and children with typical hearing, a measure of lexical organization strategies is needed. Repeated word association tasks have been used to evaluate the differences in the storage and accessibility of semantic relationships between bilingual children (bilingual in two spoken languages and bilingual in American Sign Language [ASL] and English) and monolingual children and between typically developing children and children with developmental language disorder/specific language impairment (Mann et al., 2016; Sheng et al., 2006; Sheng & McGregor, 2010). In this task, an individual prompt (e.g., the word turtle) is repeated multiple times and an individual being tested is expected to produce a different single response each time. The repeated nature of the task not only assesses storage within the lexicon, but also the accessibility of the semantically related words. In the Sheng et al. (2006) and Mann et al. (2016) studies, a child's organization was analyzed relative to responses that included synonyms, antonyms, coordinates, superordinates, or a direct negation of the prompt. In the Sheng and McGregor (2010) study, lexical–semantic organization was analyzed by coding the number of semantic, clang (phonologically related words that share alliteration or a rhyme with the target word), and error responses produced by children.

A child's ability to recognize how words relate can be an index of the lexical–semantic organization system a child has in place. Knowledge of between-words relationships is expected to change as a child ages (Nelson, 1977). In word-association tasks, for example, very young children may be expected to give responses that are not associated with the target (Nelson, 1997). As children begin to become aware of individual sounds in words, they begin to produce responses that contain the same sounds (but not always similar semantic content) as target words (Cronin, 2002). Phonological relations between words may manifest through clang responses (e.g., bat–ban), alliterations (e.g., candy–can) or rhymes (e.g., dig–fig) to the stimuli word without semantic relations (Sheng & McGregor, 2010). Across studies, typically developing children demonstrate an increase in semantically related responses and a decrease in phonologically related responses as age increases (Sheng et al., 2006; Sheng & McGregor, 2010).

When children produce words that are semantically related to a target, often those words are thematically or taxonomically related. Thematic relations describe words that coexist in the same schema (e.g., dog–bone) and taxonomic relations describe words that are hierarchically related (e.g., dog–animal). In using taxonomic organization, a word association may be classified as superordinate, basic, or subordinate. For example, Waxman and Hatch (1992) used name possible taxonomic associations for the word dog as animal (superordinate), cat (basic), and golden retriever (subordinate). Another shift that is expected to occur as a child ages is the thematic–taxonomic shift; that is, children shift from structuring their lexicon according to thematic relations to structuring according to taxonomic relations (Sheng et al., 2006). Even though this shift does not, reportedly, occur until sometime between first and third grade, typically developing children understand taxonomic organization at as early as 3 years of age (McCauley et al., 1976; Waxman & Hatch, 1992). Children master superordinate relations, the broad category that a given target word is related to, around 7 years of age (Di Giacomo et al., 2012; Lucariello et al., 1992). This shift, however, has not been typically captured or coded for in repeated word association tasks; rather, it is most reported in item-generation tasks (e.g., name animals; Nelson & Nelson, 1990).

Another age-related shift in organization relates to types of sequencing: syntagmatic or paradigmatic sequencing. In syntagmatic sequencing, an individual's response to a stimulus word is something like a word that is frequently used with the stimulus word in a sentence or phrase (e.g., cold–ice). In paradigmatic sequencing, an individual's response to a stimulus word comes from the same grammatical form class (e.g., cold–warm; Nelson, 1977; Sheng & McGregor, 2010; Sheng et al., 2006). In contrast to the thematic–taxonomic relations, which are mainly used to organize nouns, the syntagmatic–paradigmatic relations are used to organize all parts of speech and are most clearly observed with adjectives (Sheng et al., 2006). As children with typical hearing grow, the number of syntagmatic responses to word stimuli decreases as the number of paradigmatic responses increases (Lippman, 1971). Thus, as a child matures in organization strategies, researchers would expect to observe a shift from phonologically related responses to semantically related responses, a shift from more thematic responses to an increase in taxonomic responses, and a shift from syntagmatic to paradigmatic responses.

One prior study has evaluated responses on a repeated word association task for children who are DHH. Mann et al. (2016) assessed deaf participants between the ages of 6 and 10 years who were proficient in ASL (n = 12) and an age-matched comparison group of children with typical hearing (n = 49). In the group of deaf participants, there was no effect of language (ASL vs. English) on types of responses. Additionally, responses did not vary by group; children who were deaf and bilingual and children with typical hearing who were monolingual had similar rates of paradigmatic and syntagmatic responses when deaf children used ASL and children with typical hearing used English. There was, however, a difference between groups when both used English, possibly because English was a second language for the deaf group. The authors hypothesized that a lack of exposure to English may have contributed to the result.

To date, however, repeated word association has not been used to study lexical–semantic network links for children who are DHH who use spoken language as a primary mode of communication (and thus, first language). Because prior work has hypothesized that lexical–semantic organization may be underdeveloped in children who are DHH (Kenett et al., 2013; Lund & Dinsmoor, 2016; Marschark et al., 2004), this work is an important next step toward better understanding vocabulary development within that population. The purpose of this study is to compare responses of children who are DHH who use spoken language with responses of children matched for age who have typical hearing on a repeated word association task. The following research questions guided this study:

  • (a)  Do children who are DHH produce fewer semantically related responses to a repeated word association task than children with typical hearing?

  • (b)  Do children who are DHH, when they produce non–semantically related responses, produce fewer clang responses than children with typical hearing?

  • (c)  Do children who are DHH, when they produce semantically related responses, produce fewer taxonomic responses than children with typical hearing?

  • (d)  Do children who are DHH, when they produce semantically related responses, produce fewer paradigmatic responses than children with typical hearing?

  • (e)  Do children who are DHH and children with typical hearing increase semantically related, taxonomic, and paradigmatic response types as they age?

  • (f)  Does vocabulary size, after being controlled for age and regardless of hearing status, predict semantically related, taxonomic, and paradigmatic responses?

We hypothesized that children who are DHH will produce fewer semantically related responses, reflecting a lessdeveloped semantic organization system (e.g., Lund, 2016; Sheng & McGregor, 2010). When they produce non–semantically related responses, we hypothesized that children who are DHH will produce fewer clang responses than children with typical hearing, reflecting a less well-developed phonological organization system (e.g., Lund, 2022). When children who are DHH do produce semantically related responses, we hypothesized children who are DHH who use spoken language will produce fewer taxonomic and paradigmatic responses than children with typical hearing, again as a reflection of a less maturely developed organization system (e.g., Mann et al., 2016; Wechsler-Kashi et al., 2014). Additionally, we hypothesized that all children will increase semantically related, taxonomic and paradigmatic responses as they age, in the same way that all groups increase their vocabulary size over time (e.g., Lund et al., 2022). Finally, we hypothesized that vocabulary size controlled for age (as measured by the standard score on a norm-referenced assessment) would significantly predict semantically related, taxonomic, and paradigmatic responses, as these types of responses reflect organization maturation that should accompany vocabulary growth.

Method

All procedures in this study were approved by the University of South Carolina Institutional Review Board (as Institutional Review Board of Record), with Texas Christian University in agreement.

Participants

This study included 109 participants, all enrolled in the ongoing Early Language and Literacy Acquisition in Children with Hearing Loss (ELLA) Study, which is a longitudinal study of children who are DHH who use spoken language (e.g., Lund et al., 2022; Werfel et al., 2021). Children fell into two age groups: children who were 5 years old and starting kindergarten (K) and children who had completed the first grade (Grade 1).

The K group included 73 children, 30 with typical hearing (TH-K), 22 with hearing aids (HA-K), and 21 with cochlear implants (CI-K). All children were enrolled in the ongoing ELLA study around 4 years of age and completed the measures for this study around their fifth birthday (TH-K: M = 60.23 months, SD = 1.02 months; HA-K: M = 60.45 months, SD = 1.23 months; CI-K: M = 60.43 months, SD = 1.00 months). The Grade 1 group included 36 children, 16 with typical hearing (TH-1), nine with hearing aids (HA-1), and 11 with cochlear implants (CI-1). All children were enrolled in the ongoing ELLA study and completed the measures for this study following the completion of first grade (TH-1: M = 85.19 months, SD = 4.76 months; HA-1: M = 93.89 months, SD = 6.11 months; CI-1: M = 88.45 months, SD = 2.54 months). See additional demographic and descriptive information in Table 1. The children in both typical hearing groups passed a bilateral hearing screening (e.g., pure-tone thresholds of 20 dB HL or less in both ears at 1, 2, and 4 kHz). See audiological information for each group in Table 2.

Table 1.

Descriptive information and assessments (means and standard deviations) by group.

Information TH-K (n = 30) HA-K (n = 22) CI-K (n = 21) TH-1 (n = 16) HA-1 (n = 9) CI-1 (n = 11)
Gender Male: 14
Female: 16		 Male: 16
Female: 6		 Male: 10
Female: 11		 Male: 5
Female: 11		 Male: 3
Female: 6		 Male: 4
Female: 7
Race and ethnicity White: 24
Asian: 0
Black: 2
Native Hawaiian/Pacific Islander: 0
Hispanic or Latinx: 2
Prefer not to respond: 4		 White: 17
Asian: 3
Black: 1
Native Hawaiian/Pacific Islander: 1
Hispanic or Latinx: 2
Prefer not to respond: 0		 White: 19
Asian: 1
Black: 1
Native Hawaiian/Pacific Islander: 0
Hispanic or Latinx: 0
Prefer not to respond: 0		 White: 14
Asian: 1
Black: 0
Hispanic or
Latinx: 1
Native Hawaiian/Pacific Islander: 0
Prefer not to respond: 1		 White: 7
Asian: 0
Black: 2
Native Hawaiian/Pacific Islander: 0
Hispanic or Latinx: 3
Prefer not to respond: 0		 White: 10
Asian: 0
Black: 0
Native Hawaiian/Pacific Islander: 0
Hispanic or Latinx: 2
Prefer not to respond: 1
Caregiver education 18.00 years (2.18) 16.98 years (1.80) 17.48 years (2.14) 17.25 years (2.65) 17.22 years (3.11) 15.27 years (2.05)
TELD-3 Spoken Language Quotient 117.13 (9.30) 104.55 (20.66) 94.33 (18.68) N/A N/A N/A
CELF-5 Core Language N/A N/A N/A 112.00 (13.46)a 93.71 (11.15)a 95.00 (18.87)a
EOWPVT-4 120.67 (11.42) 110.50 (19.94)a 103.11 (15.20)a 121.40 (12.43)a 99.29 (17.66)a 104.09 (11.95)
PPVT-4 112.40 (13.77) 96.65 (17.11)a 89.44 (13.35)a 109.80 (11.53)a 95.57 (16.33)a 95.91 (9.90)
ESP 4.00 (0.00) 3.91 (0.42) 3.95 (0.21) 4.00 (0.00) 4.00 (0.00)a 4.00 (0.00)a
Open in a new tab

Note. TH-K = children with typical hearing, early kindergarten group; HA-K = children with hearing aids, early kindergarten group; CI-K = children with cochlear implants, early kindergarten group; TH-1 = children with typical hearing, post–first-grade group; HA-1 = children with hearing aids, post–first-grade group; CI-1 = children with cochlear implants, post–first-grade group; TELD-3 = Test of Early Language Development–Third Edition (Hresko et al., 1999); N/A = not available; CELF-5 = Clinical Evaluation of Language Fundamentals–Fifth Edition; EOWPVT-4 = Expressive One Word Picture Vocabulary Test–Fourth Edition; PPVT-4 = Peabody Picture Vocabulary Test–Fourth Edition; ESP = Early Speech Perception Test (Geers & Moog, 2012).

a

Missing scores due to testing time constraints (HA-K: two participants for EOWPVT-4, three participants for PPVT-4; CI-K: two participants for EOWPVT-4, three participants for PPVT-4; TH-1: one participant for EOWPVT-4, PPVT-4, CELF-5; HA-1: two participants for EOWPVT-4, PPVT-4, CELF-5, and one participant for ESP; CI-1: one participant for ESP).

Table 2.

Audiological information by group.

Information HA-K CI-K HA-1 CI-1
Degree of hearing loss Mild–moderate: 9
Moderate: 4
Moderate–severe: 7
Severe: 2		 Moderate–severe: 1
Severe: 3
Severe–profound: 8
Profound: 9		 Mild–moderate: 5
Moderate–severe: 4		 Severe–profound: 2
Profound: 9
Age at diagnosis in months 10.85 (15.9) 4.77 (9.58) 17.33 (20.17) 4.57 (7.48)
Age at hearing aid in months 13.84 (14.13) 8.98 (11.36) 22.33 (18.90) 6.43 (7.68)
Age at first cochlear implant in months N/A 27.64 (15.93) N/A 15.18 (8.22)
Time with hearing aid or implant (by group) in months 45.74 (15.74) 32.36 (16.32) 70.66 (20.05) 69.36 (7.85)
Number of cochlear implants N/A Two: 17
One: 4		 N/A Two: 10
One: 1
Reported etiologies of hearing loss, if known Stickler syndrome, connexin 26, Branchiootorenal syndrome, Binder syndrome, Usher syndrome, ototoxic medication, gene mutation, and Chiari malformation Auditory nerve neuropathology, cytomegalovirus (CMV), connexin 26, Pendred syndrome, bacterial meningitis, Mondini dysplasia, enlarged vestibular aqueduct Enlarged vestibular aqueducts Connexin 26
Open in a new tab

Note. HA-K = children with hearing aids, early kindergarten group; CI-K = children with cochlear implants, early kindergarten group; HA-1 = children with hearing aids, post–first-grade group; CI-1 = children with cochlear implants, post–first-grade group; N/A = not available.

Procedure

All assessments were completed by an evaluator (a master's level trained speech-language pathologist) one-on-one with each participant. Evaluators participated in several hours of testing training for all assessments with the principal investigators or study coordinator of the ELLA study. Testing setting included a quiet space with table and chairs for the participant and a parent. All assessments were audio- and video-recorded to monitor fidelity of administration and to double-score all responses.

Descriptive measures. Children participated in a series of descriptive measures to characterize their current level of language development. According to ELLA testing procedures, omnibus language was measured either via the Test of Early Language Development–Third Edition (Hresko et al., 1999) for children in the K group (TH-K, HA-K, and CI-K) or via the Clinical Evaluation of Language Fundamentals–Fifth Edition (Wiig et al., 2013) for children in the Grade 1 group (TH-1, HA-1, and CI-1). All groups participated in an expressive and a receptive single-word vocabulary test, the Expressive One-Word Picture Vocabulary Test–Fourth Edition (EOWPVT-4; Martin & Brownell, 2011) and the Peabody Picture Vocabulary Test–Fifth Edition (Dunn, 2019), respectively. Finally, all groups completed the Early Speech Perception Test (Geers & Moog, 2012) as an index of speech perception for single words that asks children to identify spoken, single-syllable words that contrast minimally from a closed set of 12 choices.

Repeated word association measure. The primary aim of this study was to evaluate the status of lexical–semantic organization in children with hearing loss via a repeated word association task (see Table 3). This task was consistent with the procedures described by Sheng and McGregor (2010). The task included 24 stimuli words, 12 nouns (foot, hat, goat, cow, frog, zipper, broom, pillow, spoon, desk, kite, and turtle) and 12 verbs (yawn, kick, cry, hide, eat, count, run, sing, read, swim, push, and squeeze), that were presented to the children three different times within the same administration time. The examiner began the task with instructions to verbally produce a single word in response to the stimuli word. The instructions were read as follows:

Table 3.

Coded repeated word association performance by group means (standard deviations).

Variable TH-K HA-K CI-K TH-1 HA-1 CI-1
Semantically related responses 33.60 (19.47) 28.46 (15.96) 25.52 (16.82) 46.94 (11.41) 39.33 (18.02) 43.18 (21.45)
 Thematic responses 27.07 (17.29) 21.55 (13.23) 20.67 (13.91) 39.56 (10.81) 33.11 (15.85) 37.82 (18.63)
 Taxonomic responses 6.53 (5.70) 6.91 (6.27) 4.86 (3.94) 7.38 (10.81) 6.22 (3.53) 5.36 (4.46)
 Syntagmatic responses 17.87 (11.63) 15.18 (9.22) 14.33 (3.94) 28.25 (8.60) 22.22 (12.00) 27.82 (13.53)
 Paradigmatic responses 15.73 (10.24) 13.73 (8.88) 11.19 (9.02) 18.69 (7.08) 17.11 (9.05) 15.36 (11.54)
Non–semantically related responses 38.40 (19.47) 43.55 (15.96) 46.48 (16.82) 25.06 (11.41) 32.67 (18.02) 28.82 (21.45)
 Errored responses 10.97 (8.65) 21.36 (14.77) 24.76 (17.18) 12.50 (8.29) 16.78 (12.29) 11.00 (5.18)
 No response 18.70 (17.57) 13.32 (9.21) 16.67 (14.63) 10.19 (8.18) 7.56 (8.86) 8.09 (5.82)
 Clang responses 8.73 (16.63) 8.86 (11.18) 5.05 (10.89) 2.38 (3.38) 8.33 (14.17) 9.73 (19.49)
Open in a new tab

Note. TH-K = children with typical hearing, early kindergarten group; HA-K = children with hearing aids, early kindergarten group; CI-K = children with cochlear implants, early kindergarten group; TH-1 = children with typical hearing, post–first-grade group; HA-1 = children with hearing aids, post–first-grade group; CI-1 = children with cochlear implants, post–first-grade group.

“We are going to play a word game. I'm going to say a word and I want you to tell me the first word that comes to mind after you hear my word. Let's practice. If I said the word ‘mom’ a word you can tell me is ‘dad,’ ‘family,’ ‘brother,’ or ‘sister.’ If I say the word ‘birthday,’ a word you can tell me is ‘cake,’ ‘candles,’ ‘presents,’ or ‘ice cream.’ Now you try some. I am going to say the same word three times, so you will try to tell me a new word each time.”

The examiner administered a practice set with the words, “moon,” “grass,” and “cut,” repeating each word 3 times to simulate the repeated nature of the task. If at any point a participant repeated a previous response, the examiner reminded the child to generate a new word or say “I don't know.” The examiner gave the participants 20 s per stimulus before counting silence as “No Response.” The task took up to 30 min for each child to complete if the child was slow to respond. The clinicians administered the task in a preplanned, randomized order for participants.

Data Preparation

All descriptive measures were scored according to procedures in the test manual. For the repeated word association task, the first author created a code book (Rush, 2022) based on procedures described in the work of Sheng and McGregor (2010) and Sheng et al. (2006). Responses were initially coded as semantically or not semantically related to the target. The not–semantically related category included clang responses, such as alliteration or rhyme responses with no semantic relation to the stimuli; error responses, such as repetition of the stimulus, inflections, and real-word responses with no semantic or phonotactic relations; and no response, such as unintelligible responses, nonsense words, and silent responses (and each type of those responses was labeled alongside the code). Then, semantically related responses were coded as thematically related to the target or taxonomically related to the target. The thematic relationship category encompassed responses that fell into the same schema as the target stimuli, including function relations (e.g., hat–wear), descriptive relations (e.g., car–fast), causal relations (e.g., eat–full), part–whole relations (e.g., hand–finger), syntactic relations (e.g., stand–up), and location relations (e.g., kite–sky). The taxonomic relationship category encompassed hierarchically related responses, including superordinate relations (e.g., dog–animal), coordinate relations (e.g., dog–cat), and subordinate relations (e.g., dog–poodle). Each semantically related response was then coded as either a syntagmatic sequence, a response that follows a different form class from the prompt, or a paradigmatic sequence, a response that follows the same form class of the prompt.

Coding fidelity. To ensure reliability for the coding of the participant responses, all were coded by two individuals. The coders had to achieve an interrater agreement of at least 80% with the first author using practice coding responses for all coding categories prior to the commencement of the coding process. The coders maintained reliability ratings of at least 80% throughout the coding process, with specific reliability percentages as follows: 86.40% for semantic versus nonsemantic codes, 89.88% for thematic versus taxonomic codes, and 90.80% for syntagmatic–paradigmatic codes. This percentage was calculated based on a point-by-point agreement formula (House et al., 1981). When coding discrepancies arose in the final analysis, the raters collectively reached an agreement because 100% of transcripts were double coded and consensus was approved by the third author.

Results

The overall purpose of this study was to evaluate lexical–semantic organization, as measured by repeated word association, in children who are DHH as compared to children of the same age with typical hearing. Prior to answering the primary research questions, a t test was completed to determine whether the performance of children with hearing aids significantly differed from children with cochlear implants. Tests compared performance of the two groups of amplification users across the K group and the Grade 1 group for proportions of responses in the following variable categories: semantic relations as a proportion of all responses, nonsemantic relations as a proportion of all responses, thematic responses as a proportion of all semantic relations, taxonomic responses as a proportion of all semantic relations, syntagmatic responses as a proportion of all semantic relations, and paradigmatic responses as a proportion of all semantic relations. The analysis yielded no significant differences between groups (p-value range: .57–.958). Because the analysis yielded no significant difference in responses based on amplification devices, all proceeding data were analyzed comparing only two hearing statuses, children with typical hearing and children who are DHH. For the descriptive performance of children across all categories of code, including a breakdown of non–semantically related codes, see Table 2.

Effects of Hearing Status on Responses

To analyze the similarities and differences of participant responses based on group and on hearing status (DHH vs. TH), analysis of variance calculations were completed for each descriptive category of responses (semantically and non–semantically related responses, thematic vs. taxonomic responses, syntagmatic vs. paradigmatic responses).

Semantically related responses. The first research question asked whether children who are DHH produce fewer semantically related responses to a repeated word association task than children with typical hearing. Following a check for all necessary assumptions of a general linear model, analysis was conducted with the dependent variable as proportion of responses in the semantically related category compared to the number of total responses and the between-subjects independent variables as hearing status (DHH or TH) and age (K or 1). The analysis yielded a main effect of age, F(1, 105) = 15.28, p < .001, but no significant effect of hearing status category, F(1, 105) = 2.89, p = .092. There was no interaction effect of age and hearing status (p = .878). See Figure 1: The Grade 1 group produced a higher number of semantically related responses than the K group and, consequently, produced fewer non–semantically related responses.

Figure 1.

A bar graph with error bars depicting semantically and non-semantically related responses for 4 groups. The description lists the mean value of the bar and the top end of the error bar. The data for the semantically related responses is as follows. Typical hearing children from the early kindergarten group: 33, 53. Deaf and Hard of Hearing children from the early kindergarten group: 27, 43. Typical Hearing children from the post first grade group: 46, 58. Deaf and Hard of Hearing children from the post first grade group: 41, 60. The data for the non-semantically related responses is as follows. Typical hearing children from the early kindergarten group: 38, 57. Deaf and Hard of Hearing children from the early kindergarten group: 45, 60. Typical Hearing children from the post first grade group: 24, 36. Deaf and Hard of Hearing children from the post first grade group: 30, 48. All values are estimated.

Open in a new tab

Semantically versus non–semantically related responses by group. Error bars reflect standard deviations. TH-K = children with typical hearing, early kindergarten group; DHH-K = children who are deaf and hard of hearing, early kindergarten group; TH-1st = children with typical hearing, post–first-grade group; DHH-1st = children with hearing aids, post–first-grade group.

Non–semantically related responses. The second research question asked whether children who are DHH, when they produce non–semantically related responses, produce fewer clang responses than children with typical hearing. Recall that non–semantically related responses could be classified as an error, as no response to the prompt, or as a clang response. To evaluate the potential differences in the types of non–semantically related responses between groups, a multivariate analysis of variance was completed with proportion of each response type (error, no response, or clang) out of total non–semantically related responses as the dependent variable, and age and hearing status as between-subjects independent variables. Following a check for all necessary assumptions (correlations between dependent variables were in an acceptable, moderate range), the multivariate analysis yielded no significant effect of age, F(1, 104) = 1.67, p = .193; Wilks λ = .97, ηp2 = .02, or overall hearing status, F(1, 104) = 2.41, p = .09; Wilks λ = .96, ηp2 = .04, on type of errored response. Between-subjects effects showed that for the “errored” response type, there was a main effect of hearing status, F(1, 105) = 3.38, p = .050, ηp2 = .03, with children who are DHH producing a higher proportion of errored responses but no main effect of age, F(1, 105) = 3.12, p = .069, ηp2 = .02, and no interaction effect of hearing status and age, F(1, 105) = 2.25, p = .14, ηp2 = .02. For the “no response” type, there was a main effect of hearing status, F(1, 105) = 4.63, p = .03, ηp2 = .04, with children with typical hearing producing a higher proportion of “nonresponses” but no main effect of age, F(1, 105) = 1.39, p = .241, ηp2 = .01, nor an interaction effect between hearing status and age, F(1, 105) = 0.118, p = .731, ηp2 = .001. Finally, for “clang” responses, no main effects were observed for hearing status, F(1, 105) = 0.42, p = .520, ηp2 = .003, or age, F(1, 105) = 0.39, p = .530, ηp2 = .003, or interaction effect, F(1, 105) = 1.61, p = .21, ηp2 = .02. See Figure 2.

Figure 2.

A bar graph with error bars depicting 3 types of non-semantically related responses for 4 groups. The description lists the mean value and the top end of the error bar. The data for the errored response is as follows. Typical hearing children from the early kindergarten group: 11, 19. Deaf and Hard of Hearing children from the early kindergarten group: 23, 38. Typical Hearing children from the post first grade group: 13, 20. Deaf and Hard of Hearing children from the post first grade group: 14, 22. The data for nonresponses is as follows. Typical hearing children from the early kindergarten group: 18, 35. Deaf and Hard of Hearing children from the early kindergarten group: 15, 25. Typical Hearing children from the post first grade group: 9, 17. Deaf and Hard of Hearing children from the post first grade group: 8, 14. The data for the Clang response is as follows. Typical hearing children from the early kindergarten group: 8, 25. Deaf and Hard of Hearing children from the early kindergarten group: 7, 18. Typical Hearing children from the post first grade group: 3, 7. Deaf and Hard of Hearing children from the post first grade group: 10, 23. All values are estimated.

Open in a new tab

Classifications of non–semantically related responses by group. Error bars reflect standard deviations. TH-K = children with typical hearing, early kindergarten group; DHH-K = children who are deaf and hard of hearing, early kindergarten group; TH-1st = children with typical hearing, post–first-grade group; DHH-1st = children with hearing aids, post–first-grade group.

Taxonomically related responses. The third research question asked whether children who are DHH, when they produce semantically related responses, produce fewer taxonomically related responses than children with typical hearing. To evaluate the potential differences in the thematic and taxonomic responses between groups, a multivariate analysis of variance was completed with proportion of each response type (thematic or taxonomic) out of the number of semantically related responses as the dependent variable, and age and hearing status as between-subjects independent variables (see Figure 3). Following a check for all necessary assumptions (correlations between dependent variables were in an acceptable range), a multivariate effect was found for age, F(2, 104) = 5.50, p = .005, Wilks λ = .90, ηp2 = .05, but not for hearing status, F(2, 104) = 0.04, p = .955, Wilks λ = .99, ηp2 < .001. Between-subjects analysis effects yielded a main effect of age on thematic responses, F(1, 105) = 9.23, p = .003, ηp2 = .08, reflecting an increased proportion from K to first grade and on taxonomic responses, F(1, 105) = 5.69, p = .019, ηp2 = .05, reflecting a decrease in proportion from K to first grade but no main effect of hearing status on thematic responses, F(1, 105) = 0.07, p = .798, ηp2 = .007, or taxonomic responses, F(1, 105) = 0.001, p = .99, ηp2 < .001. There was no interaction effect between hearing status and age for thematic, F(1, 105) = 0.08, p = .773, ηp2 < .001, or taxonomic responses, F(1, 105) = 0.64, p = .424, ηp2 = .006.

Figure 3.

A bar graph with error bars depicting thematically and taxonomically related responses for 4 groups. The description lists the mean value and the top end of the error bar. The data for thematically related responses is as follows. Typical hearing children from the early kindergarten group: 27, 43. Deaf and Hard of Hearing children from the early kindergarten group: 20, 34. Typical Hearing children from the post first grade group: 39, 49. Deaf and Hard of Hearing children from the post first grade group: 36, 52. The data for taxonomically related responses is as follows. Typical hearing children from the early kindergarten group: 5, 11. Deaf and Hard of Hearing children from the early kindergarten group: 5, 10. Typical Hearing children from the post first grade group: 9, 12. Deaf and Hard of Hearing children from the post first grade group: 5, 9. All values are estimated.

Open in a new tab

Thematically versus taxonomically related responses by group. Error bars reflect standard deviations. TH-K = children with typical hearing, early kindergarten group; DHH-K = children who are deaf and hard of hearing, early kindergarten group; TH-1st = children with typical hearing, post–first-grade group; DHH-1st = children with hearing aids, post–first-grade group.

Paradigmatic responses. The fourth research question asked whether children who are DHH, when they produce semantically related responses, produce fewer paradigmatic responses as compared to children with typical hearing. To evaluate the potential differences in the syntagmatic and paradigmatic responses between groups, a multivariate analysis of variance was completed with proportion of each response type (syntagmatic or paradigmatic) out of total semantically related responses as the dependent variable, and age and hearing status as between-subjects independent variables (see Figure 4). Following a check for all necessary assumptions (correlations between dependent variables were in an acceptable range), a multivariate effect was found for age, F(2, 104) = 4.02, p = .021; Wilks λ = .92, ηp2 = .04, but not hearing status, F(2, 104) = 0.43, p = .650; Wilks λ = .99, ηp2 = .004. Between-subjects effects indicated a main effect of age on syntagmatic responses, F(1, 105) = 8.06, p = .005, ηp2 = .07, but not paradigmatic responses, F(1, 105) = 2.46, p = .119, ηp2 = .02, and no main effect of hearing status on syntagmatic responses, F(1, 105) = 0.84, p = .362, ηp2 = .007, or paradigmatic responses, F(1, 105) = 0.37, p = .543, ηp2 = .003. There was no interaction effect between hearing status and age for syntagmatic, F(1, 105) = 0.06, p = .814, ηp2 < .001, or paradigmatic responses, F(1, 105) = 0.36, p = .552, ηp2 = .003.

Figure 4.

A bar graph with error bars depicting syntagmatic and paradigmatic responses for 4 groups. The description lists the mean value and the top end of the error bar. The data for the syntagmatic responses is as follows. Typical hearing children from the early kindergarten group: 18, 30. Deaf and Hard of Hearing children from the early kindergarten group: 15, 24. Typical Hearing children from the post first grade group: 27, 35. Deaf and Hard of Hearing children from the post first grade group: 26, 38. The data for the Paradigmatic responses is as follows. Typical hearing children from the early kindergarten group: 15, 27. Deaf and Hard of Hearing children from the early kindergarten group: 14, 22. Typical Hearing children from the post first grade group: 20, 27. Deaf and Hard of Hearing children from the post first grade group: 16, 25. All values are estimated.

Open in a new tab

Syntagmatic and paradigmatic responses by group. Error bars reflect standard deviations. TH-K = children with typical hearing, early kindergarten group; DHH-K = children who are deaf and hard of hearing, early kindergarten group; TH-1st = children with typical hearing, post–first-grade group; DHH-1st = children with hearing aids, post–first-grade group.

Effects of Age on Responses

The fifth research question considered whether children across all hearing statuses increased their productions of semantically related, taxonomic, and paradigmatic responses over time. From the models used to analyze the first, third, and fourth research questions, main effects of age can be interpreted to answer this question. A main effect of age was found for a number of semantically related responses, proportion of thematic with a slight decrease in proportion of taxonomic responses, and syntagmatic (but not) paradigmatic responses, and no interaction effects were found for hearing status. Older groups, therefore, regardless of age, seemed to grow in responses that indicate increasing use of early organization systems. Additionally, analysis of nonsemantic responses indicated a pattern relative to hearing status: The proportion of errored held steady over time for the DHH group, and were higher than for the TH group (whereas the TH group was more likely to produce nonresponse types. Notably, as both groups increased the number of semantic responses (and therefore, decreased number of nonsemantic responses), the number of nonresponse items also dropped (see Figure 2).

Effect of Vocabulary on Responses

The sixth research question asked whether vocabulary size, once controlled for age, would predict semantically related, taxonomically related, and paradigmatic responses. The standard score of the EOWPVT-4, which controls for age by virtue of being a standard score on a norm-referenced assessment, was entered as a predictor variable into three linear regressions. The first used the number of semantically related responses as an outcome variable, the second used the number of taxonomic responses as an outcome variable, and the third used the number of paradigmatic responses as an outcome variable. Results indicated a significant effect of EOWPVT standard score on semantically related responses, F(1, 99) = 10.43, p = .002; on taxonomic responses, F(1, 99) = 10.73, p = .001; and paradigmatically related responses, F(1, 99) = 13.86, p < .001.

Discussion

The overall purpose of this study was to evaluate lexical–semantic organization, as measured by repeated word association, in children who are DHH as compared to children of the same age with typical hearing. Results from a group of children of children starting K and a group of children who had just finished first grade revealed that, regardless of hearing status, children appeared to increase semantically related responses, inclusive of thematically related responses and syntagmatic responses, to target words. Increases in semantically related responses also included an increased in the proportion of thematic and syntagmatic responses overall. Relatedly, the number of non–semantically related responses of the first-grade groups was lower than for the K groups. Children in the K and first grade group who were DHH produced a higher proportion of non–semantically related, errored (rather than clang or nonresponses) responses than the groups with TH. Thus, by the end of first grade, there was no evidence of some expected organizational shifts related to taxonomic and paradigmatic responses for either group. The present data provide a starting point for additional in-depth study.

Lexical–Semantic Organization Shifts in Children Who Are DHH

As expected, age did appear to influence children's responses to a repeated word association task. Cronin (2002) found that children with typical hearing and development shifted from relating words based on phonological properties (as evidenced by clang-type responses) to relating words based on semantic properties. In our data set, children in the younger groups produced fewer semantically related responses than in the older groups, consistent with a shift in semantic responses. Perhaps more interesting were the changes in types of non–semantically related responses produced by children based on hearing status: In the younger group, children who are DHH produced more errored responses than children with TH. Recall that error responses included responses such as repetition of the stimulus (saying “read” for target read or continues to produce the same response on the second or third repetition), inflections of the stimulus (saying “pushes” for target push), and real word responses with no semantic or phonotactic relations to the target (saying “headband” for target kick). Nonresponses included children responding with only a sound (saying “ah” for target pillow), responding with nonsense words (saying “bagoo” for target goat), and silent responses (child shakes head in response to target cow). In looking at proportion distributions of responses, children with TH were more likely to not respond to the prompt than to produce an errored response at that same age. In the older group, however, both DHH and TH children substantially decreased their total number of error and no-response responses, although proportions remained similar. Contrary to the authors' hypothesis, however, there was no statistical difference between groups in clang responses (although, in general, the TH children nearly eliminated that response type in the older group and children in the DHH group did not).

There is an expected shift from thematic to taxonomic organization for children with TH between first and third grade (e.g., McCauley et al., 1976). Because the oldest group had just finished first grade, the authors took the opportunity to explore whether a repeated word task might capture any shift in response type related to taxonomic responses. Although all groups had responses that fell in the taxonomically related category, the number of those responses did not shift with hearing status or with age group. The older age groups, however, did have an increased number of thematically related responses to prompts in comparison to the younger age groups. Thus, the hypothesis that children with TH and children in older groups might have more taxonomically related responses (or shift in proportion of responses) was not confirmed. This finding could reflect the age of participants in this study: The oldest group was on the lower end of the age range at which a shift is suspected to occur. It is also possible that a repeated word association task does not capture this shift well: Other studies tend to capture a shift in naming or “slot-filler” tasks, rather than in repeated associations. Nevertheless, an increase in thematically related responses to the prompt for older groups likely signals growing maturity in semantic organization.

Extant research in children with TH has also evaluated a syntagmatic–paradigmatic sequencing shift that occurs between first and third grade and can be captured by responses to a repeated word association task (Nelson, 1977; Sheng et al., 2006). Sheng et al. (2006) assessed responses of monolingual and bilingual children between the ages of 5 and 8 years, and rather than directly assessing paradigmatic versus syntagmatic responses, considered across- and within-child language comparisons and found that children in that age range do produce paradigmatic responses. Similarly, in this study, children in DHH and TH groups did produce paradigmatic responses, but those responses did not significantly increase by the end of first grade. However, children in the older groups did have more syntagmatic responses (and, therefore, a higher proportion of responses) than children in young groups. Again, this finding could reflect the age range of the present sample: It is possible that the percentage of paradigmatic responses is likely to increase as children move from first to third grade. Across all responses types and children who are DHH and TH, it appears that growth in semantic responses (vs. nonsemantic responses) is captured by an increase in thematic and syntagmatic responses through the end of first grade.

Vocabulary Knowledge in Children Who Are DHH

This study provides additional information related to the engagement subprocess of word-learning in children who are DHH and learning spoken language. A child's organization system for vocabulary likely influences storage and retrieval of words, alongside understanding of how words are related (e.g., Lund, 2022; Sheng & McGregor, 2010). Results of this study indicate that younger children who are DHH (in this case, children who are 5 years old) may have different recognition of semantic relationships between words as compared to children with TH. Children in the younger DHH groups were more likely to produce an errored response than children in the TH groups; that is, when children in the DHH group produced nonsemantic responses, they had a higher proportion that were not clearly in the semantic network of the target but still reflected use of a real word. Children with TH, when they produced nonsemantic responses, produced a higher proportion of responses that did not involve real words (or no response at all). This difference could reflect difficulty with the engagement process during vocabulary acquisition; young children who are DHH may be learning new words but not organizing those words using similar strategies as children with TH. The finding could also reflect an early difficulty with depth of word learning: Children who are DHH may be learning new word forms but not retaining as much additional semantic information about those forms as children with TH. Numerical differences between groups based on hearing status, however, diminished in the older groups. Children who are DHH may develop their retention and semantic organization strategies as they gain additional experience with spoken language. It is also possible that school entry corresponds to increased systematic instruction related to the relationships between words: Differences in age groups could reflect positive effects of classroom participation.

Notably, hearing-status differences in the groups' errored responses do not appear to solely reflect differences in quantity of words known. On single-word measures of vocabulary, children who are DHH in this study did have lower scores than children with TH. However, the majority of children who are DHH fell in the “range of normal” for vocabulary scores on those norm-referenced assessment and would not have qualified for speech and language intervention services based on those scores. This is also true for the older groups: Children who are DHH have lower scores than children with TH, but most fall in the normal range according to the assessment. Thus, single-word measures of vocabulary do not capture changes related to depth of word knowledge, inclusive of recognition of semantic relationships between words.

Single-word vocabulary knowledge, however, did predict child use of semantically related responses, of taxonomic responses, and of paradigmatic responses. That is, children who had higher standard scores on the EOWPVT (expressive vocabulary measure) tended to produce more responses that reflect maturity in lexical–semantic organization. Because standard scores are indexed to age, this finding means that children who know more words in comparison to a normative sample of their same age also tend to have more mature responses. This finding may provide a basis for future longitudinal work: If improving overall quantity of vocabulary knowledge also improves lexical semantic organization, this could be an important implication for intervention. It is also possible, however, that the reverse is true: that having a more mature lexical–semantic organization makes it easier to build (and access) a large vocabulary.

The findings of this study contribute to the growing body of literature considering lexical–semantic organization in children who are DHH. Studies that have used other methodologies like taxonomic naming (Lund & Dinsmoor, 2016), error analysis (Ground et al., 2014), and verbal fluency (Wechsler-Kashi et al., 2014) have all found notable differences between children who are DHH and TH. Those methodologies may serve to better capture shifts in certain kinds of lexical–semantic knowledge, like the shift away from phonologically based organization (Ground et al., 2014) or the thematic–taxonomic shift (Lund & Dinsmoor, 2016; Wechsler-Kashi et al., 2014). Repeated word association, however, is more likely to capture a syntagmatic–paradigmatic shift.

Similarly to Mann et al. (2016), children in the DHH and TH groups in this study had similar rates of paradigmatic and syntagmatic responses within each age range when both groups were tested in their first language, spoken English. The Mann et al. (2016) study did find a difference between those response types when children who were DHH were tested in English, their second language, as compared to when they were tested in ASL. That study also included children from a wider age range, up to 11 years old. Findings from this study may indicate then that DHH children who learn spoken English as a first language develop similar organization knowledge as children with TH, but they may also indicate that organizational differences are not apparent when children are between 7 and 8 years old (the age range of the older group in this study).

Limitations, Future Directions, and Conclusion

Limitations of this study provide potential future research directions. The cross-sectional design of this study was not ideal for drawing conclusions across age groups. Although certainly cross-sectional work can lead to further hypotheses about age-related changes in development, a stronger design will follow children who are DHH and with TH longitudinally to track individual growth. Through the ELLA study, the principal investigators are doing exactly that: The K groups in this study will continue to participate in a repeated word association task through third grade to capture individual trajectories of change over time.

The age range of the participants in this study may not have sufficiently captured the time period when lexical–semantic organizational shifts are expected to occur. Certainly, there is value in beginning to track child performance early in development, and this study captured between-groups differences in non–semantically related responses. A future study must, then, continue to track change over time at later ages to determine whether children who are DHH and children with TH are truly similar in development, or whether those groups differ in organization strategy shifts.

Finally, future work should consider the clinically relevant, wider implications of any between-groups differences. It was not within the scope of this study to evaluate how any organization differences impacted other domains of language and literacy acquisition. Additional studies must consider how differences and changes in an experimental task like repeated word association relate to academic and linguistic tasks that children complete in ecologically valid settings.

Conclusions

This study revealed that children, regardless of hearing status, grow in their ability to produce semantically related responses to repeated word association prompts between the ages of 5 and 8 years. This growth appeared to correspond to an increase in thematically related and syntagmatic responses. Children in the younger DHH group did appear to differ from children with TH in their production of non–semantically related responses: They tended to produce errored responses at higher rates, which may indicate an early deficit in recognition of semantic relations between words. A future study should employ a longitudinal study design with a representative sample of children who are DHH and learning spoken language that tracks development over a longer period of time.

Data Availability Statement

The data sets generated and/or analyzed during this study are available from the corresponding author upon reasonable request.

Acknowledgments

This work was funded by a grant from the National Institute on Deafness and Other Communication Disorders (R01DC017173; Principal investigators: Werfel and Lund). The contributions of the ELLA study team, particularly Brittany Grey, are gratefully acknowledged.

Funding Statement

This work was funded by a grant from the National Institute on Deafness and Other Communication Disorders (R01DC017173; Principal investigators: Werfel and Lund). The contributions of the ELLA study team, particularly Brittany Grey, are gratefully acknowledged.

References

  1. Cronin, V. S. (2002). The syntagmatic–paradigmatic shift and reading development. Journal of Child Language, 29(1), 189–204. 10.1017/S0305000901004998 [DOI] [PubMed] [Google Scholar]
  2. Di Giacomo, D., De Federicis, L. S., Pistelli, M., Fiorenzi, D., & Passafiume, D. (2012). Semantic associative relations and conceptual processing. Cognitive Processing, 13(1), 55–62. 10.1007/s10339-011-0399-7 [DOI] [PubMed] [Google Scholar]
  3. Dunn, D. M. (2019). Peabody Picture Vocabulary Test–Fifth Edition. NCS Pearson. [Google Scholar]
  4. Fulcher, A., Purcell, A. A., Baker, E., & Munro, N. (2012). Listen up: Children with early identified hearing loss achieve age-appropriate speech/language outcomes by 3years-of-age. International Journal of Pediatric Otorhinolaryngology, 76(12), 1785–1794. 10.1016/j.ijporl.2012.09.001 [DOI] [PubMed] [Google Scholar]
  5. Geers, A., & Moog, J. (2012). CID Early Speech Perception Test. Central Institute for the Deaf. [Google Scholar]
  6. Ground, W., Lund, E., Werfel, K. L., & Schuele, C. M. (2014). Auditory closure performance of preschool children with and without hearing loss [Poster presentation]. American Speech-Language-Hearing Association Convention, Orlando, FL. [Google Scholar]
  7. Hoover, J. R., Storkel, H. L., & Hogan, T. P. (2010). A cross-sectional comparison of the effects of phonotactic probability and neighborhood density on word learning by preschool children. Journal of Memory and Language, 63(1), 100–116. 10.1016/j.jml.2010.02.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hornsby, B. W., Naylor, G., & Bess, F. H. (2016). A taxonomy of fatigue concepts and their relation to hearing loss. Ear and Hearing, 37(1), 136S–144S. 10.1097/AUD.0000000000000289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. House, A. E., House, B. J., & Campbell, M. B. (1981). Measures of interobserver agreement: Calculation formulas and distribution effects. Journal of Behavioral Assessment, 3(1), 37–57. 10.1007/BF01321350 [DOI] [Google Scholar]
  10. Hresko, W. P., Reid, D. K., & Hammill, D. D. (1999). Test of Early Language Development–Third Edition. Pro-Ed. [Google Scholar]
  11. Kenett, Y. N., Wechsler-Kashi, D., Kenett, D. Y., Schwartz, R. G., Ben-Jacob, E., & Faust, M. (2013). Semantic organization in children with cochlear implants: Computational analysis of verbal fluency. Frontiers in Psychology, 4, Article 543. 10.3389/fpsyg.2013.00543 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Leach, L., & Samuel, A. G. (2007). Lexical configuration and lexical engagement: When adults learn new words. Cognitive Psychology, 55(4), 306–353. 10.1016/cogpsych.2007.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Lederberg, A. R., Prezbindowski, A. K., & Spencer, P. E. (2000). Word-learning skills of deaf preschoolers: The development of novel mapping and rapid word-learning strategies. Child Development, 71(6), 1571–1585. 10.1111/1467-8624.00249 [DOI] [PubMed] [Google Scholar]
  14. Lippman, M. Z. (1971). Correlates of contrast word associations: Developmental trends. Journal of Verbal Learning and Verbal Behavior, 10(4), 392–399. 10.1016/S0022-5371(71)80038-5 [DOI] [Google Scholar]
  15. Lucariello, J., Kyratzis, A., & Nelson, K. (1992). Taxonomic knowledge: What kind and when? Child Development, 63(4), 978–998. 10.2307/1131248 [DOI] [Google Scholar]
  16. Luce, P. A., & Pisoni, D. B. (1998). Recognizing spoken words: The neighborhood activation model. Ear and Hearing, 19(1), 1–36. 10.1097/00003446-199802000-00001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lund, E. (2016). Vocabulary knowledge of children with cochlear implants: A meta-analysis. Journal of Deaf Studies and Deaf Education, 21(2), 107–121. 10.1093/deafed/env060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lund, E. (2018). Pairing new words with unfamiliar objects: Comparing children with and without cochlear implants. Journal of Speech, Language, and Hearing Research, 61(9), 2325–2336. 10.1044/2018_JSLHR-L-17-0467 [DOI] [PubMed] [Google Scholar]
  19. Lund, E. (2020). Triggering, configuration, and engagement: Spoken-word learning processes in young children with hearing loss. In Marschark M. & Knoors H. (Eds.), The Oxford handbook of deaf studies in learning and cognition (pp. 138–149). Oxford University Press. 10.1093/oxfordhb/9780190054045.013.5 [DOI] [Google Scholar]
  20. Lund, E. (2022). Phonological priming as a lens for phonological organization in children with cochlear implants. Ear and Hearing, 43(4), 1355–1365. 10.1097/AUD.0000000000001191 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lund, E., Brock, N., & Werfel, K. L. (2022). Location effects on spoken language and literacy for children who are DHH. Journal of Deaf Studies and Deaf Education, 27(1), 48–61. 10.1093/deafed/enab030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lund, E., & Dinsmoor, J. (2016). Taxonomic knowledge of children with and without cochlear implants. Language, Speech, and Hearing Services in Schools, 47(3), 236–245. 10.1044/2016_LSHSS-15-0032 [DOI] [PubMed] [Google Scholar]
  23. Mann, W., Sheng, L., & Morgan, G. (2016). Lexical–semantic organization in bilingually developing deaf children with ASL-dominant language exposure: Evidence from a repeated meaning association task. Language Learning, 66(4), 872–899. 10.1111/lang.12169 [DOI] [Google Scholar]
  24. Marschark, M., Convertino, C., McEvoy, C., & Masteller, A. (2004). Organization and use of the mental lexicon by deaf and hearing individuals. American Annals of the Deaf, 149(1), 51–61. 10.1353/aad.2004.0013 [DOI] [PubMed] [Google Scholar]
  25. Martin, N. A., & Brownell, R. (2011). Expressive One-Word Picture Vocabulary Test–Fourth Edition. Pro-Ed. [Google Scholar]
  26. McCauley, C., Weil, C. M., & Sperber, R. D. (1976). The development of memory structure as reflected by semantic-priming effects. Journal of Experimental Child Psychology, 22(3), 511–518. 10.1016/0022-0965(76)90113-2 [DOI] [Google Scholar]
  27. McCreery, R. W., & Walker, E. A. (2022). Variation in auditory experience affects language and executive function skills in children who are hard of hearing. Ear and Hearing, 43(2), 347–360. 10.1097/AUD.0000000000001098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nelson, K. (1977). The syntagmatic–paradigmatic shift revisited: A review of research and theory. Psychological Bulletin, 84(1), 93–116. 10.1037/0033-2909.84.1.93 [DOI] [PubMed] [Google Scholar]
  29. Nelson, K., & Nelson, A. P. (1990). Category production in response to script and category cues by kindergarten and second-grade children. Journal of Applied Developmental Psychology, 11(4), 431–466. 10.1016/0193-3973(90)90019-G [DOI] [Google Scholar]
  30. Nicholas, J. G., & Geers, A. E. (2007). Will they catch up? The role of age at cochlear implantation in the spoken language development of children with severe to profound hearing loss. Journal of Speech, Language, and Hearing Research, 50(4), 1048–1062. 10.1044/1092-4388(2007/073) [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Persson, A., Marklund, U., Lohmander, A., & Flynn, T. (2022). Expressive vocabulary development in children with moderate hearing loss—The impact of auditory variables and early consonant production. Clinical Linguistics & Phonetics, 36(6), 547–564. 10.1080/02699206.2021.1944321 [DOI] [PubMed] [Google Scholar]
  32. Rush, O. (2022). Lexical–semantic organization as measured by repeated word association in children with hearing loss [Master's thesis Texas Christian University]. TCU Digital Repository. https://repository.tcu.edu/handle/116099117/52664? [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sheng, L., & McGregor, K. K. (2010). Lexical–semantic organization in children with specific language impairment. Journal of Speech, Language, and Hearing Research, 53(1), 146–159. 10.1044/1092-4388(2009/08-0160) [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sheng, L., McGregor, K. K., & Marian, V. (2006). Lexical–semantic organization in bilingual children: Evidence from a repeated word association task. Journal of Speech, Language, and Hearing Research, 49(3), 572–587. 10.1044/1092-4388(2006/041) [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. U.S. Food and Drug Administration. (2022). Cochlear implants: A different kind of ‘hearing.' Retrieved December 9, 2022, from https://www.fda.gov/consumers/consumer-updates/cochlear-implants-different-kind-hearing
  36. Waxman, S. R., & Gelman, R. (1986). Preschoolers' use of superordinate relations in classification and language. Cognitive Development, 1(2), 139–156. 10.1016/S0885-2014(86)80016-8 [DOI] [Google Scholar]
  37. Waxman, S. R., & Hatch, T. (1992). Beyond the basics: Preschool children label objects flexibly at multiple hierarchical levels. Journal of Child Language, 19(1), 153–166. 10.1017/S0305000900013672 [DOI] [PubMed] [Google Scholar]
  38. Wechsler-Kashi, D., Schwartz, R. G., & Cleary, M. (2014). Picture naming and verbal fluency in children with cochlear implants. Journal of Speech, Language, and Hearing Research, 57(5), 1870–1882. 10.1044/2014_JSLHR-L-13-0321 [DOI] [PubMed] [Google Scholar]
  39. Werfel, K. L., Reynolds, G., Hudgins, S., Castaldo, M., & Lund, E. A. (2021). The production of complex syntax in spontaneous language by 4-year-old children with hearing loss. American Journal of Speech-Language Pathology, 30(2), 609–621. 10.1044/2020_AJSLP-20-00178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wiig, E. H., Semel, E., & Secord, W. A. (2013). Clinical Evaluation of Language Fundamentals–Fifth Edition (CELF-5). NCS Pearson. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data sets generated and/or analyzed during this study are available from the corresponding author upon reasonable request.

Articles from Journal of Speech, Language, and Hearing Research : JSLHR are provided here courtesy of American Speech-Language-Hearing Association

RESOURCES