Temporal preparation decreases perceptual latency: Evidence from a clock paradigm

Abstract

A clock paradigm was employed to assess whether temporal preparation decreases the time to detect the onset of a stimulus—that is, perceptual latency. In four experiments participants watched a revolving clock hand while listening to soft or loud target tones under high or low temporal preparation. At the end of each trial, participants reported the clock hand position at the onset of the target tone. The deviation of the reported clock hand position from the actual position indexed perceptual latency. As expected, perceptual latency decreased with target tone intensity. Most importantly, however, greater temporal preparation decreased perceptual latency in all four experiments, especially for soft tones, which supports rather directly the idea that temporal preparation diminishes the duration of perceptual processing.

Keywords:

• Temporal preparation
• Perceptual latency
• Warning signal
• Foreperiod
• Clock paradigm

We thank Sonja Cornelsen and Agnes Mercz for their assistance in data collection. We also thank Jeff Miller, Dirk Wentura, and two anonymous reviewers for helpful comments. This work was supported by the Deutsche Forschungsgemeinschaft (UL 116/10–1).

Notes

¹ Note that this prediction is comparable to a prior entry account (e.g., Shore, Spence, & Klein, 2001; Spence, Shore, & Klein, 2001) when one conceives of high temporal preparation as a state of increased attention at the expected time point of target tone presentation. According to the original formulation of the law of prior entry by Titchener (1908), “the stimulus for which we are predisposed requires less time than a like stimulus, for which we are unprepared, to produce its full conscious effect” (p. 251). Although this law is usually applied to temporal order judgements, presumably the same attentional mechanism underlies the perception of tones under high temporal preparation in the present experiments.

² In a constant FP design, the function relating RT to FP length is U-shaped. This function exhibits an initially sharp RT decrease up to about 200 ms FP length, followed by slow increase towards an asymptote at about 3,000 ms FP length (see Müller-Gethmann et al., 2003). We used FPs of 600 and 2,000 ms, which represent FPs that seem to produce a maximum FP effect on RT. It should be stressed that a generally lower temporal preparation level is associated with variable FPs (see Mattes & Ulrich, 1997), and therefore the FP effect might not be as strong as that with constant FPs, at least for these FP levels.

³ This avoids a number of possible confounds. Judgements about when a brief stimulus occurs can be influenced by the perceived time of its centre (“P-centre”) rather than its onset (Morton, Marcus, & Frankish, 1976). A brief target tone might also be perceived as forming a compound stimulus (which might likewise have a P-centre) with the warning signal.

⁴ We thank Dirk Wentura for bringing up this alternative explanation and for suggesting the catch trial experiment to exclude it.

⁵ We submitted the overall mean D of Experiments 1, 2, and 3 to an ANOVA with the between-subject factor experiment. The ANOVA revealed a significant effect of the factor experiment, F(2, 69) = 4.94, p = .010. Planned contrasts (Tukey test) showed that mean D differed between Experiments 3 and 1 (p = .047) and Experiments 3 and 2 (p = .012) while the means of Experiments 1 and 2 did not differ significantly (p = .860).

⁶ We submitted RT of Experiments 1 and 4 to an independent t test, which revealed RT of Experiment 4 to be longer than RT of Experiment 1, t(46) = 4.29, p < .001 (two-tailed).

⁷ We performed a correlational analysis on RT and D per participant for Experiment 1. The mean correlation across all participants and conditions was .21, which is significantly different from zero, F(1, 23) = 30.78, p < .001. Therefore RT and D share common variance, as one would expect and thus do not reflect indices associated with (completely) different information-processing streams.