Naivete refers to workers' lack of familiarity with measures, manipulations, and experimental tasks common in the social and behavioral sciences. Workers may range from completely naive to extremely familiar.
The “Naivete” Feature allows you to exclude the workers who are most likely to be familiar with research paradigms in the social and behavioral sciences. Because the most active workers complete a disproportionate number of HITs, they may be overly familiar with some research measures and manipulations and there may be instances where you may want to exclude these workers from your study.
When using the Naivete feature, you can choose which portion of the most active workers you want to exclude.
Background on Naivete
Although somewhere close to 100,000 workers complete HITs on MTurk each year, any one study is likely to draw from a much smaller pool of workers. This is because some workers are much more active than others. Research examining characteristics of the MTurk platform have shown that the most active 1,000 workers (1% of all workers) take about 21% of all HITs. In addition, the most active 10,000 workers take about 74% of all HITs. If you are interested in recruiting naive workers, you should exclude the most active participants from your study.
In some instances, there may be benefits to recruiting very active workers. Workers who have completed thousands of HITs and have a high approval rating can be trusted to follow experimental instructions and provide quality data. In addition, because experienced workers are committed to earning money on MTurk, they may be more willing to engage in long tasks, tasks that require open-ended writing, and longitudinal studies than less experienced workers.