TRASCRIZIONE 🔗
WEBVTT
00:00:00.160 --> 00:00:05.280
What if your best looking training videos are quietly making learning harder?
00:00:05.679 --> 00:00:11.279
And not because your content is wrong, but because the brain is overloaded before the lesson can land.
00:00:11.439 --> 00:00:13.039
Let's talk about how to fix that.
00:00:13.359 --> 00:00:17.519
Good morning, good evening, good afternoon, wherever you are and wherever you're watching from.
00:00:17.679 --> 00:00:19.120
Welcome to the Visual Lounge.
00:00:19.280 --> 00:00:20.960
Let's just dive in.
00:00:21.280 --> 00:00:23.760
Imagine a postcard size desk.
00:00:23.920 --> 00:00:24.800
Tiny, right?
00:00:24.960 --> 00:00:26.960
That's your learner's working memory.
00:00:27.199 --> 00:00:30.000
Put a few items on it, and it's great.
00:00:30.320 --> 00:00:34.799
Stack on a few more, and well, things are gonna start sliding off.
00:00:35.039 --> 00:00:44.560
In learning, that desk has two incoming channels: the auditory channel, the words which I say, and the visual channel, which the viewer sees.
00:00:44.719 --> 00:00:52.399
If we pour in too much, busy screens, dense narration, duplicate text, things are just gonna start falling off the desk.
00:00:52.640 --> 00:00:55.200
That's cognitive overload.
00:00:55.520 --> 00:01:01.600
Our job is to design so only the right items are on the desk at the right time.
00:01:02.000 --> 00:01:05.680
Today we'll use some proven multimedia principles to do three things.
00:01:05.840 --> 00:01:16.560
Eliminate the noise we add by accident, we're gonna respect the brain's limit with smart pacing, and find ways to guide the eye so attention lands where meaning lives.
00:01:16.799 --> 00:01:20.799
Now, extraneous load is mental effort that doesn't help learning.
00:01:20.959 --> 00:01:22.799
It's clutter on the desk.
00:01:23.040 --> 00:01:26.959
If an element doesn't point to meaning, it's probably noise.
00:01:27.200 --> 00:01:38.480
So we want to start thinking about cutting decorative backgrounds, pop-ups, extra toolbars, and sometimes our clever flourishes that really don't serve the current setup.
00:01:39.280 --> 00:01:44.239
A common overload pattern is actually reading and watching at the same time.
00:01:44.560 --> 00:01:57.200
So if you're using full sentence text on screen while narrating over a live interface, that's gonna force the visual channel to read and track the demo simultaneously while the narration competes for attention.
00:01:57.519 --> 00:01:59.040
So here's what you can do instead.
00:01:59.280 --> 00:02:02.000
First, let your voice carry the why.
00:02:02.239 --> 00:02:05.200
Let the screen, when you show it, carry the what.
00:02:05.599 --> 00:02:08.800
And keep on-screen text short and purposeful.
00:02:08.960 --> 00:02:19.199
You can think about labels, maybe keywords or steps, numbers, but put full sentence and captions on or transcripts, not floating over the demo.
00:02:19.439 --> 00:02:25.360
And while scripting or you're going through
Load full transcript
your your review process, here's three quick questions you can ask.
00:02:25.680 --> 00:02:28.879
Is this element necessary to understand this step?
00:02:29.759 --> 00:02:32.719
Is any text duplicating what I'm saying?
00:02:33.599 --> 00:02:36.960
Can a viewer tell instantly what matters on the screen?
00:02:37.520 --> 00:02:42.639
And if the answer to the last step is no, we'll solve it shortly by guiding the eye.
00:02:42.800 --> 00:02:46.159
Remember, everything has to fight for a place in your video.
00:02:46.240 --> 00:02:50.560
That's audio as well as visual as well as text.
00:02:51.199 --> 00:02:55.599
Now, intrinsic load is the complexity of the content itself.
00:02:56.400 --> 00:03:02.479
I can't make a 15-step workflow inherently simple, but I can make it learnable.
00:03:02.719 --> 00:03:03.599
How can we do this?
00:03:03.759 --> 00:03:13.199
Well, we could break the lesson into small, meaningful clusters, maybe two to five actions per cluster, and you can finish a cluster by giving it a brief breath.
00:03:13.280 --> 00:03:14.800
Then you can move on.
00:03:15.039 --> 00:03:22.639
If the platform supports it, you can let the learners click, continue, which I know gets a bad rap, but sometimes it might actually be really good.
00:03:22.879 --> 00:03:25.759
And if not, you can insert a short verbal reset.
00:03:26.000 --> 00:03:28.639
Phase one, record, do the steps.
00:03:28.800 --> 00:03:31.599
Phase two, edit, next small cluster.
00:03:31.759 --> 00:03:33.439
Phase three, export.
00:03:33.759 --> 00:03:41.680
Those two to four second resets let working memory file what it just learned before the next items hit the desk.
00:03:41.840 --> 00:03:43.360
There's probably lots of ways to do this.
00:03:43.439 --> 00:03:48.879
Another one is you could just invite the learner to pause the video and let them take the breath that they need.
00:03:49.680 --> 00:03:55.199
Now, when terms or parts are new, teach the pieces before the process.
00:03:55.439 --> 00:03:58.879
30 to 60 seconds of meet the parts plays off.
00:03:59.039 --> 00:04:05.919
For example, you'll use a timeline to cut the canvas to see changes and export to produce an MP4.
00:04:06.080 --> 00:04:07.680
That's all we need for today.
00:04:08.000 --> 00:04:14.400
Now, during the workflow, learners aren't decoding labels and following logic at the same time.
00:04:14.879 --> 00:04:18.639
Segmentation and pre-training don't dumb down anything.
00:04:18.800 --> 00:04:23.839
They actually sequence complexity so the tiny desk never overloads.
00:04:24.079 --> 00:04:31.040
Now, I realize I've covered a lot of vocabulary, and I'm about to introduce a new term which you might be saying, oh my gosh, this is a lot of learning science.
00:04:31.199 --> 00:04:32.560
And it is, but it's good to know.
00:04:32.800 --> 00:04:38.720
Germane processing is the good effort, the mental work that builds a durable mental model or schema.
00:04:38.879 --> 00:04:41.920
Both are good terms and both
well worth knowing.
00:04:42.079 --> 00:04:50.319
So we maximize our mental model or the good effort by directing attention and aligning our timing.
00:04:50.560 --> 00:04:54.480
You want to tell learners exactly where to look and why it matters.
00:04:54.720 --> 00:04:58.720
Even in a talk to camera format, we can give you an example.
00:04:58.879 --> 00:05:04.079
For instance, if I were to say, in the top right, the save button, this locks in your changes.
00:05:04.319 --> 00:05:06.639
That's gonna make immediate sense to you.
00:05:06.879 --> 00:05:23.279
Now, if I were to cut to screen occasionally, you could do something like a tight crop and a consistent highlight, give it an arrow, an outline, or halo or something, use sparingly to add to that, understanding the clarity that you're gonna provide.
00:05:23.519 --> 00:05:28.319
One thing you can do as your show a step is say the action as the action happens.
00:05:28.480 --> 00:05:33.439
You might hover a beat, then say the action, and then perform it as it's said.
00:05:34.160 --> 00:05:42.560
You know, words and visuals actually should arrive together so your viewer doesn't have to hold one thing in working memory while waiting for the other.
00:05:42.800 --> 00:05:52.160
There is a slight option there that you can say, start moving the cursor, give it a little bit of lead so the eye starts following, and then say the thing that you want to say.
00:05:52.319 --> 00:05:53.920
That's also an appropriate way.
00:05:54.000 --> 00:06:01.680
You just don't want to have such a big gap that the mouse goes up there to waiting, and you know, you wanna say the thing as it moves with it.
00:06:02.800 --> 00:06:08.959
So let's go through another example and it's just a pattern that you can follow along with.
00:06:09.040 --> 00:06:14.480
So it's pretty easy, it's gonna get very repetitive quickly, but I think it'll help illustrate the idea.
00:06:14.720 --> 00:06:16.319
So let's say you set the target.
00:06:16.480 --> 00:06:19.360
In two minutes, you'll know how to record, trim, and export.
00:06:19.600 --> 00:06:20.720
Perfect, right?
00:06:21.040 --> 00:06:26.160
Then next you can pre-train in 30 to 60 seconds, depending on what you're trying to show.
00:06:26.319 --> 00:06:35.199
Something like your timeline is going to equal cuts, and your canvas is going to equal views, and your export equals your MP4.
00:06:35.600 --> 00:06:43.279
So you're setting them up, providing them with clarity of what each of the pieces are, and then you can start to segment by your sub goals.
00:06:43.519 --> 00:06:50.480
So for instance, you might have several phases, like phase one, record a small cluster of steps and add a one-line recap.
00:06:50.639 --> 00:06:56.959
Then you'll move on to phase two, edit, which is a small cluster, and again another recap.
00:06:57.120 --> 00:07:07.600
And then you can move into phase three, and again, you never have to mention which phase you're on to the learner, you just follow the pattern where you're talking about exporting where it's a small cluster and a recap.
00:07:07.759 --> 00:07:15.120
And again, you're gonna want to signal and sync inside each cluster by naming the target clearly and saying the action as it happens.
00:07:15.360 --> 00:07:18.720
Now, at the very end of that video, you could close with the retrieval cue.
00:07:18.879 --> 00:07:24.319
This is a one-sentence summary that restates the goal and the crucial step.
00:07:24.720 --> 00:07:26.480
Whew, a lot, I know.
00:07:26.639 --> 00:07:29.839
But here's another maybe more concrete scenario.
00:07:29.920 --> 00:07:35.600
Uh so let's say that we've got a video that we're gonna make again about making an export of a video.
00:07:35.839 --> 00:07:36.959
Today's goal is simple.
00:07:37.199 --> 00:07:38.959
Record, trim, export.
00:07:39.199 --> 00:07:46.959
Before we start, the parts you'll use are the timeline for cuts, the canvas to see changes, and the export to produce your MP4.
00:07:47.360 --> 00:07:48.480
Now let's get started.
00:07:48.639 --> 00:07:49.839
First with recording.
00:07:50.079 --> 00:07:55.199
Start a capture, stop when you're done, and your clip appears in the project.
00:07:55.759 --> 00:07:57.920
For our next step, we're gonna look at editing.
00:07:58.079 --> 00:08:03.279
What you're gonna do is find the pause, make two cuts, remove the gap, and close it up.
00:08:03.839 --> 00:08:06.240
Now, the last thing we need to do is export.
00:08:06.399 --> 00:08:08.560
Choose an MP4 and confirm.
00:08:08.800 --> 00:08:12.639
The key idea is that you're explaining and showing things together.
00:08:12.800 --> 00:08:19.839
You don't need paragraphs on screen, you just want to focus on one thing at a time and help them move through the process seamlessly.
00:08:20.079 --> 00:08:24.560
If you have something going on like a continuous monologue, insert a micro reset.
00:08:24.720 --> 00:08:28.879
Something like, that completes phase one, here's what you should have now.
00:08:29.120 --> 00:08:30.560
Then you can continue on.
00:08:30.800 --> 00:08:34.240
Maybe you're reading big paragraphs on screen while narrating.
00:08:34.480 --> 00:08:35.519
Gosh, that's a lot.
00:08:35.679 --> 00:08:38.000
Replace them with labels or keywords.
00:08:38.320 --> 00:08:42.480
Put full sentences, again, in captions or in the transcript.
00:08:42.720 --> 00:08:45.360
Vague references like it's up there somewhere.
00:08:45.600 --> 00:08:48.080
You would actually want to do something like name and locate.
00:08:48.240 --> 00:08:50.000
Top right, the save button.
00:08:50.240 --> 00:08:54.720
And if you cut to a screen, crop tight and use one subtle highlight.
00:08:54.879 --> 00:09:02.080
I'd also encourage you to increase the size of your mouse cursor so it's easy to see and it's always findable on your screen.
00:09:02.320 --> 00:09:05.759
Okay, let's start recapping because that was a lot, right?
00:09:05.919 --> 00:09:07.360
So cut the clutter.
00:09:07.519 --> 00:09:10.639
If it doesn't support meaning, it steals attention.
00:09:10.879 --> 00:09:17.360
You want to chunk the challenge, teach parts first, and group steps into small, paste segments.
00:09:17.519 --> 00:09:24.080
And then you want to guide the gaze, be explicit about where to look, and say the action as it happens.
00:09:24.559 --> 00:09:29.919
Now your job isn't to merely make videos beautiful, it's to make them learnable.
00:09:30.080 --> 00:09:33.200
Control the visuals, control the timing, control the clarity.
00:09:33.519 --> 00:09:36.720
And cognitive overload, it's gonna happen at some point.
00:09:36.879 --> 00:09:39.200
You've probably experienced it as you've watched videos.
00:09:39.360 --> 00:09:46.000
Just remember that on the other side of your video is a human being who is trying to learn and trying to understand.
00:09:46.240 --> 00:09:56.080
Now you might know them or you might not know them, but your goal is to help them, regardless, to get through the complexity of whatever it is you're teaching.
00:09:56.320 --> 00:10:06.879
Here's a quick tip that we picked up a long time ago at TechSmith is if you say something like, oh, this is an easy process, just click blah blah blah, guess what?
00:10:07.200 --> 00:10:10.639
It might be easy for you, but maybe not for them.
00:10:10.799 --> 00:10:28.480
And if there's a multitude of steps, you got a lot more steps, maybe more steps than even three or four, all of a sudden you've added this complexity that you really again want to pull back on and be thinking about how can I help them take this idea, this process that they're trying to learn?
00:10:28.639 --> 00:10:34.399
Maybe it's even thought leadership, and how do I help them move it into the learning kind of process?
00:10:34.559 --> 00:10:40.639
Again, we we want to go back here as we wrap up and think about what is it that we're trying to do?
00:10:40.799 --> 00:10:58.879
Well, we're trying to make ideas, processes, all these things move from video into that working memory, into the long-term memory, so we can pull it out of the catalog, the library, and into, you know, we have a good retrieval process.
00:10:59.120 --> 00:11:04.879
Video inherently is tough to do, it's tough to move from working memory into long-term memory.
00:11:04.960 --> 00:11:17.200
So you need to be thinking about what are the things that are gonna help to reinforce, to bring back up, and so allow that person to give them time to put into long-term memory, but also encourages them to use it enough that it sticks.
00:11:17.360 --> 00:11:19.279
So, well, this has been a lot.
00:11:19.440 --> 00:11:21.200
It's a little bit different of an idea.
00:11:21.360 --> 00:11:27.039
I wanted to do something as I've been thinking a lot about video creation from a learning perspective.
00:11:27.679 --> 00:11:30.000
And here's what I'll end on.
00:11:30.159 --> 00:11:34.720
I think if you go through this, you're probably gonna say, Whew, that was a lot.
00:11:34.960 --> 00:11:42.480
There is a lot of great research out there about cognitive load, about work in memory, about the learning process, learning science.
00:11:42.559 --> 00:11:49.200
Um, if you're looking for stuff related to visuals and multimedia, Richard Mayer is a great resource to search for Dr.
00:11:49.279 --> 00:11:50.000
Richard Mayer.
00:11:50.240 --> 00:11:52.559
Uh, Jonathan Halls has some great stuff out there.
00:11:52.720 --> 00:11:56.639
If you are in the ATD ecosystem, you might know Jonathan Halls.
00:11:56.720 --> 00:11:59.039
He's written in this book creating training videos.
00:11:59.200 --> 00:12:05.840
Uh this one's about using smartphones, but it's got a lot of great backup on kind of learning sciences and getting started.
00:12:06.000 --> 00:12:07.840
Jonathan's been a guest on the show.
00:12:08.000 --> 00:12:10.639
Um, there's lots of great information out there.
00:12:10.720 --> 00:12:15.440
And I hope this is just a taste to get you going so that you want to make better videos.
00:12:15.600 --> 00:12:18.480
You want to make better, more effective learning videos.
00:12:18.720 --> 00:12:24.559
Now, the other thing I have to mention is that AI is going to play into this, right?
00:12:24.879 --> 00:12:28.559
You can take your ideas and pit them against AI.
00:12:28.879 --> 00:12:38.639
Ask it to pull on the current research, ask it to look at things, to fact-check, to make sure are there better ways to move this through.
00:12:38.799 --> 00:12:42.639
I I've just presented a series of ways, just some simple things.
00:12:42.799 --> 00:12:44.639
There's many more things that you can do.
00:12:44.799 --> 00:12:49.279
Uh, lean to your AI, just don't remember, don't let it do everything.
00:12:49.519 --> 00:12:53.440
Be the human in the process because your ideas are fantastic.
00:12:53.519 --> 00:12:59.120
And your learners will benefit from what you bring to the table in that human way.
00:12:59.360 --> 00:13:09.360
Uh, especially if you're helping them to not get overloaded cognitively where that they can actually remember the things that they need to do and apply the learning that you're providing for them.
00:13:09.600 --> 00:13:10.480
Well, that's it.
00:13:10.639 --> 00:13:14.720
I I hope that this some of this hits home for some of you.
00:13:14.879 --> 00:13:16.240
I'd love to hear from you in the comments.
00:13:16.399 --> 00:13:20.720
You can always, of course, email us at the visualounge at techsmith.com.
00:13:20.799 --> 00:13:26.000
We'd love to hear from you, and we hope that you take a little time to level up every single day.
00:13:26.240 --> 00:13:27.679
Thanks everybody.