Icons, Symbols and Cognitive Disabilities

Guide­line 1.3 Cre­ate con­tent that can be pre­sented in dif­fer­ent ways (for exam­ple spo­ken aloud, sim­pler lay­out, etc.) with­out los­ing infor­ma­tion or structure

Draft Guide­line 1.3 of the Web Con­tent Acces­si­bil­ity Guide­lines 2.0 dis­cusses the impor­tance of mak­ing infor­ma­tion avail­able in a form that can be per­ceived by the user– either directly or through an assis­tive tech­nol­ogy. For exam­ple, if a site uses an image to con­vey mean­ing, then the alt text should be present so that it can be seen (in a text-only envi­ron­ment), heard (by a screen reader) or even felt (through a refre­sh­able braille dis­play). This per­tains to both pre­sen­ta­tion and structure.

The tech­nolo­gies men­tioned above– text-only browsers, screen read­ers and elec­tronic braille– all do a good job at pre­sent­ing infor­ma­tion to the appro­pri­ate audi­ence. How­ever, what about users with cog­ni­tive dis­abil­i­ties? Text read aloud may be help­ful in some sit­u­a­tions, but what about a graph­i­cal rep­re­sen­ta­tion of con­tent? There are already some efforts in this area with the Com­mu­ni­cate: Web­wide sym­bol sup­ported browser that claims to be able to rep­re­sent over 29,000 words with sym­bols and while I applaud their efforts, with a Windows-only, pro­pri­etary, subscription-based prod­uct, I doubt there will be any wide­spread adop­tion in the near future.

Per­haps one day there will be an open sys­tem that will facil­i­tate the col­lec­tion and use of sym­bols in every­day tools. Until then, here are some other project/ideas that are explor­ing the use of symbols:

  1. Joe Clark says:

    Pro­po­nents of turn­ing every Web site into pic­tographs would need to present test results, based on a wide spec­trum of Web con­tent, prov­ing that tar­get audi­ences really under­stand the pages.

    As you’ve pre­sented them, “sym­bols” are 1:1 sub­sti­tutes for words. You haven’t accounted for mor­phol­ogy (e.g., plu­rals) or gram­mar. If this really works as well as claimed, then a German-language page could be com­pletely “trans­lated” into sym­bols. When pre­sented to an English-speaking user, that per­son could com­pletely under­stand the page.

    Right?

    Pro­po­nents also haven’t dealt with the fact that every one of these sym­bols requires an alt text. How’s that gonna work?

  2. Thanks for your thoughts Joe, regard­ing the need for test results, I know that pic­tographs are widely used in the field of spe­cial edu­ca­tion and would imag­ine that research has been done on their effec­tive­ness. I will try to track some of that down to eval­u­ate how well that same research might apply to online content.

    While, the exam­ple given above used one sym­bol to rep­re­sent one word, an effec­tive solu­tion would need to pro­vide a struc­ture wherein com­plex ideas and word pair­ing could be sim­pli­fied and pre­sented. I envi­sion a sys­tem that would more closely resem­ble Amer­i­can Sign Lan­guage than Eng­lish or Ger­man wherein there would be a sim­pli­fied mor­phol­ogy using meth­ods such as size for empha­sis and dupli­ca­tion to rep­re­sent plu­ral­ity. I don’t know whether such a sys­tem could be built that would be intu­itive enough that it could be gen­er­ated by the gen­eral pop­u­la­tion and con­sum­able by the tar­get audience.

    Regard­ing alt text for such images, they would be required for the cre­ation of con­tent, but I see them being less use­ful in the pre­sen­ta­tion of con­tent. Since the sym­bol is being used to describe a work you could think of the icon as an “alt image” with the orig­i­nal text play­ing the role of the alt text for the alt image. Then we can look at the case of a user who has a cog­ni­tive dis­abil­ity and uses a screen reader– send­ing that per­son back to the orig­i­nal text would not nec­es­sar­ily be helpful.

    I think the big­ger chal­lenge is that for any such sys­tem to work it would need to be able to take a chunk of text, ana­lyze it, sim­plify where nec­es­sary and then map the ideas from the text to sym­bols. That brings a whole new set of com­pli­ca­tions as well as pos­si­bil­i­ties– with the sim­pli­fied trans­la­tion pro­vid­ing an end solu­tion in mak­ing con­tent more acces­si­ble for many users.

    I think it is also impor­tant to acknowl­edge that there are lim­i­ta­tions as to what types and the depth of con­tent that can be rep­re­sented by sym­bols, with user input fur­ther com­pli­cat­ing the issue. How­ever, I appre­ci­ate your ques­tions and the oppor­tu­nity to think fur­ther through some of my assump­tions– thanks.

  3. Joe Clark says:

    Trans­late this arti­cle and its com­ments into “sym­bols” and let’s see how well it can be under­stood by peo­ple who have not read the original.

    The idea is a non-starter that was relent­lessly pro­moted by non-experts on the WCAG Work­ing Group.

  4. I’m going to make an attempt trans­late the arti­cle above into pic­tographs– I think it would be a chal­lenge to rep­re­sent the ideas of the post using any method of sim­pli­fi­ca­tion, there is a lot of con­tent for which basic sys­tem of icons would never work.

    There are a num­ber other bar­ri­ers to users with sig­nif­i­cant dis­abil­i­ties. One being the Catch-22 argu­ment that there is lit­tle mean­ing­ful, acces­si­ble con­tent avail­able to a per­son who can­not access aural or writ­ten communication.

    I real­ize that iconic rep­re­sen­ta­tion of ideas will never be a a uni­ver­sal solu­tion, but I do think that if/when more effec­tive meth­ods of sim­pli­fy­ing text are avail­able that a basic set of learn­able icons would be use­ful (and fea­si­ble) to present those ideas to a user who com­mu­ni­cates pri­mar­ily through symbols.

  5. Carsten says:

    Just a short hint on Incon­let, it’s sim­i­lar to Icon­finder you point to. Thanks for this post :)

  6. […] der Suche nach Icons bin ich am Woch­enende über diesen Beitrag zu Icons, Sym­bolen und kog­ni­tiven Beein­träch­ti­gun­gen gestoßen, der mich weiter zu zlango.com […]

  7. peepo says:

    Don’t let Joe put you off, he knows lit­tle enough, and shouts a lot.

    you might want to review my work, or else per­haps http://www.symbolworld.org to see exam­ples that are in cur­rent use.

Reply

or