A re-examination of “bias” in human randomness perception.

Abstract

Human randomness perception is commonly described as biased. This is because when generating random sequences humans tend to systematically under- and overrepresent certain subsequences relative to the number expected from an unbiased random process. In a purely theoretical analysis we have previously suggested that common misperceptions of randomness may actually reflect genuine aspects of the statistical environment, once cognitive constraints are taken into account which impact on how that environment is actually experienced (Hahn & Warren, Psychological Review, 2009). In the present study we undertake an empirical test of this account, comparing human-generated against unbiased process-generated binary sequences in two experiments. We suggest that comparing human and theoretically unbiased sequences using metrics reflecting the constraints imposed on human experience provides a more meaningful picture of lay people’s ability to perceive randomness. Finally, we propose a simple generative model of human random sequence generation inspired by the Hahn and Warren account. Taken together our results question the notion of bias in human randomness perception.

Inflation Forecasting: Rewarding Luck Rather Than Skill?

“…Therefore, their prediction errors cannot be completely removed. Misperception of randomness can be due to cognitive biases or not [5], but it does exist. Flipping a coin, the chance of heads three times in a row is 1 1 1 1 0.125 2 2 2 8 × × = = .…”

Order Through Disorder: The Characteristic Variability of Systems

“…Inherent bias prevents total randomness. Completely unbiased random changes in a sustainable complex system are impossible, by virtue of constraints imposed by the intrinsic organization in the system (Warren et al, 2018). Models of kinetic reaction-diffusion equations are used to describe microscopic biological populations dynamics (Cordier, 2016).…”

Generating randomness: making the most out of disordering a false order into a real one

“…In addition, common misperceptions of randomness reflect genuine aspects of the statistical environment. When cognitive constraints are taken into account they impact how that environment is experienced [108]. When people consider a series of random binary events, such as flipping a coin, they tend to erroneously underrate the probability of sequences with less internal structure [109].…”

“…Comparing human-generated sequences to unbiased process-generated binary sequences demonstrated that the constraints imposed on human experience provide a more meaningful picture of our ability to perceive randomness. A model of human random-sequence generation was thus proposed [108]. Binary sequences consist of a mixture of alternation and repetition.…”