If we are to assume that bluetooth implants and other embedded tech that can prompt a user near-instantly with information are an eventuality, I think there is an interesting question to consider:
If everyday people have near-instant access to information, how will we continue to assess expertise moving forward? Surely it’s not enough to just have access to rote information, like in the case of the cheating test-taker. We will also expect our experts to have the deep understanding that comes from experience in a domain.
Will we need different language to describe flavors of knowledge and expertise? If so, will the nature of test-taking and assessment need to evolve to identify people who actually have an understanding of the thing being tested, instead of testing for rote answers?
We already have the language to describe someone who claims an expertise in a knowledge area without the study necessary to be an actual expert: dilettante.
And that's all that we'll actually be without memorization. There is a huge gap between someone with knowledge and expertise ingrained in their head, with a solid knowledge of the gaps in their knowledge, or understanding the layout of a knowledge realm that can only come from dedicated study of a subject, vs a dabbler or whatever level of expertise another individual might have with less stringent studies.
This same problem exists in our education system and cramming. You can cram subjects and pass tests, but research had shown that the knowledge gain from this process to be extremely limited.
Without a well ingrained knowledge of a subject, it is difficult to use that knowledge in creative thought, connecting with other realms of knowledge.
If all of these human mental processes are replaced with computation, and people no longer put in the effort to learn challenging things, then I predict large amounts of mental decline. We may already be seeing this process. Perhaps I should say... "As we offload more mental processing to computers...", Because it's definitely a process that many people are going through.
That isn't too say that computerized information is all bad. My wife would probably leave me if I didn't have a calendar app.
> how will we continue to assess expertise moving forward?
This reminds me of something that I read once - allegedly Aristotle was actually against books; he believed that having ready access to books made it easy to 'fake' the type of education that requires mentorship.
I think Aristotle's perspective doesn't really make sense today because (for better or worse) we emphasize the economic utility of education - is the person actually able to do the job that they claim to be able to? We don't consider the internal changes to the person caused by their education (alas...).
> We will also expect our experts to have the deep understanding that comes from experience in a domain.
I think Chalmer's concept of the extended mind is an interesting framing here - basically it's the notion that your mind doesn't end right at your skull. For the sake of argument, let's assume in the future that we'll have BCI that's good enough to let you text with you mind (something like what Neuralink is going for). If you've got an expert system in your head/pocket that's really good at dealing with some domain and you're really good at _phrasing problems in terms that the expert system can understand_ then you + expert system might have a super-human ability to solve problems in that domain.
If I was hiring for that domain, I wouldn't particularly mind how much of your expertise is in biological tissue.
> This reminds me of something that I read once - allegedly Aristotle was actually against books; he believed that having ready access to books made it easy to 'fake' the type of education that requires mentorship.
I think that was Socrates, not Aristotle. Socrates was very firmly against the concept of writing, and we know all about this only because his student was writing down what the teacher was telling him.
>If everyday people have near-instant access to information, how will we continue to assess expertise moving forward? Surely it’s not enough to just have access to rote information, like in the case of the cheating test-taker. We will also expect our experts to have the deep understanding that comes from experience in a domain.
We've always expected professionals to have an understanding. We've just been using memorization as a proxy for this.
Any 14yo with a cell phone can go on Reddit give you caned advise about investments or why your car is making a funny noise. But we don't trust teenagers googling stuff with those sorts of things in the real world because there's a huge difference between being able to pattern match information and actually understanding what's going on hence why we don't get advice we care about from anonymous people with backgrounds that can't be vetted
Professional education has a filter in front of it so it's going to be behind the curve when it comes to reckoning with the realities of information access in the modern age but it'll have to figure something better than tests out eventually.
In my experience you can easily tell the difference between someone who knows what they're talking about and someone who just googles what they're talking about. I have a feeling that won't change even when people have implants.
Rote regurgitation is only useful for teaching the basics, not advanced stuff. I propose testing this, asking students not to cheat, and catching cheaters later, in case they screwed themselves by skipping the basics.
Why do math students need to know the sine doubling rule? Not so they can calculate with it (they could look that up) but so they can reduce certain expressions to sin(2x). That's why calculus teaches this stuff.
> Why do math students need to know the sine doubling rule? Not so they can calculate with it (they could look that up) but so they can reduce certain expressions to sin(2x). That's why calculus teaches this stuff.
Why make math students memorize random trigonometric identities that they'll use once and then immediately forget when you could teach them a more general fact that they can derive the identities from?
sin(2x) = Im(cos 2x + i sin 2x)
= Im(e^2ix)
= Im(e^ix * e^ix)
= Im((cos x + i sin x)^2)
= Im(cos^2 x + 2i cos x sin x + sin^2 x)
= Im(cos^2 x - sin^2 x + 2i cos x sin x)
= 2 cos x sin x
The way exams work is already terribly outdated for many areas of expertise. Which is why companies no longer take a single written exam when interviewing professionals, but face-to-face interviews discussing topics. Academia has yet to catch up in many domains.
Every college class I had that taught something worth knowing ended with a test that allowed access to any notes you wanted, as the point was to demonstrate that you had internalized concepts, not just memorized facts.
Isn’t this already the case? At least in my experience interviewing for software development, all the questions are meant to test understanding and the ability to explore and solve problems, never static knowledge.
If everyday people have near-instant access to information, how will we continue to assess expertise moving forward? Surely it’s not enough to just have access to rote information, like in the case of the cheating test-taker. We will also expect our experts to have the deep understanding that comes from experience in a domain.
Will we need different language to describe flavors of knowledge and expertise? If so, will the nature of test-taking and assessment need to evolve to identify people who actually have an understanding of the thing being tested, instead of testing for rote answers?