CAPTCHA is an abbreviation of Completely Automated Public Turing test to tell Computers and Humans Apart. The name says it – in short, CAPTCHA tests are used on websites to distinguish between humans and automated bots.
There are a number of different CAPTCHA’s you may have come across:
CAPTCHA is a controversial topic in the world of accessibility, and is likely to become more so when the new WCAG success criterion 3.3.7 (accessible authentication) comes into effect (which aims to tackle the issues users face when logging in to a website, including the use of CAPTCHA).
The problem is each type of CAPTCHA could be unsolvable for someone depending on their disability.
For example, image-based CAPTCHAs can’t have an alt text without defeating the purpose of the tool and allowing bots easy access a site. However, without an alt text this type of CAPTCHA will be unsolvable for blind, screen reader users who will be unable to discern images.
More broadly, image-based CAPTCHA can also create problems for all users as images can be interpreted in many ways. For example, asking users to identify a ‘Taxi’ may not be as easy as it sounds as taxis differ in appearance in different countries.
Text based CAPTCHA can also create significant problems. Characters embedded in CAPTCHA are often distorted or close together, this doesn’t only cause problems for users with other visual disabilities, its nearly unsolvable for users with cognitive impairments who have trouble processing letters or numbers.
Lastly audio CAPTCHA, and you guessed it, it also faces accessibility issues. The obvious being, deaf users will be unable to use audio CAPTCHA.
Additionally, audio CAPTCHA also tends to be distorted to avoid bots, this can cause difficulty hearing what is being said. There is also the issue of interpretation, for example, if a 9 should be input as a number or word, this could be challenging for users with cognitive disabilities.
Despite these accessibility issues, CAPTCHAs for many are indispensable. The WCAG working group believes that websites would rather keep CAPTCHA than conform with WCAG, which is why, at present, they still allow CAPTCHAs (even though they can be extremely hard for users with disabilities to use). To mitigate the inaccessibility of CAPTCHAs they recommend a few steps (as a partial solution!):
Google reCAPTCHA works by using an algorithm to follow user’s movements and determine if these are natural (or look like a bot). All the user sees is a simple ‘I am not a robot’ check box, which is accessible via keyboard and with a screen reader.
Although this is a lot more accessible than the traditional CAPTCHA, if the algorithm isn’t satisfied, users may still find that the old school CAPTCHA will return, with all the associated problems.