View in EDS HTML Full Text PDF Full Text

Time-Course of Grammatical Processing in Deaf Readers: An Eye-Movement Study

Saved in:
Bibliographic Details
Title: Time-Course of Grammatical Processing in Deaf Readers: An Eye-Movement Study
Language: English
Authors: Gómez-Merino, Nadina, Fajardo, Inmaculada, Ferrer, Antonio, Arfé, Barbara
Source: Journal of Deaf Studies and Deaf Education. Jul 2020 25(3):351-364.
Availability: Oxford University Press. Great Clarendon Street, Oxford, OX2 6DP, UK. Tel: +44-1865-353907; Fax: +44-1865-353485; e-mail: jnls.cust.serv@oxfordjournals.org; Web site: http://jdsde.oxfordjournals.org/
Peer Reviewed: Y
Page Count: 14
Publication Date: 2020
Document Type: Journal Articles
Reports - Research
Descriptors: Deafness, Eye Movements, Grammar, Accuracy, Comparative Analysis, Syntax, Vocabulary Development, Language Processing, Sentences, Hearing (Physiology), Decision Making, Reading Rate, Reading Processes, Task Analysis
DOI: 10.1093/deafed/enaa005
ISSN: 1081-4159
Abstract: Twenty participants who were deaf and 20 chronological age-matched participants with typical hearing (TH) (mean age: 12 years) were asked to judge the correctness of written sentences with or without a grammatically incongruent word while their eye movements were registered. TH participants outperformed deaf participants in grammaticality judgment accuracy. For both groups, First Pass and Total Fixation Times of target words in correct trials were significantly longer in the incongruent condition than in the congruent one. However, whereas TH students showed longer First Pass in the target area than deaf students across congruity conditions, deaf students made more fixations than their TH controls. Syntactic skills, vocabulary, and word reading speeds (measured with additional tests) were significantly lower in deaf students but only syntactic skills were systematically associated to the time-course of congruity processing. These results suggest that syntactic skills could have a cascading effect in sentence processing for deaf readers.
Abstractor: As Provided
Entry Date: 2020
Accession Number: EJ1256381
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGe8dHljpwz4k6WKF9hkLFUAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDPCKbP4JzZshnhmZaAIBEICBmlnYbu05oJIZh5TajNo1yeV5u7EuOPr-Ovy7Ed6aUX5P7uJwK3S2m-Z2B713cxet_qzRYy5o59S-ZlmAtJzDIjhrKM_tuG_Su49MAgDSD1jVKseea5F8OfvQodVKqq05ef1y94MEMWnr4jxq0xZgR5vhDuFE9WIfFNcCGdEXGBl4NP7puPG8Mx5zpaeKQsS9VYuXfo58bJbfZNI=
Text:
  Availability: 1
  Value: <anid>AN0143509754;big01jul.20;2020Jun02.06:11;v2.2.500</anid> <title id="AN0143509754-1">Time-Course of Grammatical Processing in Deaf Readers: An Eye-Movement Study </title> <sbt id="AN0143509754-2">Background</sbt> <p>Twenty participants who were deaf and 20 chronological age-matched participants with typical hearing (TH) (mean age: 12 years) were asked to judge the correctness of written sentences with or without a grammatically incongruent word while their eye movements were registered. TH participants outperformed deaf participants in grammaticality judgment accuracy. For both groups, First Pass and Total Fixation Times of target words in correct trials were significantly longer in the incongruent condition than in the congruent one. However, whereas TH students showed longer First Pass in the target area than deaf students across congruity conditions, deaf students made more fixations than their TH controls. Syntactic skills, vocabulary, and word reading speeds (measured with additional tests) were significantly lower in deaf students but only syntactic skills were systematically associated to the time-course of congruity processing. These results suggest that syntactic skills could have a cascading effect in sentence processing for deaf readers.</p> <p>Previous studies have suggested that children and adults with deafness are sensitive to morphological and syntactic cues during sentence reading ([<reflink idref="bib7" id="ref1">7</reflink>]; [<reflink idref="bib38" id="ref2">38</reflink>]). However, the time-course of grammar processing could differ in comparison to typically hearing (TH) peers ([<reflink idref="bib7" id="ref3">7</reflink>]), the focus of the present study.</p> <p>Researchers have widely documented the low reading achievement exhibited by students with a prelingual severe-to-profound deafness compared to their typically hearing peers (e.g., [<reflink idref="bib9" id="ref4">9</reflink>]; [<reflink idref="bib37" id="ref5">37</reflink>]; [<reflink idref="bib40" id="ref6">40</reflink>]; [<reflink idref="bib48" id="ref7">48</reflink>]). In this regard, [<reflink idref="bib21" id="ref8">21</reflink>]) recently conducted a longitudinal study with children with severe-to-profound prelingual hearing loss and TH children matched in nonverbal IQ. Participants in the age range of 5–7 years old were asked to complete a set of reading related tests at three time points, 1 year apart (T1–T3): single word reading, reading comprehension, English vocabulary, phonological awareness, and speech reading. Their results showed no growth in reading comprehension from T2 to T3 in children with deafness, regardless of the hearing device used (cochlear implant [CI] or hearing aids [HAs]), whereas TH children's reading progressed according to their chronological age (CA).</p> <p>However, the systematic review of [<reflink idref="bib32" id="ref9">32</reflink>]) illustrated an encouraging trend shift. They analyzed the results of 21 studies over a 20-year time period (1997–2016), on the literacy outcomes of deaf students with CIs. The review focused on two types of literacy outcomes: reading comprehension and written expression. With respect to reading comprehension, their review pointed out that the majority of deaf students with CIs (from primary to secondary school) were in the average range. Only three studies (out of 14 that included standard scores [SSs]) showed that deaf participants were significantly below the average.</p> <p>Despite this promising trend in the performance of deaf students with CIs, [<reflink idref="bib32" id="ref10">32</reflink>]) highlighted that there was a lack of research about the factors that modulate the impact of CIs in the acquisition and development of reading comprehension skills in deaf students, such as communication modality or linguistic abilities before implant.</p> <p>In this regard, linguistic factors such as phonological coding abilities or vocabulary have been found to predict the reading achievement among individuals with prelingual deafness ([<reflink idref="bib21" id="ref11">21</reflink>]; [<reflink idref="bib31" id="ref12">31</reflink>]; [<reflink idref="bib37" id="ref13">37</reflink>]). However, grammatical skills seem to play a more critical role in it (e.g., [<reflink idref="bib3" id="ref14">3</reflink>]; [<reflink idref="bib36" id="ref15">36</reflink>]). Therefore, the present study aimed to investigate the time-course of grammatical processing during sentence reading of deaf children and adolescents. By using both off-line (accuracy) and online (eye movements) measures of reading comprehension, we attempted to carry out a more comprehensive analysis of this phenomenon.</p> <hd id="AN0143509754-3">Grammatical Skills and Reading Comprehension in Deaf Students</hd> <p>Grammatical competence means the understanding and use of morphology and syntax in a sentence, that is, how words change their form and combine with other words to make sentences ([<reflink idref="bib20" id="ref16">20</reflink>]). Although morphology and syntax interact (e.g., to convert a passive sentence into an active sentence, both types of changes, morphological and syntactical are needed), the term "syntax" is typically used to refer to both terms in the literature reviewed in this study.</p> <p>[<reflink idref="bib41" id="ref17">41</reflink>]) conducted one of the first large scale studies on grammar competence in deaf students by means of a standardized test of English syntactic abilities. They found that deaf children (10–18 years old) had difficulties understanding different syntactic structures compared to their TH peers. Later, [<reflink idref="bib26" id="ref18">26</reflink>]), using a standardized test designed by [<reflink idref="bib42" id="ref19">42</reflink>]), observed a bottleneck effect of syntax in reading comprehension for deaf readers. Around 400 adolescents and young students of secondary and postsecondary programs with severe-to-profound hearing loss were asked to complete three linguistic tests, assessing respectively syntax, vocabulary and reading comprehension. The results showed that only those deaf participants in the highest quartile of syntactic skills were able to use their vocabulary competence to comprehend the text. Thus, unless participants had a sufficient level of syntax competence (between 84% and 100% of correct answers in the [<reflink idref="bib42" id="ref20">42</reflink>] test), other linguistic skills involved in reading comprehension, such as lexical knowledge, could not be brought into play for deaf students.</p> <p>More recently, Miller and collaborators ([<reflink idref="bib34" id="ref21">34</reflink>]; [<reflink idref="bib35" id="ref22">35</reflink>]; [<reflink idref="bib36" id="ref23">36</reflink>]), based on results of a series of controlled experiments, found that a structural (syntactic) knowledge deficit was the main factor related to the poor comprehension skills of prelingually deaf individuals in languages different to English, such as Hebrew and Turkish. The experimental series was aimed not only to test the syntactic knowledge deficit in deaf students, but also if they compensated such a deficit by showing a preference for a top–down processing of content words during sentence comprehension as described below.</p> <p>In the most recent study, [<reflink idref="bib36" id="ref24">36</reflink>]) asked a group of Turkish readers (from 3rd to 10th grade) with severe-to-profound hearing loss and a control group of readers with TH, matched with regard to their level of education and average CA, to complete a sentence comprehension test. The test consisted of written sentences varying in syntactic complexity (one relative clause versus two relative clauses) and semantic plausibility (plausible versus implausible), each followed by a multiple-choice comprehension question. The processing of semantically implausible sentences (e.g., the police officer who was stealing a gun) required syntactic bottom-up processing of the words (both content and function words), whereas the semantically plausible sentences (e.g., the thief who was stealing a gun) could be understood by top–down processing of content words. Hence, it was expected that the syntactic complexity would affect more prelingually deaf participants, particularly in the implausible condition where the top–down processing was not usable. As anticipated, sentence syntactic complexity affected the deaf group more than the TH group, but the semantic plausibility effect did not interact with hearing status. Overall, participants comprehended semantically plausible sentences to a higher degree than semantically implausible ones. This finding supported the hypothesis of a deficit in syntactic competence of Turkish deaf readers from 3rd to 10th grade. However, deaf students did not seem to compensate for their lack of syntactic cue understanding by using a top–down processing of content words, because their performance remained below the one of the control groups in both plausibility conditions.</p> <p>With regard to Spanish, Domínguez and collaborators (e.g., [<reflink idref="bib14" id="ref25">14</reflink>]; [<reflink idref="bib46" id="ref26">46</reflink>]) have postulated the "Key Word Strategy" hypothesis in deaf readers that is similar to the hypothesis of top–down processing strategy ([<reflink idref="bib34" id="ref27">34</reflink>]). In their experimental series with deaf participants, they used the Syntactic Strategies Assessment Test (SSAT) in which participants complete the last word of a sentence with one of four options: the target word (semantic and syntactically appropriate) or three semantic distractors (semantically plausible but syntactically incorrect). As [<reflink idref="bib18" id="ref28">18</reflink>] noted: The rationale of the test is that selecting the proper alternative (target word) requires the use of syntactic and semantic cues. If the reader selects a wrong alternative, this indicates that the sentence is being processed superficially because the reader is guided solely by global semantic cues related to the meaning of some word in the sentence. Errors are thus interpreted as a preference for using semantic strategies as opposed to an accurate analysis of the syntactic relationships among the content words. However, the test has a drawback in that the alternatives to the correct answer are all semantic foils [distractors]. (p. 156)</p> <p>Therefore, to overcome this limitation, [<reflink idref="bib18" id="ref29">18</reflink>] modified the test of [<reflink idref="bib46" id="ref30">46</reflink>]) by adding a syntactic distractor (an option syntactically correct but semantically implausible). Their sample was composed of three groups of Spanish deaf students: participants with TH, deaf participants who received a CI before 24 months of age (early implanted) and participants with deafness, who received a CI after 24 months of age (late implanted). They found that early implanted children outperformed late implanted children and behaved similarly to age-matched TH children except for long sentences with infrequent target content words. The error analyses showed that early implanted and typically hearing children selected the syntactic distractor (syntactically correct but semantically implausible) more frequently than the semantic distractors (semantically correct but syntactically inaccurate). That is, they were primarily guided by syntactic cues rather than by semantic ones. By contrast, late implanted children showed no preference for distractors. In short, early implanted children showed preferences for syntactic cues and late implanted children did not show preferences for semantic cues. As [<reflink idref="bib18" id="ref31">18</reflink>]) suggested, these findings could be interpreted as a lack of support for the Key Word Strategy and [<reflink idref="bib34" id="ref32">34</reflink>]) hypothesis of top–down processing preference in deaf children.</p> <p>Also in Spanish, [<reflink idref="bib3" id="ref33">3</reflink>]) examined the factors influencing reading comprehension skills among 47 Spanish 6- to 13-year-old students with a severe-to-profound prelingual deafness. Their analysis focused on both the impact of external variables (type of assistive technology, age at fitting, family socioeconomic status, and school age) and variables internal to the reading process (decoding, vocabulary, and receptive grammar) in reading comprehension. Contrary to the results obtained by [<reflink idref="bib18" id="ref34">18</reflink>]), the age at fitting did not predict the differences between deaf students with and without reading comprehension difficulties. Receptive grammar was the only variable that significantly contributed to explaining the differences in text comprehension performance within the deaf group.</p> <p>In a recent study, [<reflink idref="bib39" id="ref35">39</reflink>]) analyzed the relationship between the receptive syntactical skills and reading comprehension abilities in Persian. Their participants were 15 deaf students (from 3rd to 5th grade) with CIs and 15 TH controls. Regression analysis showed that in Persian, reading comprehension increases with the increase of syntactic comprehension for both deaf participants with CIs and TH participants, whereas the age of receiving a CI and the duration of speech therapy did not predict reading comprehension in the CIs group.</p> <p>To sum up, studies conducted in different languages (e.g., English, Persian, Turkish, and Spanish) suggested that the syntactic deficit of deaf students could be one of the most significant contributors to their reading comprehension difficulties. However, it is not evident whether they simply ignore grammatical cues during reading or if they process these cues in a different manner. A limitation of revised empirical studies is that they have focused on the output of grammatical understanding rather than on the process, that is, how the morphosyntactic cues are processed when they are encountered in a sentence. The next section focuses on the time-course processing of grammatical cues in deaf readers, the focus of this study.</p> <hd id="AN0143509754-4">Time-Course of Grammatical Processing in Deaf Readers</hd> <p>To our knowledge, only two studies to date ([<reflink idref="bib7" id="ref36">7</reflink>]; [<reflink idref="bib38" id="ref37">38</reflink>]) have investigated how deaf students process syntactic anomalies while they read sentences (online processing).</p> <p>Thus, in the study by [<reflink idref="bib7" id="ref38">7</reflink>]), the sample was composed of three groups: (a) deaf adolescents (users of a combination of both British Sign Language and oral English as communication modes in school), (b) a CA matched TH group, and (c) a reading age (RA)-matched TH group. In the first experiment, participants carried out a self-paced sentence reading task (with words presented one at a time, without the possibility to revisit previously read words) with two types of agreement violations (grammatical, subject-verb number agreement versus semantic, and implausible noun-verb combination). In the second experiment, participants had to explicitly detect and correct agreement violations in sentences. The results of the first experiment showed that all three groups were sensitive to semantic implausibility, presenting the same time-course of effects: increment in reading times started in the verb (V) and continued in the subsequent words, called V + 1 (first word after the verb) and V + 2 (second word after the verb). Although deaf participants were sensitive to subject-verb number agreements, they were so in a different manner than the control group. Although the CA- and RA-matched children's reading time increased immediately on the verb where the error was (with a small increment on the next word to the verb, that is, V + 1, for the CA group), deaf children's reading times did not increase until V + 1 and V + 2. The authors suggested that, although students with deafness were able to detect the agreement errors, the way they processed the agreements differed from that of TH readers, because they did it later in time (in the post-target region [V1 and V2] instead of in the target region [TAR]). In the second experiment, the same groups of participants were requested to explicitly detect and correct the grammatical violations in the sentences of experiment 1. The TH participants were able to solve the grammatical errors but not deaf participants. Thus, deaf adolescents demonstrated implicit sensitivity to grammatical agreement but less robust explicit knowledge of it. [<reflink idref="bib7" id="ref39">7</reflink>]) concluded that deaf adolescents appeared to adjust to a semantic-first account (the analysis of the semantic cues initiated earlier than the syntactic one). However, TH children, especially in the case of mature readers like adolescents, adjusted to a grammatical-first account, that is, a grammatical agreement assessment would precede or, at least, finish before the semantic analysis ends ([<reflink idref="bib25" id="ref40">25</reflink>]).</p> <p>The second recent examination of the online processing of syntactic and semantic cues during reading by deaf readers was the study by [<reflink idref="bib38" id="ref41">38</reflink>]). They asked 39 young adults (18–33 years old) who were bilingual in American Sign Language and English spoken language to read sentences with relative clauses while their eye movements were recorded. Two variables were manipulated: (a) the role of a noun phrase (NP) (subject versus object of the verb inside a relative clause) and (b) the animacy of the NP (animated NP [e.g. hiker] versus inanimate NP [e.g. avalanche]). Specifically, the following four types of relative clauses were used ([<reflink idref="bib38" id="ref42">38</reflink>], p. 983–984).</p> <p></p> <ulist> <item> Subject-animate: The hikers that fled the avalanche appeared on the six o'clock news.</item> <p></p> <item> Object-animate: The hikers that the avalanche buried appeared on the six o'clock news.</item> <p></p> <item> Subject-inanimate: The avalanche that buried the hikers appeared on the six o'clock news.</item> <p></p> <item> Object-inanimate: The avalanche that the hikers fled appeared on the six o'clock news.</item> </ulist> <p>The critical region for analysis was the relative clause region following the relative pronoun <emph>that</emph>. According to a previous finding with TH readers ([<reflink idref="bib49" id="ref43">49</reflink>]), object relative sentences are more difficult to process than subject relative sentences (as evidenced by slower reading times of the critical region in the former) but only with the animated NP (<emph>hiker</emph>), which is not expected to occupy an object position. [<reflink idref="bib49" id="ref44">49</reflink>]) suggested that such an interaction between animacy and role of the NP demonstrated the interplay of syntactic and semantic cues in written sentence processing. Consequently, [<reflink idref="bib38" id="ref45">38</reflink>]) predicted that if deaf readers made use of syntactic and semantic information as TH readers, they would show longer reading times and more eye-movement regressions in the object relative clause sentences, but only with the animate NP (example 2). Indeed, results showed that bilingual deaf adults showed longer total reading times in the TAR (the relative clause region following the complementizer <emph>that</emph>) for animate object relative clauses compared to animate subject relative clauses, whereas no differences between object versus subject conditions were found for inanimate NP sentences. Therefore, these findings suggest that deaf readers, like TH readers, are sensitive to both syntactic and semantic elements of sentences.</p> <p>To sum up, the aforementioned studies seem to confirm that children, adolescents, and young adults with prelingual severe-to-profound deafness (both sign language users and orally educated) are sensitive to morphological and syntactic cues during reading ([<reflink idref="bib7" id="ref46">7</reflink>]; [<reflink idref="bib38" id="ref47">38</reflink>]). Thus, the present study aimed to further explore the time-course of syntactic cue processing during sentence reading in children and adolescents with deafness, overcoming some limitations of these first studies. First, the procedure used in the study by [<reflink idref="bib7" id="ref48">7</reflink>]), where words were presented one at a time, does not allow drawing inferences on natural reading processes in which readers are exposed to the full text and can read back and forth. Unlike [<reflink idref="bib7" id="ref49">7</reflink>]), we used an in toto presentation mode, which allows analyzing rereading behavior (once the reader encounters a difficult word in a text or sentence) ([<reflink idref="bib2" id="ref50">2</reflink>]; [<reflink idref="bib30" id="ref51">30</reflink>]; [<reflink idref="bib33" id="ref52">33</reflink>]; [<reflink idref="bib22" id="ref53">22</reflink>]).</p> <p>Second, the lack of a TH control in the study of [<reflink idref="bib38" id="ref54">38</reflink>]) makes it difficult to determine if the interaction between semantic and syntactic cues observed in deaf readers were specific of this group of readers or, alternatively, differed in magnitude with regard to TH readers. Focusing only on syntactic cues, we have added to our design a TH control matched with regard to CA, in order to identify differences, if any, in magnitude and/or time-course between deaf and TH readers in the processing of these types of cues during reading.</p> <p>Third, in contrast to [<reflink idref="bib7" id="ref55">7</reflink>]) and [<reflink idref="bib38" id="ref56">38</reflink>]), who investigated the reading processes of adolescents and adult readers, we aimed at studying the online processing of grammatical cues in primary and secondary students who were still developing their reading strategies. Moreover, unlike prior studies, our participants were mainly CI users who, in general, received an oral education and exposed primarily to oral language.</p> <p>To recap, prelingually deaf students have reading comprehension difficulties directly related to a difficulty with grammatical information. Nevertheless, it remains unclear if this difficulty makes them use an alternative strategy in sentence processing, where they skip or "ignore" the information provided by grammatical cues in sentences, as suggested by some hypotheses (e.g., [<reflink idref="bib14" id="ref57">14</reflink>]) or if they simply put more effort into processing this information compared to their TH peers, because of their low grammatical ability.</p> <hd id="AN0143509754-5">The Present Study</hd> <p>Therefore, the purpose of this study was to examine the reading behavior of young orally educated deaf students by analyzing their eye movements while taking part in a grammatical judgment task during sentence reading.</p> <p>Our research questions were as follows: (a) Are orally educated deaf children and adolescents as accurate as TH students when detecting sentences with grammatical incongruence? (b) Do they use the same online processing behavior as TH students when reading sentences to judge their correctness?</p> <p>Based on prior studies, we made the following predictions: (a) deaf readers would be less able than TH readers to detect the grammatical incongruence of the sentences read (e.g., [<reflink idref="bib3" id="ref58">3</reflink>]; [<reflink idref="bib7" id="ref59">7</reflink>]; [<reflink idref="bib26" id="ref60">26</reflink>]; [<reflink idref="bib39" id="ref61">39</reflink>]) and (b) deaf readers' sentence processing would differ from that of TH readers for sentences with grammatical incongruences. Specifically, we expected no immediate significant effects of grammatical violations on the target area (incongruent word) for the deaf readers, but delayed effects in terms of number and duration of visits and revisits to pre- and post-target areas. This prediction of a delayed effect of grammatical congruity in the post-target area was based on previous findings by [<reflink idref="bib7" id="ref62">7</reflink>]). His results found that deaf participants were sensitive to grammar errors but later in time compared to TH readers. Whereas TH readers spent longer time reading the word that contained the grammar error (the verb region), deaf readers increased their reading time of the subsequent two words after the grammar error. These results suggested that grammatical analysis happened later in deaf readers than in TH readers. As mentioned, no previous research has explored the processing and reanalyses of pretarget areas (area before the grammar error) in deaf readers. However, earlier studies with other populations ([<reflink idref="bib2" id="ref63">2</reflink>]; [<reflink idref="bib30" id="ref64">30</reflink>]; [<reflink idref="bib33" id="ref65">33</reflink>]; [<reflink idref="bib22" id="ref66">22</reflink>]) suggest that any kind of sentence reanalysis measures (specifically, eye-movements measures), which serves as an index of what the reader does when he/she encounters a difficult word, could help to distinguish different reading patterns.</p> <hd id="AN0143509754-6">Methods</hd> <p></p> <hd id="AN0143509754-7">Participants</hd> <p></p> <hd id="AN0143509754-8">Participants who were deaf</hd> <p>Participants with deafness were recruited from different audiology services of public hospitals, educational centers (mainstream schools with special units for students with deafness) and associations of people with deafness in Valencia (Spain).</p> <p>The inclusion criteria for the study were strict. Participants with deafness were selected if they presented: (a) prelingual and bilateral severe-to-profound hearing loss (Bureau International d'Audiophonologie's criteria, BIAP, 1997); (b) spoken language (Spanish) as their primary communication mode; (c) nonverbal IQ on average (≥85) for their age, and (d) age-appropriate word and nonword decoding skills. Additionally, all deaf students showed an intact or corrected-to-normal vision and reported no additional difficulties that could interfere with their performance on tasks. All deaf students were orally educated and used spoken language as their preferred means of communication. None of them had knowledge about sign language.</p> <p>Twenty-six participants were initially selected following these criteria. Of these, two were excluded because of visual difficulties (squint), two because of comorbidities (Usher syndrome, hydrocephaly, and truancy) reported at time of testing, one for using Spanish as a second language and one due to lack of information on educational background, as consequence of a late adoption.</p> <p>After this further selection, the final sample of deaf students included 20 participants (12 females and 8 males) with a mean age of 12.05 (range 9.6–15.2 years) from 4th to 10th grade. Of the 20 deaf participants included in the study, the majority (<emph>n</emph> = 16) had profound hearing loss and four had severe hearing loss, as reported in their clinical or school records. Twelve participants were fitted with CIs (eight bilateral and four unilateral), four were equipped with HA and four received a bimodal stimulation (HA + CI). Regarding those participants who were fitted with CI, 10 received their first CI before the age of two, and six after that age, with the mean age of first implantation being 3 years and 2 months.</p> <hd id="AN0143509754-9">Participants with typical hearing</hd> <p>Twenty students (11 girls) with typical hearing (TH) and a mean age of 12.00 (range 8.10–17 years) from 3rd to 11th grade were recruited as control group. All had Spanish as their first language, normal or corrected-to-normal vision, and were not reported to present learning disabilities or developmental disorders. Both groups of participants were matched on average for CA and nonverbal IQ (see Table 1).</p> <p>Table 1. Background characteristics of deaf participants and TH participants</p> <p> <ephtml> <table><thead><tr><th>. </th><th colspan="2">Deaf. </th><th colspan="2">TH. </th><th colspan="3">. </th></tr><tr><th>Group. </th><th colspan="2">(<italic>N</italic> = 20, 12 girls). </th><th colspan="2">(<italic>N</italic> = 20, 11 girls). </th><th colspan="3">Comparisons between groups. </th></tr><tr><th>Background measures. </th><th>Mean. </th><th>SD (range). </th><th>Mean. </th><th>SD (range). </th><th><italic>F</italic>. </th><th><italic>U</italic>. </th><th><italic>p</italic>. </th></tr></thead><tbody><tr><td>CA (months) </td><td>149.7 </td><td>20.45 (114–182) </td><td>144.75 </td><td>25.25 (106–204) </td><td>.528 </td><td /><td>.503 </td></tr><tr><td>Nonverbal IQ (SS)* </td><td>103.2 </td><td>9.12 (87–118) </td><td>109.25 </td><td>10.85 (86–126) </td><td>.158 </td><td /><td>.064 </td></tr><tr><td>Word Reading Accuracy (SS)* </td><td>106.85 </td><td>8.23 (86–115) </td><td>107.07 </td><td>8.07 (80–115) </td><td /><td>198.5 </td><td>.967 </td></tr><tr><td>Nonword Reading Accuracy (SS) * </td><td>109.51 </td><td>9.41 (82–121) </td><td>108.75 </td><td>7.24 (91–121) </td><td /><td>172.5 </td><td>.456 </td></tr><tr><td>Word Reading Speed (SS) ** </td><td>107.51 </td><td>10.86 (80–123) </td><td>115.99 </td><td>7.36 (100–130) </td><td /><td>105 </td><td>.010 </td></tr><tr><td>Nonword Reading Speed (SS)* </td><td>111.78 </td><td>10.18 (82–127) </td><td>113.46 </td><td>7.69 (101–127) </td><td /><td>172.5 </td><td>.456 </td></tr><tr><td>Text Reading Comprehension (SS) </td><td>87.5 </td><td>19.18 (50–124) </td><td>103.9 </td><td>6.65 (90–115) </td><td /><td>80 </td><td>.001 </td></tr><tr><td>Receptive Vocabulary (SS) </td><td>76.86 </td><td>20.9 (55–117) </td><td>105.45 </td><td>9.1 (85–115) </td><td /><td>55.5 </td><td><.001 </td></tr><tr><td>Syntactic Ability Test (RS) </td><td>43.65 </td><td>13.55 (15–59) </td><td>52 </td><td>11.09 (27–64) </td><td /><td>125 </td><td>.042 </td></tr><tr><td>Expressive Language (RS) </td><td>35.9 </td><td>8.06 (19–48) </td><td>43.45 </td><td>5.9 (31–50) </td><td /><td>87 </td><td>.002 </td></tr></tbody></table> </ephtml> </p> <p> <emph>Note</emph>. CA = chronological age; RS = raw scores; SS = standard scores; SD = standard deviation; TH = typical hearing.</p> <p>1 *All participants of both groups fulfilled the inclusion criteria, that is, SS ≥ 70.</p> <p>2 **All participants of both groups fulfilled the inclusion criteria, that is, SS ≥ 70 but, in average, the TH group was faster or more accurate significantly than the Deaf group.</p> <hd id="AN0143509754-10">Background Assessments</hd> <p>Participants completed a battery of language and reasoning tests as part of a more general project on reading and deafness of the University of Valencia (H1469443238331). The results were used to describe the sample's baseline characteristics and to verify the inclusion and matching criteria of the study. Before each test (described below), the instructions were presented both in written and oral form. In Table 1, we report the SS for age that followed the IQ scale with a mean of 100 and a standard deviation of 15. When the test manual did not provide any kind of age-specific SSs, we used raw scores (RD).</p> <hd id="AN0143509754-11">Nonverbal IQ</hd> <p>The "Matrices" subtest from the Kaufman Brief Intelligence Test (K-BIT) ([<reflink idref="bib24" id="ref67">24</reflink>]) was used to ensure that nonverbal IQ was within normal range (SS ≥ 85). All participants of both groups fulfilled this criterion.</p> <hd id="AN0143509754-12">Reading skills</hd> <p>The word decoding skills were measured with the word and nonword test of the PROLEC battery ([<reflink idref="bib44" id="ref68">44</reflink>]; [<reflink idref="bib10" id="ref69">10</reflink>]); students completed the versions matched to their grade level (PROLEC-R for students attending up to 4th grade and PROLEC-SE for students from 5th to 11th grade). As shown in Table 1, deaf and hearing participants were in the normal range (SS ≥ 70, that is, within two standard deviations from the mean). In addition, they completed the reading comprehension subtest of the Magallanes scale of Reading and Writing TALE 2000 ([<reflink idref="bib47" id="ref70">47</reflink>]). This is an untimed reading comprehension test where students have to answer a multiple-choice format questionnaire based on the information from the text. Students are allowed to check for information in the text if they need to. The test is composed of three different texts addressed to the following grade levels: Text 1 (1st–3rd grade), Text 2 (4th–6th grade), and Text 3 (7th–10th grade). Because the students with deafness were reported to have low reading comprehension ability, they were all required to start from Text 2. Students who had obtained an average score in Text 2 and attended grades higher than 6th grade were asked to read Text 3. The TH group completed the text matched on grade level. As illustrated in Table 1, the TH group outperformed the Deaf group in reading comprehension.</p> <hd id="AN0143509754-13">Receptive Vocabulary</hd> <p>The Peabody Picture Vocabulary Test ([<reflink idref="bib16" id="ref71">16</reflink>]) measures the receptive vocabulary of individuals. The examiner asks the participant to select from four pictures the image that represents the label provided orally. To prevent the results from being affected by poor auditory perception, we adapted the task so that the student could also read the label. Table 1 shows that TH group obtained a significantly higher vocabulary level than the Deaf group.</p> <hd id="AN0143509754-14">Syntactic Abilities</hd> <p>Syntactic Ability Test (SAT) ([<reflink idref="bib13" id="ref72">13</reflink>]) measures the syntactic abilities of the participants by means of a multiple-choice task. The test consists of 64 sentences with a missing word. Students choose from among four options the correct function word for each sentence. The task was presented in a paper and pencil format. Again, Table 1 shows that the TH group obtained higher scores than the Deaf group in this test.</p> <hd id="AN0143509754-15">Expressive language</hd> <p>The subtest of Formulated Sentences from the CELF-4 ([<reflink idref="bib50" id="ref73">50</reflink>]) was used to measure expressive language skills. Participants are shown a picture (26 in total) and based on the content of the picture, they are asked to formulate one spoken sentence containing a target word or phrase (e.g., third, when, because, etc.) provided by the examiner. The task gives information about the ability of the participants to formulate spoken complex sentences, semantically and grammatically correct. To score the responses, the interpretation guidelines of the CELF-4 manual were followed. On average, the TH group obtained a higher percentage of correct answers in the Formulated Sentences subtest than the Deaf group.</p> <hd id="AN0143509754-16">Experimental Task</hd> <p>Participants silently read 24 sentences written in Spanish (12 congruent and 12 incongruent ones). A target word was manipulated in order to generate a grammatical incongruence in half of the trials (e.g., La <emph>hija</emph>/La <emph>padre</emph>—the daughter/the father). <emph>Padre</emph> (<emph>father</emph>)<emph>,</emph> which is the target word, is not congruent in gender with "la" in Spanish). Grammatical incongruence was presented in a random order and manipulated along several dimensions: article-noun gender or number agreement, verb tense, and presence/absence of preposition.</p> <hd id="AN0143509754-17">Stimulus construction</hd> <p>Taking the SSAT ([<reflink idref="bib46" id="ref74">46</reflink>]) as a basis, our material was elaborated ad hoc for the experiment. The incongruent word was associated semantically with the congruent one according to the free Spanish association norms ([<reflink idref="bib17" id="ref75">17</reflink>]). The target words were also matched in length, number of syllables (congruent: <emph>Mdn</emph> = 2; incongruent: <emph>Mdn</emph> = 2; <emph>U</emph> = 276, <emph>z</emph> = −.30, <emph>p</emph> =.758, and <emph>r</emph> =.04), number of characters (congruent: <emph>Mdn</emph> = 5.5, <emph>SD</emph> = 1.35; incongruent: <emph>Mdn</emph> = 5; <emph>U</emph> = 282, <emph>z</emph> = −.12, <emph>p</emph> =.898, and <emph>r</emph> = −.01), and written word frequency (congruent: <emph>Mdn</emph> = 37.01 per million; incongruent: <emph>Mdn</emph> = 29.04 per million; <emph>z</emph> = −.52, <emph>p</emph> =.959, and <emph>r</emph> = −.07) according to the ESPAL database ([<reflink idref="bib15" id="ref76">15</reflink>]).</p> <hd id="AN0143509754-18">Stimulus evaluation</hd> <p>Sentence congruity was checked in a pilot study carried out with 88 TH students of primary (3rd and 5th graders) and secondary school (7th and 10th graders) who did not take part in the eye-tracking experiment. Students had to rank using a 3-point Likert scale if a set of 30 sentences selected and modified from the SSAT ([<reflink idref="bib46" id="ref77">46</reflink>]) was correct or not: seemed "1 = correct," "2 = uncertain," and "3 = incorrect." Based on the results of this pilot, we selected the items that were considered congruent (those with a mean under 2) and incongruent (those with a mean equal or above 2). According to the results of this first pilot, we modified the test sentences or added new sentences. To test the plausibility and validity of the new sentences we conducted another pilot study with 180 students (from 4th, 6th, 8th, and 10th grades). We used the same Likert scale as in the first pilot to rank the congruity of the sentences.</p> <p>As a result, 24 congruent and 72 incongruent sentences were selected. To select the best incongruent sentences, we selected the ones in which the manipulated word had a higher semantic association to the sentence context according to a Spanish version of the Latent Semantic Analysis ([<reflink idref="bib29" id="ref78">29</reflink>]).</p> <p>Thus, our experimental material consisted of 24 pairs of sentences: each one with two versions: congruent or incongruent. Two lists (A and B) of 24 sentences (half-congruent and half-incongruent) were designed in such a way that only one version of each sentence appeared in each list. Lists A and B were assigned alternatively to participants in order of arrival so that each list was viewed by the same number of participants in both groups (deaf and TH). The order of sentence presentation was randomized per participant.</p> <hd id="AN0143509754-19">Interest areas for eye-movements analysis</hd> <p>Sentences were divided into three main areas of interest (AOIs). All sentences began with a pretarget region (PRE), which was composed of all the words that preceded the TAR. The TAR was the word that had been manipulated in order to generate the incongruences. Following the TAR, we defined a post-target region (POST) to capture spatially delayed effects in participants' judgments ([<reflink idref="bib7" id="ref79">7</reflink>]; [<reflink idref="bib19" id="ref80">19</reflink>]). The POST displayed the word next to the TAR, but if the word contained fewer than five characters the POST was extended to a second word in order to reduce skipping behavior of short words ([<reflink idref="bib45" id="ref81">45</reflink>]; [<reflink idref="bib19" id="ref82">19</reflink>]). Interest areas are shown in examples (1a) and (1b) (TAR in bold). The examples represent a congruent sentence (1a:"hija" is feminine in Spanish so it agrees with the feminine article "la") and an incongruent one (1b: ("padre" is masculine in Spanish so it disagrees with the feminine article "la").</p> <p>Graph</p> <hd id="AN0143509754-20">Eye-movement measures</hd> <p>Different eye-tracking measures inform about the stages of the reading processes. Although, to our knowledge, there is not a systematic list that determines the exact measures that correspond to each stage of processing, this correspondence can be partly inferred from previous literature ([<reflink idref="bib8" id="ref83">8</reflink>]; [<reflink idref="bib30" id="ref84">30</reflink>]). Measures such as First Fixation Duration or First Pass Duration are indicative of early processes, that is, when the word is first encountered (referring to lexical access). On the other hand, late measures could reflect integration processes, such as rereading patterns and reanalysis (e.g. [<reflink idref="bib30" id="ref85">30</reflink>]). Measures such as Second Pass Gaze Duration or regressions into AOI are classified into this group. Other measures that inform about the general process, such as fixation time and fixation count, could also be classified into late processing measures.</p> <p>In the target (TAR) and post-target (POST) AOI's, the following early and late processing measures were analyzed: First Pass Time or First Pass Gaze Duration (Sum of fixation durations from the first entry into an AOI until the eye leaves it in any direction), Second Pass Time or Second Pass Gaze Duration (Sum of fixation durations from the second entry into an AOI until the eye leaves it in any direction) and Fixation Count (total number of fixations within the area of interest taking into account both First and Second Pass). Trials with no fixations in TAR were excluded from the analysis.</p> <p>Focusing on the pretarget region (PRE), both grammatical congruity conditions were identical until the TAR area, so only Second Pass Gaze Duration as a reflex of late processing was considered in the analyses. Second Pass Gaze Duration in this area was calculated in a similar way to what other authors call "re-reading time" and is meant to provide an index of the time a participant spent detecting a problem and rereading the text (for a more detailed description, see [<reflink idref="bib30" id="ref86">30</reflink>]). As items differ in the number of words of the pretarget area, the number of words for each item was averaged in the Second Pass.</p> <hd id="AN0143509754-21">Procedure</hd> <p>The study was conducted in two to three sessions, depending on the child's compliance with the tasks. Before the assessment, parents or guardians gave their informed consent. During the first session, participants completed only the off-line tasks (Peabody, SAT, Reading Comprehension Test, Formulated Sentences CELF-IV, and Nonverbal IQ test). In each session, pauses were allowed to let the participants rest.</p> <p>The last session's tasks, lasting no longer than 50 min, were performed with the eye tracker. Participants were required to complete the word and nonword reading task, followed by the experimental task (the sentence judgment task).</p> <p>In the experimental task, participants were requested to judge if the sentences displayed one by one on a computer screen were correct or not while at the same time their eye movements were recorded. Before the recording, students completed four practice trials to verify that they understood the task, all the instructions were presented orally as well as in written form and, if necessary, instructions were repeated before the recording began. A fixation point in the middle-left corner of the screen preceded each sentence. Participants had to look at it and contingent on their gaze the sentence appeared ([<reflink idref="bib33" id="ref87">33</reflink>]). The students had to read in silence the sentence and immediately after reading it, the question, "Is the sentence correct?" appeared on the screen. Participants had to give a yes/no answer by clicking on the chosen option.</p> <hd id="AN0143509754-22">Apparatus</hd> <p>Stimuli were presented on a 15.6- inch ASUS connected to a 18.5″ monitor at a resolution of 1,366 × 768 pixels. Participants' eye movements during the task were tracked using an SMI eye tracker with a recording sampling rate of 60 Hz. A head-chin rest system was used to minimize head movements. Participants sat at a distance of 60 cm from the screen. In order to ensure a good recording, a 9-point calibration was used. Eye movements were recorded from the right eye only, although viewing was binocular ([<reflink idref="bib45" id="ref88">45</reflink>]).</p> <hd id="AN0143509754-23">Results</hd> <p></p> <hd id="AN0143509754-24">Accuracy</hd> <p>Because accuracy scores were not normally distributed, nonparametric tests (Mann–Whitney test—Wilcoxon rank-sum tests) were applied. Differences in the percentage of correct responses between groups (deaf and TH) and between conditions (congruent–incongruent) were examined. Our criteria for correct answers on sentence grammaticality were: "no" when sentence was incongruent and "yes" when congruent.</p> <p>Our findings confirmed that deaf students showed greater difficulties detecting grammatical incongruence than TH students. Differences between the two groups were significant for accuracy (<emph>U =</emph> 77.5, <emph>z =</emph> −3.32<emph>, p =.001,</emph> and <emph>r =</emph> −.52)<emph>,</emph> with deaf students showing a significantly lower percentage of correct responses (<emph>M</emph> = 77) than the TH students (<emph>M</emph> = 90) in the grammatical judgment task.</p> <p>The effect of grammatical congruity was not significant neither for the deaf group (<emph>z = −</emph>1.6<emph>, p =.</emph>109, and <emph>r =</emph> −.36) nor for the TH group (<emph>z =</emph> −1.54<emph>, p =.</emph>122<emph>,</emph> and <emph>r =</emph> −.34).</p> <hd id="AN0143509754-25">Time-course processing: eye movements</hd> <p></p> <hd id="AN0143509754-26">Data preparation and analysis</hd> <p>Fixations shorter than 80 ms and longer than 1,200 ms were excluded from the data set. For each eye-movement time measure, the cells that were either >2.5 <emph>SD</emph>s above or below each participant mean for each condition were excluded from the analyses (following [<reflink idref="bib33" id="ref89">33</reflink>]).</p> <p>Only trials with correct answers were analyzed. We used the lmer program of the lme4 package ([<reflink idref="bib6" id="ref90">6</reflink>]) in the R environment (R [<reflink idref="bib43" id="ref91">43</reflink>]) to perform linear mixed-effects analyses of the relationship between each time eye- movement measure, congruity, and group. We used mixed models in order to account for random variance associated with items (sentences) and participants.</p> <p>With regard to categorical fixed effects, we entered group, congruity and their interaction term into the model. Groups were matched on CA and IQ, but differed in other background measures (Table 1). Therefore, we also added as continuous control variables, word reading speed (SS), vocabulary (SS), syntactic skill (percentage of correct answers), and CA (which ranged largely within groups) and computed their interaction terms with group and congruity. The aim was to explore whether any of these continuous variables could affect variance in the eye-movements measures and highlight potential confound variables. Both groups also differed significantly in expressive language skills, however, this measure was not introduced as a control variable because it is related to language production rather than receptive language. Moreover, problems in phonetic articulation were mixed with deaf students´ expressive language capacity. Therefore, these results were disregarded, although we inform about it for the sake of transparency. Because these analyses were conducted in an exploratory manner, we did not have specific predictions about the effect of continuous variables on eye movements. All continuous variables were centered (by subtracting the mean of the variable from each variable) to reduce the effects of collinearity between main effects and interactions.</p> <p>For random effects in the mixed models, we had intercepts for subjects and items, as well as by-item and by-participants random slopes for all fixed effects (see [<reflink idref="bib5" id="ref92">5</reflink>]) when they were logical according to the experimental design. When the full random structure did not converge, we excluded the random effects that explained the least amount of variability. Each reduced model was tested against the previous one using a chi-square test of model fit with the ANOVA function in R. Only converged models are reported (see further details about the fitted models in Supplementary Material). Time measures were log-transformed and effect coding was used to contrast categorical fixed effects. Effects were considered as significant if <emph>t ≥</emph> 2. Note that in the next Result sections, we provide <emph>β</emph> and <emph>SE</emph> values in milliseconds that are computed for fixed effects on raw data to facilitate interpretation, however, the <emph>t</emph> and <emph>p</emph> values correspond to the effects computed on log-transformed measures. The <emph>β</emph>, <emph>SE</emph>, <emph>t</emph> and <emph>p</emph> values of the fixed effects computed with transformed variables are provided in Supplementary Material.</p> <p>For the noncontinuous variable (Fixation Counts), generalized mixed models were computed with the glmer function ("lme4" R package). Model fitting protocol was similar to that of time variables. Effects for noncontinuous variables were considered as significant when <emph>z ≥</emph> 2.</p> <p>Table 2 shows the means and standard deviations for eye-movement results produced for each of the three regions of interest (only correct answers) per group and congruity conditions. Note that, in order to aid the interpretation of this Table, non–log-transformed data have been reported.</p> <p>Table 2. Deaf and TH students' performance in congruent and incongruent conditions: Eye-movement measures for Target, Post-target, Pretarget AOI (means and SDs in parentheses)</p> <p> <ephtml> <table><thead><tr><th>. </th><th>. </th><th>. </th><th colspan="2">Deaf. </th><th colspan="2">TH. </th><th colspan="2">Both groups. </th></tr><tr><th>. </th><th>. </th><th>Deaf /TH. </th><th>Congruent mean (SD). </th><th>Incongruent mean (SD). </th><th>Congruent mean (SD). </th><th>Incongruent mean (SD). </th><th>Congruent mean (SD). </th><th>Incongruent mean (SD). </th></tr></thead><tbody><tr><td>Accuracy (per) </td><td /><td>20/20 </td><td>71.7 (17.7) </td><td>82.3 (15.1) </td><td>86.2 (15.3) </td><td>94 (7.3) </td><td>79 (18) </td><td>88.1 (13.1) </td></tr><tr><td colspan="9">Target </td></tr><tr><td /><td>First Pass Gaze Duration (ms) </td><td>20/20 </td><td>420.4 (36.6) </td><td>532.7 (34.1) </td><td>501.9 (47.7) </td><td>654.4 (45.8) </td><td>461.2 (42) </td><td>593.6(40) </td></tr><tr><td /><td>Second Pass Gaze Duration (ms) </td><td>20/20 </td><td>599.5 (77.9) </td><td>867.6 (85) </td><td>644.5 (144.8) </td><td>877 (138) </td><td>622 (111.3) </td><td>872.3 (111.5) </td></tr><tr><td /><td>Fixation Count </td><td>20/20 </td><td>2.57 (.07) </td><td>2.86 (.06) </td><td>2.14 (.07) </td><td>2.5 (.06) </td><td>2.36 (.07) </td><td>2.68 (.06) </td></tr><tr><td colspan="9">Post-target </td></tr><tr><td /><td>First Pass Gaze Duration (ms) </td><td>20/20 </td><td>557.2 (47.9) </td><td>422.7(47.6) </td><td>693.9 (64) </td><td>463.8 (64.3) </td><td>625.6 (56) </td><td>443.3 (56) </td></tr><tr><td /><td>Second Pass Gaze Duration (ms) </td><td>18/19 </td><td>647 (81.7) </td><td>635.6 (87.3) </td><td>671.8 (126.8) </td><td>519 (129) </td><td>659.4 (104.3) </td><td>577.3 (72.1) </td></tr><tr><td /><td>Fixation Count </td><td>20/20 </td><td>2.06 (.07) </td><td>1.9 (.07) </td><td>2.00 (.07) </td><td>1.73 (.07) </td><td>2.03 (.07) </td><td>1.82 (.07) </td></tr><tr><td colspan="9">Pretarget </td></tr><tr><td /><td>Second Pass Gaze Duration (ms) </td><td>20/20 </td><td>417.7 (32) </td><td>514 (31.2) </td><td>462 (30.8) </td><td>558.7 (30.3) </td><td>439.9 (31.4) </td><td>536.4 (30.8) </td></tr></tbody></table> </ephtml> </p> <p> <emph>Note</emph>. SD = standard deviation; TH = typical hearing.</p> <hd id="AN0143509754-27">Target AOI</hd> <p>The target area was the part of the sentence that had been manipulated to generate the incongruence. Regarding the First Pass Time (see model in Supplementary Table A1), the effect of congruity (<emph>β</emph> = −65.7; <emph>SE</emph> = 17.7; <emph>t</emph> = −2.6; <emph>p</emph> <.001), group (<emph>β</emph> = −51.6; <emph>SE</emph> = 26; <emph>t</emph> = −2.6; <emph>p</emph> <.01) and syntactic skill (<emph>β</emph> = −4.5; <emph>SE</emph> = 1.6; <emph>t</emph> = −2.6; <emph>p</emph> <.01) were significant. Participants made longer First Pass in the incongruent condition than in the congruent one (see Figure 1), TH participants made longer First Pass than deaf participants across congruity conditions (see Figure 1) and those participants with higher syntactic skills made shorter First Pass than those with lower syntactic skills in the TAR. No other main effects or interactions were significant (see Supplementary Table A1).</p> <p>Graph: Figure 1. class="chapter-para">First Pass in milliseconds in the target region (TAR) area per congruity condition and participant group. Note that nontransformed means were used for making the graphic easier to interpret.</p> <p>Also with regard to the Second Pass, the effect of congruity was significant (<emph>β</emph> = −120.1; <emph>SE</emph> = 45.8; <emph>t</emph> = −3.3; <emph>p</emph> <.01). In particular, participants made longer Second Pass in the incongruent condition than in the congruent one across groups, confirming that both deaf and TH students invested more time during their revisits to the target area with incongruence (see Figure 2). The main effect of word reading speed was also significant (<emph>β</emph> = −6.2; <emph>SE</emph> = 5.3; <emph>t</emph> = −2.1; <emph>p</emph> <.05), such that higher word reading speed was associated with a lower Second Pass in both congruity conditions and groups. No other main or interaction effects were reliable (see model, means and <emph>t</emph> values for nonreliable effects in Supplementary Material Table A2).</p> <p>Graph: Figure 2. class="chapter-para">Second Pass in milliseconds in the target region (TAR) area per congruity condition and participant group. Note that nontransformed means were used for making the graphic easier to interpret.</p> <p>Finally, regarding the Fixation Count, the model with all fixed effects did not converge so we tried different models with the categorical factors, Group and Congruity, as fixed effects, plus one continuous variable each time. Only the model with Group, Congruity and Chronological Age as fixed effects converged (see details for the model with fixed and random effects in Supplementary Table A3). This model showed that the effect of congruity (<emph>β</emph> = −.07; <emph>SE</emph> =.02; <emph>t</emph> = −2.7; <emph>p</emph> <.01) and group (<emph>β</emph> =.08; <emph>SE</emph> =.03; <emph>t</emph> = 2.2; <emph>p</emph> <.05) were significant in the target area. Participants made more visits to the incongruent condition than to congruent condition and deaf participants made more visits than TH participants to the target area (see Figure 3). Neither the effect of CA nor any interaction between Group, Congruity and Chronological age were significant (see model in Supplementary Table A3).</p> <p>Graph: Figure 3. class="chapter-para">Fixations Count (number of total visits) in the target region (TAR) area per congruity condition and participant group. Note that nontransformed means were used for making the graphic easier to interpret.</p> <p>Hence, the results in the TAR area revealed that both groups made more fixations in this region, and showed longer First and Second Pass for the incongruent condition than for the congruent one, which means that they were sensitive to the grammatical congruity manipulation at these points of the sentence processing. However, our findings also evidenced some differences between groups in this area. Whereas TH students showed longer First Pass than deaf students across congruity conditions, the deaf students made instead more fixations in this area (also across congruity conditions). Therefore, deaf students seem to show a preference for doing more but shorter fixations while TH students made fewer but longer fixations in this area. In addition, those participants (of both groups) with higher syntactic knowledge and word reading speed tended to make shorter First Pass and Second Pass respectively in this area across congruity conditions.</p> <hd id="AN0143509754-28">Post-target AOI</hd> <p>The POST area was defined as the next word with at least five characters following the TAR region. Means and standard deviations per condition and group are presented in Table 2.</p> <p>Regarding the First Pass Time, the effects of congruency (<emph>β</emph> = 92.4; <emph>SE</emph> = 17.3; <emph>t</emph> = −4.6; <emph>p</emph> <.001) and syntactic skill (<emph>β</emph> = −6.7.4; <emph>SE</emph> = 2.5; <emph>t</emph> = 2.4; <emph>p</emph> <.05) were significant. In the case of congruity, however, the pattern was different than in the TAR region, that is, participants did longer First Pass for the congruent condition than in the incongruent condition. Regarding syntactic skills, similar to the TAR region, those participants with higher syntactic skills made shorter First Pass than those with lower syntactic skills. No other main or interaction effects were reliable (see Supplementary Table A4 for details).</p> <p>With regard to Second Pass Time, neither main effects nor interaction effects were significant (Supplementary Table A5). Finally, regarding Fixation Counts, only the effect of congruency was significant (<emph>β</emph> =.1; <emph>SE</emph> =.02; <emph>t</emph> = 2.1; <emph>p</emph> <.05 (SupplementaryTable A6). Consistent with the First Pass pattern, congruent condition received more fixations than the incongruent condition.</p> <hd id="AN0143509754-29">Pretarget AOI</hd> <p>The PRE area comprised all the words before the area where the incongruence was generated (TAR area). In this case, only one late measure related to rereading (Second Pass Time) was analyzed. Second Pass Time reflected what happened in this area once the target area, which contained the incongruence, had been processed. For each item, Second Pass Time was averaged by number of words. Means and standard deviations for each condition and group for this area are shown in Table 2 (see Supplementary Table A7 for details).</p> <p>No main effects were significant, but the interaction of congruity by syntactic skills yielded a significant result (<emph>β</emph> =2.97; <emph>SE</emph> = 3.3; <emph>t</emph> = 2.7; <emph>p</emph> <.012), as in the case of the First Pass, in the TAR and Post-TAR regions. The analysis of simple effects with Bonferroni adjustments ("phia" package, [<reflink idref="bib11" id="ref93">11</reflink>]) showed that the slope of syntactic skills was higher in the incongruent condition than in the congruent one (<emph>X</emph><sups>2</sups> (<reflink idref="bib1" id="ref94">1</reflink>) = 7.2; <emph>p</emph> <.01), such that those participants with higher syntactic skills made shorter Second Pass in the incongruent condition than those with lower syntactic skills. However, the slope of syntactic skills did not affect the congruent condition.</p> <p>Moreover, the third order interaction of congruity by group and vocabulary level (SS) was significant (<emph>β</emph> = −.7; <emph>SE</emph> = 2.9; <emph>t</emph> = −2.4; <emph>p</emph> <.01). Formal statistical analyses to explore this interaction were not possible due to the reduced sample size of this study, but the visual analysis of the interaction between vocabulary and congruity per group showed that TH participants with higher vocabulary level invested longer Second Pass in the congruent condition than in the incongruent condition (see Figure 4). However, vocabulary level seems to affect deaf participants similarly in both congruity conditions. Specifically, deaf readers with lower vocabulary level (SS < 80) invested longer Second Pass in both congruity conditions. There were no TH participants with very low vocabulary levels (SS < 80) as in the case of Deaf participants, so the direct comparison cannot be performed.</p> <p>Graph: Figure 4. class="chapter-para">Second Pass (milliseconds per word) in the pretarget (PRE) region as a function of vocabulary level per each participants group (typical hearing [TH] and Deaf) and congruity condition (Congruent and Incongruent).</p> <hd id="AN0143509754-30">Discussion</hd> <p>Based on previous findings that suggested that poor grammatical skills could explain deaf children's reading comprehension deficits (e.g. [<reflink idref="bib3" id="ref95">3</reflink>]), the aim of this study was to examine how deaf students processed and used syntactic cues during reading. In this regard, deaf and TH participants matched on CA (mean age: 12 years) were asked to solve a correctness judgment task after reading sentences with and without grammatical violations. Participants' eye movements over the target area (the area that contained the grammatical incongruence), and the pre- and the post-target areas were registered and analyzed. It was predicted that: (a) deaf readers would be less capable than TH readers to detect the grammatical incongruence of the sentences and (b) deaf participants' time-course of processing sentences with grammatical incongruence would differ from that of TH participants. Specifically, we anticipated (a) no effects of grammatical violations for deaf readers on the target area but (b) we did expect delayed effects in terms of number and duration of visits and revisits to pre- and post-target areas ([<reflink idref="bib7" id="ref96">7</reflink>]).</p> <hd id="AN0143509754-31">Accuracy differences</hd> <p>Regarding accuracy in the judgment task, our first prediction was supported. TH students obtained significantly more correct responses than deaf participants across congruity conditions (90 percent and 77 percent respectively). These findings are consistent with those of [<reflink idref="bib7" id="ref97">7</reflink>]), which revealed that deaf adolescents showed less ability to detect and correct subject-verb incongruence in sentences relative to both CA and RA adolescents (Experiment 2). The contribution of our study is that this accuracy effect covers a wider age range since some of our participants were younger (range 9–17 years) than the participants in the Breadmore et al. study (11.9 to 16.3 years old). Nevertheless, it is important to highlight that our findings showed a great variability within the deaf group, with some participants scoring close to the random level (50%) (see Table 2 for standard deviation by group and condition). It is also essential to underline that, in our study, deaf participants not only detected syntactic incongruence less frequently than TH participants did, but they also made more "false positives" in congruent trials. This profile might suggest that they were answering in a more random way or felt more insecure about their grammar competence so they tended to acquiesce.</p> <hd id="AN0143509754-32">Time-course differences in grammatical processing</hd> <p>Regarding the online processing of sentence grammaticality, we predicted (a) no effects of grammatical violations for deaf readers on the target area and (b) delayed effects of congruency manipulation in terms of number and duration of visits and revisits to pre- and post-target areas ([<reflink idref="bib7" id="ref98">7</reflink>]).</p> <hd id="AN0143509754-33">No effects of grammatical violations for deaf readers on the target area</hd> <p>In contrast with our first prediction regarding the target area, both participating groups appeared to be sensitive to grammaticality manipulation, showing more fixations and longer First Pass and Second Pass Time in incongruent words than in congruent ones. These early grammar effects found in our deaf children and adolescent with deafness are consistent with the findings obtained by [<reflink idref="bib38" id="ref99">38</reflink>]) and [<reflink idref="bib49" id="ref100">49</reflink>]) with deaf adults and TH adults respectively. In particular, [<reflink idref="bib38" id="ref101">38</reflink>]) found that adult deaf readers showed longer reading times in the target areas of syntactically complex sentences (animated object relative clauses) than in syntactically simpler sentences (animated subject relative clauses) as was found for TH adults in a previous study by [<reflink idref="bib49" id="ref102">49</reflink>]). However, our results differed with the findings of [<reflink idref="bib7" id="ref103">7</reflink>]), which found early reading time effects of grammatical incongruence in the target area for the TH readers (CA and RG) but not for the deaf readers group.</p> <p>Although the effect of congruity was similar in both groups in our experiment, deaf participants differed from TH participants in number and durations of fixations across conditions. Deaf students seem to show a preference for doing more but shorter fixations while TH students made fewer but longer fixations in this area.</p> <hd id="AN0143509754-34">Delayed effects of grammatical violations in pre- and post-target areas</hd> <p>Our second prediction regarding time-course processing stated that there would be differences between groups for the pre- and post-target areas.</p> <p>With regard to the post-target area and in contrast to our prediction, both TH and deaf participants showed effects of grammatical congruity (though inverted) for both First Pass Time and fixation count. This effect for both groups contrasts again with the results of [<reflink idref="bib7" id="ref104">7</reflink>]), who found delayed effects of grammatical incongruence for deaf adolescents in the post-verb region, but not for the TH controls. Significantly, our findings revealed that congruity manipulation had a delayed effect in the post-target area for all participants. This result was contrary to what we anticipated, i.e. longer First Pass Time and more fixations for the congruent condition than for the incongruent one.</p> <p>This inverted effect could be an experimental artifact of our study, which was not present in the study by [<reflink idref="bib7" id="ref105">7</reflink>]) (Experiment 1). We asked participants to judge, explicitly, the correctness of written sentences, whereas [<reflink idref="bib7" id="ref106">7</reflink>]) used an implicit self-paced reading task to measure reading times on sentences containing grammatical errors. They assumed that if participants were sensitive to grammatical errors, they would slow down when encountering them. However, in our experiment, participants might have continued searching for incorrectness until the post-target area, resulting in delayed effects.</p> <p>Focusing on the pretarget area, no effects of congruity, group or interaction were found, so our prediction about delayed effect for deaf participants was not supported. However, we found an interesting effect of participants' syntactic skills that was also present in the target and post-target area. In particular, those participants with higher syntactic abilities made shorter Second Pass of the pretarget area in the incongruent condition than those with lower syntactic skills, whereas the readers' syntactic skills did not affect the congruent condition. Similarly, those participants with higher syntactic skills made shorter First Pass in the target and post-target areas than those with lower syntactic skills across congruity conditions. The significantly lower level of syntactic skills of the deaf readers could be a main contributor for the differences in sentence processing between deaf and TH readers (except in the case of the Fixation Count in the target area where the group effect remains after controlling for syntactic skills).</p> <p>In sum, in our experiment, both, deaf and TD students showed early detection of grammar errors in the target area (albeit a different pattern of eye movements), but a late confirmation of sentence correctness in the form of longer First Pass Time and more fixations in the post-target area. Finally, syntactic skills, significantly lower in the deaf group, seem to predict time spent in the relevant areas of the sentence.</p> <p>These findings would be in line to some extent with previous studies that indicated that deaf individuals underuse morphosyntactic cues when reading sentences ([<reflink idref="bib12" id="ref107">12</reflink>]; [<reflink idref="bib34" id="ref108">34</reflink>]). However, in our study, the deaf group did not completely ignore morphosyntactic cues as suggested by the "Key-Word strategy" of [<reflink idref="bib12" id="ref109">12</reflink>]) or the "top–down preference" hypothesis of [<reflink idref="bib34" id="ref110">34</reflink>]). In correct trials, they actually used morphosyntactic cues very early as can be inferred from the grammaticality effects in the TAR (longer First Pass in the incongruent condition). Notably, our experiment did not allow us to infer if a less efficient use of morphosyntactic markers could lead deaf students to make a more generalized use of semantic cues as previously suggested, because we did not use a control condition with only semantic violations as [<reflink idref="bib7" id="ref111">7</reflink>]) or different syntactic and semantic distractors ([<reflink idref="bib18" id="ref112">18</reflink>]).</p> <p>By contrast, our findings would be compatible with the bottleneck effect of syntactic skills proposed by [<reflink idref="bib26" id="ref113">26</reflink>]), which suggests that, unless deaf participants achieved a reasonable level of syntactic competence, they seem unable to use their vocabulary competence to comprehend the text. Again, our study design and analyses did not allow us to establish the level of syntactic competence needed for vocabulary knowledge to be used, such as Kelly did. Consequently, the bottleneck interpretation should be taken with caution. In fact, we could substitute the "bottleneck" metaphor for a "cascading" metaphor. The later term suggests the idea of a "chain of events", in such a way that if the first event (syntax) is not working properly, it would affect the functioning of subsequent events (vocabulary) but not totally block them (as the "bottleneck" metaphor would suggest). Actually, the third order interaction between hearing status, congruity, and vocabulary in the pretarget area (Figure 4) seems to suggest that revisiting the pretarget area in search for congruity or incongruity confirmation is associated to vocabulary knowledge in both, deaf and TH students.</p> <p>Finally, all other factors used as continuous variables in the linear mixed-effects analyses (CA, vocabulary, and word reading speed) did not have any systematic effect on eye-movements measures across AOI as syntactic competence did. Only word reading speed predicted the Second Pass Time in the target area. Hence, a higher word reading speed was associated with a lower Second Pass in both congruity conditions and groups. This result would suggest that the duration of the revisits in the target area was a lexical effect rather than a sentence integration effect because word reading speed (significantly lower in deaf students) is an index of lexical access. That is, those participants who were slower in lexical access, tended to revisit longer the target area presumably in search of a lexical confirmation that was not complete during the First Pass of the word. This interpretation would be compatible with the fact that deaf participants made more but shorter fixations than TH participants during the First Pass of the word.</p> <p>Regarding study limitations, there are three that stand out: (a) A CA TH control group participated in the study but not an RA control. An RA control is necessary to verify if time-course differences are typical of younger and immature readers or qualitatively different and a by-product of the particular hearing and language experience of prelingual deaf children and adolescents; (b) In our study, grammatical incongruence was manipulated along several dimensions: article-noun gender or number agreement, verbal tense and presence/absence of preposition. However, prior studies suggest that, in Spanish, different agreement mechanisms operate for gender and number, for example, which seem to affect the time-course of syntactic processing ([<reflink idref="bib4" id="ref114">4</reflink>]). Furthermore, the time-course of processing prepositions as syntactic markers in the different types of sentences (e.g. passive versus active) could vary concerning the use of within word morphological markers for gender and number. Future research should replicate this study, comparing different types of agreement violation and sentences types. Additionally, studies in other languages should be conducted in order to compare results; (c) 20 deaf students was the largest number of participants with homogeneous characteristics in terms of degree of deafness, moment of acquisition, communication mode, age range, and nonverbal IQ, which were available in the researchers' educational context. However, this is a small sample size in terms of statistical power so type II errors might have occurred (the research fails to identify a significant difference or effect that actually exists). Anyhow, our results might serve as a basis for future meta-analyses that combine similar individual studies on the topic of time-course of grammatical processing in deaf readers.</p> <hd id="AN0143509754-35">Conclusions</hd> <p>We have shown that eye-tracking techniques combined with an appropriate task paradigm can provide valuable information both on the causes of deaf readers' reading difficulties and on the nature of their strategies and reading processes as showed in previous research in TH children and adults ([<reflink idref="bib22" id="ref115">22</reflink>]; [<reflink idref="bib23" id="ref116">23</reflink>]). Specifically, we found that deaf readers make more but shorter fixations in the target area than TH participants when trying to judge for sentence correctness (regardless of the congruity of the sentence).</p> <p>A practical implication of these findings is the need of a greater focus on reading processes and strategies in instructional intervention and not only on deaf students' grammatical knowledge. By making their grammatical knowledge and grammatical parsing more accurate, their reading might become more fluent, benefitting, in addition, their reading comprehension skills. If syntactic abilities are at the heart of deaf students' problems with reading comprehension, helping students gain more efficient syntactic parsing skills would have positive cascading effects on reading processes at text level, as well as freeing up memory and attention resources required for local text processing ([<reflink idref="bib26" id="ref117">26</reflink>], [<reflink idref="bib27" id="ref118">27</reflink>], [<reflink idref="bib28" id="ref119">28</reflink>]). Second, in school settings, the incongruence detection and correction tasks could be used for both individual assessment and explicit grammar training of students with deafness. Training in grammar correction tasks would increase not only grammar knowledge but also students' self-regulation, skills that both deaf and hearing students appear to lack ([<reflink idref="bib7" id="ref120">7</reflink>]).</p> <p>Teachers might apply these practical recommendations in order to provide inclusive support for deaf students, especially if they present a similar profile than the sample of our study, that is, users of CI or HAs and spoken language as their preferred means of communication who are usually attending mainstream schools and receiving oral education ([<reflink idref="bib1" id="ref121">1</reflink>]).</p> <hd id="AN0143509754-36">Funding</hd> <p>Research funded by Universitat de Valéncia (Atracció de talent/UV-INV-PREDOC17F1-5405).</p> <hd id="AN0143509754-37">Acknowledgments</hd> <p>The authors would like to thank the students and parents who participated, and the doctors, teachers, hospitals, schools and associations that assisted with the recruitment of participants.</p> <hd id="AN0143509754-38">Conflicts of Interest</hd> <p>No conflicts of interest were reported.</p> <ref id="AN0143509754-39"> <title> References </title> <blist> <bibl id="bib1" idref="ref94" type="bt">1</bibl> <bibtext> Archbold, S., & Mayer, C. (2012). Deaf education: The impact of cochlear implantation. Deafness & Education International, 14 (1), 2 – 15. https://doi.org/10.1179/1557069X12Y.0000000003 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibl id="bib2" idref="ref50" type="bt">2</bibl> <bibtext> Au-Yeung, S. K., Kaakinen, J. K., Liversedge, S. P., & Benson, V. (2015). Processing of written irony in autism Spectrum disorder: An eye-movement study. Autism Research, 8 (6), 749 – 760. https://doi.org/10.1002/aur.1490 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibl id="bib3" idref="ref14" type="bt">3</bibl> <bibtext> Barajas, C., González-Cuenca, A. M., & Carrero, F. (2016). Comprehension of texts by deaf elementary school students: The role of grammatical understanding. Research in Developmental Disabilities, 59, 8 – 23. https://doi.org/10.1016/j.ridd.2016.07.005 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibl id="bib4" idref="ref114" type="bt">4</bibl> <bibtext> Barber, H., & Carreiras, M. (2005). Grammatical gender and number agreement in Spanish: An ERP comparison. Journal of Cognitive Neuroscience, 17 (1), 137 – 153. https://doi.org/10.1162/0898929052880101 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibl id="bib5" idref="ref92" type="bt">5</bibl> <bibtext> Barr, D. J., Levy, R., Scheepers, C., & Tilly, H. J. (2013). Random effects structure for confirmatory hypothesis testing: Keep it maximal. Journal of Memory and Language, 68, 255 – 278. https://doi.org/10.1016/j.jml.2012.11.001 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibl id="bib6" idref="ref90" type="bt">6</bibl> <bibtext> Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67 (1), 1 – 48. https://doi.org/10.18637/jss.v067.i01 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibl id="bib7" idref="ref1" type="bt">7</bibl> <bibtext> Breadmore, H. L., Krott, A., & Olson, A. C. (2014). Agreeing to disagree: Deaf and hearing children's awareness of subject–verb number agreement. The Quarterly Journal of Experimental Psychology, 67 (3), 474 – 498. https://doi.org/10.1080/17470218.2013.818702 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibl id="bib8" idref="ref83" type="bt">8</bibl> <bibtext> Clifton, C. Jr., Ferreira, F., Henderson, J. M., Inhoff, A. W., Liversedge, S. P., Reichle, E. D., & Schotter, E. R. (2016). Eye movements in reading and information processing: Keith Rayner's 40-year legacy. Journal of Memory and Language, 86, 1 – 19. https://doi.org/10.1016/j.jml.2015.07.004 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibl id="bib9" idref="ref4" type="bt">9</bibl> <bibtext> Conrad, R. (1979). The deaf schoolchild: Language and cognitive functions. London : Harper and Row. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Cuetos, F., Rodríguez, B., Ruano, E., & Arribas, D. (2007). Batería de evaluación de los procesos lectores, revisada (PROLEC-R). Madrid : TEA. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> De Rosario-Martinez, H. (2015). Phia post- hoc interaction analysis (version 0.2-1) [software package] Retrieved from https://CRAN.R-project.org/package=phia</bibtext> </blist> <blist> <bibtext> Domínguez, A., & Alegria, J. (2009). Reading mechanisms in orally educated deaf adults. Journal of Deaf Studies and Deaf Education, 15 (2), 136 – 148. https://doi.org/10.1093/deafed/enp033 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Domínguez, A. B., Alegría, J., Carrillo, M., & Soriano, J. (2013). PEALE. Pruebas de Evaluación Analítica de Lengua Escrita Número de asiento registral: 00/2013/4067. Universidad de Salamanca. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Domínguez, A. B., Pérez, I., & Alegría, J. (2012). La lectura en los alumnos sordos: aportación del implante coclear. Infancia y Aprendizaje, 35 (3), 327 – 341. https://doi.org/10.1174/021037012802238993 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Duchon, A., Perea, M., Sebastián-Gallés, N., Martí, A., & Carreiras, M. (2013). EsPal: One-stop shopping for Spanish word properties. Behavior Research Methods, 45 (4), 1246 – 1258. https://doi.org/10.3758/s13428-013-0326-1 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Dunn, L. M., Dunn, L., & Arribas, D. (2006). PPVT- III Peabody, Test de Vocabulario en Imágenes. Madrid : TEA Ediciones. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Fernández, A., Díez, E. & Alonso, M. A. (2012). Normas de asociación libre en castellano de la Universidad de Salamanca. Retrieved from <ulink href="http://inico.usal.es/usuarios/gimc/normas/index%5fnal.asp">http://inico.usal.es/usuarios/gimc/normas/index%5fnal.asp</ulink></bibtext> </blist> <blist> <bibtext> Gallego, C., Martín-Aragoneses, M. T., López-Higes, R., & Pison, G. (2016). Semantic and syntactic reading comprehension strategies used by deaf children with early and late cochlear implantation. Research in Developmental Disabilities, 49, 153 – 170. https://doi.org/10.1016/j.ridd.2015.11.020 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Godfroid, A., Loewen, S., Jung, S., Park, J., Gass, S., & Ellis, R. (2015). Timed and untimed grammaticality judgments measure distinct types of knowledge: Evidence from eye-movement patterns. Studies in Second Language Acquisition, 37 (2), 269 – 297. https://doi.org/10.1017/S0272263114000850 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Grammar. (2019). In Cambridge dictionary. Retrieved from https://dictionary.cambridge.org/es/diccionario/ingles-espanol/grammar</bibtext> </blist> <blist> <bibtext> Harris, M., Terlektsi, E., & Kyle, F. E. (2017). Concurrent and longitudinal predictors of reading for deaf and hearing children in primary school. The Journal of Deaf Studies and Deaf Education, 22 (2), 233 – 242. https://doi.org/10.1093/deafed/enw101 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Joseph, H. S., & Liversedge, S. P. (2013). Children's and adults' on-line processing of syntactically ambiguous sentences during reading. PloS One, 8 (1), e54141. https://doi.org/10.1371/journal.pone.0054141 OpenURL Placeholder Text WorldCat</bibtext> </blist> <blist> <bibtext> Joseph, H. S., Liversedge, S. P., Blythe, H. I., White, S. J., & Rayner, K. (2009). Word length and landing position effects during reading in children and adults. Vision Research, 49 (16), 2078 – 2086. https://doi.org/10.1016/j.visres.2009.05.015 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Kaufman, A. S., & Kaufman, N. L. (1997). Test breve de inteligencia de kaufman. Madrid : TEA Ediciones. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Kenney, T. J., & Wolfe, J. (1972). The acquisition of agreement in English. Journal of Verbal Learning and Verbal Behavior, 11 (6), 698 – 705. https://doi.org/10.1016/S0022-5371(72)80004-5 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Kelly, L. (1996). The interaction of syntactic competence and vocabulary during reading by deaf students. The Journal of Deaf Studies and Deaf Education, 1 (1), 75 – 90. https://doi.org/10.1093/oxfordjournals.deafed.a014283 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Kelly, L. P. (1998). Using silent motion pictures to teach complex syntax to adult deaf readers. The Journal of Deaf Studies and Deaf Education, 3 (3), 217 – 230. https://doi.org/10.1093/oxfordjournals.deafed.a014352 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Kelly, L. P. (2003). The importance of processing automaticity and temporary storage capacity to the differences in comprehension between skilled and less skilled college-age deaf readers. Journal of Deaf Studies and Deaf Education, 8 (3), 230 – 249. https://doi.org/10.1093/deafed/eng013 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Landauer, T. K., & Dumais, S. T. (1997). A solution to Plato's problem: The latent semantic analysis theory of the acquisition, induction, and representation of knowledge. Psychological Review, 104 (2), 211 – 240. <ulink href="http://dx.doi.org/10.1037/0033-295X.104.2.211">http://dx.doi.org/10.1037/0033-295X.104.2.211</ulink> Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Liversedge, S. P., Paterson, K. B., & Pickering, M. J. (1998). Eye movements and measures of reading time. In G. Underwood (Ed.), Eye guidance in reading and scene perception (pp. 55 – 75). Oxford, UK : Elsevier Science Ltd. https://doi.org/10.1016/B978-008043361-5/50004-3 Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Mayberry, R. I., Del Giudice, A. A., & Lieberman, A. M. (2011). Reading achievement in relation to phonological coding and awareness in deaf readers: A meta-analysis. The Journal of Deaf Studies and Deaf Education, 16 (2), 164 – 188. https://doi.org/10.1093/deafed/enq049 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Mayer, C., & Trezek, B. J. (2018). Literacy outcomes in deaf students with cochlear implants: Current state of the knowledge. The Journal of Deaf Studies and Deaf Education, 23 (1), 1 – 16. https://doi.org/10.1093/deafed/enx043 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Micai, M., Joseph, H., Vulchanova, M., & Saldaña, D. (2017). Strategies of readers with autism when responding to inferential questions: An eye-movement study. Autism Research, 10 (5), 888 – 900. https://doi.org/10.1002/aur.1731 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Miller, P. F. (2000). Syntactic and semantic processing in Hebrew readers with prelingual deafness. American Annals of the Deaf, 145 (5), 436 – 451. https://doi.org/10.1353/aad.2012.0116 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Miller, P. (2010). Phonological, orthographic, and syntactic awareness and their relation to reading comprehension in prelingually deaf individuals: What can we learn from skilled readers? Journal of Developmental and Physical Disabilities, 22 (6), 549 – 580. https://doi.org/10.1007/s10882-010-9195-z Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Miller, P., Kargin, T., & Guldenoglu, B. (2013). The reading comprehension failure of Turkish prelingually deaf readers: Evidence from semantic and syntactic processing. Journal of Developmental and Physical Disabilities, 25 (2), 221 – 239. https://doi.org/10.1007/s10882-012-9299-8 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Moreno-Pérez, F. J., Saldaña, D., & Rodríguez-Ortiz, I. R. (2015). Reading efficiency of deaf and hearing people in Spanish. Journal of Deaf Studies and Deaf Education, 20 (4), 374 – 384. https://doi.org/10.1093/deafed/env030 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Piñar, P., Carlson, M. T., Morford, J. P., & Dussias, P. E. (2017). Bilingual deaf readers' use of semantic and syntactic cues in the processing of English relative clauses. Bilingualism: Language and Cognition, 20 (5), 980 – 998. https://doi.org/10.1017/S1366728916000602 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Pooresmaeil, E., Mohamadi, R., Ghorbani, A., & Kamali, M. (2019). The relationship between comprehension of syntax and reading comprehension in cochlear implanted and hearing children. International Journal of Pediatric Otorhinolaryngology, 121, 114 – 119. https://doi.org/10.1016/j.ijporl.2019.03.004 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Qi, S., & Mitchell, R. E. (2012). Large-scaled academic achievement testing of deaf and hard-of-hearing students: Past, present, and future. Journal of Deaf Studies and Deaf Education, 17 (1), 1 – 18. https://doi.org/10.1093/deafed/enr028 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Quigley, S. P., & King, C. M. (1980). An invited article: Syntactic performance of hearing impaired and normal hearing individuals. Applied Psycholinguistics, 1 (4), 329 – 356. Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Quigley, S. P., Steinkamp, M. W., Power, D. J., & Jones, B. W. (1978). Test of syntactic abilities. Beaverton, OR : Dormac. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> R Core Team (2019). R: A language and environment for statistical computing (Version 3.6.1) [Software package]. Vienna, Austria : R Foundation for Statistical Computing. Retrieved from https://<ulink href="http://www.R-project.org/">www.R-project.org/</ulink>. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Ramos, J., & Cuetos, F. (1999). Evaluación de los procesos lectores en alumnos de tercer ciclo de educación primaria y ESO (PROLEC-SE). Madrid : TEA Ediciones. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Rayner, K., Chace, K. H., Slattery, T. J., & Ashby, J. (2006). Eye movements as reflections of comprehension processes in reading. Scientific Studies of Reading, 10 (3), 241 – 255. https://doi.org/10.1207/s1532799xssr1003%5f3 Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Soriano, J., Pérez, I., & Domínguez, A. B. (2006). Evaluación del uso de estrategias sintácticas en lectura por alumnos sordos con y sin implante coclear. Revista de Logopedia, Foniatría y Audiología, 26 (2), 72 – 83. https://doi.org/10.1016/S0214-4603(06)70105-X Google Scholar Crossref Search ADS WorldCat</bibtext> </blist> <blist> <bibtext> Toro, J., Cervera, M., & Urío, C. (2000). EMLE TALE-2000. Escalas Magallanes de Lectura y Escritura. Baracaldo, Bizkaia : COHS. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> <blist> <bibtext> Traxler, C. B. (2000). The Stanford Achievement Test: National norming and performance standards for deaf and hard-of-hearing students. Journal of Deaf Studies and Deaf Education, 5 (4), 337 – 348. https://doi.org/10.1093/deafed/5.4.337 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Traxler, M. J., Corina, D. P., Morford, J. P., Hafer, S., & Hoversten, L. J. (2014). Deaf readers' response to syntactic complexity: Evidence from self-paced reading. Memory & Cognition, 42 (1), 97 – 111. https://doi.org/10.3758/s13421-013-0346-1 Google Scholar Crossref Search ADS PubMed WorldCat</bibtext> </blist> <blist> <bibtext> Wiig, E. H., Semel, E., & Secord, W. A. (2006). Clinical evaluation of language fundamentals (4th Ed., Spanish). Needham Heights, MA : The Psychological Corporation. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC</bibtext> </blist> </ref> <aug> <p>By Nadina Gómez-Merino; Inmaculada Fajardo; Antonio Ferrer and Barbara Arfé</p> <p>Reported by Author; Author; Author; Author</p> <p></p> <p>Correspondence should be sent to Nadina Gómez Merino, Department of Developmental and Educational Psychology, University of Valencia, 46010 Valencia, Spain (e-mail:)</p> </aug> <nolink nlid="nl1" bibid="bib38" firstref="ref2"></nolink> <nolink nlid="nl2" bibid="bib37" firstref="ref5"></nolink> <nolink nlid="nl3" bibid="bib40" firstref="ref6"></nolink> <nolink nlid="nl4" bibid="bib48" firstref="ref7"></nolink> <nolink nlid="nl5" bibid="bib21" firstref="ref8"></nolink> <nolink nlid="nl6" bibid="bib32" firstref="ref9"></nolink> <nolink nlid="nl7" bibid="bib31" firstref="ref12"></nolink> <nolink nlid="nl8" bibid="bib36" firstref="ref15"></nolink> <nolink nlid="nl9" bibid="bib20" firstref="ref16"></nolink> <nolink nlid="nl10" bibid="bib41" firstref="ref17"></nolink> <nolink nlid="nl11" bibid="bib26" firstref="ref18"></nolink> <nolink nlid="nl12" bibid="bib42" firstref="ref19"></nolink> <nolink nlid="nl13" bibid="bib34" firstref="ref21"></nolink> <nolink nlid="nl14" bibid="bib35" firstref="ref22"></nolink> <nolink nlid="nl15" bibid="bib14" firstref="ref25"></nolink> <nolink nlid="nl16" bibid="bib46" firstref="ref26"></nolink> <nolink nlid="nl17" bibid="bib18" firstref="ref28"></nolink> <nolink nlid="nl18" bibid="bib39" firstref="ref35"></nolink> <nolink nlid="nl19" bibid="bib25" firstref="ref40"></nolink> <nolink nlid="nl20" bibid="bib49" firstref="ref43"></nolink> <nolink nlid="nl21" bibid="bib30" firstref="ref51"></nolink> <nolink nlid="nl22" bibid="bib33" firstref="ref52"></nolink> <nolink nlid="nl23" bibid="bib22" firstref="ref53"></nolink> <nolink nlid="nl24" bibid="bib24" firstref="ref67"></nolink> <nolink nlid="nl25" bibid="bib44" firstref="ref68"></nolink> <nolink nlid="nl26" bibid="bib10" firstref="ref69"></nolink> <nolink nlid="nl27" bibid="bib47" firstref="ref70"></nolink> <nolink nlid="nl28" bibid="bib16" firstref="ref71"></nolink> <nolink nlid="nl29" bibid="bib13" firstref="ref72"></nolink> <nolink nlid="nl30" bibid="bib50" firstref="ref73"></nolink> <nolink nlid="nl31" bibid="bib17" firstref="ref75"></nolink> <nolink nlid="nl32" bibid="bib15" firstref="ref76"></nolink> <nolink nlid="nl33" bibid="bib29" firstref="ref78"></nolink> <nolink nlid="nl34" bibid="bib19" firstref="ref80"></nolink> <nolink nlid="nl35" bibid="bib45" firstref="ref81"></nolink> <nolink nlid="nl36" bibid="bib43" firstref="ref91"></nolink> <nolink nlid="nl37" bibid="bib11" firstref="ref93"></nolink> <nolink nlid="nl38" bibid="bib12" firstref="ref107"></nolink> <nolink nlid="nl39" bibid="bib23" firstref="ref116"></nolink> <nolink nlid="nl40" bibid="bib27" firstref="ref118"></nolink> <nolink nlid="nl41" bibid="bib28" firstref="ref119"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1256381
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Time-Course of Grammatical Processing in Deaf Readers: An Eye-Movement Study
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Gómez-Merino%2C+Nadina%22">Gómez-Merino, Nadina</searchLink><br /><searchLink fieldCode="AR" term="%22Fajardo%2C+Inmaculada%22">Fajardo, Inmaculada</searchLink><br /><searchLink fieldCode="AR" term="%22Ferrer%2C+Antonio%22">Ferrer, Antonio</searchLink><br /><searchLink fieldCode="AR" term="%22Arfé%2C+Barbara%22">Arfé, Barbara</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Deaf+Studies+and+Deaf+Education%22"><i>Journal of Deaf Studies and Deaf Education</i></searchLink>. Jul 2020 25(3):351-364.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Oxford University Press. Great Clarendon Street, Oxford, OX2 6DP, UK. Tel: +44-1865-353907; Fax: +44-1865-353485; e-mail: jnls.cust.serv@oxfordjournals.org; Web site: http://jdsde.oxfordjournals.org/
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 14
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2020
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Deafness%22">Deafness</searchLink><br /><searchLink fieldCode="DE" term="%22Eye+Movements%22">Eye Movements</searchLink><br /><searchLink fieldCode="DE" term="%22Grammar%22">Grammar</searchLink><br /><searchLink fieldCode="DE" term="%22Accuracy%22">Accuracy</searchLink><br /><searchLink fieldCode="DE" term="%22Comparative+Analysis%22">Comparative Analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Syntax%22">Syntax</searchLink><br /><searchLink fieldCode="DE" term="%22Vocabulary+Development%22">Vocabulary Development</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Processing%22">Language Processing</searchLink><br /><searchLink fieldCode="DE" term="%22Sentences%22">Sentences</searchLink><br /><searchLink fieldCode="DE" term="%22Hearing+%28Physiology%29%22">Hearing (Physiology)</searchLink><br /><searchLink fieldCode="DE" term="%22Decision+Making%22">Decision Making</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Rate%22">Reading Rate</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Processes%22">Reading Processes</searchLink><br /><searchLink fieldCode="DE" term="%22Task+Analysis%22">Task Analysis</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1093/deafed/enaa005
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1081-4159
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Twenty participants who were deaf and 20 chronological age-matched participants with typical hearing (TH) (mean age: 12 years) were asked to judge the correctness of written sentences with or without a grammatically incongruent word while their eye movements were registered. TH participants outperformed deaf participants in grammaticality judgment accuracy. For both groups, First Pass and Total Fixation Times of target words in correct trials were significantly longer in the incongruent condition than in the congruent one. However, whereas TH students showed longer First Pass in the target area than deaf students across congruity conditions, deaf students made more fixations than their TH controls. Syntactic skills, vocabulary, and word reading speeds (measured with additional tests) were significantly lower in deaf students but only syntactic skills were systematically associated to the time-course of congruity processing. These results suggest that syntactic skills could have a cascading effect in sentence processing for deaf readers.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2020
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1256381
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1256381
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1093/deafed/enaa005
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 351
    Subjects:
      – SubjectFull: Deafness
        Type: general
      – SubjectFull: Eye Movements
        Type: general
      – SubjectFull: Grammar
        Type: general
      – SubjectFull: Accuracy
        Type: general
      – SubjectFull: Comparative Analysis
        Type: general
      – SubjectFull: Syntax
        Type: general
      – SubjectFull: Vocabulary Development
        Type: general
      – SubjectFull: Language Processing
        Type: general
      – SubjectFull: Sentences
        Type: general
      – SubjectFull: Hearing (Physiology)
        Type: general
      – SubjectFull: Decision Making
        Type: general
      – SubjectFull: Reading Rate
        Type: general
      – SubjectFull: Reading Processes
        Type: general
      – SubjectFull: Task Analysis
        Type: general
    Titles:
      – TitleFull: Time-Course of Grammatical Processing in Deaf Readers: An Eye-Movement Study
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Gómez-Merino, Nadina
      – PersonEntity:
          Name:
            NameFull: Fajardo, Inmaculada
      – PersonEntity:
          Name:
            NameFull: Ferrer, Antonio
      – PersonEntity:
          Name:
            NameFull: Arfé, Barbara
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 07
              Type: published
              Y: 2020
          Identifiers:
            – Type: issn-print
              Value: 1081-4159
          Numbering:
            – Type: volume
              Value: 25
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Journal of Deaf Studies and Deaf Education
              Type: main
ResultId 1