Is learning by trial and error a valid way of learning?

[Shiroi Tora]

Yes, it is. The question, however, is one of efficiency and efficacy. Should the child be naturally intelligent, he will be curious. He will notice all things around him and will see patterns, correlations, and anomalies. He will be able to establish causation naturally. He will rarely need to learn through trail and error as he will have foresight.

To learn something through trail and error...I don't feel is efficient...unless the child can generalize the lesson and apply it to a larger group of thought. It is, otherwise, a way to learn how to NOT do something...there are an infinite number of ways to do something wrong.

Now, if we are talking about running mental trail and error experiments...that is a part of forethought. Having to experiment on something that had adequate evidence at hand and to not be able to figure out a correct path ahead of time is only an exercise in a method of learning for future situations where adequate evidence is not at hand. Learning through trail and error is more accidental I would think (we are talking children here...not part of the scientific method).

To learn through exploration of a successful person's method and examples of successful strategies in life is an effective and efficient way to learn. As one goes through life...many lessons present themselves (so long as the child questions the process in an attempt to delve into the methods or reasons used to establish cause and effect). To ignore those lessons is to ignore experience in life. I think that learning through exploration comes about naturally. All a child needs is opportunity. Yes, playing with children of equal cognitive capability is important...however...being with children of higher capabilities would benefit the child much more I would think.

 

[eon]

I agree entirely. I was proposing this learning style to someone in the management unit at my workplace- that it is quite similar to trial and error, repetition of sequence with modifications at resistance points, and that for it to work properly, it would require access to expert examples that can be studied. I felt that the key was gaining the understanding of the scenarios calling for the specific technique. Once understood, the expert technique could be applied by the learner correctly, having gained the ability to know when to use it. So I think practice is a better description than the term trial and error because WHILE practicing, you are able to observe whether you failed to implement the correct strategy in that the fully desired outcome is not obtained. You observe the error as a component of having an understanding of what the right technique probably is based on expert example to help you identify the situation where the technique fully applies.

 

[Shiroi Tora]

Yes...and in the absence of a clear example....just as bio feedback....or the electronic tuner for the tuning of musical instruments...any way of immediate feedback would be a guide to efficiency.

 

[Geordie]

Learning by trial and error is good! If one thing doesn't work, try another way till we succeed. Cool

Waiting sucks but at least we prove to ourselves that we know one way where we failed

 

[Spinning Compass]

Well, I would say that it depends on the situation. If you are in a situation where other people's lives and/or safety could be affected, then trial and error is definitely not the way to go. Trial and error is good for when you have the luxury (time and safety-wise) of exploring alternatives and see which ones work. But if you are dealing with a problem with a known solution, it's best to stick to proven ways of solving the problem.

For example right now I am learning a new set of duties. I can't go into detail due to the confidential nature of my work, but it involves asking the computer questions that I don't know the answer to. And because I don't know the answer to what I am asking, I don't know if the answer is right or if I have asked in the right fashion. A few times when someone else checked my work they caught things I missed because I am basically learning trial and error. This bothers me because ultimately pharmaceutical companies use the data I produce as part of their decision-making process in regards to new or existing drugs. So I naturally want to do things right the first time and not have to learn by making mistakes--next time, do it this way or don't do it that way.

 

[jaws]

Excellent points. I would however argue that learning through observation does not allow for creative and new solutions to existing scenarios. Trying to use what I call aspie logic..."okay, that might work...but why would ___ not work?" Then trying it to see if there is indeed an error in the assumption and trying to correct for the error without just doing what has been done before. I would probably need more information on the various constraints to determine whether the trial and error method is feasible in a particular situation. I guess I'm trying to say that repeating an observed skill is mimicry, while trial and error is actual learning. I may be over thinking the question or under valuing the assumptions of valid versus efficient.

 

[Spinning Compass]

Ok, now here's an actual situation (it happened to me). You are sitting on a plane. You are not a frequent flier so consequently you are a little nervous. The Captain comes on the intercom and says, "Sorry for the delay, folks. We are trying a new way to start this plane. When we do we will get underway."

How do you feel about that? Personally, I'd like my pilots to know how to be able to start the plane BEFORE I get on board.

 

[Ylva]

I may be over thinking the question or under valuing the assumptions of valid versus efficient.

I am unfamiliar with the concept of overthinking something. Isn't it always best to think things through, and to do so thoroughly? If not, where goes the line for sufficient thinking? When is it enough?

But… if you are referring to us [Aspies] taking the concept too literally, I can perhaps see where you are going. I never liked the idea of error – it feels like I've done something wrong, which is true by definition but morally irrelevant, and the very word "fiasco" grates my nerves. Why not call it "trial and success" instead? I learn more about causality from my successes. Mistakes are mostly confusing.

To strike a proper balance I'd say two errors to one success is enough. More for highly challenging problems, obviously, but that seems fair. Confusion is uncomfortable, but it's not the end of the world.

I prefer learning by instruction first. Granted, I'm no longer a child, but I'm glad that when I was, I was not left to figure out algebra by myself. Or when I was even younger, that I got some help with the whole language- and motor skills part.

Children can get into magical thinking. Maybe the sense that there is such a thing as cause and effect is something we are born with, but the correctly deducing the difference between correlation and causation definitely isn't. So unless someone is there to explain, for instance, that your thoughts did not cause grandma to get hurt, the thought and the event just happened to occur close to each other in time (yeah, I'd have phrased it differently if I was speaking to a child) you might end up internalizing things like that until someone takes you aside and does explain it.

Though I suppose that is a form of trial and error, too.

 

[King_Oni]

I am unfamiliar with the concept of overthinking something. Isn't it always best to think things through, and to do so thoroughly? If not, where goes the line for sufficient thinking? When is it enough?

While I'm totally on the boat of "thinking things through"... my experience is that the predominantly NT world doesn't want that. They want you to be busy... that constitutes trial for them.

I can sit around for weeks thinking about something and not doing anything actively until I've get my stroke of genius and finish it in the same time anyone else would (including the trial and error process of maybe weeks).

 

[Aalo]

Learning styles can all be criticized about, and for me trial and error sometimes misses too much information.
If first method won't work, it's not said that second would. At least in a way it's conducted - or at the time, or given specific parameters involved. And whether that same method would universally speaking work, small details can really lead astray. As a person, who easily gets frustrated if not provided with positive feedback at times, it could get easy to begin avoiding trial and error situations. It's not always about getting upset of failure, but getting confused on what to try next because one's ability to see operations as really complex systems, where so much could go wrong.

In a way I agree saying that well planned is almost done, it has to be kept in mind that this only applies if every given altering factor is taken into account. Or at least there is realization of possible need for improvisation during the process. But of course trial and error offers more space to learn one's mistakes if applied many times enough.
It feels really unnecessary to compare this method to rote learning. Both might simultaneously be essential at some cases, one needs to have some sole base of knowledge in order to even try doing educated guesses during trial and error type activity. I also believe that it's strongest benefit is to teach person dedicated on learning, how, and when, not to do sth in some ways.
To me the key isn't do until it works, but do until I know how and why it won't work in some cases. Rote way succeeding in something once may lead to believe the same thing can be run in every working environment in every given situation. Of course this might seem more amateur-just-having-fun-type of learning, than to resemble anything efficient. But to be familiar enough with any system, it's flaws have to be known too. That is also an important measure when valuing new possible approaches.

Why get upset of errors? They won't mean that your whole life's a failure now, thy simply imply that this method doesn't work - which rises delicate question of interest about how it might. So thrilling.