Early Human Development 87 (2011) 289296

Contents lists available at ScienceDirect

Early Human Development

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev

Reach on sound: A key to object permanence in visually impaired children

Elisa Fazzi a,, Sabrina Giovanna Signorini b, Monica Bomba a, Antonella Luparia b, Jose Lanners c, Umberto Balottin d
a

Unit of Child Neurology and Psychiatry, Spedali Civili, Mother and Child Department, University of Brescia, Italy Unit of Child Neurology and Psychiatry and Child Neuro-ophthalmology, IRCCS C. Mondino Foundation, University of Pavia, Pavia, Italy Fondazione Robert Hollman, Cannero Riviera (VB), Italy d Unit of Child Neurology and Psychiatry, IRCCS C. Mondino Foundation, University of Pavia, Pavia, Italy
b c

a r t i c l e

i n f o

a b s t r a c t
Background: The capacity to reach an object presented through sound clue indicates, in the blind child, the acquisition of object permanence and gives information over his/her cognitive development. Aim: To assess cognitive development in congenitally blind children with or without multiple disabilities. Study design: Cohort study. Subjects: Thirty-seven congenitally blind subjects (17 with associated multiple disabilities, 20 mainly blind) were enrolled. Outcome measures: We used Bigelow's protocol to evaluate reach on sound capacity over time (at 6, 12, 18, 24, and 36 months), and a battery of clinical, neurophysiological and cognitive instruments to assess clinical features. Results: Tasks n.1 to 5 were acquired by most of the mainly blind children by 12 months of age. Task 6 coincided with a drop in performance, and the acquisition of the subsequent tasks showed a less agehomogeneous pattern. In blind children with multiple disabilities, task acquisition rates were lower, with the curves dipping in relation to the more complex tasks. Conclusions: The mainly blind subjects managed to overcome Fraiberg's conceptual problem i.e., they acquired the ability to attribute an external object with identity and substance even when it manifested its presence through sound only and thus developed the ability to reach an object presented through sound. Instead, most of the blind children with multiple disabilities presented poor performances on the reach on sound protocol and were unable, before 36 months of age, to develop the strategies needed to resolve Fraiberg's conceptual problem. 2011 Elsevier Ireland Ltd. All rights reserved.

Article history: Received 24 September 2010 Received in revised form 16 January 2011 Accepted 18 January 2011 Keywords: Visually impaired children Object permanence Reach on sound Cognitive development

1. Introduction In the construction of sensorimotor intelligence, vision is the most important sense [13]. Vision allows the child to develop an awareness of his external world: it is through vision that the child learns to know and recognise his mother's face and acquires object permanence and the earliest notions of causality. The sensorimotor development of the blind child [46] differs from the pattern that emerges in the normally-sighted child due to the presence, in the former, of specic neuromotor and cognitive difculties [711]. In accordance with other authors, the mobility that normally follows each postural achievement is considerably delayed in blind children without multiple disabilities. A visual function decit

Corresponding author. Tel.: + 39 0303995724; fax: + 39 0303995723. E-mail address: elisa.fazzi@gmail.com (E. Fazzi). 0378-3782/$ see front matter 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2011.01.032

implies difculties developing an awareness of physical causality, of spatial relations and, above all, of object permanence [4]. Fraiberg [12,13] maintains that the capacity to reach an object that is presented through sound is, in the blind child, the behavioural indicator of the acquisition of object permanence. Also in the normally-sighted child, sound plays a developing role in the construction of the object; since Piaget's stage 4, sound becomes a more and more powerful cue in directing the child's active search for the object [14]. While locomotion plays a necessary facilitative role in the development of object permanence [15], motor achievements are in turn supported by the progressive construction of the object through sound. In a previous study we reported that the onset of selfinitiated mobility in blind children is related to the demonstration of each child's ability to reach out and take an object presented by sound clue alone. In particular, we observed that all of the subjects of our group of blind children learned to walk independently only after gaining the ability to nd their bearings when guided by a sound [9]. Our survey provided further conrmation of the ndings of Fraiberg

290

E. Fazzi et al. / Early Human Development 87 (2011) 289296 Table 1 Socio-demographic and clinical features of the two groups. Group 1 n = 20 Gender: male n(%) Gestational age: mean(S.D.) Born: n(%) At term Moderately preterm Severely preterm Visual damage: n(%) Complete bilateral blindness Light perception Type of visual damage: n(%) Disorders of external parts of the eye Retinal disorders Leber's congenital amaurosis Joubert syndrome Retinal haemorrhage Retinopathy of prematurity Retinal dysplasia Cerebral visual impairment Optic atrophy Norrie disease Cognitive level (ReynellZinkin Scales): n(%) Normal Slight mental delay Severe mental delay n.s. = not signicant. S.D. = standard deviation. 10(50%) 38.9(0.8) 17(85%) 2(10%) 1(5%) 10(50%) 10(50%) 1(5%) 12(60%) 0 1(5%) 1(5%) 1(5%) 0 3(15%) 1(5%) 20(100%) 0 0 Group 2 n = 17 7(41.2%) 36.8(1.0) 8(47.1%) 5(29.4%) 4(23.5%) n.s. 9(52.9%) 8(47.1%) 0 0 1(5.9%) 0 0 0 16(94.1%) 0 0 0.000 0 4(23.5%) 13(76.5%) n.s. n.s. n.s. p

and Bigelow who demonstrated that reach on sound is the crucial moment in terms of the blind child's access to the world of representation, as well as being an activity that indicates the child's readiness to achieve locomotion. To deepen this aspect, Freiberg afrms that the blind child must be able to overcome a conceptual problem in order to acquire the reach on sound capacity [13]: since, for the blind child, objects do not possess the sensory and perceptible qualities which make them recognisable to sighted children, in order to conceive of grasping an object with a sound, he/she must manage without the merely visual information relating to that object and form a mental image of it [14,15]. From Fraiberg's descriptions [12,13], the acquisition of this ability emerges as a signicant moment in the development of the blind child, being at once a condition of and a catalyst for all his subsequent development. This is why it is particularly important to promote reach on sound in the blind child: it represents the solution to Fraiberg's conceptual problem, guaranteeing the attribution of substance even to objects that are presented exclusively through sound. The importance of reach on sound is analysed by Bigelow [14 16], who has developed an 11-item protocol specically designed to investigate the sequential development, in blind children, of the ability to seek and reach objects presented through sound and tactile clues. The progressive acquisition and mastering of these items appear to be related to a growing awareness of object permanence in the absence of sight. Furthermore, Bigelow establishes a correlation between the Piaget's stages in the acquisition of object permanence and the stages in the acquisition of reach on sound, suggesting that the eleven tasks in the reach on sound protocol reect the process of the acquisition of sensorimotor intelligence, and thus represent a measure of cognitive development. In the presence of multiple developmental disabilities this process of acquisition is likely to be altered. We set out to evaluate the development of the reach on sound function in a sample of congenitally blind children without and with other neuromotor and/or cognitive disabilities, paying particular attention to differences and/or similarities in their acquisition of this ability. 2. Methods 2.1. Sample description The study sample was made up of 37 congenitally blind children, consecutively referred to the Centre of Child Neuro-ophthalmology at the C. Mondino Institute of Neurology in Pavia (Italy) for diagnostic and therapeutic assessment. The Institutional Ethical Committee approved the research project. The parents of all the enrolled children signed an informed consent form in order to participate to the study. Congenital blindness was diagnosed according to the International Classication of Functioning, Disability and Health ICF [17]; cerebral visual impairment was dened according to Good [18,19]. 2.2. Procedure Visual impairment was assessed on the basis of clinical behaviour (absence of any response to light or minimal perception of light, or visual acuity less than 0.02, as evaluated using Teller Acuity Cards [20]), and a neuro-ophthalmological assessment according to previous records [21]. In addition, all the subjects underwent a neurological examination and neurophysiological investigations (VEPs, BAEPs and EEG) in order to establish whether their blindness was isolated or associated with other neuromotor or sensory disabilities. The sociodemographic and clinical features of the sample are detailed in Table 1.

All the 37 subjects underwent the assessment of developmental level through the administration of The ReynellZinkin Scales: Developmental Scales for Young Visually Handicapped Children [22]. For this study, we used the total scores and the cut-offs proposed by the authors to qualify normal mental development, slight and severe mental delay. Mental development was assessed at the followup session of 12 months of age. On the basis of the results of the clinical and instrumental examinations the subjects were divided into two groups (see Table 1): Group I of blind children (10 males and 10 females) with a normal mental development at the ReynellZinkin Scales and without associated motor handicaps or focal signs at the neurological examination. For brevity, we will call these subjects as mainly blind; they are, indeed, children in which blindness, at the moment of the evaluation, is the outstanding problem; the term mainly is used in order to not exclude the possibility of a future development of a neuropsychological problem, but this kind of handicap wasn't present at the evaluation time. We presume that, in the rst years of life, prior cognitive hitches might be present but not yet describable. Group II of children (7 males and 10 females) with associated handicaps (16 with cerebral visual impairment and 1 with Jubert syndrome). There was clinical evidence of cerebral palsy with developmental delay in 88% of the Group II subjects and of developmental delay only in the remaining 12% (p = 0.000; see Table 1). No child in either group showed hearing impairment at the BAEPs. The subjects in Group I were rst assessed at a mean age of 9.1 months (SD = 5.1; range = 424) and those in Group II at a mean age of 10.8 months (SD = 8.2; range = 628). They were followed up through longitudinal controls every three months for the period ranging from the 6th to the 36th month of age, with all the children in the two groups reaching the last follow up at 36 months. Infants were tested individually, in the presence of their mother, in a dedicated room in our Institution, by a familiar experimenter (always the same

E. Fazzi et al. / Early Human Development 87 (2011) 289296

291

operator in all the control-sessions). Each follow up comprised an interview with the parents, a neurological examination, ascertainment of the achievement, or otherwise, of the main gross motor milestones, a video-recording to explore the child's spontaneous behaviour in the presence of the mother. The latter was performed during a separate fteen minute session of observation of both the baby and his/her mother staying together.

Table 3 Correlation between the 11 tasks in the reach on sound protocol and the stages in the development of object permanence according to Piaget [16]. Reach on sound task n. 1 From 2 to 7 Piaget stages Age of acquisition (months) 4/58/9 8/911

2.3. Reach on sound protocol In this study we focused on the ability to reach objects presented through sound clue. This ability was evaluated through the Bigelow reach on sound protocol [16], the eleven tasks of which are described briey in Table 2. The sequence of tasks set out in the protocol parallels the process of the acquisition of object permanence, each task belonging, according to Piagetian thought, to a specic stage in this process (Table 3). In accordance with Bigelow's indications [16], the procedure was explained to the mother, who was asked not to give any specic cues to the child during the tasks. Nevertheless the mother was asked to paraphrase the experimenter's verbalizations when appropriate or necessary to elicit the best possible response from the child. The infant was on the oor in a sitting position (with or without back support, accordingly with the age and the motor development of the subject). Each task was tried to be presented for a minimum of 3 trials in each session, with the exception of Tasks 1, 3, 5 and 7. We used Bigelow's criterion for mastery to consider a task successful when 50% or more of the trials were correct at a certain follow-up time and also at the subsequent control. The task was judged successful if the infant reached directly to the toy used as the target and contacted it without any scanning or groping movements. Reaches directly in the direction of the toy or its sound which did not result in the child securing the toy were also judged to be successful if the reaches were off target by approximately 5 cm or less. Exceptions to this second criterion for success were when the child had continuous tactile contact with the toy (Task 1), there was continuous movement of a continuously sounding toy (Task 4), or there were based on the child's rst response to the trial presentation. All the tasks were presented in the order described in Table 2 in all the follow-up sessions. The children's responses to the tasks were scored from videotapes of the sessions.

III: Start of object permanence IV: Active searching for the object that has disappeared but without taking into account its series of visible movements 8 Transitional stage From 9 to 10 V: The child now takes into account the series of visible movements of the object, but not yet its invisible movements Completed protocol VI: The child has acquired the concept of object permanence

11/1218

1824

2.4. Reach on sound tools and targets The stimulus toys used were the ones of the experimenter's and were the same for all the subjects. Moreover, we asked the mother to bring one or two familiar child's toys in order to optimise the children's response to the tasks. As indicated by Bigelow's protocol [16], mechanical-sounding toys (such as a buzzing robot or a music box), manually operated sound toys (such as a hand bell and a giraffe squeeze toy), and toys without sound (like a doll and a rubber ball) were used. The children's toys used as targets were squeeze toys, rattles, and stuffed sounding animals. The toys used in the intermittent sound task (see Task 10 and 11) were manually operated sound toys which were manipulated to produce a sound approximately every 2 s. A metal screen, 44 cm square cloths, and small pillows were used as coverings for the stimulus objects in the cover object task 6.

2.5. Statistical analysis Continuous variables were expressed as mean and standard deviation (SD), and were compared using the parametric T-test. Categorical variables have been presented as counts and percentages and compared with the test. A p b 0.05 was considered statistically signicant. The statistical analysis was performed using the statistical package SPSS 13.0 (SPSS Inc. Corp., USA).

Table 2 Reach on sound protocol. Task n. 1 2 3 4 5 6 7 8 9 10 11 Tasks' description Reaching of a non-sonorous object that touches the child's body ( continuous tactile stimulation in the absence of sonorous information) Reaching of a sonorous object that was rst touched and then taken away while continuing to make a sound ( tactile stimulation initially, followed by continuous sonorous stimulation) Reaching of a sonorous object in a xed position ( continuous auditory stimulation in the absence of tactile information) Reaching of a moving sonorous object moved in a horizontal 180 arc around the head ( continuous auditory stimulation in the absence of tactile information) Reaching of a sonorous object and of a non-sonorous object that the child has dropped (observational test) Reaching of a sonorous object concealed by a screen Reaching of a sonorous object that designs a vertical arc ( continuous auditory stimulation in the absence of tactile information) Reaching of a non-sonorous object taken out of the child's hands and positioned in a set point close to him ( discontinuous tactile stimulation in the absence of sonorous clues) Reaching of a sonorous object pulled away from the child and moved in a horizontal arc to one side while continuing to sound ( continuous auditory stimulation) Reaching of an intermittently sounding object that designs a horizontal arc ( discontinuous auditory stimulation in the absence of tactile information) Reaching of an object that sounds intermittently at xed points in space ( discontinuous auditory stimulation in the absence of tactile information)

292 Table 4 Gross motor and ne motor abilities. Group I No psychomotor acquisition Head control Sitting Standing Walking supported Walking independently No prehension Attempts at prehension Bimanual grasp Fingerthumb grasp 0 20 (100%) 20 (100%) 20 (100%) 1 (5%) 19 (95%) 0 20 (100%) 20 (100%) 20 (100%)

E. Fazzi et al. / Early Human Development 87 (2011) 289296

Group II 4 (23%) 7 (42%) 4 (23%) 2 (12%) 0 0 11 (65%) 2 (12%) 4 (23%) 0

3. Results As regards the neuromotor development (Table 4), at the last follow-up visit, the children in Group I showed a homogeneous prole: all the subjects had acquired good head control, and the sitting and standing positions. Independent walking was achieved by 19 of the Group I subjects (95%) by the end of the follow up, at a mean age of 19 months (SD = 8.7), at which stage the one remaining subject (5%) had achieved only assisted gait. Of the 20 subjects in Group I, 9 didn't crawl (45%). Of the 11 remaining subjects, 4 (36.4%) achieved crawling at a normal time, 3 (27.3%) crawled at the same time as walking, 3 (27.3%) crawled after learning to walk independently and one crawled but was not walking at 36 months. A very different neuromotor prole, characterised by failure to achieve, or markedly delayed achievement of, almost all the functions, both postural and locomotor, emerged in the blind children with associated multiple disabilities: four children (23%) did not achieve postural control, seven (42%) acquired only control of the head (at a mean age of 15 months), four children (23%) achieved the sitting position at the mean age of 24 months, and two children (12%) affected by a less severe form of cerebral palsy were able to stand at the mean age of 26 months. Therefore, none of the Group II children had acquired the ability to walk by the age of 36 months, when the follow-up period ended. As regards the acquisition of prehension (Table 4), all the blind children (100%) of Group I achieved bimanual grasp and (index) ngerthumb grasp at the mean ages of 6 and 18 months respectively. Like the acquisition of neuromotor functions, development of manipulatory function was also found to be decient in the blind children with associated handicaps: prehension was absent in 65% (11 children) of the Group II and present, although inadequate (poorly functional attempts that did not result in the complete grasping of the
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

object) in 12% of the cases (two children). Only 4 of the Group II subjects (22%) acquired bimanual grasp, at the mean age of 24 months. No member of this group acquired (index) ngerthumb grasp. As regards mental development (Table 1), evaluated through administration of the ReynellZinkin Scales [22], all the subjects of the Group I showed normal development. In Group II, on the other hand, mental development seemed to be more impaired, with 23% (4 subjects) presenting slight mental delay and 77% (13 subjects) severe mental delay. As regards the acquisition of the items in the reach on sound protocol, 70% (14 children) of the Group I subjects had completed the protocol by the age of 36 months, as opposed to just 6% (one child) of the Group II subjects. Fig. 1 shows the percentage of acquisition at 36 months at each task in the two groups. At each task, the statistical difference between the groups is highly signicant (p b 0.001). In particular, in Group I, one child (5%) had achieved all the tasks at 12 months, 4 children (20%) at 18 months, two children (10%) at 24 months, and seven children (35%) at 36 months of age. A further 15% of the Group I subjects (3 children) achieved all the items except task n. 6, which the only two Group II children who completed a reasonable proportion of the protocol (12% of Group II) also failed to achieve. The remaining 15% of the Group I subjects (3 children) achieved only the rst few tasks in the protocol. These three children were all affected by Leber congenital amaurosis and were characterised by behavioural and relational problems. Fig. 2 illustrates the percentages of the Group I children who, at each of the ages considered (6, 12, 18, 24, and 36 months), had mastered the various tasks. It emerges that tasks n. 1 to 5 were acquired by most of Group I children by the age of 12 months, whereas acquisition of the subsequent tasks shows a less age-homogeneous pattern, partly as a result of the greater difculty of these tasks, a drop in performance coinciding, indeed, with task n. 6. This drop in performance was also seen in the Group II subjects (Fig. 3): in this group the overall trend was similar to that observed in Group I (illustrated in Fig. 2) but the task acquisition rates (percentages) were lower, with the curves dipping in relation to the more complex tasks. Furthermore, at the ages of 6 and 12 months, none of the Group II children had acquired task n. 4 or above. Conversely, as illustrated in Fig. 2, zero acquisition rates were recorded in the Group I subjects only at 6 months of age and, even then, only in relation to the very last tasks in the protocol. In Group II, two children (12% of the sample) achieved the rst three items, one child (6%) only the rst task, and eleven children (64%) none of the tasks. Of these eleven subjects, nine (52%) showed reactions indicating alertness, and two (12%) no reaction. In this group, only one child (6%) completed the protocol and the 2 children (12%) completed a reasonable proportion of the tasks

Task 1 Task 2 Task 3 Task 4 Task 5 Task 6 Task 7 Task 8 Task 9 Group 1 Group 2

Task 10

Task 11

Fig. 1. Percentage of acquired tasks at 36 months in the two groups. All the differences are highly signicant (p = 0.000).

E. Fazzi et al. / Early Human Development 87 (2011) 289296

293

6 months

12 months

18 months

24 months

36 months

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1 2 3 4 5 6 7 8 36 months 18 months 9 6 months 10 7 5% 45% 75% 80% 85% 11 8 0% 15% 35% 45% 85% 9 0% 25% 60% 60% 75% 10 0% 20% 45% 50% 75% 11 0% 15% 30% 40% 75%

TASK
1 6 months 12 months 18 months 24 months 36 months 20% 85% 95% 100% 100% 2 25% 85% 90% 95% 95% 3 15% 80% 85% 90% 95% 4 5% 55% 65% 80% 85% 5 5% 50% 65% 70% 85% 6 0% 5% 25% 35% 70%

Fig. 2. Acquisition of reach on sound function in Group I.

administered; these subjects were all characterised by cerebral visual impairment, cerebral palsy, and a slight mental delay. When observing the acquisition of the main gross motor milestones, these subjects were three of the less compromised in Group II. Only two other subjects, who didn't complete the reach on sound protocol,

were characterised by a less profound psychomotor delay: one was the subject affected by Jubert syndrome and the other one had a severe mental delay at the ReynellZinkin Scales. Of the other 12 children who didn't complete the reach on sound protocol, one had a severe psychomotor delay (he could stand seated autonomously just

6 months

12 months

18 months

24 months

36 months

100%

80%

60%

40%

20% 36 months 1 2 18 months 3 4 5 6 7 8 6 months 9 7 0% 0% 11% 17% 17% 10 11 8 0% 0% 6% 17% 17% 9 0% 0% 11% 17% 17% 10 0% 0% 0% 11% 11% 11 0% 0% 0% 11% 11%

0%

TASK
1 6 months 0% 12 months 11% 18 months 11% 24 months 22% 36 months 28% 2 6% 12% 12% 30% 30% 3 0% 11% 11% 22% 22% 4 0% 0% 11% 17% 17% 5 0% 0% 11% 22% 22%

6 0% 0% 0% 0% 6%

Fig. 3. Acquisition of reach on sound function in Group II.

294

E. Fazzi et al. / Early Human Development 87 (2011) 289296

for a few seconds at the age of 4 years) with a slight mental delay at the developmental scales, two had a less serious psychomotor delay (one reached the assisted gait at 20 months, one was able to stand at 23 months) and a severe mental delay at the ReynellZinkin Scales, 9 had both an important psychomotor and mental delay. Task n. 1 was the only one that all the members of the Group I had completed by the age of 24 months (see Fig. 2). Moreover, children in Group I mastered the tasks in sequence, except for tasks 6 and 8 that were achieved at 36 months of age by the major part of the subjects, meanwhile the tasks n. 7, 9, and 10 were reached earlier. 4. Discussion We found that the blind child's acquisition of the reach on sound function shows a characteristic pattern. Most of the children observed had completed the Bigelow protocol tasks by the age of 36 months, but from the age of 12 months onwards a peculiar trend seemed to emerge, characterised by a reduced frequency of achievement on tasks n. 6 and n. 8 that persisted at follow ups conducted at the ages of 18 and 24 months, with a recovery of these performances not appearing until the age of 36 months. Tasks n. 6 and n. 8 seemed to be particularly complex for the blind child, given that they both present a discontinuity of the tactile stimulus. In task n. 6, the presence of a cover that hides the object, impeding tactile immediate recognition and representing a new, unknown tactile substrate that has to be recognised as a cover, seems to confound the subjects, slowing the mastering of this task. In task n. 8, the difculty is represented by the presence of a discontinuous tactile stimulation in the absence of sonorous clues. It seems that the discontinuity of the tactile stimulus represents a specic difculty for the mastering of the tasks by the blind children. It is only a few months after acquisition of the ability to walk independently through space, and thus after the mean age of two years, that a child learns to know his external world and develops the strategies needed to overcome these particular challenges. This would also explain why only a small percentage of the Group II children managed to complete the protocol, and in particular to acquire tasks n. 6 and 8: the presence of a motor and/or cognitive decit challenges the child not so much by impeding his execution of the task, as by denying him the possibility of moving and gaining experience of his external world, and thus of nding a solution to Fraiberg's conceptual problem. In particular, we observed that walking in Group 1 begun when the rst 45 tasks at the reach on sound protocol were achieved. This observation is in accord with Bigelow's observations of three blind boys [14]. Children begun to walk after they entered Piaget's stage IV. Independent gait seems to be favoured by the appearance of the child's active searching for the object, activity that walking seems to support itself. This nding is in accordance with the hypothesis that the relationship between object knowledge and the advancement in locomotive skills is facilitative also in the blind child [14], as it is proposed to be for the sighted children [23]. Moreover, as observed in a recent study [9], congenitally blind children rarely crawled before walking independently and often missed out this stage altogether. In this aspect, they are similar to shufers or scooters who slide on their buttocks rather than crawl. Even though the motor responses needed in order to perform task n. 1 are the same as those required for task n. 3, from our results it would appear that blind children nd task n. 3 more difcult to achieve than task n.1. Thus it might be deduced that a task involving a continuous tactile stimulus is simpler to perform than one in which it is the sonorous information that is continuous: in short, tactile tasks are easier to perform, as suggested by Bigelow's ndings [16]. The same applies to tasks n. 2 and 8. Again, these tasks demand the same motor behaviour in order to reach the object, but the rst was seen to be more easily achieved than the second, in which the object was presented only through discontinuous tactile clues: this suggests that sound gives to the blind child important information about the

object, such as its location and direction, clues that a purely tactile stimulus that is discontinuous cannot give him. Indeed, in task n. 8, the child is deprived of all contact with and information about the object and this inuences his attempts to nd it. The fact that a substantial percentage (40%) of children were able to perform this task at 36 months could be an indication that the older child, despite being deprived of tactilesonorous information relating to the object when it is taken out of his hands, will still seek it, aware that it continues to exist even though he is no longer able to perceive it. The children's performances on task n. 8, however, demonstrate that seeking of the target object is favoured by having previously had tactile contact with it. Task n. 4 proved to be very difcult to achieve, probably for two reasons: one was the lack of tactile contact with the moving object (a further demonstration of the importance of integration of tactile and auditory information in the blind child's ability to reach and take hold of an object), and second, the fact that the blind child, in this task, had to show increased upper limb control and precision in order to reach a moving as opposed to a stationary object. The child's ability to locate an object through sound improves as demonstrated by our ndings relating to the achievement of tasks n. 3, 4 and 7, which do not include tactile clues. In these three tasks, the child is required to reach a sonorous object that is rst stationary (task n. 3) and that then moves along the horizontal plane (task n. 4), and subsequently designs a vertical arc (task n. 7). Given that psychophysically it is easier to locate sounds in the horizontal than in the vertical plane, the sequence in which these three abilities are acquired reects the psychophysical principles of location by sound. Tasks n. 5 and 9 require the child to reach a sonorous object that is sounding in a position different from the one in which it had been when he lost tactile contact with it. Children attempting these tasks thus nd themselves with contrasting tactile and sound clues, but regardless of where they heard the sound coming from, our children nevertheless sought the object in the place where they had lost contact with it. Thus, even though blind children are able to respond to sound clues, as shown by their achievement of task n. 2, in more difcult tasks like these, which introduce contrasting information, they can be seen to fall back on old reaching strategies guided by tactile clues. Thus we can afrm that sound clues facilitate the reaching of the object, but only when they do not conict with tactile information. Furthermore, task n. 5 was apparently easier for the children than task n. 9 thanks to their development and application of behavioural rules for the retrieving of objects they have allowed to fall from their hands: blind children will quickly learn that if they drop an object, then they themselves must go down to get it. Applying this rule, they do not need to postulate a spatial construct, which, instead, is precisely what they do when the object is taken from their hands as opposed to accidentally dropped (task n. 9). At task n. 9, children stopped and entered in a state of quiet listening to object movements. Their steps and gestures were guided by sound. Task n. 6, the one most similar to the object permanence tests administered to sighted children, was the one that all our blind children found hardest to achieve. The difculty of this task lies in the need to remove a cover that conceals the object, which goes on sounding, from underneath the cover, for the entire duration of the task: achievement of this task depends on the ability to separate the means from the end. According to Piaget's description of the development of object permanence, this ability is possible at 8 12 months (or the fourth stage of Piaget's model), despite the fact that, at this age, the child's perceives the object as an extension of his own action; in the subsequent stage (fth stage: 1218 months), the child will comprehend the permanence of the object only if he is able to conserve a visible map of its movements. The blind child instead has to rely on sound clues in place of this visual information, and only a small proportion of our Group I subjects were able to perform the task at the age of 18 months. Piaget indicates that at the sixth stage

E. Fazzi et al. / Early Human Development 87 (2011) 289296

295

(1824 months) the child has acquired the concept of the object as permanent, although the relatively small proportion (35%) of our children who had achieved task n. 6 by the age of 1824 months would seem to highlight the particular importance of visual information in the building of the concepts of object permanence and of the permanence of the outside world. Tasks n. 10 and 11, the ones associated with the fth stage in the development of object permanence according to Piaget, were the abilities whose acquisition was most delayed in our children. For blind children, objects that sound intermittently are particularly difcult to reach, given that, when they fall silent, their location becomes unpredictable; conversely, an intermittently sounding object that is in motion is easier to locate, because the perception of its movement allows the blind subject to predict or guess its position even when it has stopped emitting its sound (task 10). The difculties encountered by the blind children of Group I in performing the tasks of the reach on sound protocol did not seem to be related to the mental development as measured by the Reynell Zinkin Scales: in fact, all the subjects presented normal scores. The three subjects of Group I with a Leber syndrome who achieved only the rst tasks of the reach on sound protocol presented relational difculties. A particular psychological fragility is described in some children with Leber syndrome, characterised by withdrawn and stereotyped behaviour, in the absence of a diagnosis of an autistic spectrum disorder [24]. In Group II, most of the subjects possessed neither appropriate manipulatory functions, nor the ability to reach an object presented through sound; these children tended to use their hands as means of self-stimulation more than as a means through which to gain knowledge of the real world around them. These are behavioural mannerisms that have been frequently described in congenitally blind children with and without neurodevelopmental disabilities [24,25]. Upon the presentation of a sound stimulus, most of the subjects demonstrated nothing more than aspecic reactions, indicating alertness, or remained immobile; at the very most, they might make some movement indicating excitement. The prole of acquisitions in this group was found to be discontinuous and lacking in homogeneity, and the results obtained were markedly low; however, in considering these low results it has to be recalled that the presence of motor disabilities can preclude or render extremely difcult a subject's attempts to reach the target objects. For the children in Group II, touch continued to be the preferred sensory channel when seeking, and attempting to reach, the target object, as demonstrated by the fact that tasks n. 1 and 2 (involving tactile contact with the object) were the only ones that a reasonable proportion of them were able to achieve; moreover, these same tasks were also the ones most easily performed by the Group I children. The Group II children, like the Group I subjects, showed a trend of a lower performance on task n. 6. This nding, that needs to be conrmed by further studies, seems to reveal a pattern that could be specic to severe visual impairment, not being inuenced by the presence or absence of an associated motor decit. In conclusion, the subjects affected by an isolated visual decit, despite showing an initial and slight neuromotor developmental delay, were able to develop good manipulatory skills and subsequently to walk independently, thanks in part to their acquisition of the ability to reach an object presented through sound. On the basis of the results emerging from our administration of the reach on sound protocol, we are able to afrm that when comparing tasks that differed only in the type of sensory information provided (i.e., only tactile vs only auditory clues), the tactile tasks seemed to be the ones more easily achieved: in particular, blind babies presented an attentional alert to sound, but seemed to prefer touch to orientate their exploration. In fact, in the rst months of life, our children tended to reach objects that they had previously touched rather than those that they had merely heard, and in the presence of conicting tactile and auditory clues, they responded rst to the tactile

information, acquiring only subsequently the ability to use the auditory information. Only in specic situations they found to implement behaviours based on precise reaching modalities, like the ones needed to perform task n. 5. At older ages, when sensorial discrimination and multisensorial integration are more mature [26], touch and sound seemed to guide the child in reaching the object in like manner. The children whose blindness was associated with psychomotor and/or mental delay presented, in most cases, severe clinical pictures in which the delay, which usually persisted, also resulted in poor performances on the reach on sound protocol: the subjects in this group rarely achieved, or achieved only with a very marked delay, the ability to reach an object presented though sound. The exploratory and learning techniques adopted by these subjects continued to be rather primitive (the mouth playing a central role) and their ability to move in space was often impeded by the difculty they had separating themselves from environmental points of reference and contact. Thus, as we demonstrated in a previous study [9], when the visual decit is associated with other handicaps, the result is a clinical situation of extreme gravity due to the fact that each of the two pathological conditions seems to reinforce the other, giving rise to severe psychopathological pictures. On the other hand, the difculties encountered in both groups (I and II) on task n. 6 (i.e. reaching of a sonorous object concealed by a cover) may represent a characteristic pattern of severe visual impairment and highlights the particular importance of tactile information in exploring the outside world and reaching the target objects. In view of the important role that certain acquisitions play in neuropsychological development, it is crucial that the blind child be enabled to acquire through his other senses the knowledge that he is not permitted to acquire through sight. The acquisition of reach on sound, representing the solution to Fraiberg's conceptual problem, emerges as a signicant moment in the development of the blind child, being at once a condition of and a catalyst for all his subsequent development. References
[1] Piaget J. The origins of intelligence in children. New York: International University Press; 1952. [2] Piaget J. The construction of reality in the child. New York: Ballantine; 1954. [3] Piaget J. Genetic epistemology. New York: Columbia University Press; 1970. [4] Rogers SJ, Puchalski CB. Development of object permanence in visually impaired infants. J Vis Impair Blind 1988;82(4):13742. [5] Gillman AE. Handicap and cognition: visual deprivation and the rate of motor development in infants. New Outlook Blind 1973;67:30314. [6] Langley MB. Psychoeducational assessment of the multiply handicapped blind child: issues and methods. Educ Vis Handicap 1979;10:97114. [7] Fraiberg S, Freedman D. Studies of the ego development of the congenitally blind child. Psychoanal Study Child 1964;19:13369. [8] Fraiberg S. Parallel and divergent patterns in blind and sighted infants. Psychoanal Study Child 1968;23:264300. [9] Fazzi E, Lanners J, Ferrari-Ginevra O, Achille C, Luparia A, Signorini S, et al. Gross motor development and reach on sound as critical tools for the development of the blind child. Brain Dev 2002;24:26975. [10] Prechtl HF, Cioni G, Einspieler C, Bos AF, Ferrari F. Role of vision on early motor development: lessons from the blind. Dev Med Child Neurol 2001;43:198201. [11] Fazzi E. The effect of impaired vision on development. In: Dutton G, Bax M, editors. Visual impairment in children due to damage to the brain: clinics in developmental medicine. Wiley-Blackwell; 2010. [12] Fraiberg S, Siegel BC, Gibson R. The role of sound in the search behaviour of a blind infant. Psychoanal Study Child 1966;21:32757. [13] Fraiberg S. Insights from the blind: comparative studies of blind and sighted infants. New York: Plenum Press; 1977. [14] Bigelow AE. The development of the use of sound in the search behaviour of infants. Dev Psychol 1983;19:31721. [15] Bigelow AE. Locomotion and search behaviour in blind infants. Infant Behav Dev 1992;15:17089. [16] Bigelow AE. The development of reaching in blind children. Br J Dev Psychol 1986;4:35566. [17] World Health Organization. International Classication of Functioning, Disability and Health (ICF): a manual of classication relating to the consequences of disease. Geneva: World Health Organization; 2001. [18] Good WV, Jan JE, Burden SK, Skoczenski A, Candy R. Recent advances in cortical visual impairment. Dev Med Child Neurol 2001;43:5660. [19] Good WV. Cortical visual impairment: new directions. Optom Vis Sci 2009;86(6): 6635.

296

E. Fazzi et al. / Early Human Development 87 (2011) 289296 [24] Fazzi E, Rossi M, Signorini S, Rossi G, Bianchi PE, Lanzi G. Leber's congenital amaurosis: is there an autistic component? Dev Med Child Neurol 2007;49(7): 5037. [25] Fazzi E, Lanners J, Danova S, Ferrarri-Ginevra O, Gheza C, Luparia A, et al. Stereotyped behaviours in blind children. Brain Dev 1999;21(8):5228. [26] Stein BE, Perrault Jr TJ, Stanford TR, Rowland BA. Postnatal experiences inuence how the brain integrates information from different senses. Front Integr Neurosci 2009;3:21.

[20] McDonald MA. Assessment of visual acuity in toddlers. Surv Ophthalmol 1986;31: 189210. [21] Fazzi E, Signorini SG, Bova SM, La Piana R, Ondei P, Bertone C, et al. Spectrum of visual disorders in children with cerebral visual impairment. J Child Neurol 2007;22(3):294301. [22] Reynell J. The ReynellZinkin scales: developmental scales for young visually handicapped children. Windsor: Nfer Nelson; 1979. [23] Zelazo PR. Learning to walk: recognition of high order inuences. In: Lipsitt L, Rovee-Collier C, editors. Advanced in infancy research, Vol. 3. Norwood, NJ: Ablex; 1984.