Take the 2-minute tour ×
Cognitive Sciences Stack Exchange is a question and answer site for practitioners, researchers, and students in cognitive science, psychology, neuroscience, and psychiatry. It's 100% free, no registration required.

When using operant conditioning to train mice or rats, what type of stimulus is most effective? For example, does a negative reinforcement of a loud noise have a more profound effect than a negative reinforcement of pain? What about for positive reinforcement? Is pleasure more powerful than food? Which type of stimulus is considered optimal for learning?

share|improve this question

1 Answer 1

up vote 7 down vote accepted

First, the concept of optimality of a learning curve is not well defined. You can measure at least 3 different aspects of learning:

  1. Speed of learning
  2. Time before extinction
  3. Performance at peak

Of course, there may be other measures as well, and any combination of such measure may also be a legitimate measure for certain uses.

Conditioned Taste Aversion (CTA) is a form of learning in which animals (and humans) learn to avoid a type of food after consuming it and suffering symptoms resembling food poisoning shortly after. In certain cases, the learning can take a single trial, cause the animal to completely avoid the food, and last for a lifetime [1], making this an optimal learning in all of the above 3 measures.

However, CTA is a very special and specific type of learning, and won't help you in teaching your rat to walk in a maze... I haven't found a methodological comparison of different types of reinforcements, and I suspect such a comparison might be very difficult or even impossible, because the efficacy of a (positive) reinforcement is greatly affected by the animal's "appetite" for it: for example, to use food as a reinforcement, rats are usually starved (the technical term is "maintained on a food deprivation schedule") for a few days. To use water as a reinforcement, rats are kept on a water deprivation schedule. How would you equate the subjective level of starvation and thirst?

Other aspects of the learning schedule have been studied more thoroughly. For example, to produce a long lasting learning (increase measure #2) at the expanse of speed of learning (measure #1) and peak performance (#3), you can provide partial reinforcement: reinforce the animal only on some of the trials [2].


[1] Conditioned taste aversion: Memory of a special kind. Bures, Jan; Bermúdez-Rattoni, Federico; Yamamoto, Takashi; Oxford University Press, 1998 doi: 10.1093/acprof:oso/9780198523475.001.0001

[2] Partial reinforcement: a review and critique. Jenkins, William O.; Stanley Jr., Julian C; Psychological Bulletin, Vol 47(3), May 1950 doi: 10.1037/h0060772

share|improve this answer
    
Great answer, thanks! –  Josh Gitlin Aug 21 '12 at 13:18

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.