The psychophysics of society: Uncertain estimates of invisible entities

AbstractLarge-scale societies are impossible to perceive directly. Unsurprisingly, lay demographic estimates are wildly inaccurate. How should we interpret these errors? Most accounts assume these errors are evidence of topic-specific biases and prejudices. (e.g., “People overestimate immigration because immigrants threaten the status quo.”) But this glosses over the distortions that are introduced whenever underlying perceptions are translated into explicit numerical estimates. For instance, estimates are typically hedged, or ‘rescaled,’ toward an expected value — a perfectly rational strategy when information is uncertain. We show that uncertainty-based rescaling accounts for most error in individual demographic estimates. Residual errors were not even always in the same direction; populations that appeared to have been over-estimated (e.g., Asian-Americans) now appear to be under-estimated. The amount of rescaling engaged in by an individual was proportional to their uncertainty (about politics or about numbers). Perceptions of society are surprisingly good; the psychophysics of estimation gets in the way.


Return to previous page