A Randomized Controlled Trial on The Beneficial Effects of Training Letter-Speech Sound Integration on Reading Fluency in Children with Dyslexia

A recent account of dyslexia assumes that a failure to develop automated letter-speech sound integration might be responsible for the observed lack of reading fluency. This study uses a pre-test-training-post-test design to evaluate the effects of a training program based on letter-speech sound associations with a special focus on gains in reading fluency. A sample of 44 children with dyslexia and 23 typical readers, aged 8 to 9, was recruited. Children with dyslexia were randomly allocated to either the training program group (n = 23) or a waiting-list control group (n = 21). The training intensively focused on letter-speech sound mapping and consisted of 34 individual sessions of 45 minutes over a five month period. The children with dyslexia showed substantial reading gains for the main word reading and spelling measures after training, improving at a faster rate than typical readers and waiting-list controls. The results are interpreted within the conceptual framework assuming a multisensory integration deficit as the most proximal cause of dysfluent reading in dyslexia. Trial Registration: ISRCTN register ISRCTN12783279

Goal of the study is improving reading fluency in dyslexia. We will therefore deliver an intervention focussed on developing of automatic letter-speech sound integration to children with dyslexia, in order to provide a neurocognitive window on the learning mechanisms involved in developing reading fluency.

STUY OBJECTIVES
The main objective of study is to determine whether a computer-based intervention focussed on intensively training automation of letter-speech-sound associations can improve reading fluency in children with severe dyslexia.
In more detail, our objectives are:  Does the intervention improve knowledge of letter-speech sound associations, compared to the no-intervention control group?  Does the intervention improve automation of letter-speech sound integration, compared to no-intervention control group?  Does the intervention improve reading accuracy, compared to the nointervention control group?  Does the intervention improve reading fluency, compared to the no-intervention control group?  Are the gains in reading fluency associated with initial levels and/or gains in the automation of letter-speech sound associations?  Are the intervention-induced reading-related growth curves of the dyslexic children comparable to those of normal readers?  Does the intervention induce changes in brain responses in a local network related to letter-speech sound processing, as measured by ERPs of audiovisual MMN?  Does the intervention induce changes in brain responses in a more global neural network for fast visual word processing, as measured by ERPs of visual word recognition?

Study Design
The study is an open randomised controlled trial comparing an intervention addressing letter-speech sound integration to a waiting list control group. Children with severe dyslexia are randomly assigned to either an immediate intervention group or a waitinglist control group by a method of simple randomisation. Participants will be randomised using a computerised random number generator. Children in the intervention condition will receive 34 intervention-sessions in a 19 week period.
In addition, a second control group of normal readers who do not receive the intervention (same age-range, school grade, sociodemographical background, and exclusion criteria as the children with dyslexia) is included in the study.
Behavioural measures of reading, spelling, and letter-speech sound mapping are measured at baseline and at the end of a 19 week (intervention) period for all three groups. EEG measurements of visual word recognition and of letter-speech sound integration are measured at baseline and at the end of the intervention (after 19 weeks) for the intervention group, and only once, at approximately 10 weeks, for both the waiting-list control group and the normal-reader control group (used as a criterion point to compare the pre-post ERP-changes in the intervention with).

Subject Selection
Inclusion Criteria for severe dyslexia  Diagnosis of dyslexia (according to criteria of Dutch Dyslexia Foundation, and DSM-IV-TR (reading disorder))  Suffering from specific and persistent (i.e., failed to respond to remedial support at school) problems with reading at school  Percentile score ≤ 10% on a standard reading test  Age 8-9 years old  In grade 3 of primary education

Number of Participants
Planned sample size: N = 50 children with severe dyslexia; subjects fulfilling the entry criteria will be randomised to one of the two trial arms. N=25 normal reading children.

Subject Recruitment
Children with dyslexia will be recruited through the IWAL Institute. Children who (1) are referred by schools to the IWAL Institute because of specific and persistent reading problems at school, (2) received a diagnosis of dyslexia after an extensive cognitive assessment by a psychologist at the IWAL Institute, (3) had a reading score ≤ 10% on a standard reading test at the diagnostic assessment, and (4) were in grade 3 at school, are recruited.
Normal reading children will be recruited from the same sample of schools as the dyslexic children are referred from.

Informed Consent
All participants will receive a consent form. The consent form explains the aims of the study, and the study procedures. The consent form explicitly states that the participant is free to withdraw at any time without any need to give reason.

Sample size justification
Before running a sample size analysis, we first calculated the expected effect size for the intervention gains in reading fluency on the basis of the data of previous intervention studies (Tijms, in revision, Saine et al., 2009). These data revealed expected effect sizes in terms of Cohen's f in the range of 0.33 -0.40 for gains in reading fluency by intervention in comparison to a no-intervention control group. This is generally considered a medium to large effect size. Therefore, our expected intervention effect size is f = 0.33 -0.40.
As we intent to analyse the intervention effects by means of an ANCOVA with pretest scores as covariates for each corresponding posttest score (Vickers & Altman, 2001), we also calculated the R 2 for the covariate pretest score in reading fluency in the study of Tijms (in revision): R 2 = .37. By including an expected R 2 for the covariate, we were able to calculate the sample size more precisely.
Next, we run a sample size calculation (Power & Precision V4 software; Borenstein, Rothstein, & Cohen, 2001). Our aim is a power of at least 0.80 to detect an intervention effect on reading fluency, assuming the above noted expected effect size (f = 0.33 -0.40), and conducting an ANCOVA with 1 factor (intervention condition, 2 levels) and 1 covariate (pretest level, R 2 = .37), and α = 0.05. The power analysis revealed a power of at least 0.80 for n ≥ 20 per intervention condition.
Based on this power calculation and given an anticipated attrition rate of 20%, we aim to recruit a total of 50 dyslexic children.

Intervention
The intervention in this study is a computerised intervention focussed on the integration of letters and speech sounds in the context of the acquisition of reading fluency. This training is an adaptation of the treatment protocol reported in Tijms (2007). The intervention program is delivered by well-instructed junior psychologists, on a one-to-one basis for 45-minute sessions during a 19 week period. The training frequency is two sessions per week. Intervention takes place at a clinical centre.

Intervention Sites
This intervention will be conducted at the Rudolf Berlin Center in Amsterdam, and at the associated locations of the IWAL Institute in the Amsterdam area.

FUNDING
This project is funded by the Netherlands Initiative Brain and Cognition, a program of the Organization for Scientific Research (NWO), as part of the project "Fluent reading acquisition neurocognitively decomposed: The case of dyslexia" under grant number 056-14-015.

Computers
In both computers the safe mode (3 rd option "Metingen") should be selected in the initial screen when they are started. Otherwise antivirus and windows update might interfere in the analysis. The starting of the computers, preparation of files and connection check must be done before the participants come to the room. Time: ~5 minutes.

Biosemi computer
 Initiate the Biosemi Actiview software  Load the config file "gorka.cfg" by clicking on about Actiview/ load config file.

 Always check that the Biosemi AD box is connected and recognized in the Biosemi computer (USB connection)
 Always check the battery indicator and replace battery if it is not 100% full (have always a battery charging in the lab)  Keep in mind that by default the starting file, initiates the file with the "pause" button clicked so it will not start saving unless we click on "pause" again.

Presentation computer
 Insert instructions of what to do before the child enters the room.
 Prepare which experiment to load and the folders to load/save the files.
 Check all other settings (button devices, logfile naming etc)

Video computer
 Set instructions and starting point of video. We show the movie "Wall-E" in a laptop computer. The initial 5 minutes are not displayed to avoid the starting credits. The experiment is consisted of 3 parts. First you will do a task where you will hear two sounds " a" and " o" . Than the second part will have to read some words and in the third part you will again listen to the sounds. When each of the experiments starts we'll give you more detailed instructions.  Show and explain materials. Let them feel the plastic part of the syringe, the electrodes, etc.

Time distribution
 Give a written detailed description to tutor/parent (same as that sent in the informed consent letter).

Electrodes preparation steps
1. Measure head circumference + nasion-inion and choose a cap 2. Fasten the velcro band to the cap.
3. Prepare location for the external electrodes (also for frontal electrodes) : alcohol-scrub gelwater-dry.
4. Place the external electrodes in the mastoids, put extra gel.
5. Place the cap. Check that is well centred with the measurements of nasion-inion. Fast the Velcro band.
6. Fill the holes for the electrodes with gel.
7. Ask participants if he/she wants to go to the WC, etc.
8. Place electrodes in the cap.

Place external electrodes for eyes movements
10. Check on screen that the placement of electrodes is correct, the electrodes signal in the monopolar view and the electrodes offset. The offset of the electrodes should be between -20 mV and 20 mV, if this is not the case locate the electrodes out of this range and place them again, adding more gel and/or gently scrubbing the hair with the syringe.

11
. Show the Biosemi screen to the participant and do some practice to see the most common artefacts (blinking, biting, yawning, movements...)

Instructions for the participant
Before a new task/experiment the following instructions are shown on screen. The assistants also explain them and make sure they are clear to the participant. They also inform about how many blocks, how long they last and when the next pause comes.
Luister en kijk goed naar de letters en klanken.

End
"Dit was de test, je hebt het heel goed gedaan" bedankt voor het meedoen!" together with a cartoon of sponge bob.

Behavioural aspects
It is important to keep the participant motivated, relax and comfortable. The pauses should be taken as close to our schedule as possible. The assistants are important for all these aspect and must be actively interacting with the participants:

Arrival
Show the material and let them see/ interact with it (i.e. syringe). Explain in a clear way what the material is for. Make the child feel comfortable in the lab and with the people in the lab.
Explain the general aspects of the experiment: approximate length, pauses, type of tasks, importance of sitting still, say that they can stop if feeling unease, etc.
Also let him know that there will some rewards when he/she finishes the experiment and also a diploma will be given to him to acknowledge his participation . Make the child feel safe and enthusiastic about the experiment.

During the preparation
While the assistant himself and the experimenter are placing the cap and electrodes. Describe the different steps, ask him/her to say if they feel uncomfortable (just once or twice, not too often) . If the participant find it boring or by any reason is taking longer than expected, talk to him or offer him a quick drink or snack. Make the child feel comfortable with the apparatus.
After the preparation the child can make a picture with the cap and the electrodes that will be used in the diploma.

During the experiments
Before each experiment, the instructions are displayed on screen and the assistant explains them to the participant making sure he/she understands them correctly. Also ask if he/she has any doubt about it.

During the pauses
At the end of each block give always positive feedback and only if needed remind the importance of sitting still in the next block. Make clear during the whole experiment when there will be pauses. Explain always the length of the pauses and what comes next. Have always a scheme of the experiment and show it so the participant knows at all time in which step he/she is.
In the short pauses he/she can adjust his/her position in the chair, do some stretching, have a drink or snack, do some drawing, and have a talk with the assistant and experimenter.
There will be also board games available. In the long pause the participant should also stand up take a walk through the labs, go to the toilet, play a board game with the assistant, etc.

End of the experiment
Give positive feedback, let him know that what his/her participation was very important and he/she completed all the task well. Give also and immediate reward.

Biosemi recording
When the participant sits prepared (all electrodes in place, connected and still) click on PLAY button (before this the Biosemi computer/software must be also ready, see above).
Display the Biosemi software screen on the monitor visible to the participant (switch between monitors with the switch box).

General check of signal
In the monopolar view: Are there any misplaced, thick, black or noisy lines even when the subject is still and relax? If yes, check which electrodes are wrong and that the external electrodes are still on place (if these are the ones giving the distort signal).
Is the whole signal interrupted, distorted or noisy? If yes, check that the ground electrodes are well placed, connected or if there is enough gel in them.

Electrodes offset:
Keep the electrodes offset range between -20 mV and 20 mV. If an electrode is not in range, locate it and add more gel and /or gently scrub the hair with the syringe.  In the Monopolar display explore other artefacts, such as head movements, biting, yawning or laughing.
 Explain them that the more relax and still they are the clearer and nice the lines will be: "Als je gewoon rustig en ontspannen zit en het taakje uitvoert dan kunnen we hele mooie golfjes/golven meten"

Recording and pauses
Pauses between blocks  The program shows the words " Goed zo!!! even pauze" or "Prima!!even pauze!!!"  Pause the Biosemi software  The assistant interacts with the participant (see 'behavioural aspects') -In case child sits still very well -'heel goed gedaan, de golfjes zien er heel mooi uit wat kan je dat goed zeg!' In case the child moves quite a lot 'goed gedaan, als je na de pauze nog wat rustiger blijft zitten kunnen ze nog mooier worden'  Once the participant is ready to continue, check that the signal is still ok and click again in the pause button of the Biosemi to continue recording (always make sure that the file size keeps increasing)  Start the presentation file

Pauses within a block
If the participant moves a lot or by any other reason a pause is needed: press ESC in presentation, then the word ' pauze' should appear. To go further press the key set to continue.

Storage
 When starting a file keep in mind that by default the starting file, initiates the file with the " pause" button clicked so it will not start saving unless we click on "pause" again.