What Is Latent Learning?
Latent learning is a theory in psychology that describes learning without a reward. An organism learns a new concept simply from observation and without any obvious reinforcement. The organism may not be consciously aware of its new skill until it suddenly expresses that skill when it becomes useful at a later date. For instance, a person can casually observe other people using chopsticks to eat and discover much later that he or she can use them correctly without ever being taught.
A classic experiment in psychology illustrates how latent learning works. Edward C. Tolman and C.H. Honzik famously placed three groups of rats inside a maze, where the rats were allowed to wander around. One rat group always received a food reward when reaching the end of the maze, while the second group found no food at the end. The third group found no food at the end of the maze for ten days but discovered food on the 11th day.
The first group of rats learned to reach the end of the maze quickly to reach the food. The second group continued to wonder around the maze. The third group acted similarly to the second group until food was placed at the end of the maze on the last day. One day after food was placed, the third group had already learned to reach the end of the maze as quickly as the first group.
This experiment illustrates latent learning because it showed that reinforcement or a reward is not always necessary to learn. It is possible for a person — or a rat — to learn from the environment around him, without any specific incentive to do so. Once there is a reason to use that knowledge, it may be called upon and put into practice.
The latent learning theory stands in contrast to other learning theories in psychology. Proponents of the stimulus–response (S-R) association theory believed that an organism learns due to some stimulus in the environment which elicits a specific response from the organism. This viewpoint was directly influenced by Ivan Pavlov's classical conditioning theory. Pavlov famously discovered that dogs did not simply salivate to the presence of food, but eventually salivated to the sound of the bell indicating food is coming, or to the presence of the person who fed them.
@Mor - I think latent learning just needs to be described as being learning done without an obvious reward stimulus. There is always going to be some kind of stimulus going on and there's no way to completely remove that. Observational learning is something that will just happen all the time, but what scientists can measure is that latent learning can be just as effective in the long run as other kinds of learning.
@KoiwiGal - I don't actually think this is a particularly good argument against operant conditioning although it might hold some ground against classical conditioning. I just don't think that we understand a rat's motivations enough to say definitively whether or not they had prior motive to learn the maze. It might be that there was some other kind of reward.
I know that I enjoy exploring a new environment for the sake of it, and it makes sense for animals to enjoy that because learning about their surroundings would be useful.
So the rats might have already had an internal reward compelling them to learn the maze in a general sense, and they just used that information when there was another type of reward offered later on. Is it still latent learning if it was learned actively, but just not for an obvious reward? I don't know how you would design an experiment to test that.
I much prefer this idea to the one where the dogs salivate on command, because I don't think life should all be about obvious rewards. That just seems like a descent into hedonism to me, where no one will do anything that doesn't give them an immediate reward and there is no learning for learning's sake.
I know it's more likely that we are a mix of both, and that offering a reward will obviously compel people to learn something faster, but that just doesn't seem to include the idea of free will, if you don't consider that people will learn things without being bribed into doing so.
Post your comments