Visual Learning versus Classrooms and Lectures

“To hear is to forget – To see is to remember – To do is to learn!”

Most people agree with this ancient proverb.  Yet few educational systems make the most of its wisdom.

What does it mean?  Speech is forgotten – Visual material is remembered – Action and feedback allow one to learn!

A good system cannot rely on vocal instruction.  Yet most teaching is delivered by voice, in classrooms and lectures and nowadays in MOOCs.

Visual material like texts and pictures and games should convey any essential information.  This includes the bulk of what is now delivered by voice.  Then, action must follow, with feedback.

Otherwise, the speech will be forgotten, and the visual material merely remembered.  Acting, and seeing what happens, is the way to learn.

A lesson I learned 50 years ago

I was a psychology major at Columbia in 1965.  One course included a self-pace book by the leader of behavioral psychology, BF Skinner.  This was an eye-opener.

I could read faster than any teacher could talk, and go on if I understood. No need to wait for other students.

If I didn’t understand, I could review until I did. No need to worry about holding up a class.

The tests and answers were part of the book.  I found that reading and taking the tests was super-efficient, compared to sitting and listening.

In 1975, I took law courses from the University of London.  These correspondence courses enabled very fast learning.

I studied one subject at a time, whenever I had a spare hour, and took the tests when I was ready.  This too was speedy learning.

When I joined Merrill, Lynch in 1977, I found their basic training was excellent.  It was comprehensive, well-designed, and had the advantages I just mentioned.  All of it came in a box that I found on my desk, my first day.

All three of these courses were vastly more efficient than my traditional education at St Paul’s School in Concord NH, and at Columbia University in New York City.  And the cost was minimal.  Yet classrooms and lectures continue to make up most ‘teaching’ today.

Here are some telling advantages that text and visual links have, over voice, in education.

First, we can read several times faster than people talk.  The classic way to count seconds is to say “One second, two seconds, three seconds, etc.”  That is two words per second, 120 per minute.

Reading can be at least twice as fast.  This alone should make us demand written materials for every important subject.

Second, the text is our notes.  We don't have to listen and take notes at the same time.  If the text is on paper, we can write extra thoughts on it.

If the text is on a screen, we can carry it everywhere and find anything, even searching for words easily.  With an iPad, we can read anywhere and add our own notes and search for words.

Third, teachers in a classroom, and even most lecturers, are not experts like the one who wrote the book.  Vastly more time and effort and expertise can be put into a lesson that will be used by thousands of students, than can be put into almost any classroom lecture.

Fourth, text allows much greater control over what words are used.  New concepts should be taught with familiar words where possible.

Strange words, especially for younger or foreign students, will be stumbling blocks that can easily be kept out of written material.  Also, with text on a screen, we can provide an online link to a definition of every word.  This is quite impossible for voice.

Also with voice, you cannot look up the meaning without knowing the spelling.  Nothing new should ever be taught by voice – and if it isn’t new, why teach it?

Fifth, voice presentations are so slow that they usually take up a lot of time.  An hour is typical.

Yet students' attention spans are less than ten minutes on average.  Please see the MIT/Harvard analysis after my email.

Text can be written for five or ten minutes of study, providing perhaps a half hour of verbal teaching.  That can be followed by answering questions on the material, or other forms of action.

Sixth, seeing is not just remembering.  The optical power of the brain is vastly greater than the visual.  The optical nerve seems to be more wired to the other parts of the rational brain.

Seventh, text on a screen can be accompanied with links to pictures, diagrams, charts, videos, and all kinds of visual materials.  Well-designed video games are already available for learning math.  Using visuals etc will help everyone, especially those for whom text is not the perfect way to take in information.

(As an aside, many people say they learn better, or only learn, from speakers.  But after 15 years of being taught all day by speakers, one should not be surprised at this.  Younger students should be less affected. There must be studies on this, but I haven’t seen them yet.)

Finally, and maybe most important, the time saved can be used for real learning, through actions that provide feedback.  Group projects for example are excellent for learning the team skills that real life requires.

With faster intake of information from visuals, as much as half of schooldays can be spent doing projects.  I have visited such a school already!

Overall, a better system can provide much better learning with no extra time or effort.

MIT/Harvard research shows that videos of lectures are not the answer.

Speech is too slow for efficient learning.  People offering online video courses (MOOCs) are aware of these problems. Below is some research from MIT and Harvard, on their video courses.

Note the first point.  After six minutes on average, viewers tune out.  If they take a break, they have to come back to the video, and review the points leading up to where they broke off.

Stopping and starting text or other visuals is vastly easier.  And one can easily bookmark or highlight text for review.

So in those six minutes, one can read what a video might take 20 or 30 minutes to deliver – and remember it better!

The research shows that students want lecturers to speak at over 250 words per minute.  That’s over four words per second.  This is about twice the normal rate of speaking, which we saw above is only 120 wpm.  A reader who knows a little about speed reading can read four or five times that fast, at 500-600 wpm.

Finally, note the last two points, which recommend allowing breaks for the student to digest diagrams and the like.  Since breaks are not normal in lectures, say the researchers, videos should be purpose-built for ease of navigation.

Still, breaking off and restarting videos will be tedious, compared to finding one’s place on a page of text.

After these points below, you will see a link to a video called LectureScape. This sets out ways to make lectures work for online education.  But I believe that the bulk of teaching should be done with visual material in the first place, for ease of navigation etc etc.  I am including the whole edX piece just below, as many of my points are dramatized when you read and watch the video.

The implications of all this are wide-ranging.  One is that we need to build online courses on text and visuals and games, for the benefits above.

Correspondence courses are a primitive but suggestive model.  These should be our starting point.  I mentioned above the three that I took, several decades ago.

Education Re-imagined: the future

Salman Khan wrote an eye-opening book in 2012:  One World Schoolhouse.  This describes how he and others at MIT learned twice as much as normal, by avoiding classes and lectures.

Instead, they read the materials and talked among themselves.  In four years, Sal earned a BS in Math, a BS in Computer Science, and an MS in Computer Science – and was president of his class!

How much faster and more easily could he have learned, if his materials were designed for speed-learning?  How much more action and projects would he have had time for?

Khan Academy now has 20 million viewers of Khan’s online videos each month.  And he is working on adding other methods like game theory to help his pupils to learn.  He is setting such a good example!

Current thinking in education is that we need to improve the quality of teachers.  Studies show a gain in learning of perhaps 50%, when teachers are excellent instead of average.  But Khan learned 100% more with a more efficient learning system.

And even greater gains might be achieved if ‘real-learning’ was part of the design, along with extra actions and feedback and projects.

My plan is to develop an internet-base curriculum.  For maximum effect, these will have ample links and visuals.  This will leave time for much more reading, action, feedback, and group projects.

Teachers will have time to become more like Socrates, and be able to coach 30 students, with much better results than we have today.

Fast real learning will benefit everyone in society.

Please see below for the Harvard/MIT research mentioned above.  Note that most of the points apply with even greater force to classroom teachers and lecturers.

In 2016, edX, the online learning platform co-run by MIT and Harvard University, gave researchers at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) data on the second-by-second viewing habits of more than 100,000 learners perusing more than 6.9 million video sessions.  In a paper published this spring, the CSAIL team outlined some key findings on what online learners want from videos. These include:

  • Brevity (viewers generally tune out after six minutes)
  • Informality, with professors seated at a desk, not standing behind a podium
  • Lively visuals rather than static PowerPoint slides
  • Fast talkers (professors seen as the most engaging spoke at 254 words per minute)
  • More pauses, so viewers can soak in complex diagrams
  • Web-friendly lessons (existing videos broken into shorter chunks are less effective than ones crafted for online audiences)

These insights form the basis of the CSAIL team’s LectureScape – a “YouTube for MOOCs” that seeks to reinvent how online learners watch videos.  LectureScape uses data on viewing behavior — particularly the “interaction peaks” that correspond to points of interest or confusion — to present MOOC videos in a way that’s more intuitive, dynamic, and effective:

In summary, viewers can consume videos more efficiently, skipping specific sections or repeating trickier ones, without having to slog through the whole video.  A demo of LectureScape, a “YouTube for MOOCs” designed by MIT researchers that seeks to reinvent how online learners watch videos, can be found through the link.  Video courtesy of the researchers, Juho Kim, a graduate student in electrical engineering and computer science, says that the group’s previous work on the tutorial-focused platform ToolScape (PDF) demonstrated that users learn more effectively with this type of interface. He says that traditional MOOC metrics, such as completion rates, are “too simplistic,” and don’t account for the many learners seeking specific skills (versus intending to formally finish a course).

The future robotic teacher?

Please kindly find an interesting article by Chelsea Gohd, with insights on robotic teachers and predictions by Thomas Frey, an American futurist.  Frey began his career at IBM, working as an engineer.  Later, he founded the DaVinci Institute, a networking firm and think tank for technical innovation.  He predicted that "by 2030, the largest company on the internet is going to be an education-based company that we haven’t heard of yet."

Here's a link to the article.