Verbal estimation of the magnitude of time, number, and length

Abstract

AbstractPerformance similarities on tasks requiring the processing of different domains of magnitude (e.g. time, numerosity, and length) have led to the suggestion that humans possess a common processing system for all domains of magnitude (Bueti and Walsh in Philos Trans R Soc B 364:1831–1840, 2009). In light of this, the current study examined whether Wearden’s (Timing Time Percept 3:223–245, 2015) model of the verbal estimation of duration could be applied to verbal estimates of numerosity and length. Students (n = 23) verbally estimated the duration, number, or physical length of items presented in visual displays. Analysis of the mean verbal estimates indicated the data were typical of that found in other studies. Analysis of the frequency of individual verbal estimates produced suggested that the verbal responses were highly quantized for duration and length: that is, only a small number of estimates were used. Responses were also quantized for number but to a lesser degree. The data were modelled using Wearden’s (2015) account of verbal estimation performance, which simulates quantization effects, and good fits could be obtained providing that stimulus durations were scaled as proportions (0.75, 1.06, and 0.92 for duration, number, and length, respectively) of their real magnitudes. The results suggest that despite previous reports of similarities in the processing of magnitude, there appear to be differences in the way in which the underlying representations of the magnitudes are scaled and then transformed into verbal outputs.