About hemispheric differences in the processing of temporal intervals

Brain and Cognition 58 (2005) 125–132 www.elsevier.com/locate/b&c

About hemispheric diVerences in the processing of temporal intervals夽 Simon Grondin¤, Christine Girard École de psychologie, Université Laval, Que., Canada G1K 7P4 Accepted 1 November 2004 Available online 8 December 2004

Abstract The purpose of the present study was to identify diVerences between cerebral hemispheres for processing temporal intervals ranging from .9 to 1.4 s. The intervals to be judged were marked by series of brief visual signals located in the left or the right visual Weld. Series of three (two standards and one comparison) or Wve intervals (four standards and one comparison), marked by sequences of 4 or 6 signals, were compared. While discrimination, as estimated by d⬘, was signiWcantly better in the 4-standard than in the 2-standard condition when stimuli were presented in the left visual Weld (LVF), this number-of-standard eVect on discrimination varied with the diYculty levels when the signals were presented in the LVF. Moreover, the discrimination levels were constant for the diVerent base durations with stimuli presented in the LVF, but not with stimuli presented in the right visual Weld. This article discusses the implication of these Wndings for the study of hemispheric dominance for temporal processing and for a single-clock hypothesis.  2004 Elsevier Inc. All rights reserved.

1. Introduction Whether in speech or music, or in the simple fact of waiting for an event to occur, processing time is required. The purpose of the present experiment was to further elucidate the mechanisms involved in the processing of temporal information. It is proposed to study these temporal mechanisms with a classical timing task, the discrimination of short time intervals. More speciWcally, it is the discrimination of intervals presented in sequences that is going to be used.

夽 This research was made possible by a research grant awarded to SG by the Natural Sciences and Engineering Council of Canada (NSERC) and by a scholarship granted to C.G. by NSERC. We express our gratitude to Marie-Eve Roussel, who provided precious help at various stages of the project, and Howard Zelaznik, Michael Nicholls and two anonymous reviewers for their relevant comments on a previous draft of this article. * Corresponding author. Fax: +1 418 656 3646. E-mail address: [email protected] (S. Grondin).

0278-2626/$ - see front matter  2004 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2004.11.001

There is a wide variety of models developed to account for temporal processing. However, many contemporary researchers in the Weld of time perception recognize the scientiWc usefulness of the single, internalclock hypothesis for explaining temporal judgments (see Grondin, 2001a, 2001b; Helfrich, 2003; or Meck, 1996). Such a central clock is most often described as a pacemaker–counter device (Killeen & Weiss, 1987). The pacemaker emits pulses that are accumulated in a counter, and the number of pulses counted determines the perceived length of an interval. The eYciency of this accumulation is often attributed to attentional mechanisms, more attention being paid to time resulting in an increase of pulse accumulation. The pacemaker–counter view is often embedded within an information-processing perspective and is often referred to as the Scalar expectancy theory (SET: Gibbon, Church, & Meck, 1984). Within this view, the variability of temporal judgments are reported to depend not only on the processes located at the clock level, but also on those associated with memory and decision.

126

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

A central feature of most models based on a pacemaker–counter device is that the variability in a series of time judgments should increase linearly as a function of time. This property is referred to in psychophysics as the Weber law but in the Weld of timing, it is often called the scalar property: when reported on the same relative scale, the variability obtained for diVerent duration ranges should be the same. Indeed, not only is the amount of variability the same but the functions should superimpose. However, the ideas that there is one central timekeeping mechanism, and that this mechanism would have a scalar property could be challenged by two independent types of results. Both are issued from the rhythm literature, more speciWcally from experiments where sequences of intervals are presented. Challenges come in one case from the possibility that each cerebral hemisphere oVers a speciWc way for processing temporal information; and, in the second case, from a violation of the scalar property (see Ferrandez et al., 2003; Meck, 2003; Meck & Malapani, 2004; Pfeuty, Ragot, & Pouthas, 2003). The involvement and eYciency of each cerebral hemisphere for processing temporal information remains an open question. It is known that the left cerebral hemisphere has an advantage over the right in temporal processing when temporal tasks involve judgments relative to the temporal order of two sensory events (Efron, 1990; Nicholls, 1996). However, it is more diYcult to draw the same conclusion when temporal processing involves explicit judgments about time, i.e., about the relative length of two time intervals. Some recent Wndings, mostly found in the neuroscience or neuropsychology literature, argue for the existence of a right-hemisphere advantage, or for a speciWc role of the right hemisphere, for processing temporal information (Funnell, Corballis, & Gazzaniga, 2003; Handy, Gazzaniga, & Ivry, 2003; Harrington, Haaland, & Knight, 1998; Kagerer, Wittmann, Szelag, & Steinbüchel, 2002; Milner, 1962; Monfort, Pouthas, & Ragot, 2000; Pouthas, Garnero, Ferrandez, & Renault, 2000; Smith, Taylor, Lidzba, & Rubia, 2003). Other hypotheses have also been considered. For instance, Polzella, DaPolito, and Hinsman (1977) reported some evidence that a timer for very brief intervals might be located in the left cerebral hemisphere, without excluding the possibility that another mode for processing time might be used by the right hemisphere. Most relevant for the present study is the method used by Ben-Dov and Carmon (1984). These authors presented a series of brief Xashes marking 200- or 400ms intervals to both cerebral hemispheres, and a second sequence, either the same as or diVerent from the Wrst, to only one hemisphere. The participants had to say whether the sequences were the same or diVerent. Based on reaction time data and an analysis of errors, Ben-Dov and Carmon reported that the relative eYciency of the

two cerebral hemispheres depended on the number of intervals presented. Cerebral dominance shifted from the left hemisphere to the right as the number of intervals increased (in this case, from 1 to 4). The hypothesis that cerebral dominance depends on the number of intervals presented was also shown to be viable for temporal processing in the auditory mode (Alpherts et al., 2002). The second challenge is related to the scalar property. There are multiple reports in the rhythm literature involving the duration discrimination of interval sequences, but the conclusions concerning the variability to time ratio are mixed. These reports involve in most cases series of auditory signals. Some of these reports support the scalar property, i.e., the Weber’s law model (Halpern & Darwin, 1982; see Friberg & Sundberg, 1995). However, there are some cases where the scalar property does not hold. For instance, for brief intervals (<400 ms), the diVerence threshold remains constant (Schulze, 1989; ten Hoopen et al., 1995). In Drake and Botte (1993), the scalar property holds for intervals ranging from 300 to 800 ms but when intervals to be discriminated last 1.5 instead of 1 s, the variability to time ratio is much higher. 1.1. The present experiment The purpose of the present experiment was to verify, with a range of durations (from .9 to 1.4 s) diVerent from the one used by Ben-Dov and Carmon (1984), if the location of the visual source—in the left vs. right visual Welds—inXuences the discrimination of intervals marked by sequences of brief signals. Based on these authors’ Wndings, this inXuence should depend on the length of the sequence, i.e., on the number of intervals deWned by the series of signals. This potential interaction eVect between the number of intervals presented and the location of signals will be tested with diVerent diYculty levels. Moreover, the experiment provides an occasion to address another fundamental issue about temporal processing: will the discrimination levels at various base durations remain constant as would predict SET? Considering the Wndings of Drake and Botte (1993) and the range of duration under investigation, the discrimination level might vary as a function of time rather than staying constant. The present experiment addresses this issue and will test if the answer to the question depends or not on the location of the visual marker in the left vs. right visual Weld, and on the length of the sequences.

2. Method 2.1. Participants Twelve Laval University students, 9 females and 3 males, aged 20–32 (mean D 24.5) and right-handers

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

(Bryden, 1977), took part in the experiment. They received CAN$60 for their participation. 2.2. Apparatus and stimuli The intervals to be discriminated were marked by 10 ms visual signals (markers) produced by a circular, red-light-emitting diode (LED: Radio-Shack #276-088) placed 25° to the left or the right of a central Wxation point located at about 1 m from the participant.1 The LEDs subtended a visual angle of about .57°. Each observer was seated in a chair in a dimly lit room and asked to indicate whether the last interval of each sequence of intervals was “short” or “long” by pressing the left or the right button, respectively (whether observer used both hands or not was not recorded). All other aspects of the experiment were controlled by a Zenith microcomputer.

127

The experiment involved 12 sessions, one for each of 12 experimental conditions (3 base durations, 2 locations, and 2 standard intervals). Each session lasted between 23 and 43 min and was divided into four blocks of 60 trials, with a 20 s break between each block. Within each block, there were 10 repetitions of each of the six intervals (one short and one long for each of the three levels of diYculty), presented randomly but equiprobably. There were 2 s between each trial, i.e., between the response and the beginning of the next sequence of signals. The experiment did not involve any feedback. The 12 participants were assigned to three groups of four participants. Each group was then assigned to one of three permutation orders so as to balance the three standard durations (900–1125–1406, 1406–900–1125, and 1125–1406–900 ms). One participant in each group was assigned to one of four possible orders resulting from the combination of two visual-Weld conditions and two number-of-standards conditions.

2.3. Procedure There were four independent variables of interest in the experiment: the base duration, the location of the visual signals, the level of diYculty, and the number of standard intervals presented. There were three base durations: 900, 1125 (i.e., 900 £ 1.25), and 1406 ms (i.e., 1125 £ 1.25). The locations of the visual signals marking intervals were the left or the right visual Weld (LVF– RVF). There were three levels of diYculty: plus (long) or minus (short) 5.6, 16.7, or 27.7% of the base duration value, approximately. In other words, for the 900 ms base duration, short and long intervals lasted, respectively, for the three levels of diYculty, 850 and 950 ms, 750 and 1050 ms, and 650 and 1150 ms. For the 1125 ms base duration, these values were multiplied by 1.25: 1063 and 1188 ms, 938 and 1313 ms, and 813 and 1438 ms. And for the next base duration, 1406 ms, the values were once again multiplied by 1.25: 1328 and 1484 ms, 1172 and 1640 ms, and 1016 and 1796 ms. As regard the number of standard intervals presented prior to delivery of the short or of the long interval, there were two contexts: 2 or 4 standard intervals were presented. For instance, in the 2-standard condition, there were three consecutive intervals in the form of two standards followed by a short or by a long interval. The three intervals were marked by a series of 4, 10 ms Xashes from the same LED. The Xashes marked the boundaries of the intervals (e.g., 900 ms between Xashes when standards D 900 ms).

1 To verify if eye movements prevented our laterality manipulation, we asked two people, after the experiment, to perform 60 discrimination trials in the 4-standard condition at 1406 ms, in each of the visualWeld conditions. We monitored their eye movements with a Canon camera (GL1-3CCD), and veriWed each trial with a Sony VCR (GV-D 1000). No observable eye movements were detected.

3. Results Fig. 1 provides a Wrst illustration of the results of the experiment. The results are presented in such a way that the data from the three base duration conditions are placed on a common scale (ratio of the Short or Long interval to the Base duration). According to SET, the obtained functions should superimpose, which is apparently the case in the LVF condition. However, in the RVF condition, important distances between the data points for a given ratio are frequently observed. The statistical analysis of data are based on parameters of the Signal Detection Theory (SDT). There are two dependent variables of interest, sensitivity and the decisional criterion, estimated with d⬘ and c. For each diYculty level in each experimental condition, d⬘ and c were computed on the basis of two assumptions: the distributions for the short and long intervals were normal, and had equal variances. Indeed, the short and long distributions for each diYculty level are the noise and signal + noise distributions in the SDT. In the present experiment, a hit was responding “long” when the last interval of the sequence was long, and a false alarm was responding “long” when the last interval was short. d⬘ and c were computed on the basis of the presentation of 40 short and 40 long intervals in each condition, using the computer program reported by Macmillan and Creelman (1991, Appendix 6). Table 1 shows the mean d⬘ in each experimental condition. In general, the results show that d⬘ values varied as a function of the diYculty level. More importantly, there was not much diVerence between the 2- and 4-standard presentations when the visual signals were presented in the RVF, but there were quite systematic diVerences between these conditions when the stimuli were presented in the LVF.

128

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

Fig. 1. Probability of responding correctly, based on pooled data of 12 participants, as a function of the ratio of the short or long interval to the base duration in each experimental condition. The three sets of two curves on each panel correspond to the three base durations.

A 2 £ 2 £ 3 £ 3 ANOVA on d⬘ conWrmed that the diYculty level eVect is signiWcant, F (2, 22) D 209.89, p < .01. The analysis also revealed that the number-ofstandard eVect is signiWcant, F (1, 11) D 5.14, p < .05, the number-of-standard by location interaction is marginally signiWcant, F (1, 11) D 3.37, p D .094 and, most importantly, the number-of-standard by location by diYculty level is signiWcant, F (1.767, 19.438) D 6.15, p < .01. Also very important is the fact that the location by base duration interaction is marginally signiWcant, F (1.610, 17.707) D 3.34, p D .068. None of the other main or interaction eVect was signiWcant. Because the location eVect was involved in the signiWcant or nearly signiWcant interaction eVects, we conducted a 2 £ 3 £ 3 ANOVA on each location condition. For the right-visual-Weld condition, the analysis revealed, in addition to the signiWcant diYculty level eVect (p < .01), a marginally signiWcant base duration eVect F (1.741, 19.153) D 3.32, p D .055 (see Fig. 2), and a signiWcant number-of-standard by diYculty level inter-

action F (1.767, 19.441) D 5.27, p < .05. While a .289 increase is observed for d⬘ in the most diYcult condition when using 4 instead of 2 standards, .251 and .127 decreases are rather observed in the other diYculty levels. None of the other eVects was signiWcant (note the p D .86 in the number-of-standard condition). For the left-visual-Weld condition, the analysis revealed, in addition to the signiWcant diYculty level eVect (p < .01), a signiWcant number-of-standard eVect F (1, 11) D 9.71, p < .01 (see Fig. 3), but no base duration eVect (p D .347) and only a marginally signiWcant number-of-standard by diYculty level interaction F (1.913, 21.045) D 2.895, p D .079. In all diYculty level conditions, the d⬘ values increased when 4 instead of 2 standards were used: + .718, .448, and .170 from the least to the most diYcult levels, respectively. None of the other eVects was signiWcant. Finally, a 2 £ 2 £ 3 £ 3 ANOVA was conducted on parameter c. The analysis revealed that none of the eVect is signiWcant, except the diYcult level F (1.180, 12.985)

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

129

Table 1 Mean d⬘ (and standard error) for each experimental condition Base duration (ms)

DiYculty level

2-Standards

4-Standards

1 2 3

4.26 (.35) 2.55 (.25) .66 (.15)

3.96 (.28) 2.76 (.22) 1.12 (.15)

1125

1 2 3

4.89 (.37) 3.19 (.34) 1.02 (.14)

4.48 (.37) 2.87 (.28) 1.01 (.13)

1406

1 2 3

4.58 (.39) 3.07 (.25) .58 (.29)

4.54 (.46) 2.88 (.32) 1.02 (.16)

1 2 3

4.02 (.36) 2.55 (.25) .86 (.15)

5.07 (.39) 3.10 (.32) 1.36 (.31)

1125

1 2 3

3.81 (.40) 2.43 (.27) .82 (.10)

4.69 (.40) 3.02 (.31) .99 (.13)

1406

1 2 3

4.27 (.43) 2.99 (.28) .99 (.12)

4.50 (.31) 3.16 (.28) .83 (.12)

Right visual Weld 900

Left visual Weld 900

Fig. 3. Mean d⬘ values for each number-of-standard condition for leftand right-visual-Weld conditions. Bars are standard errors.

the number of intervals presented. When signals are presented in the left visual Weld, presenting 4 instead of 2 standard intervals improves discrimination. However, when signals are presented in the right visual Weld, the number-of-standard eVect depends on the level of diYculty of the task. Moreover, the results show that the overall discrimination level remained the same over the three duration ranges under investigation in the present experiment when signals are located in the LVF, but it is not possible to reach such a conclusion when signals are located in the RVF. 4.1. Number of standards

Fig. 2. Mean d⬘ values as a function of base duration for left- and right-visual-Weld conditions. Bars are standard errors.

D 4.68, p < .05. In the two easiest conditions, participants responded “short” more often (mean c D .294 and .160) than in the most diYcult condition (mean c D .056). Note that the base duration by diYculty level interaction is marginally signiWcant, F (2.542, 27.967) D 2.57, p D .082.

4. Discussion The results of the experiment show that the location of the signals in the right vs. the left visual Weld exert a signiWcant inXuence on the processing of temporal information. More speciWcally, this inXuence depends on

The present data are to some extent consistent with those of Ben-Dov and Carmon (1984) who observed, with briefer intervals and a slightly diVerent method, that the relative eYciency of each cerebral hemisphere depends on the length of a sequence of signals marking time. In the experiment, the location of the visual signal marking sequences of intervals inXuences performance, and the exact inXuence depends on the number of intervals presented in the sequence. It is clear that beneWts are gained from presenting more standards when signals come from the LVF. What is new here is that when signals are presented in the RVF, the relative beneWts depend on the level of diYculty of the task. Cases where using 4 instead of 2 standards improve discrimination are consistent with most previous Wndings on this issue (Drake & Botte, 1993; Ivry & Hazeltine, 1995; McAuley & Kidd, 1998), although such a conclusion was actually unclear in the cases of intervals marked, as in the present experiment, by a continuous sequence of brief visual signals (Grondin, 2001b, Experiment 3). For explaining why increasing the number of standards results in better discrimination, one can adopt

130

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

two diVerent perspectives: one based on measuring intervals, the other on predictability/entrainment. First, as noted in the introduction, beyond the variability produced by the clock component, variability in the context of temporal processing might also depend on the eYciency of a memory component. The multiplelook hypothesis adopted by Drake and Botte (1993) is consistent with this view: Presenting more standards allows the development of a more precise memory trace of the mean interval duration. The second perspective, the dynamic attending theory originally proposed by Jones and Boltz (1989), emphasizes a distinction between two types of temporal process. In one, called future-oriented attending, the environment is said to be structured with temporal events which are coherent to some degree. The occurrence of physical regularities within the Xow of these events makes the occurrence of future events predictable (see also Barnes & Jones, 2000; or Large & Jones, 1999). Without such regularities, an observer is forced to use some sort of internal strategy, referred to by Jones and Boltz as an analytic—attending mode of processing. The Wrst process, related to the expectation of a future signal marking an eventual interval, would come into play in the present experiment, only if a suYcient number of signals—4 standards but not 2—were presented. In the absence of a suYcient number of sensory events for generating expectancies, temporal processing may rely on the alternative mode of processing. While the analytic mode proposed by Jones and Boltz might be involving a interval-based timer such as the pacemaker–counter device described above, the future-oriented mode would rather be another type of clock that might be represented by some kind of self-sustaining, entrainable, oscillator, which may have an adaptable period (Large & Jones, 1999; McAuley & Jones, 2003). None of these perspectives explain why the 4 vs. 2 standards diVerence applies systematically when signals are located in the LVF, but not when located in the RVF. Indeed, any explanation would remain tentative. From a hemispheric viewpoint, it is reasonable to assume that two diVerent ways of processing temporal information exist. The literature proposes several large dichotomic categories for comparing hemispheric specialization (Ivry & Robertson, 1998). One such category relates to the verbal vs. non-verbal issue. In the present experiment, all conditions involved non-verbal, low-level processing. An analytic vs. holistic dichotomy seems to provide a more adequate explanation, the left cerebral hemisphere being associated with the former, and the right one with the latter (Bradshaw & Nettleton, 1981). While a global processing is characterized by a synthetic, global way of processing information (à la Gestalt), the analytic mode is based on logical, sequential, propositional, processing. This is actually the interpretation that Ben-Dov and Carmon (1984) gave to their results.

According to this dichotomy, the analytic processing (left hemisphere) would be the preferred mode for shorter sequences (for instance, only two standards), but for longer sequences, the holistic processing (right hemisphere) would be the preferred mode. However, this explanation reaches important limitations given that our results do not show that there is a LVF advantage over the RVF condition when four standards are used. They rather show that the right hemisphere (LVF) gains beneWts more systematically than the left one (RVF) when passing from a short- to a long-sequence condition. 4.2. Base duration The present experiment on hemispheric diVerences also provides new data to take into account when addressing the single-timer hypothesis from a multiple base duration perspective. Because the diVerence between the short and long intervals for the diVerent base durations increased proportionally in the present experiment, the discrimination levels were expected to remain the same. Fig. 1 shows that when plotted on the basis of the same relative scale, the functions relating correct responses and time tend to superimpose in the LVF condition, but not in the RVF condition. Indeed, there was a near signiWcant duration eVect in the RVF condition, but clearly no such eVect in the LVF condition. Strictly speaking, this could be argued to reXect two diVerent timing mechanisms. The hypothesis that there is more than a central clock was recently supported by new Wndings based on a so-called slope analysis. In this type of analysis, when temporal and non-temporal variance are sorted out properly, it is sometimes shown that the growth of temporal variance as a function of time is the same for the perception and the production of intervals (Ivry & Hazeltine, 1995). Recently Robertson et al. (1999: see also Zelaznik, Spencer, and Ivry, 2002) reported, on the basis of a slope analysis, that the growth of temporal variance as a function of time diVers across the diVerent motor tasks under investigation, a Wnding that challenges the single-clock hypothesis. In the present experiment, the main diVerence in the RVF condition was an increase of performance when passing from the 900- to the 1125-ms condition. This could be viewed like the result of a decrease, with longer intervals, of the eVect on the overall discrimination process of a non-temporal factor like the sensory noise due to the signals. Therefore, the diVerences between the LVF and the RVF related to the base duration would not reXect the use of diVerent timing processes, but would reXect diVerences in the way of dealing with the sensory signals. Finally, it is noteworthy to recall that Drake and Botte (1993) obtained larger Weber fractions at 1.5 than at 1 s. In the present experiment, there is a tendency to have better discrimination with the 1.125-ms

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

than with the 1.406-ms base duration in the RVF condition, but not in the LVF condition. 4.3. Alternative avenue It is not clearly determined, in the behavioral literature on timing, whether there is one or more internal mechanisms for processing temporal information. As well, literature on hemispheric specialization or dominance remains Wlled of uncertainty, especially as regard the case of explicit judgments of time intervals. Of course, an experiment such as the present one cannot provide a Wnal response for these issues. But the diYculty to Wnd consistent explanations indicates that a simple dichotomy between the left vs. right hemisphere for processing temporal information is not promising. Recent research in neuroscience rather proposes that each cerebral hemisphere would be involved at diVerent stages of the overall timing process. For instance, electrophysiological data (contingent negative variation) revealed that the left (and medial) frontal activity stops once the current duration reaches a memorized standard duration, but the right frontal activity continues until the end of the current duration (Pfeuty et al., 2003). In this example, portions of the left cerebral hemisphere would be associated with the memory for time, and portions of the right hemisphere would be associated with the anticipation of the end of the current duration. Our data add to the challenge for future research on timing and time perception in cognitive neuroscience where hemispheric diVerences have emerged (see Hinton, 2003; Hinton & Meck, 2004; Matell & Meck, 2004; Melgire et al., 2005). We need to know why increasing the number of standards systematically improve duration discrimination when sequences of signals are presented in the LVF, which is clearly not the case when presented in the RVF; and why the relative eYciency of the LVF vs. RVF conditions evolves diVerently as a function of base duration. What our data are revealing is that understanding cerebral hemisphere properties requires to consider carefully the exact structure of events in the environment.

References Alpherts, W. C. J., Vermeulen, J., Franken, M. L. O., Hendricks, M. P. H., van Veelen, C. W. M., & van Rijen, P. C. (2002). Lateralization of auditory rhythm length in temporal lobe lesions. Brain and Cognition, 49, 114–122. Barnes, R., & Jones, M. R. (2000). Expectancy, attention, and time. Cognitive Psychology, 41, 254–311. Ben-Dov, G., & Carmon, A. (1984). Rhythm length and hemispheric asymmetry. Brain and Cognition, 3, 35–41. Bradshaw, J. L., & Nettleton, N. C. (1981). The nature of hemispheric specialisation in man. Behavioral and Brain Sciences, 4, 51–91. Bryden, M. P. (1977). Measuring handedness with questionnaires. Neuropsychologia, 15, 617–624.

131

Drake, C., & Botte, M.-C. (1993). Tempo sensitivity in auditory sequences: Evidence for a multiple-look model. Perception & Psychophysics, 54, 277–286. Efron, R. (1990). The decline and fall of hemispheric specialization. Hillsdale, NJ: Lawrence Erlbaum Associates. Ferrandez, A. M., Hugueville, L., Lehéricy, S., Poline, J. B., Marsault, C., & Pouthas, V. (2003). Basal ganglia and supplementary motor area subtend duration perception: An fMRI study. NeuroImage, 19, 1532–1544. Friberg, A., & Sundberg, J. (1995). Time discrimination in a monotonic, isochronic sequence. Journal of the Acoustical Society of America, 98, 2524–2531. Funnell, M. G., Corballis, P. M., & Gazzaniga, M. S. (2003). Temporal discrimination in the split brain. Brain and Cognition, 53, 218–222. Gibbon, J., Church, R. M., & Meck, W. H. (1984). Scalar timing in memory. In J. Gibbon & L. G. Allan (Eds.), Annals of the New York Academy of Sciences: Vol. 423, Timing and time perception (pp. 52– 77). New York: New York Academy of Sciences. Grondin, S. (2001a). From physical time to the Wrst and second moments of psychological time. Psychological Bulletin, 127, 22–44. Grondin, S. (2001b). Discriminating time intervals presented in sequences marked by visual signals. Perception & Psychophysics, 63, 1214–1228. Handy, T. C., Gazzaniga, M. S., & Ivry, R. B. (2003). Cortical and subcortical contributions to the representation of temporal information. Neuropsychologia, 41, 1461–1473. Halpern, A. R., & Darwin, C. J. (1982). Duration discrimination in a series of rhythmic events. Perception & Psychophysics, 31, 86–89. Harrington, D. L., Haaland, K. Y., & Knight, R. T. (1998). Cortical networks underlying mechanisms of time perception. Journal of Neuroscience, 18, 1085–1095. Helfrich, H. (Ed.). (2003). Time and Mind II: Information processing perspectives. Goettingen, Ge: Hogrefe & Huber. Hinton, S. C. (2003). Neuroimaging approaches to the study of interval timing. In W. H. Meck (Ed.), Functional and neural mechanisms of interval timing (pp. 419–438). Boca Raton, FL: CRC Press. Hinton, S. C., & Meck, W. H. (2004). Frontal–striatal circuitry activated by human peak-interval timing in the supra-seconds range. Cognitive Brain Research, 21, 171–182. Ivry, R. B., & Hazeltine, R. E. (1995). The perception and production of temporal intervals across a range of durations: Evidence for a common timing mechanism. Journal of Experimental Psychology: Human Perception and Performance, 21, 3–18. Ivry, R. B., & Robertson, L. C. (1998). The two sides of perception. Cambridge, MA: The MIT Press. Jones, M. R., & Boltz, M. G. (1989). Dynamic attending and responses to time. Psychological Review, 96, 459–491. Kagerer, F. A., Wittmann, M., Szelag, E., & Steinbüchel, N. (2002). Cortical involvement in temporal reproduction: Evidence for diVerential roles of the hemispheres. Neuropsychologia, 40, 357– 366. Killeen, P. R., & Weiss, N. A. (1987). Optimal timing and the Weber function. Psychological Review, 94, 455–468. Large, E. W., & Jones, M. R. (1999). The dynamics of attending: How we track time varying events. Psychological Review, 106, 119–159. Macmillan, N. A., & Creelman, C. D. (1991). Detection theory: A user’s guide. New York: Cambridge University Press. Matell, M. S., & Meck, W. H. (2004). Cortico–striatal circuits and interval timing: Coincidence-detection of oscillatory processes. Cognitive Brain Research, 21, 139–170. McAuley, D. J., & Jones, M. R. (2003). Modeling eVects of rhythmic context on perceived duration: A comparison of interval and entrainment approaches to short-interval timing. Journal of Experimental Psychology: Human Perception and Performance, 29, 1102– 1125. McAuley, D. J., & Kidd, G. R. (1998). EVect of deviations from temporal expectations on tempo discrimination of isochronous tone

132

S. Grondin, C. Girard / Brain and Cognition 58 (2005) 125–132

sequences. Journal of Experimental Psychology: Human Perception and Performance, 24, 1786–1800. Meck, W. H. (1996). Neuropharmacology of timing and time perception. Cognitive Brain Research, 3, 227–242. Meck, W. H. (Ed.). (2003). Functional and neural mechanisms of internal timing. Boca Raton, FL: CRC Press. Meck, W. H., & Malapani, C. (2004). Neuroimaging of interval timing. Cognitive Brain Research, 21, 133–137. Melgire, M., Meck, W. H., Samson, S., Baulac, M., Penney, T. B., & Pouthas, V. (2005). Auditory/visual duration bisection in patients with left or right medial–temporal lobe resection. Brain and Cognition, in press. Milner, B. (1962). Laterality eVects in audition. In V. R. Mountcastle (Ed.), Interhemispheric relations and cerebral dominance (pp. 177– 195). Baltimore: John Hopkins University Press. Monfort, V., Pouthas, V., & Ragot, R. (2000). Role of the frontal cortex in memory for duration: An event-related potential study in humans. Neuroscience Letters, 286, 91–94. Nicholls, M. E. R. (1996). Temporal processing asymmetries between the cerebral hemispheres: Evidence and implications. Laterality, 1, 97–137. Pfeuty, M., Ragot, R., & Pouthas, V. (2003). When time is up: CNV time course diVerentiates the roles of the hemispheres in the discrimination of short tone durations. Experimental Brain Research, 151, 372–379.

Polzella, D. J., DaPolito, F., & Hinsman, M. C. (1977). Cerebral asymmetry in time perception. Perception & Psychophysics, 21, 187–192. Pouthas, V., Garnero, L., Ferrandez, A.-M., & Renault, B. (2000). ERPs and PET analysis of time perception: Spatial and temporal brain mapping during visual discrimination tasks. Human Brain Mapping, 10, 49–60. Robertson, S. D., Zelaznik, H. N., Lantero, D. A., Gadacz Bojczyk, K., Spencer, R. M., DoYn, J. G., et al. (1999). Correlations for timing consistency among tapping and drawing tasks: Evidence against a single timing process for motor control. Journal of Experimental Psychology: Human Perception and Performance, 25, 1316–1330. Schulze, H. H. (1989). The perception of temporal deviations in isochronic patterns. Perception & Psychophysics, 45, 291–295. Smith, A., Taylor, E., Lidzba, K., & Rubia, K. (2003). Right hemispheric frontocerebellar network for time discrimination of several hundreds of milliseconds. Neuroimage, 20, 344–350. ten Hoopen, G., Hartsuiker, R., Sasaki, T., Nakajima, Y., Tanaka, M., & Tsumura, T. (1995). Auditory isochrony: Time shrinking and temporal patterns. Perception, 24, 577–593. Zelaznik, H. N., Spencer, R. M. C., & Ivry, R. B. (2002). Dissociation of explicit and implicit timing in repetitive tapping and drawing movement. Journal of Experimental Psychology: Human Perception and Performance, 28, 575–588.