PERCEPTUAL ORGANIZATION AND PSYCHOPHYSICAL REFINEMENT OF A

Pictorial rating scales may reduce dependence on language-specific verbal anchors, but their internal perceptual structure must be demonstrated before images can be treated as ordered scale levels. We evaluated thirteen pictorial stimuli derived from the GOAL Scale in 85 physically active adults. Each image was judged twice using free magnitude estimation and verbal-category scaling with ten difficulty-based descriptors. Magnitude estimates were analysed on the logarithmic scale as geometric-mean ratios, and category responses were represented on an ordered 1–10 continuum. The two methods ordered the images almost identically (Spearman ρ = 0.99) and separated 70 of 78 image pairs in the same direction; 15 of 20 planned geometric-mean ratios (11 after Holm correction) and 17 of 20 category contrasts were significant. Cross-method integration with equivalence testing identified two pairs of perceptual duplicates (4B–5B and 6B–7A), a proportional plateau spanning 6B–8A, and a convergent inversion of 4A below 3B. A chain analysis requiring each adjacent pair to be separated by both methods and by at least 70% of observers yielded eight stimuli: 1B, 2B, 4A, 5A, 6A, 6B, 8A, and 8B. These images form the proposed Pictorial Exertion Scale, 8 Levels (PES-8), with adjacent geometric-mean ratios of 1.19–1.56. The PES-8 is supported as a psychophysically refined ordinal pictorial structure with estimated, non-uniform group-level proportional spacing; equal intervals and individual-level ratio properties are not established.