Target velocity effects on manual interception kinematics

Target velocity e�ects on manual interception kinematics

Adam Dubrowski, Jerry Lam, Heather Carnahan *

Department of Kinesiology, University of Waterloo, Waterloo, Ont., Canada N2L 3G1

Received 19 October 1998; received in revised form 8 November 1999; accepted 22 November 1999

Abstract

Participants generated manual interception movements toward a target cursor that moved

across a computer screen. The target reached its peak velocity either during the ®rst third, at

the midpoint, or during the last third of the movement. In Experiment 1 the view of the target

was available for either the ®rst 316, 633, 950, or 1267 ms, after which it disappeared. Results

showed that for all viewing conditions, the timing of the interception velocity was related to

the temporal properties of the targetÕs trajectory. In Experiment 2, when the portion of the

target trajectory that was viewed was reversed (such that participants did not see the ®rst 316,

633, 950, or 1267 ms of the trajectory, but instead saw only the later portions of the trajectory),

there was no clear relationship between the target trajectory and the timing of the aiming

trajectory. These results suggest that participants use visual information early in the targetÕstrajectory to form a representation of the target motion that is used to facilitate manual

interception. Ó 2000 Elsevier Science B.V. All rights reserved.

PsycINFO classi®cation: 2330

Keywords: Reaching; Vision; Motor control; Moving targets

1. Introduction

When manually intercepting moving targets, it has been shown that the charac-teristics of the target motion in¯uence how the arm reaches toward the target. In

Acta Psychologica 104 (2000) 103±118

www.elsevier.com/locate/actpsy

* Corresponding author. Fax: +1-519-885-0470.

E-mail address: [email protected] (H. Carnahan).

0001-6918/00/$ - see front matter Ó 2000 Elsevier Science B.V. All rights reserved.

PII: S 0 0 0 1 - 6 9 1 8 ( 9 9 ) 0 0 0 5 6 - 6

computer simulations of reaching van Donkelaar, Lee and Gellman (1992) haveshown that the duration of a reaching movement is shorter when the target is movingfaster, both in a full vision condition, and when vision of the hand is restricted. Thisrelationship between interception speed and target speed was also demonstrated bySmeets and Brenner (1995) in a study where participants were required to hit targetsmoving across a computer screen with a handheld rod, as quickly as possible. In spiteof the instruction to hit all the targets as quickly as possible, it was found that fasttargets were hit with a higher velocity than slow targets. In another study, whengenerating reaching and grasping movements to a three-dimensional target rollingdown a ramp, the speed of the reach has also been shown to increase as target speedincreases, and when the target was moving very slowly (<40 mm/s), participantsreached even more slowly than they did for a stationary control target (Carnahan &McFadyen, 1996). Conversely, when participants self-selected the location wherethey would intercept a moving target, their aiming speed decreased as the targetspeed increased (Chie�, Fogassi, Gallese & Gentilucci, 1992). However, this wasprobably due to the observation that participants grasped the faster moving targetscloser to their body. The shorter movements and related lower peak velocities thatwere observed when intercepting the fast targets are consistent with the predictionsof FittsÕ Law (see Mackenzie, Marteniuk, Dugas, Liske & Eickmeier, 1987).

While this relationship between the speed of the interceptive motion and the speedof the moving target exists, it has not been clearly established which characteristics ofthe target motion are most important. As a target moves faster between two points,two movement characteristics can change. Not only is the velocity of the faster targetgreater, its movement time is shorter. It is not clear whether both of these two targetmotion characteristics can in¯uence the interceptive movement. One of the mostcommonly cited models to explain the visuomotor control of target interception hasbeen the Tau model originally proposed by Lee (1976). While there have been manyvariations or updates to the model over recent years (Tresilian, 1995), the basicpremise of this model is that a time variable (time to contact) is used to trigger theappropriate interceptive action. For example the Tau model has been shown tomodel actions such as jumping up to intercept a falling volleyball (Lee, Young,Reddish, Lough & Clayton, 1983), or the hitting of a table tennis ball (Bootsma &Van Wieringen, 1990). This model does not take into account the changes in thevelocity characteristics of the target, only its temporal characteristics. However, in areview by Bootsma, Fayt, Zaal and Laurent (1997) the possibility is entertained thatwhen intercepting accelerating targets a second-order variable may be used.

In a recent study in which participants intercepted targets on a computer screen,targets moved either at a constant acceleration, a constant deceleration, or at aconstant velocity (Port, Lee, Dassonville & Georgopoulos, 1997). It was proposedthat the response time of the interceptive movement is composed of a constantprocessing time, and the time required for the stimulus to travel a threshold distance,thus response time is velocity dependent. It was also suggested that some participantsuse an estimate of time to contact to determine when to initiate their interceptivemovements. It was further shown that the interceptive movements toward slow-moving targets were produced by generating a series of sub movements that were

104 A. Dubrowski et al. / Acta Psychologica 104 (2000) 103±118

produced to keep the displacement of the hand proportional to the target positionand that this process is independent of the acceleration of the target (Lee, Port &Georgopoulos, 1997). Thus, these authors have shown that participants are not ableto utilize a higher order variable (acceleration) to control their interceptive actions,but instead, base the initiation and production of their interception movements onthe temporal characteristics of the moving target.

It might not be surprising that individuals in the Lee et al. (1997) study were notable to successfully utilize target acceleration information. The general consensus inthe psychophysical literature is that accelerations cannot be directly perceived fromtarget motion. Instead it has been widely argued that target accelerations are com-puted indirectly from the comparisons of two instantaneous velocities acquiredacross a speci®c length of time (Babler & Dannemiller, 1993; Gottsdanker, 1952;Werkhoven, Snippe & Toet, 1992). However, Rosenbaum noted that ``the perceptualsystem is primarily responsive to changes in stimulation, (and) it is noteworthy thatacceleration involves more changes than does constant velocity'' (Rosenbaum, 1975,p. 402). Thus, our visuomotor system could be viewed as one that searches for theoptimal set of information that will allow it to perform the required task. Mostobjects in real life move with non-constant velocity pro®les. Therefore it would bebene®cial for the visuomotor system to be able to retrieve and use visual informationabout changes in object velocity to produce successful interception movements.While the majority of the evidence collectively suggests that only temporal infor-mation about target motion, or information about instantaneous velocity, is used toguide interceptive actions, in the present study, changes in the velocity characteristicsof the targets were manipulated.

The purpose of the present study was to examine if the changing velocity char-acteristics of a target have any in¯uence on the temporal and kinematic character-istics of manual interceptive movements. Participants intercepted targets withcommon movement times and magnitudes of peak velocity. However, the temporalproperties of the targetÕs trajectory were varied across conditions. If participants areable to use information about the changing velocity of the targetÕs trajectory then itis expected that the kinematics of the interception movement will vary as a functionof target velocity condition. A related question is how fast can that information beretrieved from the trajectory of the moving target? As well, how much of the in-formation is needed to precisely intercept the target and when is it needed? The nexttwo experiments were designed to investigate these issues.

For tasks where the target disappears for more than one second before contact,the interception hand trajectory cannot be controlled using continuous time tocontact information, but rather the visuomotor system must rely on an internalmodel or representation of the future target trajectory, built during the initialplanning phase. This view was experimentally supported by Whiting's (1968) studyof baseball catching. With decreased amounts of visual information, the authorconcluded that constant visual monitoring of the ball ¯ight was not necessary forprecise interception. Whiting showed that performance did not decrease signi®cantlywith a decreased amount of visual information about the ball trajectory. Based onthese ®ndings, one can suppose that the visuomotor system is able to create an

A. Dubrowski et al. / Acta Psychologica 104 (2000) 103±118 105

internal model of the target trajectory and use this model to produce successfulmanual interception. In such a case, the performance of the interception task shouldsu�er signi®cantly when the visual information about the target trajectory is re-moved from the planning period. This hypothesis was tested by Whiting and Sharp(1974) where participants were required to catch tennis balls in a dark room. For abrief period of time the ball was visible to the participant during its ¯ight. It wasfound that when the visual information about the ball trajectory was blocked untilvery late in the ¯ight, the catching performance su�ered, emphasizing the impact ofvisual information early in a targetÕs trajectory.

2. Experiment 1

Participants were required to intercept targets with di�ering temporal properties.The target reached its peak velocity either early (®rst third), at the mid point, or late(®nal third) in the targetÕs trajectory. The viewing time of the trajectory, where theparticipants viewed only portions of the target trajectory, was also manipulated. Iftarget information is used early in the trajectory, then the timing of the participants'hand velocity pro®les should be in¯uenced by the temporal properties of the targetÕsvelocity pro®les, even when the view of the target trajectory is only available for ashort time. If this information is not used early in the target trajectory, it is expectedthat the participants will be in¯uenced by the target's velocity pro®le only when it isavailable for viewing for the entire trajectory.

2.1. Method

2.1.1. ParticipantsTwelve undergraduate students (®ve males and seven females) with a mean age

19.8 yr (range 19±22 yr) participated in the study. They were all right-handed asdetermined by self report, with normal, or corrected to normal vision.

2.1.2. ApparatusA Summasketch II Professional graphics tablet (30.5 cm ´ 46.0 cm) with a mouse

attachment was connected to an IBM 386 computer. The computer monitor wasplaced directly behind the tablet. The mouse controlled the participantÕs cursor�1:5 cm� 1:0 cm� that was positioned at the bottom of the 29� 42 cm computerscreen, (11 cm below the interception zone). A rectangular target �1:5 cm� 1:0 cm�was located at the top of the computer screen. The target traveled 11 cm down thescreen in a straight line toward a 1.5 cm interception zone which was located in themiddle of the screen. The interception zone remained visible to the participants at alltimes. Both the target cursor and the participantÕs cursor had only one degree offreedom and could move only in a straight line toward the interception zone. Therewas a one to one ratio between movements of the mouse and the participantÕs cursoron the screen. Position of the mouse on the graphics tablet was recorded with aprecision of 0.5 mm and data were sampled at 60 Hz.


2.1.3. ProceduresUsing the preferred hand, the participant's task was to move the cursor so that it

terminated in the interception zone simultaneous with the arrival of the target. Theonly constraint was that the movement of the participantÕs cursor could not startbefore the target cursor started moving. There were three target trajectory conditionsin this study, in which the time that peak velocity was reached was manipulated. Forthe early peak condition, the time to peak velocity was 400 ms, for the middle peakcondition the time to peak velocity was 767 ms, and in the late condition it occurredat 1150 ms. For all conditions the target traveled the same distance to the inter-ception zone (11 cm), took the same amount of time to arrive at the interceptionzone (1583 ms), had the same average velocity (6.9 cm/s), and the same peak velocity(10 cm/s). Refer to Fig. 1 for a schematic of the target trajectory conditions.

The amount of time that each of the target trajectories was viewed by the par-ticipant was also manipulated. As seen in Fig. 1, for each target trajectory conditionthe target disappeared at either 316 ms (condition A), 633 ms (condition B), 950 ms(condition C), or 1267 ms (condition D). In addition there was a full exposure of thetarget trajectory (condition E). There were 10 trials for each combination of target

Fig. 1. Target velocity pro®les (upper panel) and visual exposure conditions for Experiments 1 (middle

panel) and 2 (bottom panel). For the middle and bottom panels, the thick line denotes the trajectory

viewing period, dashed lines indicate the time period when the target trajectory was blocked. The vertical

lines indicate the instantaneous target velocity at the moment of the end of the viewing period. Note that

the target trajectory visual exposures for Experiment 2 are the reverse to those used in Experiment 1.


velocity timing (three levels) and exposure (®ve levels). Thus, each participant per-formed 150 trials. All trials were presented in the same random order for all par-ticipants, and no type of trial was presented more than twice in a row. Before theactual data collection, each participant was given 30 practice trials in a randomorder, which provided two exposures to each combination of the velocity and ex-posure conditions. No augmented feedback was provided regarding the accuracy orprecision of the interception.

2.1.4. AnalysisDependent measures included the following: initiation time which was de®ned as

the time from the onset of the target motion to the onset of the participant's handmotion; movement time, de®ned as the duration between the onset of reaching to theend of the interception movement (movement end was de®ned as the moment theparticipantÕs cursor reached a zero velocity within the interception zone); peak ve-locity, de®ned as the maximum value of hand (participantÕs cursor) interceptionresultant velocity; time to peak velocity, which was the duration from the initiationof hand movement to peak velocity; time from peak velocity, de®ned as the timefrom peak velocity to the end of movement (deceleration time); and temporal error,which re¯ected the time duration between the arrival of the center of the targetcursor into the center of the interception zone and the termination of the motion ofthe center of the participantÕs cursor into the interception zone (which was alwayssuccessfully accomplished). Separate 3 (velocity: early, middle, late) X 5 (exposure:316, 633, 950, 1267 and 1583 ms) repeated measure analyses of variance (ANOVAs)were used to analyze each dependent measure. ANOVAs signi®cant at P < 0.05 werefurther analyzed using the Tukey method for post hoc comparison of means.

2.2. Results

Initiation time. There was a signi®cant main e�ect for velocity when initiation timewas analyzed, F �2; 22� � 37:1; P < 0:001. Initiation time was shorter for the earlypeak velocity condition, compared to the other two conditions that did not di�er.Plots of all signi®cant e�ects for Experiment 1 are seen in Fig. 2.

Movement time. There were signi®cant main e�ects for velocity, F �2; 22� � 116:4;mP < 0:001, and exposure, F �4; 44� � 14:8; P < 0:001, in addition to a velocity byexposure interaction, F �8; 88� � 3:0; P < 0:05. For exposure durations 633 and950 ms, and full exposure, movement time was shorter when intercepting the earlypeak velocity target, intermediate for the middle target, and longest for the late peakvelocity target. A similar pattern was observed for the 316 ms exposure, however thesteps from early to middle velocity and from middle to late were not statisticallydi�erent. Also, a similar pattern was observed in the 1267 ms exposure condition,although the step from the early to middle peak velocity target was not signi®cantlydi�erent.

Temporal error. There were signi®cant main e�ects for velocity, F �2; 22� � 123:2;P < 0:001, and exposure, F �4; 44� � 17:0; P < 0:001, in addition to a signi®cantinteraction, F �8; 88� � 3:6; P < 0:001. For exposure durations of 950 and 1267 ms,


and full exposure, the participants arrived early in the interception zone for the earlypeak velocity target, nearly on time for the middle peak target, and late for the latepeak velocity target. A similar trend was observed for the 633 ms viewing timecondition, however the participants arrived in the interception zone late for themiddle peak velocity target and they were even later for the late peak velocity target.When the targets were viewed for the initial 316 ms, the steps from the early tomiddle, and from the middle to late peak velocity conditions, were not statisticallydi�erent.

Peak velocity. There was a main e�ect for target velocity, F �2; 22� � 110:8;P < 0:001, in addition to a signi®cant velocity by exposure interaction,F �8; 88� � 5:0; P < 0:001. For the four exposure conditions, 633, 950 and 1267 ms,and full vision, interception movements to the early peak velocity target resulted in

Fig. 2. The interaction of exposure time and timing of target peak velocity for all temporal and kinematic

measures in Experiment 1. (IT � initiation time, MT � movement time, PV � peak velocity, TPV � time

to peak interception velocity, TFPV � time from peak interception velocity to movement end,

TE � temporal error.)


signi®cantly greater hand peak velocity compared to the middle peak velocity target,which was signi®cantly larger compared to the late peak velocity target. A similarpattern was observed in the 316 ms condition, but the magnitude of the di�erenceswas not signi®cant for the early to middle, and middle to late velocity targets. Thepeak velocity of the participantÕs cursor (hand) did not di�er across the exposureconditions. Typical hand velocity pro®les are presented in Fig. 3.

Time to peak velocity. There were main e�ects of velocity F �2; 22� � 35:3;P < 0:001, and exposure F �4; 44� � 4:8; P < 0:05 when time to peak velocity wasanalyzed. The interaction between velocity and exposure was not signi®cantP > 0:05. Thus, the e�ect of target velocity shape on hand velocity timing did notdi�er as a function of exposure time. Time to peak velocity was smallest for the earlypeak velocity target, intermediate in the middle peak velocity target and largest in the

Fig. 3. The hand velocity pro®les for one participant in response to the three di�erent target velocity

pro®les. (a) Experiment 1. Left panel shows the hand velocity pro®le when the initial 316 ms of target

trajectory was visually available. The right panel shows the hand velocity pro®le when the entire target

trajectory was visually available to the participant. (b) Experiment 2. The left panel shows the hand ve-

locity pro®le when the ®nal 316 ms of target trajectory was visually available. The right panel shows the

hand velocity pro®le when the entire target trajectory was visually available to the participant.


late peak velocity target. Only the exposure condition of 1267 ms signi®cantlydi�ered from the 316 and 633 ms conditions, which did not di�er from the otherexposure conditions.

Time from peak velocity. There were main e�ects of target velocity, F �2; 22� � 8:3;P < 0:05, and exposure, F �4; 44� � 16:7; P < 0:001 when time from peak velocitywas analyzed. Time from peak velocity was smaller in the early velocity condition incomparison to the middle and late conditions. Also, time from peak velocity wassigni®cantly larger for the 316 and 633 ms exposure conditions in comparison to the950 and 1267 ms, and full vision conditions.

2.3. Discussion

The initiation time data suggest that information about the properties of thetargetÕs trajectory was retrieved early in the target's trajectory. In this study, thedurations of movement initiation were consistent with the notion that in situationswhere participants must rely on on-line visual information, responses are quickerto targets moving at greater velocity because a critical threshold distance of targetmovement is achieved sooner with a faster moving target (van Donkelaar et al.,1992). The consistency in the movement time pattern observed across all exposureconditions suggests that the motor program for the hand reach movement wascompleted during the movement initiation time period. Even in the 316 ms ex-posure condition the in¯uence of the target velocity conditions on the pattern ofmovement time was similar. Thus, it was not necessary for participants to see thetarget after the initiation of hand movement to scale the temporal properties ofthe targetÕs trajectory. The temporal errors for the three target velocity conditionsdid not vary as a function of exposure duration for the 950, 1267 and 1583 msexposures. This suggests that the additional visual information present at the endof the target trajectory was not used by the participants to correct the timingerrors.

The relationship between the temporal characteristics of the target and the timingof the hand interception kinematics (as evidence by time to peak velocity) was seeneven in the 316 ms exposure condition. This suggests that a temporal template orrepresentation of target trajectory is made during the preparation time and containschanges in the target's velocity information during that period. The absence of aninteraction e�ect between the target peak velocity timing and the exposure durationfor the deceleration time analysis suggests that the ®nal interception phase was notaugmented with additional visual information about the targetÕs trajectory. Inconjunction with the temporal error data, it can be concluded that the ®nal phase ofthe interception did not rely on the visual information regarding the targetÕs tra-jectory, instead the hand velocity was prede®ned at this point based on target tra-jectory information that was available earlier in its movement. The pattern ofdeceleration was roughly a mirror image of the pattern to reach peak velocity,presumably to ensure that regardless of the internal timing structure of the move-ment, the overall time of the interception movement remained relatively constant topreserve aiming temporal accuracy.


3. Experiment 2

The general conclusion from Experiment 1 was that the information about thetarget's trajectory characteristics was acquired by the participants and used early inthe planning and generation of the interception movement. The patterns that de-scribed the temporal and kinematic characteristics of the hand movement weresimilar in all exposure conditions, even when the information about the target'smotion was available only during the initiation period. However, it is not clear ifinformation about the target motion must be available during the planning process.The second study intended to examine how delays in visual information regardingthe target's trajectory in¯uence the temporal and kinematic patterns of manual in-terception movements. It was expected that when target information was delayedbeyond this initial planning period participants would change their strategy ofmanual target interception. Thus, the main purpose of the second study was to in-vestigate the characteristics of manual interception movements that are a�ected bythe early visual occlusion of target motion. A primary concern was to investigatehow much of a delay could be tolerated by the observer before modi®cations in theinterception kinematics took place. That is, does the visual information about thetarget motion have to be present during the movement initiation period, or can it beintegrated into the hand movement in an on-line fashion after the hand movementhas been initiated?

3.1. Methods

3.1.1. ParticipantsTwelve undergraduate students (2 males and 10 females) with a mean age of

19.3 yr (range 19±21 yr) participated in the second study. They were all right-handed,with normal or corrected to normal vision.

3.1.2. Apparatus and proceduresThe second experiment was performed using the same apparatus and procedures

as the previous study, with the only procedural di�erence being the presentation ofvisual information about the target trajectory. Schematics that demonstrate whenvision of the target was available are shown in Fig. 1. In Experiment 2, the onset oftarget movement was indicated by its disappearance from the screen. The target thenreappeared with a varied duration of time before the interception zone. There were®ve conditions presented in the same random order to all participants. In conditionA, the target reappeared 316 ms before the target reached the interception zone, incondition B the target reappeared 633 ms before the interception zone, in conditionC, 950 ms before the interception zone, in condition D, 1267 ms before the inter-ception zone, and ®nally in condition E, the target was available for 1583 ms, or theentire trajectory of the target.

3.1.3. AnalysisThe analysis was identical to that of Experiment 1.


3.2. Results

Initiation time. There was a signi®cant main e�ect of exposure, F �4; 44� � 4:5;P < 0:01 in addition to an interaction of exposure and velocity, F �8; 88� � 3:0;P < 0:01. In the full vision condition, initiation time was shorter for the early peakvelocity target, in comparison to the late peak velocity target, while initiation time tothe middle target did not di�er from the other two. In the other exposure conditions,there were no di�erences in initiation time as a function of target trajectory char-acteristics. All signi®cant e�ects for Experiment 2 are plotted in Fig. 4.

Movement time. There was a signi®cant velocity by exposure interaction,F �8; 88� � 9:5; P < 0:001 when movement time was analyzed. For the 950 and

Fig. 4. The interaction of exposure time and timing of target peak velocity for all temporal and kinematic

measures for Experiment 2. (IT � initiation time, MT � movement time, PV � peak velocity,

TPV � time to peak interception velocity, TFPV � time from peak interception velocity to movement

end, TE� temporal error.)


1267 ms, and the full exposure conditions, movement time was shortest for the earlyvelocity target, intermediate for the middle velocity target, and longest for the latevelocity target. A similar pattern was observed for the 633 ms exposure condition,however the di�erences between the early and middle peak velocity targets, and themiddle to late peak velocity targets were not statistically di�erent and only early andlate velocity conditions di�ered. An opposite pattern was observed in the 316 ms tointerception zone exposure condition, but this e�ect was not statistically signi®cant.

Temporal error. In the analysis of temporal error, there were signi®cant maine�ects for velocity, F �2; 22� � 36:4; P < 0:001, and exposure, F �4; 44� � 17:1;P < 0:001, in addition to a signi®cant interaction e�ect, F �8; 88� � 12:1; P < 0:001.For the 1267 ms, and full exposure conditions, the participants were early in theearly peak velocity condition, late in the middle condition, and even later in the latepeak velocity condition. A similar trend was observed when the last 950 ms of thetarget trajectory were viewed, however the participants were late in the early targetvelocity condition with no signi®cant changes in the other two conditions. In the633 ms exposure condition, again, the pattern was similar to the other exposureconditions, however the di�erence between the early and middle velocity targets wasstatistically signi®cant. Finally, with vision available only in the last 316 ms beforethe interception zone, this pattern was no longer present. Overall, this exposurecondition had the highest positive temporal error values for the three target trajec-tories and they were not statistically di�erent from each other.

Peak velocity. There were main e�ects for velocity F �2; 22� � 31:4; P < 0:001,and for exposure, F �4; 44� � 11:9; P < 0:001, in addition to a signi®cant interactionof velocity and exposure, F �8; 88� � 11:3; P < 0:001 when peak velocity was ana-lyzed. In the three exposure conditions of 950 ms, 1267 ms, and full vision, a clearstep-like pattern was found. In all conditions, the early velocity condition was sig-ni®cantly higher than the middle and late velocity conditions. Interception velocityto the middle velocity target was signi®cantly di�erent compared to the late velocitytarget only in the 1267 ms exposure condition. In the other two visual exposureconditions (316 and 633 ms) the pattern was not present and none of the velocityconditions di�ered signi®cantly.

Time to peak velocity. When time to peak velocity was analyzed, there were maine�ects of velocity F �2; 22� � 7:9; P < 0:01, and exposure, F �4; 44� � 6:6; P < 0:001,in addition to a velocity by exposure interaction, F �8; 88� � 2:8; P < 0:01. The earlyvelocity target in the 316 ms exposure condition, was reached later than the middlevelocity target, which did not di�er from the late velocity target. While not statis-tically signi®cant, it was observed that in the full vision condition the pattern wassimilar to that obtained in the full vision condition in Experiment 1 (see Fig. 2). Thatis, peak velocity was reached sooner for the early peak target, and later for the latepeak target.

Time from peak velocity. When time from peak velocity was analyzed, there weremain e�ects of velocity F �2; 22� � 47:2; P < 0:001, and exposure, F �4; 44� � 11:9;P < 0:001 in addition to an interaction of velocity and exposure, F �8:88� � 4:7;P < 0:001. Time from peak velocity was shorter for the early peak velocity target incomparison to the late peak target, for all exposures except for the 316 ms condition.


Only in the 1267 ms exposure condition was deceleration time for the three targetsstatistically di�erent. Also, for the 316 ms exposure condition, the deceleration timewas statistically di�erent for the early peak and middle peak targets, but not the latepeak velocity target.

3.3. Discussion

The results of Experiment 1 were replicated only for the full vision condition inExperiment 2. Since the e�ect of velocity condition on initiation time emerged onlyin the full vision condition, these ®ndings suggest that the information about targetvelocity (or its change) is retrieved early in the target's trajectory. More speci®cally,the peak velocity ®ndings for the full exposure, 1267 and 950 ms conditions repli-cated the ®ndings from Experiment 1. However, the step-like pattern seen in theseconditions was not observed in the 633 and 316 ms conditions in Experiment 2. Thisindicates that the hand kinematics were speci®ed on the basis of visual informationavailable in the period prior to the last 950 ms of the target trajectory. It is possiblethat the participants can plan hand trajectories early in the movement, after themovement initiation time period, however this plan can be modulated on-line afterhand movement initiation.

Of particular interest was the reversal of the pattern in time to peak velocityobserved in the 316, 633, 950, and 1267 ms visual exposure conditions compared tothe pattern observed in the full vision condition. This ®nding indicates that the ki-nematics for the interception movement was determined as early as 316 ms into thetrajectory. The second study showed that when the information was withheld duringthis crucial early time window the temporal characteristics of the hand trajectorywere not consistent with the timing of the target kinematics.

4. General discussion

The present results collectively support the proposal that during the movementplanning phase, changes in the timing of the target's trajectory are visually detectableand used in the preparation of manual interception movements. In both experiments,the targetÕs peak velocity was reached at three di�erent times, while the totalmovement times and distance traveled always remained the same. The experimentswere designed to examine whether changes in the timing of the targetÕs trajectoryin¯uenced interception kinematics. The most important conclusion from these twoexperiments was that individuals were able to re¯ect changes in the timing of targetvelocity in the generation of their interception movements. Despite never being in-structed to copy the trajectory of the target, only to intercept it, the timing of thevelocity of the manual interception movements resembled the temporal properties ofthe targetÕs trajectory. It is unlikely however that participants were actually con-trolling the time to peak velocity of their movements. Instead they were probablycontrolling the initial acceleration impulse of their hand movement, and because ofthe constraint that all movements had the same travel distance and movement time,


participants had to couple the braking of their initial impulse with the magnitude ofthe initial impulse to arrive in the interception zone at the appropriate time. Thiscoupling is re¯ected in the time to peak reaching velocity, and the time from peakvelocity to the end of movement.

Experiment 1 was designed to investigate how much visual information about thetargetÕs trajectory was necessary to in¯uence hand kinematic characteristics duringmanual interception and Experiment 2 aimed to determine when this informationwas needed. That is, what are the consequences of the removal of the visual infor-mation about the target trajectory during movement planning? It was demonstratedthat in order to produce this relationship between hand and target peak velocitytiming, the visual information about target velocity must be present in approxi-mately the ®rst 300 ms of the target trajectory. Any target exposure after this timedid not further alter the kinematics of the interception movement. Since the target'smovement time and distance traveled were constant in all conditions, participantscould have chosen to adopt a common interception strategy based on the optimi-zation of movement time. Clearly this was not the case. In Experiment 1, even withshort exposure periods showing the initial part of the target's motion (316 ms) theparticipants were able to acquire a su�cient amount of visual information about thetiming of the targetÕs trajectory to produce kinematically altered hand trajectories. Ifthe initial change in the target's velocity was not processed by the performer, nochanges in time to peak velocity would be expected across the three target velocityconditions. It was expected in the ®nal experiment that when visual informationabout target trajectory was delayed beyond the planning period, participants wouldplan their interception movements with an alternative strategy, perhaps based on thetargetsÕ common movement times. In this case, the production of the interceptionmovement would not be related to the target kinematics. This was precisely what wasfound in Experiment 2 when vision was blocked for the initial portion of targettrajectory; hand peak velocity timing remained the same across all target velocityconditions.

The data from both experiments show that the initial and ®nal phases of themanual interception tasks were driven by functionally di�erent mechanisms. Such adi�erential control of the initial and ®nal phases of the reach movement has previ-ously been suggested by other researchers (Arbib, 1981, 1985; Jeannerod, 1981, 1984;Woodworth, 1899). What is common among many models of visually directedaiming and prehension is that there is an initial phase of the reaching movement thatis preprogrammed and had mostly a ballistic character. The ®nal phase, as the handapproaches the target, is subject to visually based corrections. However, contrary tothese models, in the present studies, the kinematic pro®le of the reaching movementis mediated by visual information about the target's trajectory gathered very earlyduring the trajectory (during the initial 316 ms). The representation of target motionthat is formed on the basis of the early part of the trajectory is then used to guide therest of the movement such that visual information about target motion at the time ofmovement termination is not necessary. The di�erences between existing models ofvisuomotor control (Arbib, 1981, 1985; Jeannerod, 1981, 1984; Woodworth, 1899)and the present data may be related to the fact that participants were aiming toward


moving targets in the current work whereas prior models were based on aimingtoward stationary targets.

However, another possibility that must be acknowledged is that the pattern of our®ndings stems from the fact that these movements were generated via a mousecontrolling a cursor on a computer screen. The control processes involved may havedi�ered from those involved in making three-dimensional movements to real targets.It has been shown that there are fundamental di�erences in how compliant andconstrained computer-aided drawing movements, and unconstrained movements arecontrolled (Desmurget, Jordan, Prablanc & Jeannerod, 1997). Desmurget et al.(1997) concluded that compliant and unconstrained movements involve di�erentcontrol strategies based on the observation that unconstrained movements followeda more curved trajectory path compared to compliant hand paths, which tend to bestraighter. Thus, it cannot be assumed that the previously observed relationshipbetween target speed and interception speed will naturally transfer between para-digms. However, experiments from our lab which have involved grasping real three-dimensional moving targets, and those involving intercepting targets on a computerscreen with a mouse, have shown similar patterns of results (Carnahan & McFadyen,1996; Carnahan, Mason & Sinden, 1997; Mason & Carnahan, 1999).

The control of interception movements relies upon several strategies, employingdi�erent visual information sources depending on the situation (Michaels & Oude-jans, 1992). The results from the present studies suggest that it is highly possible thatdi�erent sources of visual information are used in di�erent phases of the manualinterception of a moving target. The initial, ballistic portion of hand movementseems to be guided by the estimated target arrival time to the interception zone,derived by processing the timing of the target trajectory characteristics during themovement planing period. The visual update of targetÕs velocity for the guidance ofthe hand in the ®nal movement stage was not necessary when the location of theinterception was known to the participants. Thus, the hand motor system must relyon the temporal estimates made earlier in the target trajectory.

References

Arbib, M. A. (1981). Perceptual structures and distributed motor control. In V. B. Brooks (Ed.),

Handbook of physiology: Section I. The nervous system: (Vol. 2) Motor control (pp. 1449±1480).

Bethesda, MD: American Psychological Society.

Arbib, M. A. (1985). Schemas for the temporal organization of behavior. Human Neurobiology, 4, 63±72.

Babler, T. G., & Dannemiller, J. L. (1993). Role of image acceleration in judging landing location of free-

falling projectiles. Journal of Experimental Psychology Human Perception and Performance, 19, 15±31.

Bootsma, R. J., Fayt, V., Zaal, F. T. J. M., & Laurent, M. (1997). On the information-based regulation of

movement What Wann (1996) may want to reconsider. Journal of Experimental Psychology Human

Perception and Performance, 23, 1282±1289.

Bootsma, R. J., & van Wieringen, P. C. W. (1990). Timing an attacking forehand drive in table tennis.

Journal of Experimental Psychology Human Perception and Performance, 16, 21±29.

Carnahan, H., Mason, A. H., Sinden, K. (1997).The contributions of target movement time and velocity

to manual aiming velocity. Journal of Sport and Exercise Psychology, 19, S40.


Carnahan, H., & McFadyen, B. J. (1996). Anticipatory control and coordination when reaching toward

and grasping moving targets. Acta Psychologica, 92, 17±32.

Chie�, S., Fogassi, L., Gallese, V., & Gentilucci, M. (1992). Prehension movements directed to

approaching objects in¯uence of stimulus velocity on the transport and the grasp components.

Neuropsychologia, 30, 877±897.

Desmurget, M., Jordan, M., Prablanc, C., & Jeannerod, M. (1997). Constrained and unconstrained

movements involve di�erent control strategies. Journal of Neurophsiology, 77, 1644±1650.

Gottsdanker, R. M. (1952). The accuracy of prediction motion. Journal of Experimental Psychology, 43,

26±36.

Jeannerod, M. (1981). Intersegmental coordination during reaching at natural visual objects. In J. Lolong

& A. Baddeley (Eds.), Attention and performance IX (pp. 153±168).

Jeannerod, M. (1984). The timing of natural prehension movements. Journal of Motor Behavior, 16,

235±254.

Lee, D., Port, N. L., & Georgopoulos, A. P. (1997). Manual interception of moving targets II. On-line

control of overlapping submovements. Experimental Brain Research, 116, 421±433.

Lee, D. N. (1976). A theory of visual control of braking based on information about time to collision.

Perception, 5, 437±459.

Lee, D. N., Young, D. S., Reddish, P. E., Lough, S., & Clayton, T. M. H. (1983). Visual timing in hitting

an accelerating ball. Quarterly Journal of Experimental Psychology, 35A, 333±346.

MacKenzie, C. L., Marteniuk, R. G., Dugas, C., Liske, D., & Eickmeier, B. (1987). Three-dimensional

movement trajectories in ®tts task implications for control. Quarterly Journal of Experimental

Psychology, 39A, 629±647.

Mason, A. H., & Carnahan, H. (1999). Target viewing time and velocity e�ects on prehension.

Experimental Brain Research, 127, 83±94.

Michaels, C. F., & Oudejans, R. R. D. (1992). The optics and action of catching ¯y balls zeroing out

optical acceleration. Ecological Psychology, 4, 199±222.

Port, N. L., Lee, D., Dassonville, P., & Georgopoulos, A. P. (1997). Manual interception of moving

targets I. Performance and movement initiation. Experimental Brain Research, 116, 406±420.

Rosenbaum, D. A. (1975). Perception and extrapolation of velocity and acceleration. Journal of

Experimental Psychology Human Perception and Performance, 1, 395±403.

Smeets, J. B. J., & Brenner, E. (1995). Perception and action are based on the same visual information

distinction between position and velocity. Journal of Experimental Psychology, 21, 19±31.

Tresilian, J. R. (1995). Perceptual and cognitive processes in time to contact estimation: analysis of

prediction motion and relative judgment tasks. Perception and Psychophysics, 57, 231±245.

vanDonkelaar, P., Lee, R. G., & Gellman, R. S. (1992). Control strategies in directing the hand to moving

targets. Experimental Brain Research, 91, 151±161.

Werkhoven, P., Snippe, H. P., & Toet, A. (1992). Visual processing of optic acceleration. Vision Research,

32, 2313±2329.

Whiting, H. T. A. (1968). Training in a continuous ball throwing and catching task. Ergonomics, 11,

375±382.

Whiting, H. T. A., & Sharp, R. (1974). Visual occlusion factors in a discrete ball catching task. Journal of

Motor Behavior, 6, 11±16.

Woodworth, R. S. (1899). The accuracy of voluntary movement. Psychological Reviews, 3 (Monograph

Supplement), 1± 119.


Target velocity effects on manual interception kinematics

Documents

Transcript of Target velocity effects on manual interception kinematics