Avoidance of Repetition in Produced Sequences and the Concept of Randomness

Abstract

Avoidance of repetition in randomized response sequences has repeatedly been attributed to biased concepts of randomness. The few studies directed toward testing of this hypothesis did not present supporting evidence. However, the manner in which concepts of randomness have been measured in previous studies, namely, selection of one random series from a number of series varying in percentage of repetitions, may be inadequate. It is argued that judgements in these studies might easily be affected by bottom-up perceptual processes. A different measure, probability assignment, is used to measure the concept of randomness. The results show that the concept measured in this way is not biased towards nonrepetition and that there are no correlations with avoidance of repetition in randomization of responses. This is considered evidence against the explanation of avoidance of repetition in production of sequences by biased concepts of randomness.