Purchase this article with an account.
Yu Wang, Yu Luo, Alejandra Echeverri, Jiaying Zhao; Visual and numerical representations of dynamic systems. Journal of Vision 2016;16(12):1106. doi: https://doi.org/10.1167/16.12.1106.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Most natural systems are dynamic, involving a set of relationships among measurable quantities, where mathematical models describe how the quantities evolve over time. For example, water flow in the river and the number of fish in a lake both follow fixed mathematical rules. Thus, the challenge for the visual system is to integrate multiple sources of quantities over time to represent the overall changes in the system. Here we examine a dynamic system where an inflow tube is connected to a tank where objects enter. The tank is also connected to an outflow tube where objects leave. In Experiment 1, there were two conditions. In the visual condition, participants viewed objects flowing into and out of the tank in each trial, and estimated the net change of objects in the tank. In the numerical condition, participants viewed a chart showing the inflow rate of objects, a chart showing the outflow rate, and a chart showing the number of objects in the tank. Based on the charts, they estimated the net change of objects. There were three types of trials: depletion (inflow less than outflow), equilibrium (inflow = outflow), and replenishment (inflow greater than outflow). We found that estimation error of the net change was reliably lower in the visual condition than in the numerical condition. Estimation errors were greater in depletion and replenishment trials than in equilibrium trials. This pattern of results remained the same when participants could view only the inflow and outflow information (Experiment 2), or only the tank information (Experiment 3). The current findings suggest that visual representation is more veridical than the representation derived from only numerical information. The state of equilibrium produces a more accurate representation than the state of change. Finally, the visual system can form an efficient representation of the dynamic system based on partial information.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only