1MODALITY & INTENSITY EFFECTS ON TIME REPRODUCTION
Supplementary Materials
Combined effects of motor response, sensory modality,
and stimulus intensity on temporal reproduction
Allegra Indraccolo1,2,*,ŧ, Charles Spence1, Argiro Vatakis3, & Vanessa Harrar1,4, ŧ
1Department of Experimental Psychology, University of Oxford, GB
2Department of Human Science, UniversitàEuropea di Roma, Italy
3Cognitive Systems Research Institute, Athens, Greece
4School of Optometry, Université de Montréal, Canada
ŧA.I. and V.H. contributed equally to this work
(*) Correspondence should be addressed to:
Allegra Indraccolo
Department of Human Science,
Via degli Aldobrandeschi, 190 00163 Roma, Italy
Email:
- ANOVA results
- RT analysis
The repeated-measures analyses of variance (RM-ANOVAs) on the reaction time (RT) data revealed a significant main effect of Intensity [F(1,19)=21.499; p<.001; η2p=.531] and Modality [F(1,19)=100.698; p<.001; η2p=.841], but no significant interaction [F(1,19)=1.369; p=.257]. The difference between the RTs to low and high intensity stimuli was larger for auditory (difference=25 ms; SE=4.24) than for visual stimuli (difference=19 ms; SE=6.52), but both differences were significant (tauditory(19)=5.971, p<.001;tvisual(19)=2.918, p=.009; see Figure S1). An ANOVA was also conducted on the variability of RTs (standard deviation) but no significant effects were found.
1.2.Temporal Reproduction analysis
In order to determine the effects of stimulus modality, intensity, and sample duration on the temporal reproduction task, three RM-ANOVAs were conducted on the three performance measures: median reproduced durations, accuracy (the duration of the comparison stimulus divided by the duration of the sample stimulus), and coefficient of variation (CV; standard deviation divided by the median reproduced duration calculated separately for each condition interval). Each of these RM-ANOVAs had three factors: Modality (two-levels: auditory, visual), Duration (three-levels: 600, 800, and 1000 ms), and Intensity (four-levels corresponding to the four intensity combinations).
Median reproduced duration (see Figure S2): There were significant main effects of Modality [F(1,19)= 13.170; p=.002, η2p=.409], Duration [F(2,18)=159.395; p<.001, η2p=.947], and Intensity [F(3,17)=9.627; p=.001, η2p=.629], as well as significant interactions between Modality and Duration [F(2,18)=12.429; p<.001, η2p=.580], and Intensity and Duration [F(6,14)=7.986; p=.001, η2p=.774]. Most importantly, there was also a significant 3-way interaction [F(6,14)=4.356; p=.011, η2p=.651]. This 3-way interaction was appropriately followed up with three 2-way RM-ANOVAs, one for each duration.
600 ms duration: There was a main effect of Intensity [F(3,17)=10.510; p.001; η2p=.650] and Modality [F(1,19)=38.554; p<.001; η2p=.670], but no interaction [F<1].
800ms duration: There was a significant main effects of Intensity [F(3,17)=10.095; p.001; η2p=.640] and Modality [F(1,19)=9.260; p=.007; η2p=.328], as well as a significant interaction [F(3,17)=4.750; p=.014; η2p=.456].
1000ms duration: There was a significant main effects of Intensity [F(3,17)=8.984; p=.001; η2p=.631]. The condition “sample high/comparison low” was reproduced with the longest duration for both auditory and visual stimuli.
Precision: The analysis for the CV revealed a significant main effect of Modality [F(1,19)=6.739; p=.018, η2p=.262] and a significant main effect of Duration [F(2,18)=17.465; p<.001, η2p=.660], since participants’ responses were more variable in the 600 ms duration condition thanin either of the longer durations. There was also significant interaction between Modality and Intensity [F(3,17)=4.539; p=.016; η2p= .445]. No other significant effects were observed.
Accuracy: There were significant main effects of Modality [F(1,18)=16.459; p=.001, η2p=.478], Intensity [F(3,16)=8.776; p<.001, η2p=.622], and Duration [F(2,17)=59.883; p<.001, η2p=.876], and significant interactions between Modality and Duration [F(2,17)=17.827; p<.001, η2p=.677], Intensity and Duration [F(6,13)=8.126; p=.001, η2p=.789], but no significant three-way interaction (see Figure S3).
1.3.Discussion
The results, for the most part, simply replicate previous studies by demonstrating effects of intensity and modality on duration reproduction. However, most previous studies have looked at the effect of modality or intensity in isolation. Our results demonstrate an interaction between these factors, which might help explain previous differences in the literature (see also Walker & Scott, 1981). Further, the interaction of these factors with duration suggests multiplicative effects of intensity and modality on duration perception (see review of additive versus multiplicative effects in time research in Gil & Droit-Volet, 2011). The complex interaction effects between duration, intensity, and modality suggests that the results obtained by investigating these factors in isolation cannot be generalisedeasily.
The fact that different studies have reported different effects of stimulusmodality, might be attributable to the different methodologies used for stimulus presentation (see above Table 1). For example, how conditions are blocked can affect “memory mixing” (Szelag et al., 2002). When exposed to a stimulus of some duration, we might create a duration representation that is a kind of average of all previous durations (including visual and auditory) in order to estimate the interval on the current trial (Penney, 2003). Since we presented auditory and visual stimuli in separate blocks of trials, effects of modality are unlikely to have been affected by memory mixing. It would be interesting for future research to compare the results obtained here, with those obtained using a different temporal task, and at different stimulus intensities (cf. Gil & Droit-Volet, 2011).
Figure S1. Reaction times (RTs) as a function of modality and stimulus intensity. Low intensity refers to the dark-grey square on a black background (for the visual stimuli) and to the 40dB white noise (for the auditory stimuli); high intensity refers to the white square and the 70dB noise. The standard errors are shown with error bars. (*** p<.001; ** p<.01)
Figure S2. Reproduced durations for auditory and visual stimuli, at multiple durations and stimulus intensities. Visual: low and high refers to the dark-grey or white squares on a black background, respectively. Auditory: low or high refers to a 40dB or 70dB white noise stimulus.Error bars show standard errors.
Figure S3. Accuracy for reproduced durations for auditory and visual stimuli, at multiple durations and stimulus intensities.Accuracy calculated bydividingthemedian reproduced duration by the sample duration. Accuracy scores greater than one demonstrate reproduced durations that are longer than the sample durations; less than one demonstratestimuli are reproduced with durations that are shorter than the sample duration. Visuallow and visual high refers to the dark-grey or white stimulus square that were presented on a black background, respectively. Auditory low or high refers to a 40dB or 70dB white noise stimulus.Error bars show standard errors.