Frequently asked questions

The Pilotest website is the result of 7 years of labor and of a fruitful cooperation between airline pilots, psychologists specialized in pilot assessment, and IT engineers. You will find test preparation questions specifically designed to allow you to carry out an efficient preparation for the psycho-technical and psycho-motor tests that you will face during the screenings in a flight school or at an airline company. The Pilotest website was born from the will to democratize the training for psycho-motor and psycho-technical tests. It has become necessary to focus on personal preparation rather than to rely on chance and allow everyone to access the same quality of preparation material independently of their financial means. Besides, in the global competitive industry many expensive training solutions have emerged - some of which useless - over the past years. By choosing Pilotest you will be sure to access the best up-to-date information and unlimited access to the training materials.

Different more or less qualitative preparation tools for the psycho-technical tests have emerged over the past years and and today take substantial advantage of the competition between students and less experienced pilots. These schools or websites benefit from applicants' stress and give them the illusion that paying will allow them to access extra information which could make the difference at the screening. Indeed that's an illusion! Keep in mind that Pilotest centralises all the information you may need to pass your selections and that it is kept up-to-date everyday. Because of its free access, it works like a community and our users are quick to report to us any test modifications. We can then adapt the website almost immediately. Bear in mind that paying won't help you to make the difference!

We can only keep the tests up-to-date and relevant thanks to your feedback (be gentle please ;-)...If you have any information about pilot aptitude tests, do not hesitate to contact us.

The site currently covers the ENAC-EPL and Air France selections - both cadet and experienced pilot streams - along with the PILAPT®, COMPASS, CUT-E and DLR batteries.
We are now moving on two fronts. First, broadening the offer: the ADAPT battery and the Mollymawk tests, of which a first building block, the memory test, is already online.
Second, modernising what already exists, which takes just as much effort: rebuilding the oldest tests, updating their content as the real selections evolve, and steadily improving how they feel to play.
We would rather announce a test once it is ready than commit to a date. All of it stays free.

Your test results are given not only via a standard score summarizing correct and incorrect answers but also via a "stanine mark" ranging from 1 to 9. This stanine mark expresses your performance compared to that of other users. Users are divided as follows:
█ 1 : 4% (0% to 4%)
█ 2 : next 7% (4% to 11%)
█ 3 : next 12% (11% to 23%)
█ 4 : next 17% (23% to 40%)
█ 5 : next 20% (40% to 60%)
█ 6 : next 17% (60% to 77%)
█ 7 : next 12% (77% to 89%)
█ 8 : next 7% (89% to 96%)
█ 9 : best 4% (96% to 100%)
If the test is not being calibrated, the scale of the stanine marks is displayed on the test page. Note that for the same test and with the same result (%) you may obtain 2 different stanine marks at 2 different times. The test was certainly recalibrated in the meantime and the average reference of the other users has changed.
At the bottom of your results page, you can download your results as a CSV file.
You can also visualise your results for every selection or test battery Pilotest prepares for: tests are grouped by aptitude, and you can follow how each aptitude evolves over time.

Since a majority of candidates use Pilotest, stanine marks are all the more relevant.

For some tests, the stanine width may be very narrow or identical. For example and for a given test, stanine 7 means an 80% score while stanine 8 is for 81% and stanine 9 for 81% too. This puzzles some of our users. In fact, we must come back to the stanine definition, that is to say a position relative to other users' performance. Just imagine a reward or a podium. Stanine 1 is the bronze medal and 9 the gold. A 81% result is sure to be in the best 11% (stanine 8) but uncertainty remains as to whether it is in the best 4% (stanine 9). In this case, the website will give a stanine 8 - the only one we can be sure of. Nevertheless, this situation is rarely seen. We are doing our best to correct the tests in order to distinguish the users and spread their Gaussian results.

For each selection Pilotest prepares for, your results are organised by aptitude where you have an average mark. It is worked out from your latest known result on each of the tests making up that aptitude, weighted by how much that test counts within it. An aptitude therefore only appears once all of its tests have been taken at least once, and its history reads as a curve. Results that start to age are flagged: they still count, but they no longer necessarily reflect where you stand. Depending on the selection and other criteria, the minimum result for each aptitude varies. Nevertheless it is often between 4 and 5 for a professional pilot selection and about 6 or 7 for an ab initio pilot screening.

Your results (expressed as stanine classes) are computed using your correct and incorrect answers as well as the time you take to answer questions. Depending on the test that is considered, other data may be taken into account. Wrong answers may or may not be detrimental from one test to another. However, we strongly recommend that you NEVER answer at random. This will always be unfavourable to your overall score not only on Pilotest but overall at your selection.

The test calibration phase is aimed at ensuring the reliability of the test and its accuracy. During this phase, all user test results are saved in order to build a reference Gaussian curve and a message is displayed to warn about the stanine mark inaccuracy. Once the calibration is finished, the test is published with a result stanine scale on Pilotest.
When your results are displayed, they are adjusted with the calibration curve that has been previously computed.
Note: tests on Pilotest are continuously re-calibrated with our users' results.

In order to allow you to prepare efficiently and successfully, Pilotest develops online tests as close as possible to the ones you will be taking during your screening. Some of our users have reported to us that the tests on Pilotest are for some of them harder than those at the exam. Keep in mind that it is done on purpose and that we design the test to optimise your chances of success. Training on slightly harder tests will make you comfortable enough to excel at your selection.

Little by little, we are trying to add test corrections when it wasn't initially done. The correction is either available at the end of the test or after each question (do not forget to select the training mode).

Yes. Practising is the key to success. As a consequence, your Pilotest subscription gives you full and unrestricted access to all tests. Furthermore, most of the tests are built with random question generators, so you will almost never see the same question twice!

Your account is personal. Sharing it distorts the calibration of the tests, which relies on comparing distinct candidates.
For that reason, only one session can be active at a time: if you sign in from another device or another browser, the previous session is closed automatically. You may use as many devices as you like, just not simultaneously.

Your personal data remains your property. You have the right to access and modify it: your profile for your account details, this page for your pilot background.
Deleting your account. The "Cancel my account" link at the bottom of your profile triggers an immediate and permanent erasure, constituting the exercise of your right to erasure under the GDPR: account, profile, sign-in dates and IP addresses are erased or anonymised. One single element remains, a cryptographic fingerprint of your email address, which does not contain the address itself and serves one purpose only: preventing a new account from being opened with that same address. Your results are kept in anonymised form, severed from any identity, so as not to distort the calibration of the tests.
The confirmation screen shown once deletion completes is your proof of execution: seeing it means your request has been carried out in full. There is therefore no need to write to us to confirm or repeat it - we hold no remaining data about you to consult. See also our terms of use.

Pilotest cannot be held liable if the user fails the assessment exam, and no compensation will ever be provided for such a case.

Any reproduction or copy, partial or complete, of the Pilotest website by any means is strictly prohibited, even for personal use. In the event of reproduction, even partial, of the website, Pilotest will systematically lodge a complaint against the author of such reproductions and will undertake all required legal procedures. In such a situation, we reserve the right to communicate with any other parties about this infringement and its author. The Pilotest website is protected by international copyright laws.

Pilotest targets the three most recent major versions of the main browsers - Chrome, Firefox, Safari and Edge - with JavaScript and cookies enabled. It may well work elsewhere, but we cannot guarantee it. No plug-in is required.
Some tests have particular needs, indicated on their own page: a compatible game controller for the flying exercises, sound for the auditory tests, or 3D rendering for spatial reasoning.
Any device will do - Windows, Mac or Linux computer, tablet, phone - but a few tests are simply unrealistic on a small screen. The ENAC dexterity test is the typical case: we recommend a screen of at least 8 inches diagonal for comfortable practice.

Dexterity tests are designed to be played with a gamepad or a joystick. Every modern browser now supports the Gamepad API: Chrome, Firefox, Safari and Edge. Only Internet Explorer, which Microsoft stopped maintaining in 2022, is unsupported.

One caveat, and it only applies to Safari on Mac. Safari only exposes devices that the system declares as "GamePad". A flight stick - Saitek, Thrustmaster, most aviation joysticks - declares itself as "Joystick", so Safari simply does not see it, and nothing you do will change that. Console-style gamepads work normally. If your stick does not show up in Safari, use Chrome or Firefox: both recognise either family.

If you use real gamepads, you must first carry out a calibration phase which is done once and for all, per test. The calibration purpose is to link axis and buttons to the test. Let's calibrate together your gamepad to play the PILAPT® - Total.

You know you are supposed to move the ILS laterally with the left/right gamepad axis(1). Move the joystick laterally. The blue rectangle (2) moves which represent joystick 1/axis 3. In the select menu, you will select joystick 1/axis 3 for axis (1). Do the same for the vertical axis and link the relevant axis. Note that if the sliders are not greyed, you can adjust the axis sensitivity with the sliders (A).
To link the trigger button (3), just pull on the trigger on your gamepad. A blue rectangle will blink (4) which is joystick 1/button 5 on our example. Enter joystick 1/button 5 for the trigger in the select menu (B).
Normally, if you release your gamepad, the blue rectangles of the axis should all be halfway. If it is not the case, just release all your gamepads and press the TRIM button. It will add an offset that will recenter your axis.

If you use Windows (but why? ... :-) ), a Windows gamepad calibration is done upstream of your browser. Most of the time, it works well. But if it doesn't, our website will receive incorrect gamepad position values.
In such a case, please go to (1) Control panel > All control panel items > Device and printers. (works with Windows 10 - for former versions, press Windows key + "R" and enter "joy.cpl" + Enter). Select the relevant gamepad (2) with right click + "parameters" to display the calibration tools (3). Click on the game controller and open the properties windows (4).

a) Axis are all centered.

b) Single axis (Z, Rotation or Thrust lever) are centered HALFWAY which is 0 on the picture and not -1 or 1.

c) No button is pressed.

d) Any point-of-view controllers are centered.

Once the calibration phase is done, click the button "Save the calibration". Your calibration for this test is now saved as a cookie and will stay available as long as you don't erase it.

Every gamepad recognised by your operating system is also recognised by the tests, on Chrome, Firefox and Edge. Safari is the exception and ignores flight sticks: see "Gamepads and browser compatibility" above. If yours is not detected, move its axes or press one of its buttons to wake it up.
If you use two controllers, remember to act on each of them: for privacy reasons, browsers only reveal a controller to the page once you have used that specific device. Moving one stick reveals that one only.
If that is not enough, unplug it and plug it back in: some browsers only recognise gamepads connected AFTER they were launched. It also happens that a browser only detects gamepads in a single tab - in that case close every tab but the test one. As a last resort, this page lets you check online whether your gamepad is correctly seen by your browser.

Spatial reasoning tests draw actual three-dimensional objects in your browser, using a standard technology called WebGL, which relies on your computer graphics card. Eight tests are concerned: Cars basic and sequential, Chairs, Stacked cubes, 3D objects, the DLR tube test, the PILAPT flying test and the CUT-E complex control test. When the browser cannot use the graphics card, the test either refuses to start, or stops in the middle of a session. Here is how to fix it yourself, in order.
1. What usually does the trick

• Reload the page.
• Close the other tabs showing 3D, maps or videos. Every such page reserves a share of the graphics card, and your browser drops the oldest ones when it runs short.
• Quit the browser entirely, then open it again. If your computer just woke up from sleep, restart it.

2. Check whether WebGL works at all
Start with our 3D display diagnosis page: it draws four control pictures with the same technology as the tests and tells you whether your browser actually produced them, which separates a browser that does not draw from a browser that draws without displaying. You can also open get.webgl.org. If a cube is spinning, your browser can display 3D and the problem lies elsewhere: go back to step 1, then to the extensions in step 3. If the page shows an error message, follow the section for your browser below.
3. Chrome, Edge, Opera and Vivaldi

• Graphics acceleration. Open chrome://settings/system (or edge://settings/system), turn on "Use graphics acceleration when available", then restart the browser.
• Diagnosis. Open chrome://gpu (or edge://gpu). At the very top, the WebGL line must read Hardware accelerated. If it reads Software only, Disabled or Unavailable, the reason is given right below, under Problems Detected.
• Extensions. Open a private window (Ctrl+Shift+N, or Cmd+Shift+N on a Mac) and run the test again: extensions are disabled there. If everything works, turn them back on one by one to find the culprit. Ad blockers and anti-tracking add-ons are the usual suspects, as some of them neutralise graphics drawing to prevent fingerprinting.
• Experimental options. If you have ever changed settings in chrome://flags, use the Reset all button at the top of that page, then restart the browser.

4. Firefox

• Graphics acceleration. In Settings > General > Performance, untick "Use recommended performance settings", then tick "Use hardware acceleration when available". Restart Firefox.
• Diagnosis. Open about:support and scroll down to the Graphics section. The WebGL 2 Driver Renderer line must name your graphics card. If you read Blocked or Failed there, Firefox is refusing your driver.
• Advanced settings. Open about:config, accept the warning, and check that webgl.disabled is false. If your card is being refused, set webgl.force-enabled to true.
• Tracking protection. The privacy.resistFingerprinting setting, in about:config, restricts graphics drawing when set to true. Strict enhanced tracking protection can have the same effect: add pilotest.com to the exceptions.

5. Safari, on Mac, iPad and iPhone

• WebGL is on out of the box and cannot be switched off from the ordinary settings. Start by installing the macOS or iOS updates.
• On a Mac, if you have enabled the Develop menu (Settings > Advanced > Show features for web developers), check under Develop > Feature Flags that WebGL has not been turned off by mistake.
• On iPad and iPhone, close the other tabs before starting a 3D test: graphics memory is very limited there, and Safari silently drops the pages it finds too demanding.

6. Brave
Click the Shields icon, the lion in the address bar, and set fingerprint blocking to Standard or Disabled for pilotest.com: in strict mode, Brave switches WebGL off. Everything else is set up like Chrome, in step 3.
7. Drivers and hardware

• Windows. Install the graphics driver from the manufacturer website, NVIDIA, AMD or Intel, rather than through Windows Update, which often lags one or two versions behind. An unstable driver gives exactly this symptom: the picture vanishing mid-session.
• Laptops with two graphics cards. In the NVIDIA or AMD panel, check that your browser is not assigned to a disabled card, and plug in the power supply: some battery saving modes switch the dedicated card off.
• Mac. Drivers come with the system updates, there is nothing to install separately.

8. Virtual machines and remote desktops
VirtualBox, Parallels, VMware, Windows Remote Desktop, Citrix and most locked-down corporate computers do not provide usable 3D. No browser setting will change that: run these tests from an ordinary machine.
9. If the picture vanishes during a test
It means the browser lost access to the graphics card during the session: waking from sleep, a driver restarting, or graphics memory taken over by other tabs. Reload the page and the test starts afresh. If it happens again, apply steps 1, 3 and 7.
If none of this helps, write to us with your browser and system, and if possible a screenshot of chrome://gpu or about:support.

You must be able to access the internet to launch a test. However, you can then play the test without an internet connection (metro, tunnel...). Internet access is needed to upload your result and to convert it into a stanine mark. You must also access the internet to load a new set of questions.

For most tests, 2 modes are available. In exam mode, you just answer the questions one after another and no correction is given. In training mode, a correction is given after each question while the chrono is set to pause.

Don't hesitate to contact us. We will try to reply within 48 hours (two working days).

Eight tests display actual three-dimensional objects and therefore need WebGL, your browser 3D display feature. If those tests stay blank for you, or if you simply want to check your hardware before a session, open our 3D display diagnosis page.
It draws four control pictures with exactly the same technology as the tests, then reads back what your browser actually produced. It can therefore tell a browser that does not draw 3D from a browser that draws it without managing to display it, which no screenshot can reveal. Each of the four pictures isolates one setting: plain rendering, cast shadows, transparent background and antialiasing, so that a failure points at the culprit.
The page also shows your graphics card, your WebGL version and a full technical report. If the problem persists, send us a screenshot of that page and we will know what to look for. What to try, browser by browser, is detailed in the question about 3D display problems.

Pilotest access is free of charge and that's our aim to keep it such. Light advertising will soon be added in order to financially support the website and pay the salaries of our 2 developers. In such a context you will understand that we refuse access to users with an ad blocker. By the way, we are very interested in any kind of partnership or financial support. Do not hesitate to contact us on this matter!