Fig 1.
A graphical illustration of how the psychometric function for body size estimation can be used to separate out a participant’s self-estimated BMI (indexed by the point of subjective equality, PSE) from their sensitivity to notice differences in body size (indexed by the just noticeable difference, JND).
On the left, participants A, B, and C might all have the same actual BMI of 25. However, participant A under-estimates their body size whilst participant C over-estimates their body size. On the right, participant A is more sensitive to notice change in body size than participant B, and therefore has a steeper psychometric function, with a smaller JND (redrawn from Groves et al., [22]).
Fig 2.
Examples of the CGI images used in the study.
Table 1.
Participant characteristics, n = 172.
Table 2.
Re-analysis of 2 laboratory studies and the current online study using three-quarter view stimuli only.
Table 3.
Pearson correlations between outcome and predictor variables.
Fig 3.
Scatter plots to illustrate the modelled outcomes from top half of Table 3.
Table 4.
Linear mixed effects model for estimated BMI comparing responses from three-quarter and front view avatar presentation.
Fig 4.
Scatter plots to illustrate the modelled outcomes from bottom half of Table 3.
Table 5.
Mean (SD) scores and intra-class correlation coefficients for self-estimated BMI derived from 3Q and front view stimuli at each timepoint.