Contributed by: Birger Johansson, Lund University

Introduction
To engage in a continuous interaction with a dynamic world, it is essential to anticipate how the world will change to select and control actions depending, not on the past, but on the future. One of the earliest interactions to develop in infants is the ability to look at moving objects.

Smooth Pursuit
Smooth pursuit occurs when the eyes track a moving target with a continuous motion, which ideally is centered directly on the target and makes the image of the target stationary on the retina. Smooth pursuit is complicated by the fact that the initial visual processing in the human brain delays the stimulus by approximately 100 ms before it reaches the visual cortex (Wells & Barnes, 1998, Fukushima et al., 2002). If smooth pursuit movements were solely controlled by the position error on the retina, the eye would constantly lag a moving target.

Target prediction
To overcome this problem, the brain makes use of prediction (Deno et al., 1995, Mehta & Schaal, 2002, Poliakoff, Collins & Barnes, 2004). Because eye control is based on predicted target location rather than the actual target position which is not yet known, it is possible for the gaze to overshoot when the target disappears unexpectedly or changes direction. This does not happen when the disappearance of the target is controlled by the subject, for example by a button (Stork, Neggers & Müsseler, 2002). In this case, the gaze velocity slows down before the target disappears which shows that their expectations control the velocity of the smooth pursuit. Smooth pursuit movements cannot normally be generated without a moving stimulus, except that it can appear a short moment before a target is expected to appear (Poliakoff, Collins & Barnes, 2005, Wells & Barnes, 1998). Subjects can learn to anticipate the velocity a target will have when it appears, and in the case of several targets, subjects can produce predictive eye movements of appropriate velocity when one of the targets is cued (Poliakoff, Collins & Barnes, 2004).

Smooth Pursuit in Infants
Infants as young as one month can exhibit smooth pursuit, but only at the speed of 10 degrees/s or less and with low gain (Roucoux et al., 1983). A three month old infant does not follow if a target abruptly changes its direction of movement. Instead it continues in the original direction for a quarter of a second before adjusting its eye movements (Aguiar & Bailargeon, 1999). However, at five months of age, the infant learns the abrupt turn and its lag is reduced. The ability to smoothly track a target thus develops very rapidly, and at five month of age this ability approaches that of adults (von Hofsten & Rosander 1997).

Before the infant can use smooth pursuit, it follows moving targets using small saccade movements that rapidly move the gaze from one position to another (Dayton & Jones, 1964). As the smooth pursuit system develops, these saccades become less frequent, but are still used to catch up if the lag becomes too large.

Target Occlusion in Infants
Of particular interest is how infants behave when the target disappears, for example behind an occluder. According to Piaget (1937), the child is able to predict that a train that disappears at one end of a tunnel will appear at the other end. This can either be explained by a tracking mechanism that continues to track the motion of the train when it has disappeared, or as form of event learning where the child learns to predict that the disappearing train predicts the subsequent reappearance of the train. There exits evidence for both types of mechanisms. For example, Wentworth and Haith (1998) found that three-month-old infants could learn spatiotemporal expectations. Other researchers have mainly studied the ability of infants to smoothly track a moving target using predictive models of the motion (von Hofsten & Rosander & 1997, Rosander & von Hofsten, 2004).

Contrary to earlier speculations, infants do not continue to track an occluded objects with smooth pursuit. Instead the tracking stops and one or two saccades are made to the other side of the occluder (Rosander & von Hofsten, 2004). These saccades are made to anticipate when the object reappear not when it disappears. Infants that are 7-9 weeks old continue to look at the edge of the occluder where the object disappears for 1 second before finding it again (Rosander & von Hofsten, 2004). Infants that are 12 weeks old move their eyes as soon as the target becomes visible again. This delay decreases with each trial which indicates that the infant starts to anticipate where the objects will reappear. Some of these effects have been seen in younger infant as well, but they have not been reliable. It is possible that the younger infant would have performed better if the object was made invisible instead of occluded since the occluder distracts attention from the target (Jonsson & von Hofsten 2003).

References
Aguiar, A. & Baillargeon, R. (1999). 2.5-month-old infants’ reasoning about when objects should and should not be occluded Cognitive Psychology, 39, 2, 116–157.

Dayton, G. O. & Jones, M. H. (1964). Analysis of characteristics of the ?xation re?ex in infants by use of direct current electro-oculograpy. Neurology,14,1152-1156.

Deno, D. C., Crandall, W. F., Sherman, K. & Keller, E. L. (1995). Characterization of prediction in the primate visual smooth pursuit system. BioSystems, 34, 107–128.

Jonsson, B. & von Hofsten, C. (2003). Infants’ ability to track and reach for temporarily occluded objects. Development Science, 6, 88–101.

Mehta, B. & Schaal, S. (2002). Forward models in visuomotor control. J Neurophysiol, 88, 942–953.

Poliakoff, E., Collins, C. J. S. & Barnes, G. R. (2004). Target selection for predictive smooth pursuit eye movements. Exp Brain Res, 155, 129-133

Poliakoff, E., Collins, C. J. S. & Barnes, G. R. (2005). Attention and selection for predictive smooth pursuit eye movements. Cognitive Brain Research, 25, 688-700.

Roucoux, A., Culee, C. & Roucoux, M. (1983). Develop of ?xation and pursuit eye movements in human infants. Behavioural brain research, 10, 133–139.

Rosander, K. & von Hofsten, C. (2004). Infants’ emerging ability to represent occluded object motion. Cognition, 91, 1, 1–22.

Stork, S., Sebastian, F. W., Neggers, S. F. W, & Müsseler, J. (2002). Intentionally-evoked modulations of smooth pursuit eye movements. Human Movement Science, 21, 335–348

von Hofsten, C. & Rosander, K. (1997). Development of smooth pursuit tracking in young infants. Vision Research, 37, 13, 1799–1810.

Wells, S. G. & Barnes, G. R. (1998). Fast, anticipatory smooth-pursuit eye movements appear to depend on a short-term store. Exp Brain Res, 120, 129–133.

Wentworth, N. & Haith, M. M. (1998). Infants’ acquisition of spatiotemporal expectations. Developmental Psychology, 34, 2, 247–257.