Keywords
learning disability - hearing - evoked potentials auditory - cognition - learning
Introduction
The initial process of reading and writing acquisition is directly related to phonological
awareness ability (i.e., the ability to recognize, decompose, compose, and manipulate
speech sounds).[1]
[2]
[3] Phonological awareness is the ability to understand that words are made up of smaller
components that can be separated and manipulated. It is the ability to analyze speech
in its phonological components and their combinations based on acoustic recognition.[4]
[5]
The proper development of phonological awareness is vital for the child to correlate
the aspects of speech sounds with the writing code by converting phoneme to grapheme,
properly developing the foundations of reading and writing.[6] Studies show that students with learning disabilities present impairments in cognitive,
linguistic, visual processing, and auditory information processing. When the activation
of cognitive mechanisms to analyze, synthesize, manipulate, store, and recall linguistic
information is altered, impairments in phonological awareness and phonological working
memory occur, causing difficulties in the perception and production of speech whether
orally or in reading.[7]
[8]
In Brazil, behavioral tests of auditory processing are applied in schools and highlight
the association between phonological awareness abilities and auditory processing,
confirming that low performance on phonological awareness tests may be due to low
performance on auditory processing tests.[7]
[8]
[9] In addition to behavioral tests to assess hearing, the auditory middle latency auditory
response (AMLR),[10]
[11] an objective test for assessing central auditory system, has also been employed
in children with learning disorders. Several studies have demonstrated the diagnostic
value of this potential to assess the injuries of central auditory nervous system
(CANS); however, little is known about the value of AMLR diagnosis for cases of auditory
processing disorder and its effectiveness to investigate the occurrence of CANS impairments.
Auditory evoked response also has advantages over behavioral tests, as it helps in
differentiating expressive or receptive problems.[12]
Studies reported that AMLR in students with learning disorders show typical wave morphologies,
such as elongated latency for Na wave and decreased amplitude for Nb wave, consistent
with perceptual difficulty at the cortex level.[12]
[13] Currently, research describing AMLR and phonological awareness in students with
learning disorders analyze the tests separately[14]
[15]; therefore, the literature does not describe any studies directly associating phonological
awareness and AMLR, evidencing the need to invest in scientific research in this field.
Based on these points, this study aimed to characterize the AMLR potentials and the
phonological awareness tests and to investigate correlations between responses in
a group of students with learning disorders.
Methods
This study was conducted after the institution's Ethics Committee reviewed and approved
it under submission number 1512/2007, case number 118/2007, after subjects signed
the informed consent. This research is characterized as an observational, cross-sectional,
nonrandomized study.
The investigation included 25 children of both sexes (14 boys and 11 girls) with learning
disorders; 72% were between 8 and 10 years of age ([Table 1]), and average age was 9 years and 9 months. The children were enrolled in the third
grade of elementary school.
Table 1
Distribution of participants by age group
|
Age (y)
|
n
|
%
|
|
8
|
7
|
28
|
|
9
|
5
|
20
|
|
10
|
6
|
24
|
|
11
|
2
|
8
|
|
12
|
2
|
8
|
|
13
|
1
|
4
|
|
14
|
2
|
8
|
|
Total
|
25
|
100
|
The participants were volunteers from a learning disorders clinic. They showed no
medical history of chronic diseases, epilepsy, motor developmental disorders, motor
or sensory impairment disorder, attention deficit hyperactivity disorder, or autism
spectrum disorder.
The diagnosis of learning disorders was considered when, during assessment by the
multidisciplinary team (neurologist, neuropsychologist, and speech-language pathologist)
of the students' institution, any of the following was found: discrepancy between
verbal IQ and performance in psychological assessment; Wechsler Intelligence Scale
for Children-III[16] changes in the memory, reading, and writing in the neuropsychological tests[17]; difficulty in oral reading and writing under dictation of words and pseudowords;
phonological disorders in speech and writing; changes in syllabic and phonemic skills
in phonological awareness tests[18]; significant changes in syntactic and semantic language skills and in other areas
of learning such as mathematical reasoning.[19]
All participants in this study were submitted to the Phonological Awareness Test:
Instrumento de Avaliação Sequential—CONFIAS (phonological awareness sequential evaluation
instrumentl),[20] which is composed of two parts. The first part of the test corresponds to syllabic
awareness and consists of nine items: synthesis, segmentation, identification of initial
syllable, rhyme identification, production of a word with a given syllable, identification
of medial syllable, rhyme production, exclusion, and transposition. The second part
of the test corresponds to phoneme awareness and consists of seven items: production
of a word that begins with the given sound, identification of initial phoneme, identification
of the final phoneme, exclusion, synthesis, segmentation, and transposition. The test
was scored in a specific protocol. Each correct answer given by the student is scored
1 point and each incorrect answer is scored 0 points. The highest possible score is
70 (syllabic tasks = 40 points and phonemic tasks = 30 points).
The evaluations were conducted individually, in a silent room, recorded in MP3 audio
using Sony Ericsson, Model K79, São Paulo, Brazil. The average duration of the phonological
awareness application test was 40 minutes,[21] and average duration for the AMLR (Auditory Middle Latency Response) test was 20
minutes; both were performed in a single session. To record the AMLR, right and left
ears were stimulated and electrodes were placed on the right and left hemispheres.
The students were positioned in a recliner and instructed to remain with their eyes
open and alert. The exam environment was protected acoustically and electrically.
The electrodes were fixed with microporous tape after cleaning the skin with abrasive
paste. Electrolytic paste was used to improve the conductivity. The impedance of each
electrode did not exceed 5 kΩ, and impedance between the electrodes did not exceed
2 kΩ.[22]
For data collection, the electrodes were placed at C3 and C4 (left and right hemisphere),
in reference to the ears A1 and A2 (left and right ear), ipsilateral and contralateral
matched, and ground at Fz (forehead). As stimuli, rarefaction monaural filtered clicks
at 80-dB hearing level were used, with presentation rate of 11 stimuli per second,
analysis time (window) of 100 milliseconds, acoustic filter of 10 to 100 Hz, and sensitivity
of 75 μV.
Results were interpreted based on the latency of Na and Pa waves and Na-Pa amplitude,
parameters that were classified into normal or abnormal (prolonged) according to the
normality recommended in the literature, which states that in normal conditions, Na
appears as the first highest negative peak between 12 and 27 milliseconds; followed
by Pa, which is the highest positive peak after Na, between 25 and 35 milliseconds;
and then Na-Pa amplitude, with approximate value of 1.0 μV.[22]
Statistical analysis was performed using the Statistica 7.0 software, São Paulo, Brazil.
We first performed descriptive statistics (mean, confidence intervals, and standard
deviation), then verified data normality using the Shapiro-Wilk test, identifying
the study variables as nonparametric.
To analyze the correlation between measures of MLR and phonological awareness, we
used the coefficient of linear Spearman correlation, which can vary between −1 and
+1. A value of −1 is a perfect negative correlation and the value +1 is a perfect
positive correlation. The value of 0 represents no correlation. The significance level
was ≤0.05.
Results
The results of the phonological awareness test—CONFIAS, including the scores on the
subtests of syllabic and phonemic awareness, are presented in [Table 2].
Table 2
Distribution of participants as for CONFIAS score in syllabic and phonemic subtests
|
CONFIAS
|
Research group (n = 25)
|
|
Mean score
|
Standard deviation
|
Minimum
|
Maximum
|
|
Subtests syllabic
|
28.64
|
6.06
|
16.00
|
38.00
|
|
Subtests phonemic
|
17.16
|
6.47
|
6.00
|
28.00
|
Abbreviation: CONFIAS, phonological awareness sequential evaluation instrument.
[Table 3] shows the descriptive statistics (mean, confidence interval, and standard deviation)
of the AMLR measurements and combinations of ear and hemisphere in relation to the
variables Na latency, Pa latency, Nb latency, and Nb and NaPa amplitude.
Table 3
AMLR Statistics according to ear and hemisphere
|
Ear
|
Hemisphere
|
Variable
|
Mean
|
Mean 95% CI
|
Standard deviation
|
|
LL
|
UL
|
|
Right
|
Right
|
lat_na
|
19.16
|
17.73
|
20.58
|
3.45
|
|
lat_pa
|
35.29
|
32.67
|
37.92
|
6.36
|
|
lat_nb
|
49.31
|
46.3
|
52.32
|
7.3
|
|
Ampl_Na_Pa
|
1.36
|
1.13
|
1.6
|
0.56
|
|
age
|
9.92
|
9.15
|
10.69
|
1.87
|
|
Left
|
lat_na
|
19.1
|
17.75
|
20.45
|
3.27
|
|
lat_pa
|
35.35
|
32.91
|
37.79
|
5.91
|
|
lat_nb
|
50.40
|
47.59
|
53.31
|
6.93
|
|
Ampl_Na_Pa
|
1.55
|
1.02
|
2.09
|
1.29
|
|
age
|
9.92
|
9.15
|
10.69
|
1.87
|
|
Left
|
Right
|
lat_na
|
20.7
|
18.83
|
22.57
|
4.53
|
|
lat_pa
|
35.42
|
32.45
|
38.39
|
7.19
|
|
lat_nb
|
49.34
|
45.92
|
52.77
|
8.31
|
|
Ampl_Na_Pa
|
1.52
|
1.18
|
1.86
|
0.83
|
|
age
|
9.92
|
9.15
|
10.69
|
1.87
|
|
Left
|
lat_na
|
19.9
|
18.25
|
21.55
|
3.99
|
|
lat_pa
|
34.46
|
31.84
|
37.07
|
6.33
|
|
lat_nb
|
49.69
|
46.53
|
52.85
|
7.65
|
|
Ampl_Na_Pa
|
1.25
|
0.9
|
1.59
|
0.84
|
|
age
|
9.92
|
9.15
|
10.69
|
1.87
|
Abbreviations: Ampl_, amplitude; CI, confidence interval; lat_, latency; LL, lower
limit; AMLR, auditory middle latency response; UL, upper limit.
[Table 4] shows the correlation measure between the results of phonological awareness and
AMLR tests according to ear and hemisphere. Measurements of the Spearman correlation
coefficient negative showed weak to moderate correlation, indicating that with increased
scores of the phonological awareness subtests, the latency values of Na, Pa, and Nb
waves decreased in both ears and hemispheres, especially those obtained via the contralateral
ascending right ear left hemisphere and left ear right hemisphere.
Table 4
Spearman correlation (R) between the CONFIAS and AMLR subtests according to ear (right
and left) and hemisphere (right and left)
|
CONFIAS
|
Right hemisphere
|
Left hemisphere
|
|
Na
|
Pa
|
Nb
|
Amp Na-Pa
|
Na
|
Pa
|
Nb
|
Amp Na-Pa
|
|
Right ear
|
|
|
|
|
|
|
|
|
|
Syllabic
|
−0.27
|
−0.42[a]
|
−0.02
|
0.14
|
−0.22
|
−0.61[a]
|
−0.28
|
−0.30
|
|
Phonemic
|
−0.08
|
−0.27
|
0.19
|
0.21
|
−0.16
|
−0.26
|
−0.05
|
0.24
|
|
Left ear
|
|
|
|
|
|
|
|
|
|
Syllabic
|
−0.54[a]
|
−0.52[a]
|
−0.25
|
0.49[a]
|
−0.58[a]
|
−0.45[a]
|
−0.03
|
0.40[a]
|
|
Phonemic
|
−0.35
|
−0.24
|
0.05
|
0.32
|
−0.07
|
0.12
|
0.26
|
0.38
|
Abbreviations: Amp, amplitude; CONFIAS, phonological awareness sequential evaluation
instrumentl; AMLR, auditory middle latency response.
a
p < 005.
A negative moderate correlation was observed in the Pa component and syllabic phonological
awareness. Results are presented in [Fig. 1]. The opposite effect was found for amplitude Na-Pa measures, where a weak positive
correlation was observed, indicating that the phonological awareness scores increased
as Na-Pa amplitude also increased.
Fig. 1 Pa component and syllabic phonological awareness. Abbreviations: RE, right ear; LE,
left ear; RH, right hemisphere; LH, left hemisphere.
Discussion
The altered phonological awareness abilities result from difficulty in cognitive mechanisms
to analyze, synthesize, manipulate, store, and recall linguistic information.[11]
[23] Such changes were identified in the sample of this study, revealing that students
with learning disabilities and average age of 9 years and 9 months showed low scores
for the age group in phonological awareness subtests, specifically in phonemic awareness.[18]
[20]
[21] The phonological awareness abilities and auditory processing have been investigated
in students with learning disorders, and these studies have revealed the relationship
between low performance, both in phonological awareness and auditory processing.[7]
[8]
[11]
[19]
[24]
Studies investigating phonological awareness abilities associated with auditory abilities
in students with dyslexia and learning disorders revealed that coding and organization
of acoustic spectra impairments may be responsible for changes in metalinguistic abilities
observed in this population.[7]
[9]
[19] The deficits in information auditory processing have been studied in populations
of children with learning disorders, and ∼80% of students with complaints related
to reading and/or writing also present auditory processing alterations, which often
interfere with the acquisition of reading and writing.[23]
The AMLR examination revealed, from the descriptive statistics in [Table 3], that the prolonged Pa latency in the contralateral pathway in children with learning
disorders does not reach values higher than 30 ms in Pa latency in pediatric populations
with typical development.[10]
[13]
[25] The neural origin of the Pa wave is attributed to the medial area of Heschl gyrus,[26] which is responsible for acoustic recognition and discrimination abilities of the
auditory cortex.[27] In this study, when the correlation between the Pa component and the syllabic awareness
was performed, there was an association between AMLR and phonological awareness in
students with learning disorders, confirming the relationship between auditory and
phonological abilities. Thus, the operation of the ascending auditory pathways is
essential for decoding sound information because any changes affect the phonemic structure
and auditory-linguistic association in the primary auditory cortical area as in the
case of students with learning disabilities.[28]
Recognition and manipulation of syllabic structure during the syllabic awareness tests
are not only based on the acoustic-perceptual characteristics of intensity and duration
of the syllable. Speech perception can also be understood as an extralinguistic system
of the auditory system related to the symbolic and abstract phenomenon of cognition.[29] During the syllabic awareness tests, structures are recognized from the lexical
knowledge that recalls their meaning through previous linguistic experiences,[29] which could explain the better performance of students in this study of syllabic
tasks.
The CANS is a net composed of innumerous nerve fibers, and the majority of these fibers
cross and uncross this system at some point. Thus, the left auditory cortex is dominant
for the perception of linguistic stimuli, and the right auditory cortex is more functional
for the perception of tonal sounds. Such interpretation justifies the fact that a
specific association has been observed between syllabic awareness and AMLR (Pa wave)
in students with learning disorders stimulated with nonverbal sounds in this study.
The opposite effect was observed for Na-Pa amplitude measures, where a positive weak
correlation was found. The local electrical activity measurement of these components,
described in microvolts, results from the auditory stimulation of the primary auditory
cortex and secondary areas involved in linguistic processing and tends to be reduced
in patients with learning disorders[30]; these values may increase as auditory performance and phonological awareness improve.
Further studies with different methodologies should be conducted, such as case–control
studies and randomized clinical trials, to help understand the findings in this study
(i.e., whether the findings constitute specific manifestations of students with learning
disorders or whether linguistic or educational environment may influence directly
or indirectly on the occurrence of such events observed in this study). Further studies
that address assessment and intervention to analyze the variation of AMLR measures
concerning decrease in latency and increase in amplitude of students' waves may favor
the observation of the relationship between the measures and confirm the positive
effect of simultaneous interventions under hearing and phonological awareness, thus
reducing the risks of generic interpretations and possible limitations found in this
study.
Conclusion
This study allowed a better characterization of AMLRs and phonological awareness tests
in children with learning disabilities and allowed us to understand the correlation
between the tests in the study group. When phonological awareness subtests were performed,
specifically the phonemic awareness tests, the students showed a low score for the
age group. Prolonged Pa latency was observed in contralateral pathway in the AMLR
test.
In the correlation between the measurements, we observed a weak to moderate negative
correlation for the latency of wave Pa and a weak positive correlation for the Na-Pa
amplitude, which may indicate a relationship between the measures and confirm the
positive effect of simultaneous interventions under hearing and phonological awareness.
Thus, this study showed that auditory and metaphonological training can provide changes
in the neurophysiological response in the auditory pathway, and consequently in results
for AMLR and phonological abilities.
