Binocular vision and prehension in middle childhood

Abstract

Binocular cues have been shown previously to make an important contribution to the control of natural prehensile movements in adults [Visual Cognition 4 (1997) 113, Vision Research 32 (1992) 1513, Neuropsychologia 38 (2000) 1473]. The present study examined the role of binocular vision in the control of prehension in middle childhood. Fourteen children aged 5–6 years, and 16 children aged 10–11 years reached out and grasped different sized objects at different distances, in either binocular or monocular viewing conditions. In contrast to adult data, many of the principal kinematic indices of the children’s reaches were unaffected by the removal of binocular information. The older children, like adults, spent an increased amount of time in the final approach to the object when only monocular information was available. However, both peak wrist velocities and peak grip apertures were unaffected by the removal of binocular information and continued to scale with object properties in the normal way. These results suggest that the use of binocular cues to control prehensile movements is not yet mature at the age of 10–11 years.

Introduction

It is clear that vision plays a key role in the control of natural adult prehension. Prehensile movements are thought to consist of two independent, though coupled, components: the transport component refers to the positioning of the hand in a suitable spatial location and the grasp component refers to the appropriate posturing of the fingers to pick up the object [16], [27]. Various kinematic ‘markers’ of these components which occur well before the object is actually grasped vary closely as a function of the distance and size of the object, indicating that these properties are recovered from visual information. The peak velocity of the wrist, for example, increases as a linear function of the distance of the object from the starting position of the hand [15], [16], and the peak opening of the grasp (before it is closed onto the object) increases as a linear function of the object’s size along the axis on which it is to be grasped [7], [15], [16].

Analysis of the kinematic parameters of infants’ reaches shows that they begin to demonstrate similar properties to those of adults early on in life. By just 9 months of age, infants make prehensile movements in which the grasp opening is pre-configured to reflect the size and orientation of the target object [13], [20] and they begin to resemble the ‘stereotyped’ reaches of adults (with smooth, bell-shaped velocity profiles) by 2 years of age [17]. However, the control of prehension continues to be refined and developed through early and middle childhood. Kuhtz-Buschbeck et al. [19] examined prehensile movements made by children aged between four and 12 years. They found that younger children open their grip aperture proportionally wider than older children (which may allow for a greater ‘margin of error’) and that the closing of the grasp and the approach of the hand to the object became increasingly closely co-ordinated with increasing age (see also [18], [23], [24]). They also found that only the oldest children were able to scale their grip apertures appropriately with object size when on-line visual feedback of the hand and object was unavailable (i.e. when object properties, recovered from vision prior to movement onset, had to be retained; see also [26]).

These findings suggest that visually guided prehension becomes increasingly ‘fine-tuned’ throughout middle childhood. However, very little is known about the nature of the visual information used to control reaching in this period, and whether this also changes as prehension develops. The results of several studies suggest that in adults binocular information is particularly important for the control of natural prehensile movements [14], [31], [34]. Binocular cues (including binocular disparities, angle of convergence and possibly vertical disparities) are a strong candidate for supporting accurate prehension. They can be used to specify the full metric properties of the visual scene, including absolute distance and size [3], [6], [25], and can provide very precise information about relative depth which might be particularly useful in the on-line control of movements [4]. Also, Sakata and co-workers have shown that many disparity sensitive ‘manipulation-related’ cells in the posterior parietal cortex of primates are selective for 3D surface orientation and for an object’s 3D axis orientation [28], [29], [33]. Moreover, binocular cues have been shown to be of critical importance for the control of prehension in the neurological patient d.f., who despite being a profound visual form agnostic, is able to make normal visually guided prehensile movements [5], [21].

In human adults, Servos, Goodale and Jakobson [31] found that the removal of binocular cues (by covering one eye) had significant effects on many of the kinematic parameters of participants’ reaches—including slower peak wrist velocities, smaller peak grip apertures and longer deceleration phases—and concluded that binocular information was critical in the control of the transport and grasp components. More recent studies have questioned the generality of this conclusion, but suggest that binocular information may be selectively involved in the on-line control of the movement, and in the control of the grasp [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34]. Watt and Bradshaw [34], for example, reported that peak wrist velocities were unaffected by the removal of binocular information, whereas under monocular conditions participants opened their grasp wider, and spent more time in final approach to the object (see also [14], [21]). However, the role of binocular information in the control of prehension in middle childhood remains to be determined. To address this question, we determined the effects of removing binocular information (cf. [31]) on the kinematic parameters of prehensile movements of children in two age groups: 5–6 years old and 10–11 years old. These age groups should provide an interesting developmental cross-section because in 5–6 year olds prehension has been shown not to be developed fully, whereas by 11 years of age, prehensile movements resemble more closely those of adults [19]. The precise nature of the information available from binocular vision means that its contribution might be expected to increase as reaching and grasping movements become increasingly refined.