HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J Neurophysiol 97: 3148-3151, 2007. First published February 7, 2007; doi:10.1152/jn.01339.2006
0022-3077/07 $8.00

This Article Free upon publication Abstract Full Text (PDF) All Versions of this Article: 97/4/3148    most recent 01339.2006v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by van Donkelaar, P. Articles by Woollacott, M. Search for Related Content PubMed PubMed Citation Articles by van Donkelaar, P. Articles by Woollacott, M.

REPORT

Multiple Saccades Are More Automatic Than Single Saccades

Department of Human Physiology and Institute of Neuroscience, University of Oregon, Eugene, Oregon

Submitted 19 December 2006; accepted in final form 2 February 2007

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
In this paper, we demonstrate that when a peripheral object is foveated by a sequence of multiple saccades, the initial saccade in the sequence is initiated markedly faster than a single accurate saccade to the same object. We suggest that multiple saccades represent a more automatic form of oculomotor planning that may be the result of a reduced influence from the cerebral cortex. To test this, we compared single and multiple saccade characteristics across development. We find that in contrast to the reduction in the latency of single saccades that is observed across development, the latency of initial saccades in multiple saccade sequences is remarkably stable across all age groups. Moreover, the longer the latency of this initial saccade, the more accurate it is, suggesting that there is a relation between the degree of procrastination and the accuracy of the response. Finally, the frequency with which multiple saccades occurred within each age group was positively correlated with the tendency to generate erroneous saccades during a fixation control task. Taken together, the present data suggest that multiple saccades are generated in a more automatic manner than single saccades.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Saccades are high-velocity eye movements that allow targets of interest in peripheral vision to be foveated. A typical saccade to a peripheral target consists of a single primary saccade that covers all or most of the distance to the target, followed shortly thereafter by a small secondary corrective saccade if required. A much less common form of output consists of a series of at least two smaller saccades (Salapatek et al. 1980)—termed a multiple saccade. Such behavior may reflect a more automatic form of saccade execution that occurs before planning is fully complete. Such planning entails a number of sensory, attentional, and motor processes each of which take a finite period of time between target appearance and subsequent saccade generation. Thus modulations in saccadic reaction time reflect changes in the speed with which these processing steps are completed, or in the extreme case, whether they are included at all. In the current study, we have examined the differences in planning single versus multiple saccades across development. We show that the latter are much more frequently observed in children, as has previously been shown in infants (Salapatek et al. 1980), yet the planning processes as inferred from reaction time are remarkably constant across development. We suggest that this implies that multiple saccade sequences are controlled in a more automatic manner than single saccades.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Participants

Forty-one subjects participated—11 adults (6 female; 20–29 yr old); 8 10- to 16-yr olds (3 female), 12 7- to 9-yr olds (6 female), and 10 4- to 6-yr olds (5 female). All had normal or corrected-to normal vision and no sensory, motor, cognitive, or attentional deficits that would affect saccadic eye movements. All subjects or their parents/guardians signed informed consent forms prior to participation and the local university human subjects compliance committee approved the experimental protocol.

Experimental tasks

The saccade task described in this paper was a control condition in a larger study examining the influence of trunk postural control on eye-hand coordination in children (Saavedra et al. 2006). Two components of the task were constrained by this larger study. First, the participant was seated at arms length in front of a display screen on which visual targets were presented; and, second, the main data set was from trials in which the target jumped to the side of the dominant hand (because this was the hand used for pointing). At the onset of each trial a central fixation target (subtending 0.5°) appeared. After a variable delay (500–1,500 ms), a second target was presented in the periphery. On most trials (66%), this target was 9.2–10.8° (depending on arm length) to the side of the dominant hand. In the remaining trials, it appeared with equal probability 4.8–5.4 or 12.3–14.8° to the dominant or 9.2–10.8° to the nondominant side. These trials served as infrequent alternates to keep the participant from anticipating target direction and amplitude. When the peripheral target appeared, the participant was required to make a saccadic eye movement as quickly and accurately as possible to refoveate the target. In a control task, the participants were instructed to maintain central fixation when the peripheral target appeared. Participants completed six blocks of 18 trials for each task with the different target directions and amplitudes pseudorandomly interleaved. Because of this interleaving procedure, there was no predictable sequence of target presentation.

Data recording and analysis

Eye movements were monitored at 60 Hz with a head-free video-based system (ASL Model 5400). We used this system because of our interest in examining the influence of trunk postural control on eye-hand interactions. Although the temporal resolution (16 ms) was relatively course, the latency differences that we observed were substantially larger (e.g., 150 ms). Thus we feel confident that this system accurately captures the saccadic latency effects in which we were interested. It is possible that there are interesting details in the reaction time differences between single and multiple saccades that were hidden by the low temporal resolution of this system. Subsequent studies in our lab designed to further clarify these differences will use a system with higher temporal resolution. The system was calibrated prior to the experiment by having the participant make saccades with the head stable to nine targets forming a rectangle (3 x 3 targets) covering most of the horizontal and vertical range of the display screen. When calibrated the system had a spatial resolution of 0.25° through a horizontal range of ±50° (including head rotation). The eye movement recordings were analyzed using a graphical user interface implemented in Matlab. Saccadic reaction time was the main dependent variable of interest. It was defined as the period of time from the appearance of the peripheral target to the onset of a detectable change in eye position. Single saccades were defined as one discrete change in eye position covering 90% of the distance to the target with no corrective saccade (gain range: 0.91–1.03). Multiple saccades were defined as two or more changes in eye position in which the initial saccade covered <90% of the distance to the target. Trials with large artifacts due to excessive head motion, blinking, or equipment malfunction accounted for >2% of the data and were removed from analysis. The final data set consisted of a total of 3,024 trials (4- to 6-yr-olds: 798 trials; 7–9: 912 trials; 10–17: 566 trials; adults: 748 trials). ANOVAs and linear regression analyses were used to test for statistical significance across the relevant conditions. Only the experimental trials (i.e., targets 9.2–10.8° to the dominant side) were included in the initial analysis. The main effects of interest were then confirmed qualitatively for the trials with alternate targets.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Figure 1, A and B, shows example eye movement trajectories during single and multiple saccades for an 8-yr-old child and a typical adult participant. Whereas reaction time was substantially slower in the child than in the adult during single saccade responses, it was substantially faster and remarkably similar in both the adult and child participant during multiple saccade responses. Moreover, the child produced more trials that consisted of multiple saccades than the adult. In all cases, because the target was visible throughout the response, the final eye position was equally accurate.

View larger version (14K): [in this window] [in a new window]   FIG. 1. Example saccadic trajectories in an 8-yr-old child (A) and an adult (B). Eye position is plotted as a function of time with the change in target position occurring at the beginning of each trace. Black lines show single saccades and gray lines show multiple saccades. Black vertical line, average single saccade reaction time in the 8-yr-old child; gray vertical line, average multiple saccade reaction time in the 8-yr-old child. Notice that the adult initiates single saccades much more quickly than the child but initiates multiple saccades at about the same latency as the child. Group means for frequency of multiple saccade trials across the different ages (C). Group means for the median reaction times during single (filled circles) and multiple (open circles) saccades across the different ages (D). Error bars, 1 intersubject SE. Numbers above data points represent the number of trials making up the mean.

View larger version (15K): [in this window] [in a new window]   FIG. 2. A: group means for frequency of saccade intrusions during the fixation condition across the different ages. Error bars, 1 intersubject SE. Scatter plot of percentage of trials with multiple saccades versus percentage of fixation trials with saccadic intrusions for the 4- to 6-yr olds (B), 7- to 9-yr olds (C), 10- to 17-yr olds (D), and adults (E). Line of best fit is from linear regression analyses.

Most of the multiple saccade trials consisted of only two saccades with only a small percentage of trials consisting of three or more saccades (4- to 6-yr-olds: 4 trials; 7–9: 4 trials; 10–17: 1 trial; adults: 0 trials). In addition, single saccades (covering 90% of the distance to the target) followed by a subsequent corrective saccade [4- to 6-yr-olds: 7.14% (57 trials); 7–9: 2.97% (27 trials); 10–17: 6.17% (39 trials); adults: 1.07% (8 trials)] displayed reductions in reaction time across development that were similar to those observed for single saccades (4- to 6-yr-olds: 301 ± 56 ms; 7–9: 289 ± 42 ms; 10–17: 271 ± 66 ms; adults: 257 ± 49 ms). It should be noted that previous studies investigating the incidence of corrective saccades have shown them to be much more common when target amplitude exceeds 10° (Becker and Fuchs 1969; Prablanc and Jeannerod 1975). Finally, single saccades that covered <90% of the distance to the target were rarely observed.

Qualitative analysis of the saccades made to the alternate targets demonstrated that multiple saccades were still produced with a frequency similar to that for the main target [4- to 6-yr-olds: 18% (32 trials); 7–9: 11% (24 trials); 10–15: 6% (9 trials); adults: 1.5% (3 trials)]. Furthermore, whereas the single saccades to the alternate targets demonstrated a similar reduction in latency across development (4- to 6-yr-olds: 326 ± 46 ms; 7–9: 299 ± 37 ms; 10–15: 285 ± 45 ms; adults: 280 ± 47 ms), the multiple saccades were again remarkably consistent in the children and adults (4- to 6-yr-olds: 195 ± 31 ms; 7–9: 196 ± 27 ms; 10–15: 185 ± 35 ms; adults: 189 ± 41 ms). Thus based on this pattern of results, we feel confident that multiple saccades do not simply result from the differential application of predictive strategies by children relative to adults.

Analysis of the reaction times of the second saccade in the multiple saccade sequences demonstrated that they were substantially longer than those of single saccades across all four subject groups [mean single saccade RT: 316 ± 23 ms; mean 2nd saccade in multiple sequence RT: 414 ± 35 ms; F(1,35) = 41.106, P < 0.0001]. This implies that it is neither a normal corrective saccade nor the originally planned but preempted single saccade but rather represents a separately planned voluntarily saccade generated to foveate the target after the inaccurate initial saccade in the sequence.

To gain a better understanding of how latency and amplitude were related, the difference on a trial-by-trial basis in reaction time for the initial saccade in a multiple saccade sequence relative to the median latency for single saccades and the analogous difference for saccade amplitude was compared. Regression analyses demonstrated that there was a significant linear correlation between the latency and amplitude differences in the three groups of children (4–6: r = 0.29; 7–9: r = 0.51; 10–17: r = 0.48; P < 0.002) but not in the adults (r = 0.23). This demonstrates that as the reaction time period progresses, the amplitude of the saccade gets more fully specified. A similar trend was observed for the alternate targets—suggesting that it is not simply due to larger amplitude saccades taking longer to prepare.

Data from a control condition in which participants were required to maintain fixation demonstrated a developmental progression in this ability [F(3,37) = 18.133, P < 0.0001]. Post hoc Tukey's tests showed that there were progressively fewer breaks from fixation as age increased. Linear regression analyses revealed significant correlations between the frequency of multiple saccades and the frequency of saccadic intrusions during the fixation condition within the two youngest age groups (4–6: r = 0.80; 7–9: r = 0.62; P < 0.02) but not the older children (r < 0.01) or adults (r = 0.49).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
In the present study, we have demonstrated a substantial difference in the time required to prepare a single saccade to a peripheral target compared with the initial saccade in a sequence of multiple saccades to the same target. The multiple saccades were much more frequent in young children compared with older children and adults; however, the reaction time of this class of saccades remained remarkably invariant across age; whereas that for single saccades showed a typical developmental improvement (Fukushima et al. 2000; Salman et al. 2005). Previous studies that have investigated voluntarily planned sequences of multiple saccades have shown that reaction time increases with the number of saccades in the sequence (Zingale and Kowler 1987); whereas, we have shown a reduction in reaction time for multiple saccades. This is most likely due to the difference in purposefully planning a sequence of saccades compared with eliciting them in a more involuntary manner as the result of incomplete preparation. Saccadic reaction time also increases with saccade amplitude; however, this is only true for extremely small or large changes in target position (Kalesnykas and Hallett 1994; Wyman and Steinman 1973), so it is unlikely to account for the current results.

It is possible that multiple saccades represent express saccades (Pare and Munoz 1996) that are initially hypometric during early development and become adapted as the CNS matures. If this is the case, then one would expect to see some tendency for express saccades to be hypometric in adulthood. However, the metrics of express saccades are generally indistinguishable from those of visually guided saccades with normal latencies (Rohrer and Sparks 1993). Alternatively, multiple saccades could represent normal primary-corrective saccade output that is simply more hypometric during early development. However, we are not aware of any evidence which demonstrates that normal primary-corrective saccades have markedly faster latencies than single saccades. In fact, we explicitly examined responses containing truly corrective saccades and found that they had reactions times similar to those for single saccades. Furthermore, when we directly examined the latency of the second saccade in the multiple saccade responses, we found that they occurred much later in time than a normal corrective saccade would have. Thus we feel confident that this behavior is neither an immature form of express saccades nor an extreme version of normal primary-corrective saccadic output.

Saccades occur as the result of a release from inhibition from higher-level decision-making centers in the cortex on lower level saccade generation circuits in the brain stem. In particular, the balance of activity between saccade and fixation cells in the intermediate and deep layers of the superior colliculus determines whether the eye is released from fixation. This balance is controlled by direct input from decision-making activity in the frontal eye field (FEF) (Schall and Thompson 1999) and lateral intraparietal area (Gaymard et al. 2003), and indirect input from the FEF via the caudate nucleus and substantia nigra pars reticulata in the basal ganglia (Schiller and Tehovnik 2005). Although we can only speculate as to the nature of the neural underpinnings of multiple saccade behavior, one possibility is that it reflects a release from fixation within the brain stem prior to the completion of planning and decision-making at the level of the cortex. Indeed, the former is more fully developed and myelinated than the latter by the end of the first decade (Barkovich 2000; Klingberg et al. 1999; Matsuzawa et al. 2001; Olesen et al. 2003). Moreover, our demonstration of a positive relation between the latency and amplitude of the initial saccade in a multiple saccade sequence in children is consistent with the idea that saccades become more fully planned when the cortically mediated inhibition is stronger. In fact, it has been suggested that the purpose of the descending input from the cortex is to procrastinate so that accurately planned responses are elicited (Carpenter 2004; Neggers et al. 2005; Reddi and Carpenter 2000). This lack of control in the release from fixation is much more apparent at younger ages and is consistent with the development of the ability to maintain fixation (Gowen and Abadi 2005). This is supported by the fact that the frequency of multiple saccade occurrence is highly correlated with the (lack of) ability to maintain fixation in the younger children but not the older children or adults.

In conclusion, we have shown that the initial saccade in a multiple saccade sequence is released from fixation markedly sooner than a single saccade directed to the same target. Although the frequency of this type of saccade decreases substantially with age, the rapidity with which it is generated remains remarkably invariant. This suggests that multiple saccades are controlled in a more automatic manner than single saccades and may be the associated with a reduced influence from oculomotor centers in the cerebral cortex.

	GRANTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
This work was supported by National Institute of Neurological Disorders and Stroke Grant 5R01NS-038714-07 awarded to M. Woollacott.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
The authors thank C. Boydston for providing computer support.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Address for reprint requests and other correspondence: P. van Donkelaar, Dept. of Human Physiology, 122C Esslinger Hall, Eugene OR 97403-1240 (E-mail: paulvd{at}uoregon.edu)

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Barkovich AJ. Normal development of the neonatal and infant brain, skull, and spine. In: Pediatric Neuroimaging, edited by Barkovich AJ. Philadelphia: Lippincott Williams andWilkins, 2000, p. 13–69.

Becker W, Fuchs AF. Further properties of the human saccadic system: eye movements and correction saccades with and without visual fixation points. Vision Res 9: 1247–1258, 1969.[CrossRef][ISI][Medline]

Carpenter RHS. Contrast, probability, and saccadic latency; evidence for independence of detection and decision. Curr Biol 14: 1576–1580, 2004.[CrossRef][ISI][Medline]

Fukushima J, Hatta T, Fukushima K. Development of voluntary control of saccadic eye movements. I. Age-related changes in normal children. Brain Dev 22: 173–180, 2000.[CrossRef][ISI][Medline]

Gaymard B, Lynch J, Ploner CJ, Condy C, Rivaud-Pechoux S. The parieto-collicular pathway: anatomical location and contribution to saccade generation. Eur J Neurosci 17: 1518–1526, 2003.[CrossRef][ISI][Medline]

Gowen E, Abadi RV. Saccadic instabilities and voluntary saccadic behaviour. Exp Brain Res 164: 29–40, 2005.[CrossRef][ISI][Medline]

Kalesnykas RP, Hallett PE. Retinal eccentricity and the latency of eye saccades. Vision Res 34: 517–531, 1994.[CrossRef][ISI][Medline]

Klingberg T, Vaidya CJ, Gabrieli JD, Moseley ME, Hedehus M. Myelination and organization of the frontal white matter in children: a diffusion tensor MRI study Neuroreport 10: 2817–2821, 1999.[ISI][Medline]

Matsuzawa J, Matsui M, Konishi T, Noguchi K, Gur RC, Bilker W, Miyawaki T. Age-related volumetric changes of brain gray and white matter in healthy infants and children. Cereb Cortex 11: 335–342, 2001.[Abstract/Free Full Text]

Neggers SF, Raemaekers MA, Lampmann EE, Postma A, Ramsey NF. Cortical and subcortical contributions to saccade latency in the human brain. Eur J Neurosci 21: 2853–2863, 2005.[CrossRef][ISI][Medline]

Olesen PJ, Nagy Z, Westerberg H, Klingberg T. Combined analysis of DTI and fMRI data reveals a joint maturation of white and grey matter in a fronto-parietal network. Brain Res Cogn Brain Res 18: 48–57, 2003.[CrossRef][Medline]

Pare M, Munoz DP. Saccadic reaction time in the monkey: advanced preparation of oculomotor programs is primarily responsible for express saccade occurrence. J Neurophysiol 76: 3666–3681, 1996.[Abstract/Free Full Text]

Prablanc C, Jeannerod M. Corrective saccades: dependence on retinal reafferent signals. Vision Res 15: 465–469, 1975.[CrossRef][ISI][Medline]

Reddi BA, Carpenter RH. The influence of urgency on decision time. Nat Neurosci 3: 827–830, 2000.[CrossRef][ISI][Medline]

Rohrer WH, Sparks DL. Express saccades: the effects of spatial and temporal uncertainty. Vision Res 33: 2447–2460, 1993.[CrossRef][ISI][Medline]

Saavedra S, Woollacott, M, van Donkelaar, P. Effects of postural support on eye hand interactions across development. Exp Brain Res In press

Salapatek P, Aslin RN, Simonson J, Pulos E. Infant saccadic eye movements to visible and previously visible targets. Child Dev 51: 1090–1094, 1980.[CrossRef][ISI][Medline]

Salman MS, Sharpe JA, Eizenman M, Lillakas L, Westall C, To T, Dennis M, Steinbach MJ. Saccades in children. Vision Res 46: 1432–1439, 2005.[CrossRef][ISI][Medline]

Schall JD, Thompson KG. Neural selection and control of visually guided eye movements. Annu Rev Neurosci 22: 241–259, 1999.[CrossRef][ISI][Medline]

Schiller PH, Tehovnik EJ. Neural mechanisms underlying target selection with saccadic eye movements. Prog Brain Res 149:157–171, 2005.[ISI][Medline]

Wyman D, Steinman RM. Letter: latency characteristics of small saccades. Vision Res 13: 2173–2175, 1973.[CrossRef][ISI][Medline]

Zingale CM, Kowler E. Planning sequences of saccades. Vision Res 27: 1327–1341, 1987.[CrossRef][ISI][Medline]

This Article Free upon publication Abstract Full Text (PDF) All Versions of this Article: 97/4/3148    most recent 01339.2006v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by van Donkelaar, P. Articles by Woollacott, M. Search for Related Content PubMed PubMed Citation Articles by van Donkelaar, P. Articles by Woollacott, M.