If an infant reaches for a toy hidden under a cloth, what does this suggest?

D How Biology Has Shaped the Cognitive Theory

The neurobiological and neuropsychological evidence we have presented supports a new interpretation of Piaget’s observations regarding the development of object permanence. Specifically, performance on the object permanence task is seen to be closely related to the ability to perform delayed-response tasks in general. Extensive neurobiological evidence implicates the principal sulcus in the performance of delayed-response tasks, and although the precise function of prefrontal cortex is not yet clear, the development of performance on delayed-response tasks is closely related to the state of maturation of this region of the brain. Thus, we can begin to view Piaget’s notion of object permanence in more general psychological terms as a marker for the state of an infant’s working memory capacity and in more particular biological detail as a reflection of the state of development of the infant cortex.

Taking a neurobiological approach to this set of problems provides developmental psychology with tools to move beyond the use of rigid stages to describe behavioral consistencies (Case, 1985), and thereby to understand the processes of change more thoroughly. It might be tempting to reinterpret the biological findings in discrete, stagelike terms, by arguing simply that the prefrontal cortex is “immature” prior to 712months, and once it “matures” at 712months, working memory capacities come into operation. However, the neurobiological and behavioral evidence points in another direction. Delayed-response tasks (and the object permanence task in particular) tap multiple interacting and possibly overlapping neural systems that are still poorly understood, and the available evidence suggests that brain change is continuous, not discrete. Thus, the emergence of the ability to reach for hidden objects is unlikely to be the result of a discrete change to a single neural system. Instead, the development of working memory capacities as indexed by object permanence tasks is a gradual process, first indicated in behavior by habituation and preferential looking measures, later by the oculomotor delay paradigm and visual versions of the object permanence task, and still later by performance in object retrieval (Munakata et al., 1994). Thus, the ability to respond to the persistence of objects in time and space appears to emerge gradually, first in simpler responses, later in more complex ones. A complete understanding of this pattern of results, and the mechanisms of neural and cognitive change that drive them, demands investigation of the function and maturation of the multiple brain regions and pathways that underlie the infant’s expanding capacity to demonstrate object permanence.

To illustrate this notion, consider that the information about the location of an occluded object may be maintained in one location useful for making both looking and reaching responses, or this information may be distributed into systems that maintain internal representations in a form particular to looking or reaching (Munakata et al., 1994). In either case, developmental differences in the pathways that underlie these behaviors may explain why one form of response precedes the other. In short, a biological perspective, and in particular an ethological one, helps researchers to resist the notion of a unitary object permanence concept that emerges at some prespecified developmental stage, but instead to explore how patterns of brain development contribute to the emergence of behaviors and capacities consistent with the continuous and gradual development of object permanence.

Object Permanence

Knowledge about the properties of objects shows further advances in children 6–12 months of age. Infants begin to display object permanence, or the realization that objects continue to exist and can be retrieved even when out of sight. For example, at about 8 months of age, infants actively implement searching for an attractive object that they have just been shown but which has then been hidden behind a cloth. For many developmental psychologists, this search behavior is theorized to signal the onset of a more sophisticated level of representation or a more active type of memory than that associated with object identity for which 3-month-olds demonstrate competence.

Object permanence undergoes a protracted developmental progression. Eight-month-olds may search for a toy that has been hidden under a single cloth. However, if they retrieve it from that location several times and it is then hidden under a second cloth, they are not likely to initiate a correct search for it, returning instead to the original location in which they had repeatedly found it during their previous efforts to retrieve the toy. This incorrect search, described as the ‘A-not-B’ error, usually is no longer made by infants as they approach 1 year of age and may be due to an increasing ability to inhibit an inappropriate response or the development of greater memory strength for the more recent event relative to repeated earlier events.

Nested timescales

People do new things when faced with new challenges, often changing their behavior flexibly in the moment, when faced with some new input from the world. People learn new skills after hours, days and months of practice. Developmental change — the emergence of reaching, or walking or talking — occurs over weeks, months and years. If we are to explain how development creates we must understand processes of change over multiple timescales. Accordingly, we present one more task first introduced by Piaget [1954] to illustrate the importance of nested times scales of change in understanding the creative force of development.

Piaget [1954] invented the A-not-B task to answer the question of “when do infants acquire the concept of object permanence?” He devised a simple objecthiding task that works as follows: The experimenter hides a tantalizing toy under a lid at location A and the infant reaches for the toy. This A-location trial is repeated several times. Then, there is the crucial switch trial: the experimenter hides the object at new location, B. At this point, 8- to 10-month-old infants make a curious ‘error’. If there is a short delay between hiding and reaching, they reach not to where they saw the object disappear, but back to A, where they found the object previously. This ‘A-not-B’ error is especially interesting because it is tightly linked to a highly circumscribed developmental period: infants older than 12 months of age search correctly on the crucial B trials. Thelen, Smith, Schöner, Spencer and colleagues1 have offered a formal theory, the dynamic field model to explain the A-not-B error as the emergent product of multiple causes interacting over nested timescales. Here we use the ideas from this formal theory to highlight the importance of nested times scales.

What might cause an infant to reach — at any one moment — to A or to B? Figure 5 provides the key task analysis: One influence on behavior is the continually present box and hiding wells. A second influence is the specific or transient input (top row) that occurs when the experimenter hides the toy. This is a momentary event that will be remembered in the system for some time, the endurance of this memory is, of course, critical to infants' success in the task. Then the infant reaches, an act that will create its own memories. Moreover, the infant sees the transient event and reaches repeatedly to A before the shift trial at B. These are events and memories at another timescale. The dynamic field theory account of the error is at its core an account of the dynamics of how all these processes at different times scales change over the course of the trials in the tasks and how they integrate to yield a decision to reach to A or to B.

Figure 6a illustrates the evolution of activation on the very first A trial. Before the infant has seen any object hidden, there is activation in the field at both the A and B locations from the two covers. As the experimenter directs attention to the A location by hiding the toy, it produces a high, transient activation at A. Then a decision emerges in the field over time. When the activation at a location stabilizes, the infant reaches to that location. Most crucial for this account is that once infants reach, a memory of that reach becomes another input to the next trial. Thus, at the second A trial, there is some increased activation at site A because of the previous activity there. This combines with the hiding cue to produce a second reach to A. Over many trials to A, a strong memory of previous actions builds up. Each trial embeds the history of previous trials. Now consider the crucial B trial (Fig. 6b). The experimenter provides a strong cue to B. But as that cue decays, the lingering memory of the actions at A begins to dominate the field, and indeed, over time, to shift the decision back to the habitual, A side.

The model clearly predicts that the error is time dependent: there is a brief period immediately after the hiding event when infants should search correctly, and indeed they do [Smith, Thelen, Titzer and McLin, 1999; Bremner and Bryant, 2001]. Using this model as a guide, experimenters can experimentally make the error come and go, almost at will. This is achieved by changing the delay, by heightening the attention-grabbing properties of the covers or the hiding event, and by increasing and decreasing the number of prior reaches to A [Smith et al., 1999; Marcovitch, Zelazo and Schmuckler, 2002; Marcovitch and Zelazo, 2006]. Researchers have made the error occur (and not occur!) even when there is no toy to be hidden [Smith et al., 1999]. Directing attention to an in-view object (A) heightens activation at the location and, in the experiment, infants reach to that continually in-view object. Subsequently, when the experimenter directs attention to a different nearby in-view object (B), infants watch, but then reach back to the original object (A).

Experimenters have also made the error vanish by making the reaches on the B trials different in some way from the A trial reaches. In the model, these differences decrease the influence of the A trial memories on the activations in the field. One experiment achieved this by shifting the posture of the infant [Smith et al., 1999]. An infant who sat during the A trials would then be stood up, as shown in Fig. 7, to watch the hiding event at B, during the delay and during the search. This posture shift causes even 8- and 10-month-old infants to search correctly, just like 12-month-olds. These results underscore the highly decentralized nature of error: the relevant causes include the covers on the table, the hiding event, the delay, and the past activity of the infant and the feel of the body of the infant. This multicausality demands a rethinking of what is meant by knowledge and development. Intelligence is creative — as a unified system — over multiple timescales, including the timescales of individual action as well as the timescales of learning and developmental change. The A-not-B error has been of most interest to developmental theorists, however, because of the changes over developmental time. This is an error tightly linked to a few months in infancy. Or is it?

Spencer, Smith and Thelen [2001] invented an A-not-B task that was suitable for 2-year-olds by hiding toys in a sandbox. The surface of the sand presents a uniform field so there are no markers to indicate the two possible hiding locations. Experimenters gave toddlers many trials at location A, and then hid the toy at location B. With a delay of 10s, the toddlers, having watched the toy being hidden at location B, still returned to the A location to dig in the sand for the toy. Indeed there are many other situations in which both children and adults fall back on a habit despite new information [Butler, Berthier and Clifton, 2002]. Nonetheless, in the standard A-not-B task, infants change their behavior over 2 months. In the field model increasing the resting activation of the field simulates this by making it easier for the input from the hiding cue to form a self-sustaining peak at B and thus compete with the A memory.

If self-sustaining memories drive the successes of older children, then we must ask where they come from: What are infants doing everyday that improves their location memory? One possibility is their self-locomotion. Crawling and self-locomotion appears to improve the spatial memories of infants and to predict success in the A-not-B task [Bertenthal and Campos, 1990]. Indeed, teaching babies to self-locomote (in walkers) causes them to succeed at a younger age than usual in the A-not-B task [Kermoian and Campos, 1988]. Like sticky-mittens and visual cliffs, the causes of developmental change may lie in seemingly disparate achievements that involve overlapping component processes.

Dynamic systems approaches to cognition have been criticized as being suitable for the study of perception and action, but not for higher level human cognition [Abrahamsen and Bechtel, 2006; Dietrich and Markman, 2003] — the cognition that, in these views, underlies such abstract aspects of human thought as mathematics, science, allegory, and poetry. Accordingly, in the next section we outline how the coupling of sensory-motor processes to the perceivable consequences of those processes may make higher-level cognition.

Separation Anxiety

Sometime in the middle of the first year, when infants understand that people exist even when they are out of sight (person permanence), they react to the everyday recurring disappearances of their parents by attempting to maintain proximity through the behaviors available to them, including crying, cooing, and crawling. In manifesting these responses, infants not only indicate their desire to stay in proximity with the caregiver but also the development of ways to control distance and separation. During this stage, infants increasingly initiate interaction with their parents and actively protest when their primary caregiver departs, even for a moment. By the first birthday, behaviors that indicate separation distress are even more clearly detected, with infants tending to become agitated and upset upon separation.

The Normative Course of Separation Anxiety

Separation distress, signaled by crying in response to parental separation, may be observed as early as 4 or five months of age (Stayton et al., 1973) but most accounts identify 8 months as the age when separation anxiety emerges. Distress from brief separations continues to characterize toddlers' behavior well into the second year of life (Jacobson and Wille, 1984); the normative response typically peaks around 12 to 18 months and then fades after two years of age. In diverse cultural contexts, such as the Kalahari Bushmen (Ellis, 1991), the Israeli Kibbutz (Sagi et al., 1985) and Guatemala (Lester et al., 1974), infants display distress in response to separation from their mothers; this is considered a normative part of development and its emergence is viewed as a major milestone in the formation of the emotional tie between the child and primary caregiver. The reaction to separation from the mother appears to be a universal phenomenon, however specific parenting practices and cultural experiences may impact the timing and the intensity of the response. For example, in cultural settings where infants experience constant physical contact with their mothers (i.e., as in Japan), distress to separation was observed before 8 months (Mizukami et al., 1990). The use of an inanimate companion such as a blanket or doll (also known as a transitional object) is one of the ways toddlers attempt to alleviate separation distress (Winnicott, 1971). While separation anxiety gradually fades for most children after the second birthday, some children will continue to express extreme distress in the face of parental separation. In many cases, these children will be subsequently diagnosed as suffering from separation anxiety disorder (SAD), a psychological disorder briefly discussed in the final section of the chapter.

TOILET TRAINING AS A DEVELOPMENTAL MILESTONE

Toilet training, like every developmental milestone, is the compilation of numerous neurobiological processes affected by social opportunities, cultural expectations, and temperamental tendencies. Infants develop object permanence, the basis for the toddler's fascination and fear of stool's disappearance down the toilet. As toddlers pass through the sensorimotor phase, they busily explore the world, entering the bathroom to investigate. The child younger than 2 years may have phases of separation difficulties and is eager to mimic. This creates opportunities to watch parents use the bathroom and imitate them by sitting on the potty chair at the same time. As the child enters the preoperational stage, language develops rapidly. Now the child can indicate when diapers are wet or when he or she needs to defecate. In this egocentric phase, the child is highly focused on himself or herself. Along with “It's mine” and “I do it,” the child might interrupt anything and everything when the urge to defecate or urinate comes.

The toilet training period is a time of increased autonomy and initiative. Two- to 3-year-old children often want to pick their own clothes, feed a new baby, color the picture themselves, and explore every item in the room without parents' help. The developmental stage is set for toilet training, the ultimate skill of toddler independence. Toilet training requires the cognitive understanding of where stool and urine go, the motor skills to get there, and the desire to do it without help: skills that finally consolidate between 2 and 3 years of age in the typically developing child.

On the other hand, toilet training is like no other developmental milestone. Few childhood skills bring similar joy to parents when accomplished and frustration when delayed. Although stooling and urinating are universal bodily functions, they are nonetheless emotionally charged tasks. When a child uses the bathroom independently, parents are freed from the inconvenient and time-consuming task of regular diaper changes. The child who struggles with training can bring far more burden to the family than do children with other delayed milestones. For example, late talkers may communicate with pointing or word approximations; late walkers still can move from one place to another by crawling, cruising, or scooting. A child who is not toilet trained has no compensatory options. More so than those delayed in meeting other milestones, children who remain toilet untrained can be a source of enormous frustration and embarrassment to their families. Children with a persistent need for diapers or training pants (Pull-Ups®) can cause a significant financial, emotional, and time burden.

What does it mean when a child develops object permanence?

At what age does object permanence develop?

What is the difference between object permanence and representational thought?

When an infant understands that objects and events continue to exist even when they Cannot be seen heard or touched the infant has achieved?