Open Collections

Interference in the perception of correlation in two population scatterplots

University of British Columbia

Rensink & Baldridge (2010) first operationalized a methodology to study perceptual performance for single population scatterplots. Rensink (2014) emphasized that this methodology could be extended to understand the perception of more complex displays. In this paper, the original methodology is extended to examine the perception of two-population scatterplots, containing target and distractor populations. In three experiments, we investigated the nature of performance for a discrimination task used to measure viewer precision (defined by parameter k in our analysis) and accuracy (defined by parameter b in our analysis). The results show that perception for two-population scatterplots is non-trivially different from the perception of single population scatterplots. Namely, there is a significant degree of interference for selecting and discriminating the target populations in each display. This interference occurs due when the target and distractor populations are featurally distinct from one another, violating assumptions from Feature Integration Theory and Guided Search. Discrimination performance also degrades as a function of the density of the distracting population. Findings from this work not only help solidify a methodology to study the mechanisms underlying ensemble selection and inhibition in the presence of multiple populations, but also guide design choices for the creation of more complex scatterplot displays.

Text

2016-08-25

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/58967

Psychology

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/