Plain English Summary
Background and study aims
One consequence of a stroke can be aphasia – difficulty understanding and producing language. It limits a person’s ability to take part in conversations. There has been considerable progress in vocabulary interventions for aphasia, but less in therapies for sentence processing. This is a problem as everyday talking involves sentences, and understanding and using sentences are critical for taking part in conversations. We apply a new theory of sentence processing to aphasia: usage-based construction grammar. In this intervention study, we test the effectiveness of a new computer therapy. It aims to improve understanding of spoken language and to stimulate flexible production of common phrases in everyday talking. The treatment involves listening to high-frequency phrases, and subsequent practice in producing those phrases with increasing flexibility as new vocabulary is inserted into the phrase. We will explore the effects of combining computer therapy with a form of non-invasive brain stimulation (transcranial direct current stimulation, or tDCS).
We will measure language abilities twice before therapy starts to discover if behaviour is stable before intervention. All participants will then take part in 12 sessions of computer therapy spread over a period of 4 weeks. In addition to the computer therapy, participants will receive either active tDCS or sham tDCS, to determine whether tDCS enhances the effects of computer therapy. We then re-measure language abilities immediately after the therapy period and again after an 8 week period to determine the longer-term effects of the therapy.
Who can participate?
People with aphasia who have had a stroke at least 6 months ago, who previously were competent in speaking English. Participants must have sufficient visual and auditory acuity to interact with a computer.
What does the study involve?
Participants have to travel to University College London and complete 12 sessions of therapy in a 4 week period. Each session lasts for approximately one hour. All participants will receive a new computer therapy aimed at improving sentence comprehension and production abilities, together with either active or sham tDCS for 20 minutes in each therapy session. Participants will also complete a further 4 assessment sessions, two before the intervention and two after intervention. Language ability, linked cognitive skills and attitudes and feelings to life after stroke will be measured. Audio recordings of participants’ speech will be made. Participants who meet MRI safety criteria will have an MRI brain scan at the end of the study.
What are the possible benefits and risks of participating?
The possible benefits of participating are that individuals will receive 4 weeks of intensive therapy which may lead to improved sentence understanding and production.
There are no known side effects associated with computer therapy. However, participants might not benefit from this therapy. During the intervention, there is the potential for some distress, as a
participant is confronted with the loss of linguistic competence. tDCS and MRI are not suitable for all people (e.g. people with metal implants in the head or chest). tDCS can cause slight itching or irritation which usually fades. There are no known medium or long-term risks of tDCS. MRI may be unsuitable for people who are anxious in confined spaces, and some people do not like the sound of the scanner when it is in operation.
Where is the study run from?
The study is run from the Bloomsbury campus (Chandler House) of University College London (UK).
When is the study starting and how long is it expected to run for?
The study begins in September 2019 and will run until February 2022.
Who is funding the study?
The Stroke Association.
Who is the main contact?
Professor Rosemary Varley
Reconstructing sentence processing in aphasia: Unification Therapy Integrating LexIcon and Sentences
1. Sentence therapy is more effective than usual care.
2. Outcomes of behavioural therapy are enhanced by active-anodal tDCS over left inferior frontal gyrus.
3. The effects of therapy are retained over an 8-week no-treatment period.
4. Maintenance is enhanced by active-anodal tDCS.
Approved 12/06/2019, UCL Research Ethics Committee (Office of the Vice Provost Research, 2 Taviton Street University College London; +44 (0)20 7679 8717; firstname.lastname@example.org), ref: 8123/001.
Single centre randomised controlled trial
Primary study design
Secondary study design
Randomised controlled trial
Patient information sheet
After the initial baseline assessment, participants are randomised to ‘immediate’ or deferred treatment conditions, and active or sham transcranial direct current stimulation (tDCS). Randomisation is performed by an external randomisation service. Researchers and participants are blind to tDCS condition. Participants in the ‘immediate’ condition complete a second baseline assessment four weeks later and then immediately start the intervention. The intervention consists of a computerised behavioural intervention combined with active or sham tDCS. Stimulation is applied to the left inferior frontal region. Active stimulation parameters are 20 minutes at 1.5 mA. Sham stimulation comprises the same electrode montage, but, after initial ramp-up to 1.5 mA, stimulation is slowly decreased over 30 seconds. The computerised intervention comprises three tasks: attention to auditory sentence stimuli; attention to key components of sentence structure in a word monitoring paradigm; sentence production with gradual increase in sentence length and creativity. All stimuli include frequent phrase/sentence frames. The treatment is delivered in 12 x one hour sessions over a four week period.
Participants in the deferred condition will act as a waiting list control. They complete a second baseline assessment at eight weeks after the first baseline evaluation and then enter the intervention phase with randomisation to either active or sham tDCS.
For both study arms, outcomes are measured immediately at end of the treatment phase and again after an eight-week no treatment/maintenance period.
Primary outcome measure
1. Sentence comprehension is measured using the Test for Reception of Grammar (TROG-2, Bishop, 2003).
2. Sentence production is measured using narrative speech evaluated by automated language analysis with the key variable of combination ratio (number of 3-word combinations/total number of words).
Primary outcome measures are completed twice at baseline to evaluate pre-intervention stability of behaviour. Assessments will be repeated immediately on cessation of treatment and after an 8-week no-treatment maintenance period.
Secondary outcome measures
1. The capacity to produce treated and matched control sentences is measured using a study-specific story completion test at Baseline 1 and immediately after the intervention.
2. Perceptions of quality of life are measured using SAQOL-39 at Baseline 2 and second/maintenance outcome (Hilari et al., 2003).
3. Sentence production is also evaluated using connected speech measures of mean length of utterance, function word ratio and frequency characteristics of word combinations at all four assessment time-points.
Overall trial start date
Overall trial end date
Reason abandoned (if study stopped)
Participant inclusion criteria
1. Premorbid competence in English
2. Premorbidly right-handed
3. Single left hemisphere stroke
4. At least six-months post-stroke onset
5. Presence of aphasia, characterized by spoken and written comprehension impairment and spoken sentence production difficulties
6. Capacity to give informed consent
7. Sufficient auditory and visual capacity to interact with the behavioural therapy
Target number of participants
Participant exclusion criteria
1. Significant other neurological disorder (e.g., neurodegenerative illness)
2. History of speech and language disorder prior to stroke (e.g., developmental dyslexia)
3. Metal implants on/in head causing tDCS risk
4. Recurring headaches
5. History of seizure in the past 3 months
6. Current involvement in another brain stimulation or behavioural therapy trial
Recruitment start date
Recruitment end date
Countries of recruitment
Trial participating centre
University College London
Language & Cognition, Chandler House, 2 Wakefield Street
Funding Body Type
private sector organisation
Funding Body Subtype
professional associations and societies
Results and Publications
Publication and dissemination plan
Protocol will be published in open science source by 01/12/19
Pilot case series will be submitted for publication by 01/03/20
Trial outcomes will be submitted for publication by 01/07/22
IPD sharing statement: the data sharing plans for the current study are unknown and will be made available at a later date.
Intention to publish date
Participant level data
Stored in repository
Basic results (scientific)