The View From Where We Are

Early in these readings we talked about how infants come to reach for and grasp objects. In a study from my lab group and collaborators, caregivers filmed their infants picking up a pencil from a high-chair tray every two weeks from 6 to 13.5 months of age (Campbell et al., 2026). Over that period, whole-hand grasps were gradually joined by grasps in which the hand takes shape before it reaches the object, with the thumb and index finger leading and the other fingers tucked out of the way. Almost every infant moved in that direction, but plotted one by one, no two infants took quite the same route, and the timing varied a great deal.

One familiar way to explain an outcome this reliable, reached along paths this variable, is to say that a genetic or evolved template provides a basic plan, while specific differences in genes and in experience account for the variation. This assigns the familiar, reliable pattern to one kind of source and the differences between infants to others. But an infant doesn’t first develop a generic grasp and then acquire an individual version of it. The reliable development and the individual variation can’t be separated: there are not two different developmental processes at work, there is just one.

These readings still ask for causes, but of a particular kind: not what single factor produced an outcome, but what enabled it to happen. For the grasp, that means asking which conditions, present at that moment and carrying their own histories, made this way of reaching possible for this infant, and how the infant’s own activity changed those conditions as it went. This is a question about constraints, in the sense developed in The Riddle of Form and Matter. Constraints limit what can happen next, and by doing that they give development its structure.

Put as plainly as I can, the answer these readings give is that development is reliable because the conditions it depends on are themselves reliably rebuilt in each generation, and because each step of development constrains what can happen next. Those conditions include DNA, but also the organized egg, the physical properties of cells and tissues, bodies and nervous systems developing together, the surroundings and practices a young organism grows up within, and the organism’s own activity.

Earlier readings approached this from different directions: as constraints in a space of possibilities in The Riddle of Form and Matter, as self-organization in DNA as a Developmental Resource, as a developmental system in Developmental Systems Theory, and as a life cycle reconstructed in each generation in Life Cycles and the Direction of Development. These are parts of one account, not separate explanations. The same account covers variability: the conditions recur, but never exactly, and each individual meets them from its own history and changes them as it goes.

Constraints and Their Histories

Some of the constraints on a single reach belong to the moment, such as whether the infant is alert and how stably they are sitting. Some belong to the infant’s own history, such as how seeing an object has come to shape the hand before contact over the preceding months. The high chair and the pencil are designed objects, and seating an infant with something to pick up is a caregiving practice with a history across generations. The hand itself, with its opposable thumb, and the nervous system that controls it have an evolutionary history much older than our species.

Explaining how that hand evolved and explaining how it comes to grasp in this infant are related tasks, but neither can substitute for the other. All of these constraints are present at the moment of the reach. What the frames in Explanatory Frames distinguish is the different history through which each of them came to be there.

A study of reaching might reasonably take the high chair and the pencil as given. If we want to explain why infants repeatedly encounter them, we need to follow the history through which people have maintained or changed those practices. What one explanation takes as its starting point can become the subject of another. We don’t need to pursue every history in every study, but we do need to recognize which conditions our explanation depends on.

Development Carries a History

Early in the readings, the life cycle was introduced as a key biological foundation of development. We used the example of the frog, whose life cycle recurs across generations because it is reconstructed from resources and relations that earlier generations help make available. DNA is part of that continuity, along with the organization of the egg and the conditions in which it develops.

An infant reaching for an object begins from a body and nervous system shaped by earlier development. As the infant acts, its coordination can change, along with what it encounters and can do next. Explaining the grasp means following how these relations change through time, and that is what I mean by a historical view of development.

When I teach a class based on this material, I think of it as a history class. That doesn’t mean listing what happened in what order, which doesn’t explain much. Historians ask what made something possible and what could have gone differently, and a developmental explanation has to ask the same. We need to know how earlier development changed the individual, and how those changes affected what could happen next.

As development proceeds, the individual changes in ways that affect what further changes are possible. Connections established through earlier activity help make some responses reliable, while making other patterns harder to establish. So the organization produced through development becomes part of the conditions for development to continue, and to understand what can change now, we need to consider how that organization came about.

Removing cataracts that developed in adulthood can restore good vision, while a child whose congenital cataracts are treated late may continue to have substantial visual difficulties. The child has lost opportunities for visual development that surgery may not fully recover. We cannot explain the continuing difficulty by looking only at the condition of the eyes after surgery. Restoring patterned input does not erase the history of the visual system, and that history affects what the restored input can do.

Reliability and Variability

There is a more careful version of the familiar answer, one that gives experience a part in the reliable pattern too. In Plasticity and Constraint we met the distinction between experience-expectant development, which depends on conditions that members of a species ordinarily encounter, and experience-dependent development, which reflects an individual’s particular history. Patterned light reaching the eyes in early infancy is the classic case of expected experience: nearly every infant receives it, and the developing visual system depends on it.

But read as an account of reliability and variability, the distinction divides the work in the same way, with expected experience producing what infants share and particular experience producing how they differ. As that reading suggested, it tends to tell us when experience has its effects, not how.

Some of what development “expects” is supplied by the developing organism itself. Before patterned light reaches the eyes, waves of activity generated within the retina are already helping to organize the visual system. Gilbert Gottlieb’s ducklings showed the same thing in behavior. Ducklings that had never heard a mother duck still preferred the maternal call of their own species. But embryos call inside the egg, and Gottlieb found that removing this ordinary auditory activity disrupted the later preference. The apparent absence of relevant experience had concealed experience generated by the duckling itself. Biologists often describe development like this as canalized, and the stability is real. But calling something canalized, like calling it innate or expected, still leaves us to explain how that stability comes about and under what conditions it might change.

Human infants also help supply the conditions of their own development. An infant who picks up a toy and holds it out to a caregiver helps create an occasion for shared attention, and once infants begin to walk, they carry objects across the room to other people, changing what happens next through what they can now do with their bodies (Karasik et al., 2011).

This is also why it is misleading to picture the developing brain as a processor waiting for input from outside. What reaches a developing nervous system is partly produced by the organism’s own activity, from retinal waves and fetal movement to an infant’s reaches and looks, so a stimulus meets activity that is already under way. It can help to think of constraints as the banks of a river that channel where the water can flow, but a river also carves and moves its own banks. A developing infant is more like that river, because what it does changes the constraints it develops within.

Under ordinary conditions, contributions like these are easy to miss because they happen so reliably. Conditions that recur can support familiar outcomes, while differences in those conditions, and in how an organism meets them, can produce different paths. Every infant in the grasping study started with a human hand, but each had a different body and a different history of attempts.

Looking back from a familiar outcome can make all of this seem more settled than it was. Similar outcomes can be reached along different paths, and an early difference can be amplified or absorbed as development goes on. A precision grasp gives us a useful reference point for studying earlier hand movements, but it does not follow that those movements were directed toward that grasp.

Human Lives

The grasping study already involves more than the infant. The high chair and the caregiver who responds when a toy is held out belong to practices that people have built up and passed on, and a human life is organized through practices like these from the start. Most of these readings have concerned how bodies and brains develop. As Life Cycles and the Direction of Development pointed out, when we turn to how people live, we also bring in ideas about what a life should become, and those ideas affect what we consider to be a developmental problem and what we think would count as an improvement.

Human children are born into a world that other people have already done a great deal to organize. The ways they are cared for and taught depend on practices that were there before they were born. Of course, children also have a part in what happens, and anyone who has cared for a child knows that things don’t always go as planned. But what children can do depends partly on the opportunities that other people make available, and these differ sharply between families and communities.

We often think about the effects of experience in terms of what it has changed inside the child, perhaps in the brain. But other people continue to be involved. A child may be able to take part in an activity because someone makes time for it, provides help, or changes how it is done. Schools organize these opportunities in particular ways, and families differ in what they can provide. People who are no longer present can still have a part in a child’s life through the practices and institutions they helped establish, and children learn to participate in those practices and may eventually help change them.

In developmental research, we often describe a child’s trajectory by plotting measurements taken at different ages. This can tell us something useful about how the child is changing relative to other children. But understanding the trajectory also means asking what the child has been doing, how their abilities have changed, and how those changes affect what happens next. The measurements describe aspects of that history; they don’t give us the history itself.

Annette Karmiloff-Smith (1998) made a related point about how we interpret developmental disorders. A capacity that seems relatively spared in an older child or adult should not automatically be treated as a normally developed module sitting alongside impaired ones. An early difference can change how several systems develop and how they relate to each other, so a similar score on a task may conceal a different developmental route. A diagnosis can help identify difficulties and obtain support, but it does not tell us how the pattern developed in this particular person.

From the point of view of these readings, a developmental difference is a different path through a space of possibilities. That does not make it any less real. A path can involve suffering or limit what a person can do, and some differences call for substantial support. But distance from an average is not, by itself, an account of what has gone wrong or what would help.

We often assess children in terms of how close they are to doing something that we expect them to do later. There are good reasons for this, but a child is also living a life now. What would make that life better? The answer has to include what the child can do and what matters to them, which may not be captured very well by their position on a developmental scale.

If we want to intervene, we are entering a life that is already underway. What happens will depend on what the person can already do, as well as the opportunities they have to use any new abilities. It is one thing to help a child do something in an assessment or a treatment session, and another to make it possible for them to do it in everyday life, where the circumstances may be quite different. A historical account turns this into practical questions to investigate: what changes in the person’s activity, and what would sustain that change? Sometimes the answer will involve working directly with the individual; sometimes it will require changing how other people respond or how an activity is organized.

What Is Not Settled

I have spent these readings making the case that a developmental systems approach, with the developing organism at its center, gives us a better account of how organized lives come about, but you should ask whether I am right. Some researchers think the gene-centered view explains more than I have given it credit for. Others think the approach taken here is too vague to guide empirical work. That second worry deserves a straight answer, because a developmental systems explanation can be just as empty as a genetic program if naming the system takes the place of showing how a particular organization is produced.

One way of working that follows from the approach is the strategy used by Kuo and Gottlieb and discussed in Developmental Systems Theory: change a condition that ordinarily recurs so reliably that nobody notices it, and see what happens to development. Another is to study infants and children while they are moving and acting, rather than only in setups where a stimulus is delivered and a response recorded, and to measure the coupling between what an individual does and what it encounters instead of measuring each separately. Tools that can track the hand from ordinary home video, like those used in the grasping study, make this kind of study possible at home and across many months.

Following the same individuals closely over time, as the grasping study did, lets us treat variation within a person as information about how their activity is organized rather than as noise around an average (Molenaar, 2004; Ram & Gerstorf, 2009).

If reliable development comes from conditions that are reliably reconstructed, and those conditions are reconstructed because the life cycle recurs, the account of reliability can look circular. Brain, Body, World accepted the loop: follow the organizer back and there is no first step. My answer is that the loop has a history. It was stabilized over evolutionary time and can be altered, and in particular cases we can find out which of its conditions matter by changing them. So the loop is something we can investigate, although I do not think this fully answers the objection.

Every explanation starts from something it does not itself explain, and in development that starting point is an organization that already exists. Gene regulatory networks and molecular gradients can be legitimate starting points for explaining later changes. We still need to understand mechanisms, and explaining how a capacity works now is a legitimate task: a study can concentrate on it without also explaining how the capacity developed. The difficulty comes when we place the later organization at the beginning, described as a program or as a mechanism defined only by the job it will eventually do (Marshall, in press). We then risk assuming what we set out to explain.

These readings have given more attention to how development is reliable than to how individuals come to differ, because the idea of a plan draws most of its support from reliability. I have argued that the two cannot be separated, but the case for that rests more on the grasping study and the cortical maps than on anything worked out across the readings as a whole. Most of the brain material concerned sensory and motor cortex, and how far the account extends to capacities such as language requires further work.

Developmental Systems Theory argued that the organism is the center of the developmental system rather than one resource among others, because it is a living whole that maintains itself and engages with its surroundings. The readings have relied on that claim more than they have explained it. Part of what makes the organism the center is that things can go better or worse for it. For a person, what counts as better or worse has a history of its own, and these readings have not tried to explain it.

Cycles of Contingency

The infant reaching for the pencil does so from a body shaped by earlier development, and picking up the pencil makes further activity possible; holding it out can draw another person into what happens next. We call what the infant is developing “the precision grasp,” as if there were an ideal grasp somewhere that each infant gradually approaches. But there is no ideal grasp, only the acts of grasping by actual infants.

Susan Oyama, Paul Griffiths and Russell Gray called their edited book on developmental systems theory Cycles of Contingency (2001). If something is contingent, it depends on certain conditions being present, and it could be otherwise. Life cycles can be remarkably stable without being inevitable, because the conditions that sustain them keep being reconstructed.

A single human life is just as contingent, though saying that it could have gone differently doesn’t mean we can now change it in any way we like, because what has already happened matters. But we can ask which of the current difficulties depend on conditions that are still present, and whether changing those conditions would open new possibilities. The historical view lets us ask what is being reconstructed in a particular life, and what it would take for that life to go differently.

At the start of my course on development, I ask students whether there are instructions for a frog somewhere in its fertilized egg. Almost all of them say yes, in the frog’s DNA. That answer is reconstructed in each generation of students through what they are taught, and like any other cycle it could be otherwise. You might ask which assumptions about development you were taught, and whether you still hold them.

References

Gottlieb’s duckling studies and Kuo’s experiments are discussed in Developmental Systems Theory, and the cataract research and the experience-expectant distinction in Plasticity and Constraint.

Campbell, K., Horger, M. N., Pavlakos, G., Parma, V., & Marshall, P. J. (2026). Dynamics of infant preshaping: Longitudinal analysis using automated hand pose tracking from home video. Infant Behavior and Development, 84, Article 102232. https://doi.org/10.1016/j.infbeh.2026.102232

Karasik, L. B., Tamis-LeMonda, C. S., & Adolph, K. E. (2011). Transition from crawling to walking and infants’ actions with objects and people. Child Development, 82(4), 1199–1209. https://doi.org/10.1111/j.1467-8624.2011.01595.x

Karmiloff-Smith, A. (1998). Development itself is the key to understanding developmental disorders. Trends in Cognitive Sciences, 2(10), 389–398. https://doi.org/10.1016/S1364-6613(98)01230-3

Marshall, P. J. (in press). Psychology in 2125: A developmental science of living sense-making. Perspectives on Psychological Science. Preprint and background: https://peterjmarshall.com/psychology-2125.html

Molenaar, P. C. M. (2004). A manifesto on psychology as idiographic science: Bringing the person back into scientific psychology, this time forever. Measurement: Interdisciplinary Research and Perspectives, 2(4), 201–218. https://doi.org/10.1207/s15366359mea0204_1

Oyama, S., Griffiths, P. E., & Gray, R. D. (Eds.). (2001). Cycles of contingency: Developmental systems and evolution. MIT Press.

Ram, N., & Gerstorf, D. (2009). Time-structured and net intraindividual variability: Tools for examining the development of dynamic characteristics and processes. Psychology and Aging, 24(4), 778–791. https://doi.org/10.1037/a0017915