Time-order Effects for Aesthetic Preference

Participants compared successive color patterns (Exp. 1) or jingles (Exp. 2), selecting the preferred one. Results were well described by Hellström’s sensation-weighting model, with a greater weight for the second stimulus than for the first. Mean time-order errors were negative, which can be explained as a consequence of this stimulus weighting and of a reference level for aesthetic attractiveness, lower than that of the average stimulus; this level seems to reflect the low aesthetic value of the visual or auditory stimulus background. Fechner (1860) was the first to notice the Zeitfehler, that is, time-order error (TOE) in comparisons of successive stimuli. Also, he introduced experimental aesthetics (Fechner, 1876). Apparently Fechner did not combine those two subjects. However, as we shall see, the results of this combination should have been of interest to him. Koh (1967) had different participants compare the pleasantness of paired successive musical excerpts, which had been rated for pleasantness by similar participants. Out of two equally pleasant excerpts participants preferred the second (a positive TOE); out of two equally unpleasant excerpts, the first (a negative TOE). On average, there was a slight negative TOE. These effects resemble findings for, for instance, lifted weights (Hellström, 2000). Hellström (1979) studied in detail the effects of the stimulus magnitude on the TOE for loudness; this led to the explanation of the TOE as a side effect of sensation-weighting: the subjective difference, d, between two compared stimuli is not the simple difference between their magnitudes; instead it is the difference between two weighted compounds, one for each stimulus, where the stimulus and a reference level (ReL) enter with weights s and (1-s): d = k {[s1 ψ1 + (1 s1) ψr1] [s2 ψ2 + (1 s2) ψr2]}, (1) where d is the scaled subjective difference, k a scale constant, ψ1 and ψ2 the sensation magnitudes of the stimuli, s1 and s2 weighting coefficients, and ψr1 and ψr2 the subjective magnitudes of the ReLs (possibly different for the two stimuli). The weighting-in of the ReLs substitutes averaged magnitudes for stimulus magnitudes that are missing or noisy due to, for instance, memory loss; this improves stimulus discriminability (Hellström, 1985, 1989). The TOE can be defined in subjective units as the value of d in a pair of equal stimuli (Hellström, 1985). Setting ψ1 = ψ2 = ψ and simplifying by assuming ψr1 = ψr2 = ψr yields (TOE =) d = k (s1 s2) (ψ ψr). (2) In earlier research (with ISIs of several seconds) the TOE was generally negative, more so the higher the stimulus magnitude level in the series, and positive only for stimuli of low magnitude. Eq. 2 explains this as a consequence of the weight relation s1 < s2 and the stimulus magnitudes in the pair being below the ReL (ψ ψr < 0) (cf. Hellström, 2000). For brief stimuli and ISIs Hellström (1979, 1992) found the opposite effect of stimulus magnitude, and interpreted this as being due to the weight relation s1 > s2. In Koh's (1967) experiment the stimuli were of rather long duration, and the ISI was kept constant at 6 s. No account was taken of individual differences. The present study aimed at studying the phenomenon by using the SW model and individual scaling of the stimulus magnitudes, and investigating the possible effect of the ISI for relatively brief stimuli. Experiment 1: Color patterns