Nondeterministic experiment

from Wikipedia, the free encyclopedia

A nondeterministic experiment in psychology is an experiment in which the participants (humans or animals) only seem to have to cope with learning tasks , but in reality the outcome of the experiment (e.g. the reward) is independent of the behavior of the participant.

Structure of the experiment

It is z. For example, an apparently interactive test set-up is designed so that the test subject's answers (mostly yes / no decisions) have no influence on the displayed result. However, the test subject believes in a real interactive test setup and feels directly responsible for the displayed result.

If the displayed result now follows the desired result expected by the test person (the learning curve ), the test person assumes that he has “learned correctly”. This also applies if the results displayed happen to follow the learning curve.

After evaluating the experiments, it was found that the test subjects develop the most adventurous theories about what they have "learned". These theories have two amazing properties:

  1. They are resistant , i. H. in some cases the test subject continues to believe that he has found a successful system; even after she was shown that e.g. B. a switch was not connected at all and thus the experimental setup was only apparently interactive. (see gambling addiction , superstition )
  2. You are contagious . In some experimental setups, the test subjects were brought together in the final interview with people who had really mastered the learning task (in some cases, the result of these people was always displayed as the correct result, they then automatically move on the learning curve). Often the "real learners" followed the eloquent and delicate statements of the "false learners" because they mistrusted their own, often simple but correct statements.

Some nondeterministic experiments have been described by Paul Watzlawick (e.g. in “How Real is Reality?”).

Experiments with animals

The superstitious rat

A variant of these experiments is the so-called “The Superstitious Rat” experiment . The experimental setup is designed in such a way that a rat can cover the path from a movable grid to a feeding trough in approx. 2 seconds. The feeding trough is only filled if it arrives at the feeding trough after 4 seconds at the earliest and 5 seconds at the latest.

Result: At some point after opening the grate, the rat will not run directly to the feeding trough, but will carry out some activity that means that it only arrives at the feeding trough after exactly 4 to 5 seconds and thus fulfills the condition by chance. The food reward increases the likelihood that the rat will perform the same activity again in the next test run, that is: to reinforce it . Humanizing this behavior can be interpreted to mean that she now "believes" (therefore superstitiously ) that this activity resulted in the reward, and therefore repeats the activity.

Such experiments can result in the conditioning of bizarre movements. Burrhus Frederic Skinner has documented such "superstitious behavior" in domestic pigeons in film recordings .

The neurotic horse

Experiments (e.g. with a horse) in which a punishment takes place are also similar. For example, by a power surge with a simultaneously triggered warning tone. The learned behavior (avoiding the punishment), including the reaction to the signal tone, is retained, even if the punishment by the current surge no longer occurs ( classic conditioning ). Only with the onset of the forgetting curve do regressions of this behavior occur again. But the stronger the punishment, the longer the regression takes.

literature

  • Paul Watzlawick: How Real is Reality? Munich 2005. ISBN 3-492-24319-3
  • Burrhus Frederic Skinner: Superstition in the Pigeon. Journal of Experimental Psychology 38, 1947, pp. 168-172, full text

See also

Web links

Individual evidence

  1. ^ Skinner and his superstitious doves. (PDF) At: ewi-psy.fu-berlin.de , accessed on September 16, 2015