ARIA: figure role

The ARIA figure role can be used to identify a figure inside page content where appropriate semantics do not already exist. A figure is generally considered to be one or more images, code snippets, or other content that puts across information in a different way to a regular flow of text.

A figure is a perceivable section of content that typically contains a graphical document, images, code snippets, or example text. The parts of a figure MAY be user-navigable. Any content that should be grouped together and consumed as a figure (which could include images, video, audio, code snippets, or other content) can be identified as a figure using role="figure".

<div role="figure" aria-labelledby="caption">
  <img src="image.png"
      alt="put image description here">
  <p id="caption">Figure 1: The caption</p>
</div>

In the above example, we have a figure that consists of two separate content items — an image and a caption. This is wrapped by a <div> element that identifies the content as a figure using role="figure".

For HTML, use the <figure> and <figcaption> elements. The figcaption will serve as the accessible name for the figure. When not using HTML, or when retrofitting legacy HTML, use the aria-labelledby on the figure, pointing to the figure's caption. If there is no visible caption, aria-label can be used.

<div role="figure" aria-labelledby="figure-1">
  ...
  <p id="figure-1">Text that describes the figure.</p>
</div>

• Use aria-labelledby when the text is a concise label.
• Use aria-describedby when the text is a longer description.
• Use aria-label when there is no visible figure caption.

This can be done semantically, without ARIA, with HTML's <figure> element along with <figcaption>.

<figure>
  <img src="image.png"
      alt="put image description here">
  <figcaption>Figure 1: The caption</figcaption>
</figure>

aria-describedby

The id of an element containing reference text serving as a caption.

aria-labelledby

The id of an element containing text serving as a label.

aria-label

If there is no element containing text that could serve as a label, you can add the label directly as a value on the aria-label on the element with the figure role or on the <figure> element.

No role specific keyboard interactions.

No role specific JavaScript requirements. If you don't have control over the HTML semantics, you can improve the accessibility of HTML by adding these roles and properties with JavaScript.

We could extend the initial example on the page to also identify a paragraph that provides a descriptive label for the figure by referencing its ID in aria-labelledby:

    <div role="figure" aria-labelledby="figure-1">
      <img src="diagram.png"
           alt="diagram showing the four layers of awesome and their relative priority order —
                music, cats, nature, and ice cream">
      <pre>`
        let awesome = ['music', 'cats', 'nature', 'ice cream'];
      `</pre>
      <p id="figure-1">Figure 1: The four layers of awesome.</p>
    </div>

Only use role="figure" if you have to — for example if you don't have control over your HTML but are able to improve accessibility dynamically after the fact with JavaScript.

If at all possible, you should use the appropriate semantic HTML elements to mark up a figure and its caption — <figure> and <figcaption>. For example, our above example should be rewritten as follows:

<figure>
  <img src="diagram.png"
       alt="diagram showing the four layers of awesome and their relative priority order —
       music, cats, nature, and ice cream">
  <pre>`
    let awesome = ['music', 'cats', 'nature', 'ice cream'];
  `</pre>
  <figcaption>Figure 1: The four layers of awesome.</figcaption>
</figure>

• Accessibility Object Model
• ARIA in HTML
• HTML <figure> element
• HTML <figcaption> element