Avatars in eLearning: Pros and Cons

To use avatars, or not to use avatars. That is a conversation I’ve had too many times and the decision is often made based on cost or time restraints. I’ve been challenged with projects where the decision to use, or not to use, avatars has taken place before being brought in on the project. As a result, client expectations have to be reset, or budgets and timelines renegotiated, in order to deliver a high effective eLearning experience.

Avatars have their place in eLearning. Done right, they can improve retention and application of materials. Used incorrectly, they can actually hamper learning.

“Should I Use Avatars?”

Honestly, that is really up to you. It is true that the cost of creating, implementing, an effective avatar can be cost prohibitive and time consuming. If it doesn’t fit in the budget, or you can’t meet the delivery date for the course because of the extra time the avatar will take, then don’t do it.

If you choose to use avatars, or your client is pushing for their use, this article will take you through the science behind using avatars – what makes them effective and when are they harmful to learning. I’ll also give you some pointers and cost-effective alternatives.

Benefits of Avatars

Is there ever any benefit for using avatars? In short, yes!

First, avatars are a great way of engaging the learner and putting a face to the narration.  It gives a face to the narration, helps define the relationship between the narrator and learner, and can add humor or personalization to the lesson.

Second, you can use the avatar as part of the content. Even when the avatar is an illustrated character, it’s posture, clothing, expression, etc. tell more of a story than a disembodied voice. It can help shape your culture and actively or passively reinforce behaviors (like dress code, use of safety gear, etc.) taught in the current or past lessons.

And third, and most importantly, avatars can improve learning! Academic studies have shown that avatars have the ability to increase retention and improve the learners ability to apply the knowledge on the job. Ultimately, isn’t that our goal of any eLearning course? But for this to work, the avatars must possess these specific characteristics.

Avatars as Educators

To realize the educational benefits of using avatars, the avatar must function as an educator. Our brains are wired to pickup on many physical cues from a human educator: facial expression, hand gestures, body language, etc. For the avatar to improve the learning in any eLearning course, the avatar must enhance the learning experience by offering the learner the cues normally provided by a human instructor. That not only requires the avatar to be animated, but it must also achieve a level of realism. Let’s look at these two aspects in a bit more detail.

Learning and Cognitive Processing

The human brain, like your computer, has a limit to its capacity to process input for learning. When the brain reaches its processing capacity, information begins to be missed. When you are learning, there is information coming in that is intrinsic, or germane, to the topic – information that is important to learning – and extraneous to the topic – information that is not important to the learning. Minimizing the extraneous information allows more of the brain’s learning capacity to be focused on the information the end user needs to learn. There are simple ways you may do this every day. You may put headphones on to block out noise when you are working. Maybe you prefer less commotion. TV on, or off? These are strategies for eliminating extraneous input into your brain while you work on learn.

Extraneous input from your eLearning lesson takes up brain capacity that could be used for learning. That’s why providing content that is complementary to, or beyond, what the learner must know can hinder the understanding and retention of key information in a lesson. For eLearning, including audio and visual input that is not germane to the topic can hamper the learner’s ability to process, retain and apply the information they learn in the lesson.

Avatars, if done incorrectly, are extraneous information, but if done correctly, they improve retention and the ability to apply the knowledge on the job.

The difference is in the “realism” of the avatar. As I mentioned earlier, our brains innately recognize and understand physical cues of a human educator. These human gestures, facial expressions, and body language enhance the learning experience by aiding in the brain’s ability to process and retain the information. But because our brains are wired to process this visual input, this visual cues do not take up any of the brain’s processing power during learning. By creating avatars that can mimic a human instructor with a level of realism that the learner’s brain can “read” the visual cues as he or she would a human educator, then the avatar can enhance the learning experience. Otherwise, an avatar may negatively impact the learning experience.


The human brain is programmed to pick up cues from other humans, so the level of realism in the avatar must be human-like. Realism in an avatar comes in two forms: behavioral and visual.

To increase retention and the learners’ ability to apply the knowledge on the job, the avatar must be animated to behave like a human educator. Using hand gestures to point to key concepts on the screen, head gestures, body language, facial expressions, and timing the mouth to the corresponding narration. The more life-like the animation of the avatar, the better the human brain is at taking the cues as if they were being presented by a human avatar. Otherwise, the brain must work to understand the avatar, taking processing capacity away from learning.

The same is true with the visuals. The more realistic the avatar, the easier it is for the brain to process the cues. The brain is able to treat the avatar as a human educator rather than having to dedicate processing capacity to watch and process the avatar as an additional source of input. There is a very interesting catch to this. You can actually make avatars too realistic. If you graph the effectiveness of the realism of avatars you’ll find an interesting valley as you reach the most realistic avatars. It’s called the “uncanny valley.” When an avatar reach a certain level of realism, the brain shifts from viewing them as realistic avatars to humans with something weird going on. When avatars reach this level of realism, the brain begins to focus on the avatar in an attempt to decipher what is wrong with the avatar. Thus, avatars that fall into the “uncanny valley” become huge distractors from learning. There’s a great study by Karl MacDorman and Hiroshi Ishiguro that studies emotional responses to different robotic designs. This study provides visuals to help you see the transition to and from the “uncanny valley.” There are a number of researchers in the field of psychology, lead by Richard E. Mayer, that are applying this research to avatars and learning. Check them out if you would like additional information.


Not everyone is going to have the skills, resources, time or budget to make highly-animated, realistic avatars. That is a given. And the struggle is that if we fall short of this goal with animated avatars, we risk adding extraneous information and interfering with learning. Granted, if you are not using a highly-animated, realistic avatar, you will not achieve the promise of increasing the learners’ ability to apply the knowledge on the job, but you can improve engagement and even comprehension. Here are some recommendations based on my experience and research.

When and Where

If you have avatars that do not meet the level of realism that is necessary to enhance learning, be selective about when and where you use them. These avatars can still be engaging. Use them at the beginning of the lesson to put a face to the narrator. Have them on transition slides to keep the learner grounded in the world of the eLearning lesson. Follow up with bringing the avatar back to conclude the lesson. This allows the avatar to be the navigator through the learning. And while you are leveraging the benefits of the avatar on engagement, by not putting the avatar on slides where the learner needs to focus on learning, the avatar does not hamper learning by being extraneous information.

Still Images

There are plenty of places you acquire visually realistic avatars. Stock image or eLearning resource web sites have collections of avatars, many not even designed to be animated, but visually on target for supporting eLearning. Instead of animating the avatar, take these stock avatars, create several positions and create still images of the avatar you can place into the lesson. This also works with photographs of models. Find collections of the same person in multiple poses and knock out the backgrounds. Whether it is the avatar or photos of models, these “anthropomorphic virtual agents” are an effective way to guide the leaner through the lesson, increase engagement, and still have a positive impact on learning.

Again, these still images won’t increase retention and application of knowledge as a highly-animated avatar would, they can still increase engagement and enhance learning.

That’s the skinny on using avatars in eLearning. Use them wisely and your learners will reap the benefits. And if you have any questions, you know where to find me.


MacDorman, Karl F., and Ishiguro, Hiroshi. 2006. The uncanny advantage of using androids in cognitive and social science research. Interaction Studies, 7:3, 297-337.

Strategist, author, and Senior Instructional Designer, Michael is passionate about organizational transformation and the Associate Experience.

Dancor Solutions is a communication strategy company that specializes in elevating the Associate Experience through strategic and compelling communication. Follow @DancorSolutions or visit www.DancorSolutions.com.


Leave a Comment

Your email address will not be published. Required fields are marked *