An interventional study of Grammatical Reasoning and Card Pairs in Healthy Aging, sponsored by University of Exeter. Completed at 1 site in United Kingdom. Open to participants aged 50 Years to 100 Years, including healthy volunteers. Per ClinicalTrials.gov, last updated 2026-05-04.
Sponsored by University of Exeter · Not applicable, Interventional, and Prevention
The START study will investigate the impact of playing a Grammatical Reasoning brain training task on overall brain function (cognition) and day-to-day function in people over 50.
.
The study builds on existing work that has shown very promising results with brain training approaches in older adults. There is good evidence that playing brain training games leads to improvements in brain functions including memory and reasoning. There is also a suggestion that regular involvement in brain training games might help reduce the risk of cognitive decline and dementia later in life. It appears that reasoning, or problem-solving, is particularly important as it is one of the first abilities to decline with age. This study will use this evidence to target reasoning in a specific brain training task. It will also explore the potential role of genetic factors in how people perform in the task.
The study will compare the impact of a Grammatical Reasoning Task (START) with a control task consisting of simple picture-matching. Over 7000 participants will be randomly allocated to either START or the control and will be asked to play the task as often as they wish for a period of six weeks. Both tasks will be delivered entirely online so people will access the study from home on their computers. This will be achieved through our dedicated online research platform, PROTECT, which hosts a cohort of over 20,000 older adults who have provided DNA samples as part of a longitudinal study.
This study will measure the impact of the training on cognition and Instrumental Activities of Daily Living. Assessments will be completed online at baseline, two/six weeks and six months. Overall, this study aims to generate exciting new data about how brain training could be included in guidance on healthy ageing
University of Exeter is the lead sponsor of 138 studies on the registry; 25 are open to participants now.
Counted across the registry records on this site, refreshed daily.
Exclusion Criteria:
This group will complete a short-term Intensive Grammatical Reasoning cognitive task delivered online by Wesnes Cognition Ltd (START). Participants will be encouraged to complete the START training once a day for six-weeks.
Other: Grammatical Reasoning
The control group will complete a basic picture-matching task that will provide the same level of engagement, but without the learning effects.
Other: Card Pairs
START measures the ability to determine the relationships among different shape combinations which are assigned to grammatical statements about them which can be either correct or incorrect. It is a measure of attention, working memory and executive control.
Also known as: START
A basic picture-matching task.
Change in Executive Function From Baseline
Measured by the validated task-switching test of Trailmaking B. This is a well established computerised test of executive function. On a computer screen participants are presented with 24 circles with either numbers (0-13) or letters (A-L). They must select the circles in the correct numerical and alphabetic order, alternating numbers and letters each time (e.g. 1, A, 2, B, 3, C etc). The test is completed when all the circles have been correctly selected. The test outputs a total score. A higher score indicates a better performance. A negative mean score indicates a decline in performance. The range of scoring for this test is 0 - 300,000. Data reported here are for mean change per group form baseline to six weeks
Time frame: Baseline to Six weeks
Change in Numerical Working Memory From Six Weeks
Digit Span test for numerical working memory. Partifcipants complete a computerised task. They must remember increasingly long strings of digits and repeat them back. Scores are based on correct responses. Higher score means better performance. Score range is 0 - 20
Time frame: Baseline to six weeks
Change in Working Memory From Baseline
Measured through a computerised version of the Paired Associate Learning task in which participants must remember the location of a symbol behind a series of boxes on a screen, and accurately identify the correct location when prompted. Scores are based on accuracy. Higher score indicates better performance. Data reported here shows mean change from baseline. Range of scores is 0 - 16
Time frame: Baseline to six weeks
Change in Reaction Time From Baseline
Measured through a Simple Reaction time test as part of the Wesnes Cognition Ltd CogTrackTM system. Participants are shown a target symbol on the screen and must react by pressing the keyboard / touching the screen (depending on the device being used) when the target appears. Output is measured in milliseconds as a reaction time variable. Lower scores are better; higher scores are worse. Range of scores is 100 - 30,000
Time frame: Baseline to six weeks
Change in Episodic Memory From Baseline
Measured with the Delayed Picture Recognition task in which participants recall a series of images shown earlier in the cognitive test battery. The test outputs percentage of correct answers as the measure. A higher score shows a better outcome. Range output is 0 - 100
Time frame: Baseline to six weeks
Change in Overall Composite Measure of Global Cognition
This is a composite measure of overall cognition that is created by normalising the outputs from all other cognitive tests to give a total score out of 100. The data shown here is the change in total score from baseline to six weeks. Range is 0 - 100. A higher score indicates a better outcome
Time frame: Baseline to six weeks
| Milestone | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Started | 3279 | 3265 |
| Completed | 1739 | 1736 |
| Not completed | 1540 | 1529 |
Measured by the validated task-switching test of Trailmaking B. This is a well established computerised test of executive function. On a computer screen participants are presented with 24 circles with either numbers (0-13) or letters (A-L). They must select the circles in the correct numerical and alphabetic order, alternating numbers and letters each time (e.g. 1, A, 2, B, 3, C etc). The test is completed when all the circles have been correctly selected. The test outputs a total score. A higher score indicates a better performance. A negative mean score indicates a decline in performance. The range of scoring for this test is 0 - 300,000. Data reported here are for mean change per group form baseline to six weeks
| score on a scale | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Executive Function From Baseline | 5435.9 ± 331.80 | 9671.19 ± 330.46 |
Digit Span test for numerical working memory. Partifcipants complete a computerised task. They must remember increasingly long strings of digits and repeat them back. Scores are based on correct responses. Higher score means better performance. Score range is 0 - 20
| score on a scale | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Numerical Working Memory From Six Weeks | 0.55 (0.41 to 0.69) | 0.29 (0.14 to 0.42) |
Measured through a computerised version of the Paired Associate Learning task in which participants must remember the location of a symbol behind a series of boxes on a screen, and accurately identify the correct location when prompted. Scores are based on accuracy. Higher score indicates better performance. Data reported here shows mean change from baseline. Range of scores is 0 - 16
| score on a scale | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Working Memory From Baseline | 1.90 (1.68 to 2.1) | 1.09 (0.87 to 1.31) |
Measured through a Simple Reaction time test as part of the Wesnes Cognition Ltd CogTrackTM system. Participants are shown a target symbol on the screen and must react by pressing the keyboard / touching the screen (depending on the device being used) when the target appears. Output is measured in milliseconds as a reaction time variable. Lower scores are better; higher scores are worse. Range of scores is 100 - 30,000
| time (milliseconds) | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Reaction Time From Baseline | 10.04 (6.13 to 13.95) | 3.66 (-0.26 to 7.58) |
Measured with the Delayed Picture Recognition task in which participants recall a series of images shown earlier in the cognitive test battery. The test outputs percentage of correct answers as the measure. A higher score shows a better outcome. Range output is 0 - 100
| score on a scale | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Episodic Memory From Baseline | 57.9 (54.77 to 61.03) | 45.4 (42.27 to 48.53) |
This is a composite measure of overall cognition that is created by normalising the outputs from all other cognitive tests to give a total score out of 100. The data shown here is the change in total score from baseline to six weeks. Range is 0 - 100. A higher score indicates a better outcome
| score on a scale | Grammatical Reasoning | Control - Card Pairs |
|---|---|---|
| Change in Overall Composite Measure of Global Cognition | 0.22 (0.18 to 0.28) | 0.08 (0.04 to 0.12) |
Collected over 6 weeks. Non-serious events are listed at a 0% frequency threshold.
| Group | Deaths | Serious | Other |
|---|---|---|---|
| Grammatical Reasoning | 0/3,279 (0%) | 0/3,279 (0%) | 0/3,279 (0%) |
| Control - Card Pairs | 0/3,265 (0%) | 0/3,265 (0%) | 0/3,265 (0%) |
| Age, Continuous(years) | Grammatical Reasoning | Control - Card Pairs | Total |
|---|---|---|---|
| Mean | 62.4 ± 7.23 | 62.5 ± 7.08 | 62.5 ± 7.15 |
| Sex: Female, Male(Participants) | Grammatical Reasoning | Control - Card Pairs | Total |
|---|---|---|---|
| Female | 2617 | 2601 | 5218 |
| Male | 662 | 664 | 1326 |
| Educational Attainment(units on a scale) | Grammatical Reasoning | Control - Card Pairs | Total |
|---|---|---|---|
| Mean | 3.28 ± 1.40 | 3.23 ± 1.36 | 3.25 ± 1.38 |
| Race and Ethnicity Not Collected(Participants) | Grammatical Reasoning | Control - Card Pairs | Total |
|---|---|---|---|
| Count of participants | — | — | 0 |
Documents are hosted by the registry — open the source record to download them.
Plan to share: Yes — De-identified individual participant data (IPD) that underlie results in a publication or for secondary data analysis purposes.
Supporting information: Study protocol, Sap
This study is completed, as verified in Apr 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
University of Exeter