The experimenter expectancy effect first came to be recognized at the Hawthorne plant of the Western Electric Company in Chicago in 1927. Industrial researchers were studying worker productivity. They observed workers over two years through a study design that altered their rest periods and refreshments. What the study found was that all variations in rest periods and refreshments increased worker output by 30 percent. It was then concluded that being watched and experimented upon had the greatest effect on the workers, not the changes in working conditions.
Unconscious physical movements by the experimenter often have a great effect on the results produced by the experiment's participant. Muscular tension in the throat or pitching the voice higher or lower can have surprising effect on participant response. Even how a question is constructed or the order of questions on the page can influence a participant in favor of the experimenter's expectation.
Since experiments have a tendency to be biased, one result of experimenter bias is the need for peer reviews and follow-up studies. Follow-up studies often extrapolate and reverify previously tested variables, and peer reviews analyze a study's conclusions, research and methods, to validate good conclusions. It's a kind of checks and balances system.
When making judgments about animals and animal behavior, humans have a particular tendency to conclude what their own view of reality tells them. A pioneering example of this occurred when a group of undergraduates was instructed to pass judgment on a group of rats that they were told were genetically superior to normal rats. Proving the experimenter expectancy effect, most students found these rats to be smarter than normal rats.