Abstract: There are two competing views that explain the influence of repetition priming on temporal order perception. One perspective assumes an enhanced perceptual processing; the other suggests a decisional response bias. Burnham et al. (2006) found that repetition priming caused a primed word to be perceived as having occurred earlier, and that a decisional response bias caused subjects to guess the primed word as the correct response in a temporal order judgment task. In order to investigate further the influence of repetition priming on temporal order perception, three experiments with figure stimuli were conducted in the present study.
The target pair in all the experiments consisted of a pair of figure stimuli: a square and a diamond. The target pair was presented either above or below the center of the screen. One of the targets was preceded by a prime appearing at the center of the screen. The prime was a smaller replica of the primed target, and the side length of this prime was 2 cm. There were 18 participants (10 females; mean age: 21 years) in Experiment 1 and another 18 participants (eight females; 21 years) in Experiment 2. In one block, they made temporal order judgments by choosing which of the two shapes appeared first (1st instruction); in another block, they judged which appeared second (2nd instruction), in both Experiment 1 and Experiment 2. The two experiments were conducted using a 2 (prime: present versus absent) × 2 (instructions: 1st versus 2nd) × 9 (SOA separating the onset of the primed and unprimed figures: ±112 ms, ±84 ms, ±56 ms, ±28 ms, and 0 ms; positive numbers indicated the primed figure preceded the unprimed figure, whereas negative numbers indicated that the unprimed figure appeared first and 0 ms meant the primed figure and unprimed figure appeared simultaneously) within-subject design. The duration of the prime was 70 ms in Experiment 1 and 14 ms in Experiment 2. Seventeen volunteer participants (nine females; mean age: 20 years) took part in Experiment 3 and responded only with the 1st instruction. However, the maximum SOA was increased to 210 ms, and the interval between the prime and the first-presented target figure was increased to 196 ms in Experiment 3.
The results illustrate two main findings. First, there were significant main effects for instruction and priming—as well as a significant interaction between them—whereas the effect of the duration of the prime did not reach significance in the conditions of the unprimed target figure appearing first in Experiments 1 and 2. Second, the repetition priming effect on temporal order was reversed under the condition where the primed target figure appeared first, in Experiments 1, 2, and 3.
Taken together, the results of the three experiments show that (1) the influence of repetition priming on temporal order is due to both the enhanced perceptual processing of primed target figure and the decisional response bias, (2) the duration of the prime has no influence on the instruction effect, (3) there is a reversal of the repetition priming effect when the primed stimulus appears first—an effect that has not been observed in previous studies, and (4) a new hypothesis about dual processes and representation matching modulation was proposed to integratively explain the present research results

Key words: repetition priming, temporal order perception, hypothesis about dual processes and representation matching modulation