SENSATION WEIGHTING IN PREFERENCE JUDGMENTS CREATES A GOODNESS-LEVEL DEPENDENT WORD-ORDER EFFECT
HellstrÃ¶m (2003) found a goodness-level dependent word-order effect (WOE) for preference judgments. However, HellstrÃ¶m used preference scales adapted to each stimulus pair, and goodness was rated for whole stimulus pairs. Therefore, participants might not have judged preference, and statistical testing of the stimulus weights was not possible. In the present study, 211 participants indicated within-pair preferences for 25 stimulus pairs by choosing one of six written (same for all stimuli) preference expressions (e.g., â€œapple I like more than pearâ€). Within-pair presentation order was reversed for half of the participants. Participants also rated each stimulusâ€™s goodness by choosing one of seven written expressions (e.g., â€œApple I generally likeâ€). Results replicated HellstrÃ¶mâ€™s results; there was a greater weight for the left stimulus and a positive correlation between WOE and goodness level. These results strengthen the evidence that the WOE is due to differential sensation weighting rather than, for instance, semantic congruity.