Brain and Cognition 64 (2007) 68–73

www.elsevier.com/locate/b&c

Optokinetic stimulation aVects temporal estimation in healthy humans

Carmelo Mario Vicario a, Carlo Caltagirone b, Massimiliano Oliveri a,b,¤

a Dipartimento di Psicologia, Università degli Studi di Palermo, Viale delle Scienze, EdiWcio 15, 90100 Palermo, Italyb Fondazione “Santa Lucia” IRCCS, Roma, Italy

Accepted 7 December 2006Available online 29 March 2007

Abstract

The representation of time and space are closely linked in the cognitive system. Optokinetic stimulation modulates spatial attention inhealthy subjects and patients with spatial neglect.

In order to evaluate whether optokinetic stimulation could inXuence time perception, a group of healthy subjects performed “time-comparison” tasks of sub- and supra-second intervals before and after leftward or rightward optokinetic stimulation.

Subjective time perception was biased by the direction of optokinetic stimulation. Rightward optokinetic stimulation induced an over-estimation of time perception compared with baseline and leftward optokinetic stimulation. These results indicate a directional bias intime perception induced by manipulation of spatial attention and could argue for a mental linear representation of time intervals.© 2007 Elsevier Inc. All rights reserved.

Keywords: Space; Time; Perception; Optokinetic stimulation

1. Introduction

Mental representations of space and time represent ele-mentary aspects of cognition and play a critical role inplanning action and forming decisions in everyday life.Converging evidence from behavioral, neuroimaging andneuropsychological studies suggests the presence of closelinks between these diVerent cognitive systems (Walsh,2003).

At a behavioural level, spatial factors do aVect the per-ception of time and other magnitudes. When healthy sub-jects are asked to carry out tasks in environments built to1/6, 1/12, or 1/24 of actual size and to stop when 30 minhad passed, the ratio of time passed to time estimatedscales according to environmental scale (De Long, 1981).The apparent duration of a dynamic stimulus can bemanipulated in a local region of visual space by adaptingto oscillatory motion or Xicker (Johnston, Arnold, &Nishida, 2006).

* Corresponding author. Fax: +39 0917028431.E-mail address: maxoliveri@tiscali.it (M. Oliveri).

0278-2626/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.bandc.2006.12.002

The importance of visual factors is also supported byrecent Wndings, showing that short intervals of timebetween two successive perisaccadic visual (and not audi-tory) events are underestimated, indicating a compressionof perceived time (Morrone, Ross, & Burr, 2005).

These data implicate spatially localised temporal mecha-nisms in duration perception. The link between spatial andtemporal estimation has been also documented at a neurallevel. Neuropsychological studies show the occurrence ofspatial as well as temporal deWcits following right hemi-spheric damage (Critchley, 1953; Koch, Oliveri, Carlesimo,& Caltagirone, 2002). Harrington, Haaland, and Knight(1998) showed that brain regions more frequently associ-ated with in alterations of temporal perception tasks are thesuperior and middle frontal gyrus in the right frontal cor-tex, and the angular and supra-marginal gyri in the rightparietal cortex.

An important lesson comes from patients with contrale-sional spatial neglect. These patients overestimate dura-tions of stimuli presented in the neglected space andunderestimate durations of stimuli presented in the goodright Weld (Basso, Nichelli, Frassinetti, & di Pellegrino,1996).

C.M. Vicario et al. / Brain and Cognition 64 (2007) 68–73 69

Neuroimaging studies document an activation of theright parietal cortex in temporal as well as spatial tasks(Rao et al., 2001) and studies using transcranial magneticstimulation show that disrupting the function of right pre-frontal cortex transiently disrupts time perception (Koch,Oliveri, Torriero, & Caltagirone, 2003).

A critical role in mediating the link between spatial andtemporal perception could be made by attention. In fact,attention is known to inXuence perceived duration andtemporal order (Enns, Breahaut, & Shore, 1999; Rose &Summers, 1995; Tse, Intriligator, Rivest, & Cavanagh,2004). This suggests that procedures that manipulate spa-tial attention should also aVect temporal estimation tasks.A rich literature documents that manipulation of spatialattention towards the right or the left Weld by optokineticstimulation is able to directionally inXuence spatial tasks.

Fast optokinetic stimulation towards the neglectedhemispace (Pizzamiglio, Frasca, Guariglia, Inoccia, &Antonucci, 1990), and slow horizontal background motionto the left (Mattingley, Bradshaw, & Bradshaw, 1994) havebeen shown to reduce the horizontal line bisection errorprobably by allocating selective attention to stimuli in theneglected space sector.

Mattingley et al. (1994) using the OKS paradigm atspeeds that did not elicit optokinetic nystagmus or percep-tual after eVects, reported that control subjects were accu-rate in all conditions, and showed minimal eVects ofbackground conditions. By contrast, patients with left uni-lateral spatial neglect were sensitive to leftward back-ground motion, showing a signiWcant leftward shift inbisection error, relative to neutral, static, and rightwardmoving backgrounds.

Other studies (KerkhoV, Schindler, Keller, & Marqu-ardt, 1999) provide evidence that coherent visual back-ground motion towards the neglected hemispacesigniWcantly modulates the perceived horizontal object size,hence the size distortion, in neglect patients when compar-ing two stimuli separated in both hemispaces. This Wndingindicates that background motion not only aVects the sub-jective straight ahead (Karnath, 1996) and subjective mid-line in line bisection or line extension (Bisiach, Pizzamiglio,Nico, & Antonucci, 1996; Mattingley et al., 1994; Pizzami-glio et al., 1990), but also facilitates the accuracy of spatialjudgments concerning two stimuli separated in left andright space. On the same theoretical position is Schindler(Schindler & KerkhoV, 2004), suggesting that visual motionstimulation in neglect alleviates both egocentric deviations(the subjective straight ahead) and allocentric perceptualdistortions in neglect with or without visual Weld disorders.

The combination of these studies conWrm that the para-digm of optokinetic stimulation is able to transiently movethe “attentive focus” towards a speciWc side of space Weld(left vs. right), depending on the direction of optokineticmotion.

The rationale behind the present study is that manipula-tion of spatial attention by means of optokinetic stimula-tion could be able to inXuence the performance not only in

spatial tasks but also in temporal tasks, thus arguing for theuse of spatial codes in time perception.

2. Subjects and methods

Thirty-Wve right-handed graduate students (17 male and18 female, aged 22–31 years) with normal or correctedvision participated in the studies after providing writteninformed consent.

2.1. Optokinetic stimulation

Optokinetic stimulation (OKS) was performed by pre-senting 384 series of a sequence of 17§2 white line seg-ments, presented on a black background, each with aduration of 50 ms. Stimuli drifted either leftward or right-ward, across the full width of a 14� computer monitor, sothat the direction of the apparent movement was from leftto right and vice-versa in diVerent block of trials. The stro-boscopic movement velocity was established functionallyfor individual subjects on the basis of subjective saccadicresponse, monitored by visual inspection by one experi-menter. The average velocity was of 38°/s.

In the Wrst experiment, the subjects were asked to Wxate acentral mark on the screen, with the head immobilised overa chin-rest. Central Wxation was monitored by one experi-menter. In the second and third experiment, the same para-digm of optokinetic stimulation was done without a centralWxation mark.

2.2. Time estimation task

2.2.1. Experiment 1Fifteen subjects (8 men, 7 women, aged 21–26 years) par-

ticipated in this experiment. They were positioned 60 cmopposite a white computer screen (the same used for OKSparadigm) and were required to determine whether a testcue (a red circle) had been presented for a time intervallonger or shorter than a memorized standard duration of areference cue (a blue circle).

The duration of the reference cue was 300 ms in all trials.The test cue was presented immediately after the referencecue and its duration was selected randomly using intervalsnear the standard duration, ranging from 250 to 350 ms (insteps of 10 ms for a total of 10 intervals without presentingthe average interval of 300 ms). The computer programselected interval lengths pseudo-randomly, so that eachsubject received Wve presentations of each interval lengthwithin a 50 trials block.

The intertrial interval was of 2000 ms. Reference and testcues were presented as pairs at the centre of the screen.Total duration of a block was 130 s (50 trials¤ (0.3 s refer-ence cue duration + 0.3 s test cue average duration + 2 sintertrial interval)).

Subjects indicated their judgment of the test cue’s dura-tion by pressing either of two response buttons with theirright and left index Wngers. In half of the trials, a right

70 C.M. Vicario et al. / Brain and Cognition 64 (2007) 68–73

response indicated a smaller duration and a left response abigger duration of the test cue and vice-versa in the otherhalf of the trials.

Time estimation task was performed in three sessions(on three consecutive days) to avoid carry over ofaftereVects (optokinetic after-nystagmus) betweenblocks: a baseline session and a session immediately fol-lowing a rightward or leftward block of optokinetic stim-ulation. The order of the sessions was randomised acrosssubjects.

In all experiments, subjects performed a 10-trials train-ing on the time estimation task, before entering the mainexperiment.

2.2.2. Experiment 2Ten subjects (5 men, 5 women, aged 20–24 years) diVer-

ent from those recruited for experiment 1, participated inthis experiment. OKS was performed without a central Wxa-tion mark. All the other parameters of OKS and time esti-mation were identical to those of experiment 1.

2.2.3. Experiment 3Ten subjects (4 men, 6 women, aged 20–23 years) partici-

pated in this experiment. None of them had participated inthe previous experiments. The paradigm of OKS was iden-tical to that of experiment 2 (OKS without a central Wxa-tion mark).

The time estimation task was similar to that of experi-ment 1 and 2 with the exception of the duration of referenceand test cues. The duration of the reference cue was of2000 ms. Test cue was presented immediately after the refer-ence cue and its duration was selected randomly usingintervals near the standard duration, ranging from 1500 to2500 ms (in steps of 100 ms for a total of 10 intervals with-out presenting the average interval of 2000 ms). The com-puter program selected interval lengths pseudo-randomly,so that each subject received Wve presentations of eachinterval length within a 50 trials block.

The intertrial interval was of 2000 ms. Reference and testcues were presented as pairs at the centre of the screen.Total duration of a block was 300 s (50 trials¤ (2 s referencecue duration + 2 s test cue average duration + 2 s intertrialinterval)).

Response parameters and the other aspects of the exper-imental protocol were identical to those described forexperiment 1.

2.3. Data analysis

In all experiments responses were analysed by assigninga value of 0 to correct responses, ¡1 to time underestima-tion (duration of an interval >300–2000 ms judged asshorter than the reference cue), 1 to time overestimation(duration of an interval <300–2000 ms judged as longerthan the reference cue).

The average responses were analyzed using ANOVA forrepeated measures with the factor condition (baseline vs.

leftward vs. rightward optokinetic stimulation) as within-subject factor.

Duncan’s post-hoc comparisons were made whereappropriate. The level of signiWcance was set at p < .05.

3. Results

3.1. Experiment 1

Fig. 1 shows subjects’ performance in the three mainexperimental conditions. There was a signiWcant eVect ofcondition [F(2,14)D4, pD .01]. Rightward optokineticstimulation induced a signiWcant time overestimation com-pared with both baseline (pD .04) and post-leftward opto-kinetic stimulation blocks (pD .008); the diVerence betweenleftward optokinetic stimulation and baseline was not sig-niWcant.

3.2. Experiment 2

Fig. 2 shows subjects’ performance in the time estima-tion tasks following OKS without a central Wxation mark.There was a signiWcant eVect of condition [F(2, 9)D5.24;pD .005]. Rightward optokinetic stimulation induced a sig-niWcant time overestimation compared with both baseline(pD .01) and post-leftward optokinetic stimulation blocks(pD .003); the diVerence between leftward optokinetic stim-ulation and baseline was not signiWcant.

3.3. Experiment 3

Fig. 3 shows subjects’ performance in the time estima-tion task of supra-second intervals. OKS did not induce sig-niWcant eVects in the time task [F(2, 9)D 0.78; pD .47].

4. Discussion

The main result of the present study is that manipulationof spatial attention towards the right or left space by means

Fig. 1. Experiment 1. Percentage change vs. baseline in the time estimationtask across the diVerent OKS conditions (grey bars: leftward OKS; whitebars: rightward OKS). Error bars indicate 1 SE of mean.

C.M. Vicario et al. / Brain and Cognition 64 (2007) 68–73 71

of optokinetic stimulation induces a bias in a standard timeestimation task performed immediately after the cessationof optokinetic stimulation. Moving attention towards theright hemispace induces time overestimation, while movingattention towards the left hemispace induces a trendtowards time underestimation compared with baseline.This result was present when OKS was done with a centralWxation mark, but it was even more evident in the absenceof OKN suppression, with the consequent increase in atten-tion diversion.

Optokinetic stimulation has been described to aVecthealthy subjects’ exploration of space (Sandor, Bachtold,Henn, & Brugger, 2000). In a standard line bisection task,the bisection marks of both hands are shifted signiWcantlyto the left during leftward stimulation, introducing apseudoneglect for the left and right hands. On the otherhand, rightward stimulation does not inXuence line bisec-tion tasks.

In contrast with these observations, in the present proto-col leftward stimulation failed to signiWcantly aVect perfor-

Fig. 2. Experiment 2. Percentage change vs. baseline in the time estimationtask across the diVerent OKS conditions (grey bars: leftward OKS; whitebars: rightward OKS). Error bars indicate 1 SE of mean.

Fig. 3. Experiment 3. Percentage change vs. baseline in the time estimationtask across the diVerent OKS conditions (grey bars: leftward OKS; whitebars: rightward OKS). Error bars indicate 1 SE of mean.

mance in the time estimation task, although a trendtowards time underestimation was observed. DiVerences inthe experimental protocol of OKS could partly explain thisdiVerence.

Both phases of standard and test cue duration estima-tion seem to imply an explicit process of duration encoding.However, in the present protocol a peripheral sensoryaccount based on the time of activation of a population ofvisual neurones cannot be totally excluded (Efron, 1970). Inthis case, an implicit process of time perception based oninspecting the time at which neural or perceptual systemssupporting visual iconic memory become active would playsome role (Dennett & Kinsbourne, 1992; Johnston & Nish-ida, 2001).

Temporal duration can range from sub-second intervalsto days and weeks. The temporal scale is critical when con-sidering the neural mechanisms involved in time judge-ments. In the present study, modulation of time estimationby OKS was selectively observed when using sub-secondintervals. In fact, the OKS paradigm did not induce a sig-niWcant eVect on time estimation of supra-second intervals.This result could likely be explained with reference to theneural pathways mediating the eVects of OKS. In fact, OKSinduces a widespread activation of the cerebellum, a neuralstructure that has been reported to be selectively involvedin timing sub-second intervals (Koch, Oliveri, Torriero, Sal-erno, & Caltagirone, 2006).

The cerebellum has also been associated with spatialtasks (Townsend et al., 1999). Thus, the brain pathwaysconnecting the cerebellum with the contralateral cerebralregions could be critical for the integration of spatial andtemporal information. Critical regions in this system are theparieto insular vestibular cortex (PIVC) to which otherareas of the central sulcus (area 3a) and the prefrontal andfrontal cortex are connected to form a “vestibular corticalsystem” (Faugier-Grimaud, Baleydier, Magnin, & Jeann-erod, 1997; Guldin & Grusser, 1998).

Galvanic-vestibular stimulation (exciting both semicir-cular canals and otolitic eVerents) (Goldberg, Smith, & Fer-nandez, 1984) determines activations in the temporalparietal junction, central and intraparietal sulci and in thepremotor regions of the frontal lobe (Lobel, Kleine, Bihan,Leroy-Willig, & Berthoz, 1998). Lesion studies in the mon-key (Ventre & Faugier-Grimaud, 1986) show that unilateralsurgical ablations of the convexity of the inferior parietallobule (area 7a) reduce the gain of the slow phases of theVOR (vestibular–ocular response) directed ipsilesionally.

The strong anatomo-functional links between vestibularand auditory systems from one hand, and between vestibu-lar system and superior colliculi from the other hand(Crowder, Dawson, & Wylie, 2003) suggest that theobserved manipulation of timing by OKS could also extendto the auditory domain.

Distortions of duration perception have recently beendemonstrated around the time of a saccadic eye movement(Morrone et al., 2005). Making a saccade can also deliveran apparent reversal of temporal order. Eye movement-

72 C.M. Vicario et al. / Brain and Cognition 64 (2007) 68–73

dependent changes in duration perception and spatial dis-tortion (Ross, Morrone, Goldberg, & Burr, 2001) have beenrelated to predictive remapping of receptive Welds in lateralintraparietal area (LIP; Duhamel, Colby, & Goldberg,1992). Another process that has been associated with sac-cades is the suppression of magnocellular activity (Burr,Morrone, & Ross, 1994; Ross et al., 2001). This mechanismwould be speciWc for the visual modality (Johnston et al.,2006; Morrone et al., 2005).

Although the paradigm adopted in the present study isnot optimal for inducing saccades, our results could partlybe in line with these observations.

Whatever the mechanism is, the Wndings of this work arein line with previous work showing that for the brain timeand space are not processed separately, but inXuence eachother (Burr, 2000). Our study provides another example ofthe interactions between the two dimensions.

One could argue that a bias in temporal estimationcould reXect a bias in motor responses. In fact, left-handresponses could be facilitated following leftward- andright-hand responses following rightward optokineticstimulation blocks. However, this bias cannot accountentirely for the present Wndings given that right/leftresponses for the time estimation task were counterbal-anced in the single subjects during each session.

An alternative explanation for these results could be theuse of a linear representation of time intervals in healthysubjects performing time estimation tasks. According tothis hypothesis, an attentional bias towards the right hemi-space could also have biased the temporal judgment of thetest cue towards overestimation. Thus, the relation linkingtime and space perception could be similar to that betweennumbers and space processing, exempliWed by the meta-phor of mental number line positing that low numbers areassociated with left side space and higher numbers withright-side space (Dehaene, Dupoux, & Mehler, 1990).Future studies with diVerent experimental paradigms couldfurther address this issue.

References

Basso, G., Nichelli, P., Frassinetti, F., & di Pellegrino, G. (1996). Time per-ception in a neglected space. Neuroreport, 7(13), 2111–2114.

Bisiach, E., Pizzamiglio, L., Nico, D., & Antonucci, G. (1996). Beyond uni-lateral neglect. Brain, 119, 851–857.

Burr, D. C. (2000). Motion vision: are ‘speed lines’ used in human visualmotion? Current Biology, 10, 440–443.

Burr, D. C., Morrone, M. C., & Ross, J. (1994). Selective suppression of themagnocellular visual pathway during saccadic eye movements. Nature,371, 511–513.

Critchley, M. (1953). The parietal lobes. Hafner Press.Crowder, N. A., Dawson, M. R. W., & Wylie, D. R. W. (2003). Temporal

frequency and velocità-like tuning in the pigeon accessory optic system.Journal of Neurophysiology, 90, 1829–1841.

Dehaene, S., Dupoux, E., & Mehler, J. (1990). Is numerical comparisondigital? Analogical and symbolic eVects in two-digit number compari-son. Journal of Experimental Psychology. Human Perception and Per-formance, 16, 626–641.

De Long, A. J. (1981). Phenomenological space-time: towards an experien-tial relativity. Science, 213, 681–683.

Dennett, D. C., & Kinsbourne, M. (1992). Time and the observer: thewhere and when of consciousness in the brain. The Behavioral andBrain Science, 15, 183–247.

Duhamel, J. R., Colby, C. L., & Goldberg, M. E. (1992). The updating ofthe representation of visual space in parietal cortex by intended eyemovements. Science, 255, 90–92.

Efron, R. (1970). EVect of stimulus duration on perceptual onset and oVsetlatencies. Perception & Psychophysics, 8, 231–234.

Enns, J. T., Breahaut, J. C., & Shore, D. I. (1999). The duration of a briefevent in the mind’s eye. The Journal of General Psychology, 126, 355–372.

Faugier-Grimaud, S., Baleydier, C., Magnin, M., & Jeannerod, M. (1997).Direct bilateral cortical projections to the vestibular complex inmacaque monkey. In P. Thier & H. O. Karnath (Eds.), Parietal lobecontributions to orientation in 3D space (pp. 57–76). Springer.

Goldberg, J. M., Smith, C. E., & Fernandez, C. (1984). Relation betweendischarge regularity and responses to externally applied galvanic cur-rents in vestibular nerve aVerents of the squirrel monkey. Journal ofNeurophysiology, 51, 1236–1256.

Guldin, W. O., & Grusser, O. J. (1998). Is there a vestibular cortex? Trendsin Neurosciences, 21, 254–259.

Harrington, D. L., Haaland, K., & Knight, R. (1998). Cortical networksunderlying mechanism of time perception. The Journal of Neuroscience,18(3), 1085–1095.

Johnston, A., Arnold, D. H., & Nishida, S. (2006). Spatially localized dis-tortions of event time. Current Biology, 16, 472–479.

Johnston, A., & Nishida, S. (2001). Time perception: brain time or eventtime? Current Biology, 11, R427–R430.

Karnath, H. O. (1996). Optokinetic stimulation inXuences the disturbedperception of body orientation in spatial neglect. Journal of Neurology,Neurosurgery, and Psychiatry, 60, 217–220.

KerkhoV, G., Schindler, I., Keller, I., & Marquardt, C. (1999). Visual back-ground motion reduces size distortion in spatial neglect. Neuroreport,10(2), 319–323.

Koch, G., Oliveri, M., Carlesimo, G. A., & Caltagirone, C. (2002). SelectivedeWcit of time perception in a patient with right prefrontal cortexlesion. Neurology, 59(10), 1658–1659.

Koch, G., Oliveri, M., Torriero, S., & Caltagirone, C. (2003). Underestima-tion of time perception after repetitive transcranial magnetic stimula-tion. Neurology, 60(11), 1844–1846.

Koch, G., Oliveri, M., Torriero, S., Salerno, S., & Caltagirone, C. (2006).Repetitive TMS of the cerebellum interferes with millisecond timeprocessing. Experimental Brain Research, in press, [Epub ahead ofprint].

Lobel, E., Kleine, J. F., Bihan, D. L., Leroy-Willig, A., & Berthoz, A. (1998).Functional MRI of galvanic vestibular stimulation. Journal of Neuro-physiology, 80, 2699–2709.

Mattingley, J. B., Bradshaw, J. L., & Bradshaw, J. A. (1994). Horizontalvisual motion modulates focal attention in left unilateral spatialneglect. Journal of Neurology, Neurosurgery, and Psychiatry, 57, 1228–1235.

Morrone, M. C., Ross, J., & Burr, D. (2005). Saccadic eye movementscause compression of time as well as space. Nature Neuroscience, 8,950–954.

Pizzamiglio, L., Frasca, R., Guariglia, C., Inoccia, C., & Antonucci, G.(1990). EVect of optokinetic stimulation in patients with visual neglect.Cortex, 26, 541–554.

Rao, S. M. et al. (2001). The evolution of brain activation during temporalprocessing. Nature Neuroscience, 4, 317–323.

Rose, D., & Summers, J. (1995). Duration illusion in a train of visual stim-uli. Perception, 24, 1177–1187.

Ross, J., Morrone, C. M., Goldberg, M. E., & Burr, D. C. (2001). Changesin visual perception at the time of saccades. Trends in Neurosciences,24, 113–121.

Sandor, P. S., Bachtold, D., Henn, V., & Brugger, P. (2000). EVects of opt-okinetically induced rotatory self-motion on spatial perception andrepresentation. Neuropsychiatry Neuropsychology and Behavioral Neu-rology, 13(3), 188–194.

C.M. Vicario et al. / Brain and Cognition 64 (2007) 68–73 73

Schindler, I., & KerkhoV, G. (2004). Convergent and divergent eVects ofneck proprioceptive and visual motion stimulation on visual space pro-cessing in neglect. Neuropsychologia, 42, 1149–1155.

Townsend, J., Courchesne, E., Covington, J., WesterWeld, M., Harris, N. S.,Lyden, P., et al. (1999). Spatial attention deWcits in patients withacquired or developmental cerebellar abnormality. The Journal of Neu-roscience, 19(13), 5632–5643.

Tse, P., Intriligator, J., Rivest, J., & Cavanagh, P. (2004). Attention and thesubjective expansion of time. Perception & Psychophysics, 66, 1171–1189.

Ventre, J., & Faugier-Grimaud, S. (1986). EVects of posterior parietallesions (area7) on VOR in monkeys. Experimental Brain Research, 62,654–658.

Walsh, V. (2003). A theory of magnitude: common cortical metrics of time,space and quantity. Trends in Cognitive Sciences, 7(11), 483–488.