What’s in an emoji smile
Apple has given me the option to choose between two emojis, but how do I know which one will be understood?
In the early days of mobile text based conversation the lack of facial or audio channels often lead to misunderstanding. A simple joke taken out of context due to that missing smile. Thus the birth of the emoticon, a simple tool for conveying the sender’s mood. As we moved forward and mobile devices became more capable of handling rich and complex imagery this brought on the birth of the emoji. However unlike their early cousins, emojis don’t simply clarify intention they can change the context. At Play we are very interested in how that affects our behaviour when communicating with others and how that communication is interpreted.
So we have had the release of the new iOS 10 and one update that had everyone in the office experimenting with was the emojis within iMessage. Apple has made it even easier to use emojis with a new replacement and predictive system that swaps out keywords and phrases with visually relevant emoji characters.
Basically it gives you the ability to turn text messages into emoji-filled sentences as opposed to the basic green and blue blocks of text.
Some emojis have many potential keywords, and for these you are shown a pop-up box where you choose the emoji you want. For example, the word “thinking,” is associated with both the “thinking face” and “thought balloon” emoji.
This got me pondering, Apple has given me the option to choose between two emojis, but how do I know which one will be understood? What if it’s a group text, will all the recipients understand the same emoji in the same way? My girlfriends first language isn’t english, she might take my selection of emoji the wrong way… now this is just getting complicated!
A can of worms
One thing to remember is we are only talking about iOS emojis at this point. If I was to bring in the fact that emojis across different platforms differ greatly, I would be opening up a can of worms. This was evident from the research carried out by the GroupLens Research team at the University of Minnesota.
Source: Gizmodo
GroupLens researchers asked subjects to rate 22 anthropomorphic emoji from five platforms by sentiment, using a scale that ranged from strongly negative (-5) to strongly positive (5). And here’s where you start to see where “grinning face with smiling eyes” goes so very wrong. Apple’s average sentiment ranking was almost -1, while Microsoft, Samsung, LG, and Google all were 3 or above.”
Now this is just one extreme example, they tested a number of emojis and confusion amongst platforms was rife. This should be a concern when you consider emojis make up a growing proportion of our written language, about half of the characters used on Instagram are emojis for example! Co-author Hannah Miller wrote an essay about the findings where she states;
“A number of scholars have argued that emoji represent a fundamental shift in language use. As such, fully understanding emoji’s role in human communication will be an important step in developing the next generation of language technologies.”
Source: Grouplens
How do I know they will all be perfectly understood
One thing is for sure, since the latest iOS update I am finding myself using more and more emojis as they are predicted and presented to me. For example, just yesterday I wrote the words ‘busy bee’ only to be shown a “bee” emoji. Obviously I went ahead and used that emoji, but this only reminds me of my previous concerns. If I am to continue with the increasing use of emojis, how do I know they will all be perfectly understood and mistakes won’t be made?
The guys at GroupLens also found that Apple users were most likely to be confused by emojis on their own platform:
“Apple has the highest average within-platform sentiment misconstrual (1.96); Google has the lowest (1.79).”
Source: Duncanstephen
With more and more people being promoted to use emojis in place of traditional words due to the highlighting of key words, surely there’s a risk that more misunderstandings will occur. I will give you another example; within my whatsapp chats I have a group chat setup with my closest male friends. I also have an individual chat with my girlfriend (the ‘english as a second language’ one).
If I was to use the “wink” emoji in the first ‘close friends’ group, it would be universally understood to represent a jovial, fun piece of communication. Yet if I was to use this same emoji for the same reasons with my girlfriend, her first impressions would be that I am being sarcastic and even rude towards her! You see my dilemma…
“I think some people thought that they could use [emojis] with little risk and what we found is that it actually is at high risk of miscommunication,” Hannah Miller, a Ph.D. student at the University of Minnesota
Source: Npr.org
Based on this, we decided to run a quick internal test with the team at Play. We took a quote from the late great Steve Jobs; “Design is a funny word. Some people think design means how it looks. But of course, if you dig deeper, it’s really how it works.” and asked everyone to rewrite it using their emoji keyboard. Once the quote is written, iOS 10 does suggest an emoji edit. However we asked everyone to ignore that and write their own using as many or as few emojis as they felt comfortable with.
The aim was to get a feel for:
- How many emojis are suitable to still make sense of the sentence.
- How the emojis were used, were they selected individually, in groups etc?
- The most important question; which emojis did people use to replace what words?
- Were there any patterns, or huge variations?
Initial findings
This is a selection of the results from the test. You can see the original text received from the participant, followed by the actual text description highlighting the ‘real world’ names of each emoji used.
- Some people use emojis to emphasize words or points they are trying to make as opposed to replacing words. So they caveat a point with an additional emoji i.e. ‘I hate it when it rains :rage emoji:’
- Others will use less literal interpretations of words via emojis. They will take the sound of the word and replace it with something that looks like it sounds.
- All of the participants were consistent with recurring words. They will use the same emoji more than once, possibly with the hope that it becomes easier to understand the more it’s seen and read.
- The hacking of emojis is common amongst the group. When the right emoji doesn’t exist people will merge 1 or more emojis together to make the point.
- Some of the group will group emojis to make sentences, whereas others will only use one emoji at a time to replace a word. Grouping of emojis vs singular use is very distinct across individual posts, one person doesn’t combine grouping with individual use.
- When some of the participants are unsure that the reader will understand their use of the emoji, they will purposely put an emphasis on the emoji, either using quotation marks or brackets for example.
- Across all posts, the words ‘dig’ and ‘funny’ were the only two words that were universal represented with the same emojis, in this case the ‘pick’ and the ‘joy’ emoji.
- In terms of age, the younger the participant the more fluid they appear to be when using emojis. They dwell less on the detail and use emojis that they ‘expect’ to be understood with less consideration than the older participants.
- Gender didn’t seem to play a role, with both males and females acting in the same manner.
- Finally, when people viewed the actual definitions/names of the emojis they used (for example an emoji they used to represent ‘design’ was actually called ‘writing_hand’), there was a ‘WTF’ moment with people showing surprise when presented with the emoji definition.
Closing thoughts
Based on the findings, it’s clear there is a long way to go before emojis used as a form of language are universally understood. This begs the question, is Apple’s move towards the predictive emoji actually a new way of standardising the use and thus simplifying the emoji landscape? This may be a start, but it would still be difficult for even the likes of Apple to cover all bases when it comes to standardising emojis.
For example, how does the use of emojis differ between groups of people with differing levels of emotional connection, i.e. a close partner may understand you more than a work colleague and therefore your use of emojis with that person could be more varied and off the cuff. In order to get someone you know formally to understand your use of emojis, do you need to consider their use more carefully?
If this is the case, how on earth does someone like Apple even begin to standardise this behaviour? Will we get preemptive emojis based on the contact we are communicating with? That might work. If we use specific emojis with targeted groups or individuals does the recently used emoji feature need to do more work to understand the context and audience of the conversation.
Other things to consider
- Does the visual complexity and variance of the current emoji landscape makes it a poor tool for understanding subtlety in a conversation than it’s simpler emoticon cousin. From memory I struggled less when using and receiving emoticons back in the good old days of the Nokia 3210. This is not proof in anyway, however it does raise another consideration.
- How else will we begin to make our communications more understandable? For example, will sound effects such as Dubsmash be the next step in layering meaning and depth onto our messages alongside emojis? Would this take away the confusion, or just add to it?
- Is constraining what emojis we can use on certain platforms the same as removing our ability to share our opinions freely.
- Screen readers and those with visual impairments need further consideration if the emoji is to become a significant part of our written conversation.
We don’t claim to have found the answers in this very short internal study, but we do feel we are asking the right questions and pointing in the right direction….Should we revisit an established design principle, keep it simple stupid?
P.s within the office we have been trying to sign off on this article using emojis and we can’t agree on the right ones to use!
