Experiment 2 was a replication of Experiment 1 using auditory, instead of visual, stimuli. It was expected that the click train would shift participants’ psychophysical function to the left, manifesting as an overestimation of durations (e.g., Penton-Voak et al., 1996; Wearden et al., 1999). Some studies with individuals with schizophrenia have shown underestimation (e.g., Elvevåg et al., 2003), but others have not reported any difference between those with and without schizotypals (e.g., Lee et al., 2006).
Method
Participants and Stimuli
A new sample of 174 naïve participants (102 females, 42 males), mean age 25.7 (+8.6; range = 18-58) years, were recruited as described in Experimetn 1. The stimulus used for presenting the durations (in both training and test phrases) was a single 144Mhz tone. The click-train for the clicker conditions was a 144Mhz tone, that lasted for 10ms, with 190ms of silence, over 5000ms, which was perceived as a 5s click train. Schizotypy scores were determined using the O-LIFE questionnaire as in Experiment 1.
Design and Procedure
The design and procedure of Experiment 2 were the same as Experiment 1, apart from the fact that the stimuli were auditory, as opposed to visual. There were 78 participants in the lower schizotypy group (mean = 8.51 + 3.25; range 1-13), and 96 in the higher schizotypy group (mean = 19.22 + 3.51; range 14-30).
Results and Discussion
Mean percent of ‘long’ responses per condition appears in Figure 2. A three-factor mixed-mdeol ANOVA (clicker x duration x Schizotypy revealed a significant main effect of duration, F(6,1020)=1385.58, p<.0001, ηp2=.89[.000:.983]. There was no significant main effect of clicker, F(1,170)=1.33, p=.25, ηp2=.008[.000:.239], but a significant clicker x duration interaction, F(6,1020)=2.92, p=.008, ηp2=.017[.000:.109]. Pairwise comparisons (Table 2) revealed the clicker conditions led to higher percentage of ‘long’ responses at the 400ms duration. There was no significant main effect of Schizotypy, F(1,170)=1.52, p=.219, ηp2=.009[.000:.275], but there was a Schizotypy x duration interaction, F(6,170)=4.68, p<.001, ηp2=.027[.000:.164]. There were no other significant interactions [clicker x Schizotypy, F(1,170)=1.28, p=.26, ηp2=.007[.000:.242]; three-way, F(6,1020)=1.934, p=.072, ηp2=.011[.000:.075]).
The Schizotypy x duration interaction is illustrated in Figure 3. Pairwise comparisons (Table 3) confirmed that the higher schizotypy group had lower percentage of ‘long’ responses, at 200, 300 and 400ms durations, suggesting they were more likely to underestimate a duration. Although not significant, the higher schizotypal group tended to overestimate durations above the arithmetic mean.
Experiment 2 found that the effect of the clicker on duration judgments for auditory stimuli was smaller and more tenuous than for visual stimuli (Experiment 1). This is consistent with previous evidence of a smaller effect of click trains on increasing perceived durations of auditory stimuli (e.g., Penton-Voak et al., 1996; Wearden et al., 1999). Higher schizotypy showed a steeper gradient in the psychophysical function for auditory stimuli, with a tendency to underestimate durations below, and overestimate durations above, the arithmetic mean. This is consistent with results reported by Carroll et al. (2008) who noted that, compared to healthy controls, schizophrenic patients were more likely to underestimate auditory durations. This pattern of results may be explained by a lengthening memory distortion of the short and long anchors. That is, if the anchors are lengthened in long-term memory of individuals with higher schizotypy levels, then short stimuli are less likely to be judged as ‘long’, and longer stimuli are more likely to be judged as ‘long’, compared to low schizotypals. This finding is consistent with previous reports using the temporal generalisation task (Hopkins et al., 2024), where higher schizotypy is associated with a lengthening in memory for visual stimuli compared to low schizotypals. Carroll et al. (2009) also suggested that misremembered anchor durations could deviate duration judgements, which was also noted in the current report.
Between-Experiment Analysis
To assess differences between experiments, a four-factor mixed-model ANOVA with duration (200, 300, 400, 500, 600, 700, & 800ms) and clicker (absent x present) as within-subject factors, and experiment (1 x 2) and Schizotypy (lowerversus higher) as between subject-factors was conducted. This revealed significant main-effect of clicker, F(1,293)=27.175, p<.001, ηp2=.085[.000:.798], duration, F(6,1758)=2452.201, p<.001, ηp2=.893[.000:.984], and experiment, F(1, 293)=9.87, p=.002, ηp2=.033[.000:.590], with higher percentage of ‘long’ responses for auditory (Experiment 2) than visual (Experiment 1) stimuli. The main effect of Schizotypy was not significant, F<1, ηp2=.001[.000:.005]. There was a significant clicker x experiment interaction, F(1,293)=14.73, p<.001, ηp2=.048[.000:.682], a significant duration x experiment interaction, F(6,1758)=7.42, p<.001, ηp2=.025[.000:.153], and clicker x duration, F(6,1758)=10.90, p<.001, ηp2=.04[.000:.210]. Those two-way interactions were further qualified in a significant experiment x schizotypy x duration interaction, F(6,1758)=3.175, p=.004, ηp2=.011[.000:.072], and a clicker x duration x experiment interaction, F(6,1758)=4.218, p<.001, ηp2=.014[.000:.093]. The remaining interactions were not significant[1]. The clicker x duration x experiment interaction was expected given the observation of a clicker x duration interaction in Experiment 2 (auditory) but not Experiment 1 (visual).
Of particular interest was the significant experiment x schizotypy x duration interaction, illustrated in Figure 4. To explore the interaction, two separate mixed-model ANOVAs (experiment x duration) were conducted on percent of ‘long’ responses for lower and higher schizotypy participants. For lower schizotypy, there were significant main effects of duration, F(6,864)=966.56, p<.000, ηp2=.870[.700:.979], experiment, F(1,144)=8.85, p=.003, ηp2=.058[.000:.723], and a significant interaction, F(6,864)=4.76, p<.001, ηp2=.032[.000:.190]. Pairwise comparisons (Table 4) confirm that lower schizotypy participants overestimated auditory durations between 400-600ms, suggesting the classic effect that auditory stimuli are perceived longer than visual stimuli (e.g., Carroll et al., 2008; Wearden, 2016). For high schizotypalparticipants, there was a significant main effect of duration, F(6,894)=1640.767, p<.001, ηp2=.917[.800:.987], not of experiment, F(1, 149)=1.85, p=.176, ηp2=.012:[.000:345], but there was a significant interaction, F(6,894)=6.164, p<.001, ηp2=.040[.000:.227]. Pairwise comparisons (Table 4) confirmed that higher schizotypal participants significantly underestimated durations of auditory stimuli compared to visual stimuli at durations between 200-300ms. This underestimation switched to an overestimation of auditory compared to visual stimuli at durations betweenf 500-800ms.
[1] [Clicker Presence X Schizotypy, F(1, 293) = 1.57, p = .211.,ηp2 = .005, [.0000, .1866]; Duration X Schizotypy, F(6, 1758) = 1.52, p = .166, ηp2 = .005, [.0000, .0358]; and Schizotypy X Experiment, F(1, 293) = 1.89, p = .170ηp2 = .006, [.0000, .2158]; Clicker Presence, Schizotypy and Experiment, F(1, 293) = .009, p =.924, ηp2 .000, [.0000, .0013]; Clicker Presence, Duration and Schizotypy, F(6, 1758) = 1.834, p = .127, ηp2 = .006, [.0000, .0428]; Clicker Presence, Duration, Schizotypy and Experiment, F(6, 1758) = .745, p = .614, ηp2 = .003, [.0000, .0178].