Science and method
What it measures, and what it does not.
SYNSIQ trains attention and reaction with established task types and makes your progress visible over time. It is feedback for training, not a medical measurement. This page sets out the whole method, including the point where it stops.
01
Reaction-time and Go/No-Go tasks are long-established tools in attention and performance research.
02
Stroop, task switching and N-Back are widely used to study attention, flexibility and working memory.
03
SYNSIQ applies these task types for training feedback, not for clinical measurement.
Foundations
Six areas, and the networks the literature associates with them.
Choose an area to see what its exercises ask of you, and which systems the literature associates with those demands.
The percentage is the area’s weight in the composite index. Areas without enough data are dropped and the remaining weights are rebalanced, so a missing area never drags the index down.
drag to rotate
- Brainstem
- Prefrontal cortex
- Parietal cortex
- Temporal cortex
- Occipital cortex
- Precentral / supplementary motor cortex
Reaction
Weight 20%Answering a cue quickly is a chain: be ready, see it, choose, move. Tasks like these rely on processes commonly associated with alerting, early visual and sensorimotor systems.
- Cognitive processes
- Alerting · Stimulus detection · Response selection · Motor execution
- Associated networks
- Alerting / arousal · Visual network · Sensorimotor network
- Exercises that feed it
- Simple Reaction · Choice Reaction · Start Lights · Audio Reaction
- What one of them asks of you
- Detect one signal and answer it as fast as you can. The task is a standard measure of simple response time and depends on being ready, seeing the cue and moving.
SYNSIQ measures what you do in the exercises. It does not measure brain activity. This model illustrates the functional networks that the cognitive processes behind a task are commonly associated with in the scientific literature.
Fourteen tasks, described by what they demand.
For every exercise: what it asks of you, what the app records, and the functional networks the literature commonly associates with those demands. SYNSIQ measures behaviour in the exercise. It does not measure brain activity.
Reaction
04 exercises- Simple ReactionReact as fast as you can.
Detect one signal and answer it as fast as you can. The task is a standard measure of simple response time and depends on being ready, seeing the cue and moving.
Recorded
Mean, median and best response time · False starts · Omissions
Processes
Alerting · Stimulus detection · Motor execution
Associated networks
Alerting / arousal · Visual network · Sensorimotor network
- Choice ReactionLeft or right — decide fast.
Two possible signals, two possible answers. Choosing costs time, and that cost is what separates this from simple reaction.
Recorded
Response time · Accuracy · Error count
Processes
Stimulus discrimination · Response selection · Motor execution
Associated networks
Visual network · Dorsal attention network · Frontoparietal control network · Sensorimotor network
- Start LightsFive lights out — go!
React the moment the lights go out, and not before. Timing an anticipated moment is a different problem from reacting to a surprise.
Recorded
Response time · False starts · Consistency
Processes
Temporal anticipation · Response preparation · Withholding a premature response
Associated networks
Motor timing circuits · Sensorimotor network · Alerting / arousal
- Audio ReactionReact to the tone.
Respond to a sound instead of something on screen. The value is measured inside the app and depends on the device's audio path.
Recorded
Response time · False starts · Omissions
Processes
Auditory detection · Alerting · Motor execution
Associated networks
Auditory network · Alerting / arousal · Sensorimotor network
Executive control
06 exercises- Go/No-GoAct — but only at the right moment.
Act on one cue and hold back on the other. Stopping a response you had already prepared is the demanding part.
Recorded
Response time on go trials · Commission errors · Omissions · Accuracy
Processes
Sustained attention · Response inhibition · Conflict monitoring
Associated networks
Cingulo-opercular network · Frontoparietal control network · Sensorimotor network
- StroopColour beats word.
Report the colour, not the word. The two answers disagree on purpose, and resolving that disagreement is the task.
Recorded
Response time · Accuracy · Error count
Processes
Selective attention · Interference resolution · Conflict monitoring
Associated networks
Frontoparietal control network · Cingulo-opercular network · Visual network
- N-BackRemember what happened 2 steps ago.
Hold a running sequence in mind and report a match two steps back. Everything has to be kept current while new items keep arriving.
Recorded
Accuracy · Response time · Error count · Omissions
Processes
Working memory maintenance · Updating · Match decision
Associated networks
Frontoparietal control network · Dorsal attention network
- Task SwitchingTwo rules, instant switches.
Two rules, alternating on cue. Every switch has a cost, and the task is built to expose it.
Recorded
Response time · Accuracy · Error count
Processes
Rule representation · Set shifting · Response selection
Associated networks
Frontoparietal control network · Cingulo-opercular network
- FlankerFocus beats distraction.
Answer the middle one and ignore its neighbours. Filtering out what is beside the target is the whole difficulty.
Recorded
Response time · Accuracy · Error count
Processes
Selective attention · Interference resolution · Response selection
Associated networks
Visual network · Dorsal attention network · Frontoparietal control network · Cingulo-opercular network
- Target MatchRespond only to your target symbol.
One symbol at a time, and only one of them is yours. You have to hold the target in mind while rejecting everything else at speed.
Recorded
Response time · Commission errors · Omissions · Accuracy
Processes
Target template maintenance · Rapid discrimination · Response inhibition
Associated networks
Visual network · Frontoparietal control network · Cingulo-opercular network
Attention
04 exercises- Peripheral VisionSee more than you look at.
Keep your eyes on the centre and catch what appears at the edge of the screen. A handheld device covers only the central part of the visual field, so this is not a test of true peripheral sight.
Recorded
Response time · Accuracy · Omissions
Processes
Covert attention shifts · Divided visual attention · Detection
Associated networks
Visual network · Dorsal attention network
- Sustained AttentionStay sharp for minutes.
Rare signals over several minutes. Staying ready while nothing happens is harder than reacting when it does.
Recorded
Response time · Omissions · Lapses · Fatigue index · Consistency
Processes
Vigilance · Sustained attention · Detection of rare events
Associated networks
Cingulo-opercular network · Dorsal attention network · Alerting / arousal
- Visual SearchSpot the target fast.
Find the one that matters among the ones that do not. Search combines seeing with deciding where to look next.
Recorded
Response time · Accuracy · Error count
Processes
Visual feature analysis · Attentional selection · Target decision
Associated networks
Visual network · Dorsal attention network · Frontoparietal control network
- Multiple Object TrackingKeep several objects in view.
Keep several moving objects in mind at once and pick them out again at the end. Attention has to be split and held across all of them.
Recorded
Accuracy · Tracking stability · Error count · Missed targets
Processes
Divided attention · Sustained object tracking · Spatial updating
Associated networks
Visual network · Dorsal attention network · Frontoparietal control network
Nothing hidden in the arithmetic.
Every finished exercise produces raw values: response times, accuracy, errors, omissions, and a reading for focus, consistency and fatigue. Each of those maps onto a 0–100 scale through one documented window, and the results are averaged into the areas that exercise can speak to.
The composite index is a weighted blend of the areas that actually have data. Missing areas are dropped and the remaining weights are rebalanced, so an area you have not trained yet never drags the number down.
| Area | Weight | Reads |
|---|---|---|
| Reaction | 20% | How quickly you respond to relevant cues. |
| Focus | 20% | How steadily you hold your attention under distraction. |
| Decision | 20% | How reliably you tell cues apart and make decisions. |
| Visual awareness | 15% | How well you spot and track visual information. |
| Fatigue resistance | 10% | How stable your performance stays under sustained load. |
| Consistency | 15% | How evenly your performance holds across runs. |
- Sessions before an index
- 3
- Areas before an index
- 2
- Reaction window
- 200–600ms
- Fatigue needs at least
- 40s
A mean response time at or below the fast end of the window scores 100, one at or above the slow end scores 0, and everything between the two moves linearly. A run too short for a second-half slowdown to mean anything is left out of fatigue resistance rather than scored generously.
Three signals you see after every run.
Focus
A combined reading of how attentive your responses were through the session. Higher values usually mean calmer, steadier attention.
Consistency
How even your response times were. A high value means few outliers, and points to a steady rhythm rather than one lucky hit.
Fatigue
Whether you slowed down in the second half of a session. A low value means you stayed fresh. A higher one can be a sign that a shorter session would suit you better.
The part most products leave out.
A screen is not your field of view
Peripheral vision begins well outside the area a handheld screen covers. The exercise that asks you to hold the centre and catch what appears at the edge trains a shift of attention inside the central field. It cannot test true peripheral sight, and SYNSIQ does not claim that it can.
Transfer to a sport is an open question
Whether training these tasks changes what you do on a track, a court or a slope has not been shown for SYNSIQ. Exercises are described by the processes they demand, never as preparation for a particular sport.
Small movements are often noise
Response times vary from day to day, and the first repetitions of any task improve through practice alone. A difference of a few percent over a couple of days is usually variation rather than progress. SYNSIQ gives you the number and leaves the reading to you.
No study has been run on SYNSIQ
The task types are long established in research. The product itself has not been the subject of a published trial, so you will not find effect sizes, percentages or citations suggesting otherwise anywhere on this site.
Not a medical test
SYNSIQ is a training tool. It does not replace medical assessment, therapy or diagnosis.
Personal comparison
SYNSIQ compares you with your own history, not with clinical norms. That helps you spot patterns. It does not diagnose.
Local data in guest mode
In guest mode your training stays on your device. Nothing is uploaded until you choose an account.
Transparent metrics
Every value comes from your real responses, and the scale behind it is explained. There is no hidden scoring.
Trust
The model
The brain shown on this site is a real anatomical atlas, simplified for the web. It is not a drawing, and not a shape generated from a formula. What sits on top of it is a different matter, and worth being exact about. The points and connecting lines are derived from the geometry of that surface: the points are not neurons, the lines are not synapses or fibre pathways, and the light travelling along them is not a signal anyone recorded. All the picture takes from the literature is which regions are joined, and which are lit for a given area. The rest is illustration.
In the cognitive map on this page, the model reports a selection. Everywhere else it does not. The model in the home-page header and on the sign-in screen is atmosphere: the bright patches inside it wander on a timer. Neither the places they visit nor the order they visit them in corresponds to an exercise, to a cognitive process, or to anything the app measures.
Geometry: BodyParts3D, © The Database Center for Life Science, licensed under CC Attribution-Share Alike 2.1 Japan. Selected, simplified and re-labelled for this site; the resulting file is under the same licence.