Put the mouse in a drawer and try to finish one of your own courses. Many developers get stuck within a few slides: a drag-and-drop that will not move, a hotspot that cannot be reached, a pop-up layer that swallows the focus and never gives it back. Keyboard navigation in e-learning is easy to overlook because the people building courses almost always use a mouse. Captions have a similar blind spot, since most developers can hear the narration perfectly well. Can learners complete your course without a mouse, and are your captions actually good enough to rely on?
A learner should be able to reach, see and operate every control in a course with the keyboard alone, in an order that matches the slide's meaning, and every piece of prerecorded audio in synchronized media should have accurate captions. You check the first with a tab-through test using Tab, Shift+Tab, Enter, Space, the arrows and Escape.
You check the second by watching with the sound off and confirming the captions are complete, in sync and identify speakers and important sounds. Drag-and-drop and hotspot interactions usually need deliberate keyboard alternatives. Transcripts are a worthwhile addition to captions, not a replacement.

Why Does Keyboard Navigation in E-Learning Matter?
Keyboard navigation matters because a significant group of learners cannot, or do not, use a mouse. That includes people with motor disabilities, people who use screen readers (who navigate by keyboard), people using switch devices or voice control that emulates keys, and anyone with a temporary injury. Keyboard operation is also the foundation that much assistive technology builds on.
WCAG, the Web Content Accessibility Guidelines, covers this in Success Criterion 2.1.1. The W3C's explanation of the keyboard criterion says all functionality must be operable through a keyboard interface without requiring specific timings for individual keystrokes, with an exception only for input that depends on the path of movement, such as freehand drawing. A related criterion, 2.1.2, says the keyboard must never get trapped in one component.
Whether any law or contract obliges your organization to meet these criteria is a different question, and the answer depends on where you operate and who you train. Put that one to your counsel or the official source. The test below only tells you whether the course works from the keyboard.
How Do You Test a Course With Only the Keyboard?
Work through every slide with the mouse out of reach and record what happens. These are the keys to use:
- Tab and Shift+Tab move forward and back between controls.
- Enter and Space activate buttons, links and checkboxes.
- Arrow keys move within groups, such as radio buttons, tabs or sliders.
- Escape should close pop-ups, layers and menus.
On each slide, check these points in order:
- Can you reach every interactive element, including player controls such as play, pause, menu and next?
- Can you always see where the focus is? A visible focus indicator is itself a WCAG AA requirement (2.4.7, Focus Visible).
- Can you activate each element and get the same result as a click?
- When a layer or pop-up opens, does focus move into it, and does it return to the triggering button when it closes?
- Can you answer every quiz question and submit?
- Are you ever trapped, unable to Tab out of something?
- Does anything time out before a keyboard user could finish?
Note each failure with the slide number and the element. A simple three-column table (slide, element, problem) is enough for the developer fixing it.
What Should Focus Order Look Like on a Busy Slide?
Focus order should follow the order in which a sighted learner would read and use the slide. Success Criterion 2.4.3, explained in the W3C's note on focus order, requires an order that preserves meaning and operability. In practice that usually means top to bottom and left to right, with the slide's main task before the navigation buttons.
Take a made-up slide with a title, a photo, three tabs, a content panel that changes with each tab, and Back and Next buttons. A sensible order would be:
- The slide title (if the player makes it focusable).
- The photo, if it is informative and has alt text.
- Tab one, tab two, tab three.
- The content panel for the selected tab.
- Back, then Next.
The common fault is an order based on when each object was added to the slide. If the Next button was placed first, it receives focus first, before any content. Most authoring tools have a focus or tab order panel; check it on every slide that has more than a few objects. Also confirm that hidden objects, such as content on a closed layer, do not receive focus while they are invisible.
How Do You Make Drag-and-Drop and Hotspots Keyboard Friendly?
Give them a keyboard path that tests the same knowledge. Dragging is a mouse gesture by nature, so accessible drag and drop needs a deliberate alternative.
Two approaches work well:
- Select and place. The learner tabs to an item, presses Space to pick it up, tabs or arrows to a target, and presses Space again to drop. Announce each step so screen reader users know what is selected.
- An equivalent format. Offer the same task as a matching question with drop-down menus, which asks the learner to make the same decisions without dragging.
WCAG 2.2 added a related criterion at level AA, 2.5.7 Dragging Movements, which asks for a single-pointer alternative to dragging. A select-and-place design usually serves both keyboard and pointer users.
Hotspots need each clickable area to be a focusable element with a label. Be careful with "find the hazard" activities, where searching is the point. Labelling a hotspot "Frayed cable" gives away the answer. Generic labels such as "Area 1" through "Area 6", with the feedback naming what was found, keep the challenge intact for every learner.
What Makes Captions Good Enough?
Captions are good enough when a learner who cannot hear the audio gets the same information as one who can. Success Criterion 1.2.2 requires captions for prerecorded audio in synchronized media at level A. The W3C's easy checks for captions list what to look for: captions are present, in the language of the audio, synchronized with it, identify who is speaking, and include important sounds. The same page notes that automatically generated captions alone are not sufficient.
For e-learning, add a few practical checks:
- Narration over slides counts. If a slide has voice-over and changing visuals, caption it.
- Captions should not cover on-screen text or the part of the screen the learner needs.
- Technical terms, product names and numbers must be spelled correctly. This is where automatic captions often go wrong.
- The caption toggle must work from the keyboard.
Watch each video or narrated sequence once with the sound off. If you can follow it from the captions alone, they pass.
Do You Need Transcripts as Well as Captions?
Usually yes, because they serve different needs. Captions run in time with the media. A transcript is the full text available all at once, which suits learners who read faster than the narration, people using braille displays, anyone reviewing later, and anyone who wants to search for a term.
Audio-only content, such as a narrated slide with no meaningful visuals or a podcast-style segment, generally calls for a transcript under WCAG. For video where important information is only visual, consider audio description too, so learners who cannot see the screen are not left guessing. The W3C's media guidance treats captions, transcripts and audio description as complementary parts of accessible media.
A practical habit: keep the narration script as the source for both. If the script is final and matches the recording, captions and transcripts become an editing job rather than a transcription job.
Unplug the Mouse Once and You Will Never Skip This Test Again
Keyboard navigation in e-learning and good captions both come down to the same principle: every learner should get the whole course, whatever way they use it. Test with the keyboard on every slide, set focus order deliberately, give drag-and-drop and hotspots a keyboard path, and check captions with the sound off.
The visual side, from text colors to image descriptions, is covered in color contrast and alt text. Do the tab-through once on your own course, with the mouse genuinely out of reach. The slides where you get stuck are the ones to fix first.
