One topic I want to dive into particularly deeply coming out of the Study Group in Reggio Emilia is that of toddlers making representations: choosing to draw or paint or otherwise create a representation of another object.
On a visit to a preschool in Reggio Emilia I observed four children in the infant-toddler room (so, <18 months old) sitting on chairs around a table. On the table in front of each child was an orange, a sheet of paper, a paintbrush, and a set of paints in different shades of orange. The children were painting with orange paint (because that’s the only color choice they had) onto the paper in a variety of lines, blobs, and squiggles. In a Q&A session after we looked around the school I asked the Pedagogista (the school’s head teacher/learning coordinator) whether she thought the children were in fact representing oranges in paint on their paper and she said unequivocally “yes”: “What the child paints may not look like an orange but it represents that to them.”
About a year ago – long before I learned about the Reggio approach to children’s learning – I read Pennie Brownlee’s booklet Magic Places, a guide to supporting young children’s creative art work. Brownlee states: “Children working in the scribbling stage [age 0-6 years] are not depicting anything. Because they’re not depicting anything, it is inappropriate to ask a scribbler, “What are you drawing?” It is equally inappropriate to ask a scribbler to draw any particular thing: they are making marks and patterns” (Magic Places p.27). Brownlee does go on to note that the “symbol stage” lasts from 4 to 10 years so apparently may be some overlap between the two stages, but according to Brownlee’s definition children younger than four are unlikely to be depicting anything.
Based on this reading of Brownlee’s work I was so surprised to hear the Pedagogista say that the 18 month-olds were indeed “painting oranges” that I queried Daniela Lanzi, another Pedagogista who lectured in many of the learning sessions of the Study Group, during a different Q&A session. Her response (in Italian via an in-person translator) was as follows:
“Representation is a concept that only belongs to the human mind. Children may make a gesture (on paper) by chance but in doing so they realize that they can make a sign. Maybe in the early years we can’t describe what they are drawing but the child is representing something. Toddlers are fully aware of the representation process. Through representation you give a concrete, or visible, shape to an action or thought – neuropsychologists agree on this. Out of a single subjective representation we later achieve a collective representation [Lanzi then specifically referenced the video of the “Road of Notes” project we’d just watched, in which children invented their own musical notation], when different minds agree on a common shared language or code through which they can formalize what they think.”
Lanzi encouraged me to read Piaget on this topic which, of course, I did – as well as some more contemporary thinkers. The most direct comment on Piaget’s work about this that I have found is the following:
“One important aspect in the development of cognition is the appearance of the symbolic function. This refers to the fact that from 2-4 years the child begins to develop the ability to make something – a mental symbol, a word, or an object – stand for or represent something else which is not present. For example, the child can use a mental symbol of a bicycle, or the word “bicycle,” or a small schematic toy to stand for the real bicycle when it is not in immediate view” (Ginsburg & Opper p.73).
So Piaget – on whom the Reggio approach very heavily relies – says that symbolic function begins to develop between ages 2-4 – and even so, no specific mention is made of taking the activity one step further: from forming a mental symbol of an object to making marks on paper to depict that object. Piaget later explains, as an aside to a description about how children don’t see names as a label for an object but rather as an intrinsic quality of that object (e.g. the word “sun” isn’t a label for the sun but a quality of the sun itself):
“This phenomenon is analogous to the “intellectual realism” which M.[Monsieur] Luquet has so clearly demonstrated in children’s drawings. They draw what they know about an object at the expense of what they see, but they think they are drawing exactly what they see” (Piaget p.70).
Taken together, these statements to some extent explain the Reggio Emilia approach to symbolic representation for the over-twos: to paraphrase, children understand that a symbol can represent a thing, and children make drawings that may not look like the thing but still symbolize the thing. Two facets of this require further digging: Piaget offers no evidence that under-twos can symbolize a thing, and Brownlee says they cannot symbolize until much older – perhaps even as old as age four.
Digging into the research
I did a lot of reading to see if I could get to the bottom of this topic, eventually working my way to the book “Children’s Understanding and Production of Pictures, Drawings and Art,” a large chunk of which is available on Google Books, including probably the most helpful chapter for this topic: “The origins and development of pictoral symbol functioning” by Tara Callaghan of St. Francis Xavier University. Tara was kind enough to engage in a fairly lengthy Q&A with me so I could try to sort out what’s going on here. I’m going to draw from different parts of our email conversation, as well several of her published papers, to try to understand what Carys understands and how I can support her in developing her understanding of symbols.
When do children understand that pictures are symbols for a real object?
“A host of studies have now addressed when toddlers and young children “understand” the symbolic function. Age estimates for this insight vary from 13 to 30 months old. A major issue in resolving the discrepant estimates centers on defining what constitutes “understanding.”” (Callaghan, Rochat and Corbit 2014 p.348) Callaghan conducted a study where one experimenter showed a series of Canadian children some toys of two kinds (say, balls and whisks) and drew a simple line drawing to represent each kind. The toys were divided between two boxes according to type and the pictures used to label the appropriate boxes. After the first experimenter left the room, a second experimenter asked the child if the child wanted to play a trick on the first experimenter by switching the labels on the boxes; the child agreed and the labels were switched. A child who understands that the pictures are representations should “understand that the [first experimenter] will hold a false belief about the contents and will look in the wrong box for their toys” (ibid). The children passed this test (i.e. understood that the first experimenter would look in the wrong box) in the 3- to 5-year age range. In other words, it seems highly unlikely that an 18-month-old would understand that the picture represents the contents of the box and thus understands the symbolic function.
In another study (described in greater detail later on), Callaghan asked toddlers to draw simple objects that can be represented using lines and circles – an important choice because in their scribbles the children were clearly already using these forms. Callaghan thus says that if we don’t understand what a child’s drawing represents it’s because they aren’t yet capable of representing something, not because they lack the physical skills to represent what they understand.
At what age does your research show that can children make their own symbols (i.e. see an object, form an intention to draw it, and then draw it)?
“Taken together, the results from infancy and toddler studies that explore when children appreciate representational intention in the drawings of others, and instantiate it in their own drawings, show that by approximately 2 years children identify representational intentions and are beginning to appreciate something about the normative practices governing the use of pictoral symbols. From approximately 30 months children begin to be able to muster the intention to represent and produce their own pictoral symbols” (Callaghan & Corbit 2014 p.278).
So do you think it’s possible that the 18 month-olds I saw in Reggio Emilia could possibly have been painting the oranges on the table? (Or is it possible they might actually have been painting representations of the containers of orange paint?)
My hunch is that perceptual matching would be in play in what you saw, and definitely not an intention to symbolize, especially since they only had different shades of orange paint to choose from (it would have been interesting to see what happened if the children had been given a choice of paint colors…).
Parents often label the things children see in books, and most parents ask children what their scribbles are (except those who think that asking will stifle the child’s creativity), so children have a clear indication from those around them that pictures go with labels. [Jen’s note: This is supported by Jolley & Rose (2008): “…even babies in the first months of life can notice a similarity between a three-dimensional object and its picture” (p.208).] This knowledge is not the same as knowing that pictures represent things, people, events in the world or in the imagination. Even if the children had announced the intention to draw an orange and then produced a scribble I don’t believe that’s sufficient evidence of intention to represent the orange. You will know that a child is representing an orange when you offer them a choice to paint an orange or something that looks totally different (e.g. a whisk) and the image painted looks like an orange and not like a whisk.
So you are saying that children who understand symbolic representation will decide to represent an object with a drawing (or painting) – and then draw the thing in a way that others can recognize it. Luquet (on whom Piaget relies) seemed to think something different was going on.
“According to Luquet, children initially discover resemblance between their scribbles and a potential referent in a process he called fortuitous realism. [Jen’s note: this is exactly the process that Daniela Lanzi described: “Children may make a gesture (on paper) by chance but in doing so they realize that they can make a sign.”] A child who makes a wedge-shaped scribble and names it “sailboat” captures Luquet’s notion of fortuitous realism. Subsequently, children progress through two stages that are primarily distinguished by the intention first to depict what they know about an object in their drawings (intellectual realism) and then to depict what they see (visual realism)…From Luquet’s classical view of drawing development, children’s scribbling was considered to be a sensorimotor activity where the child initially explored the medium and, once a similarity of form between the scribble and referent was noticed, then attempted to create meaning in increasingly more successful attempts toward visual realism…[but] contemporary research does not convincingly support this claim… Social engagement with adults who use pictures as symbols in their interactions with children [has appeared] as a dominant influence.” (Callaghan & Corbit 2015 p.272-278).
In other words, children model the things they see in the people around them: for example, a 12-month-old will primarily look at a picture of a toy if she first sees an experimenter looking at the same picture. If the experimenter attempts to touch and manipulate the picture of the toy, the baby will as well (Rochat & Callaghan 2005).
I was out hiking with Carys when she was about 17 months old and one day she spontaneously pointed to a picture of a bike on a “no bikes” sign and said “bike.” (I was pretty surprised.) More recently (23 months) she saw a picture of a row of beach huts from a distance and said “train” – upon reflection it actually does look quite a bit like a train. Is this evidence of “symbolic insight” and if not, what connection am I missing here? It seems as though she must have some understanding that the picture represents the thing since these were spontaneous utterances and not coerced by me in any way, right?
While “Preissler and her colleagues make the controversial claim that infants know a picture is a symbol around 15 months” (Callaghan & Corbit p.276), I believe the examples you give about your daughter are of her seeing perceptual similarities – a row of huts has a similarity to a row of train cars. The insight that something represents is different from verbalizing a resemblance. It is understanding what the culturally-shared function of an item is, which is a complex concept to understand and emerges for most children around 3.5 or 4 years and becomes further refined based on the cultural support for its further development.
It seems as though a child probably doesn’t just flip a switch one day from not being able to engage in symbolic representation to fully understanding it. Do children go through stages in their development of understanding related to symbolic representation?
Yes. In the presymbolic stage, infants often confuse pictures and their referents – attempting to grasp objects depicted in high-quality photographs. There are also anecdotal reports of infants moving drawing tools in a way that represents the referent (e.g. bouncing the pencil across the page to represent a rabbit hopping), which indicates that the child has confused the symbol and referent.
Later in this stage, children can differentiate between 2D depicting artifacts and the depicted 3D reality. So your daughter knows the picture of the bike is a picture, not a real bike. Children begin to move a pencil in a relatively controlled fashion within the confines of a page, contemplate their pictures and refer to them, but are still oblivious of the symbolic nature of pictures.
Next comes the emergence of symbolic representation, in which the child can appreciate the resemblance as well as the difference between the 2D artifacts and the depicted 3D reality. By approximately their third year children may label their scribbles but they do this afterward, dressing their scribbles with symbolic meaning – and likely noticing similarities between their scribbles and objects existing in the environment. Children in this stage do not intend in advance to make a form stand for a particular thing even though they are physically adept at producing a rich array of graphic features (lines, circles, dots, etc.) that would be sufficient to produce symbolic drawings.
The child develops a conceptual understanding of the link between the picture and the object. When presented with a realistic photo of an apple the infant realizes that “it is an apple” but also that “it is not an apple.” The child has become capable of “dual representation,” the ability to hold in his head two meanings of an object, a critical precursor for symbolic representation. Children begin to show planning and intentionality in their drawings, announcing beforehand what they intend to draw. This is the time when children begin to produce simple forms such as tadpole drawings (a round shape with lines for limbs) to stand for people. They also understand that a pencil drawing and a color painting and a photograph of an apple all represent the apple.
Beginning between age 3.5-4 years, the child uses a variety of forms to symbolically depict a referent, depending on the purpose of the drawing. She may add wild hair and big feet to distinguish herself from her parent in a drawing of her family. Again, this stage typically appears at around 3.5-4 years and is unlikely to be found in a child of 18 months or
Later on, around 5 years, the child refines his understanding of symbolic representation and begins to construe the idea that a person intentionally produced the picture to convey a message, and can attempt to understand the content of that message and for whom it is intended. The children can convey the emotions of happy, sad, excited, and calm in their drawings.
[Jen’s note: this section draws heavily on Rochat & Callaghan 2005, but refines the levels discussed in that paper per Callaghan’s developed thinking in the intervening decade. The revised approach, conflating the five levels into three stages of presymbolic, emergent, and refinement stages, is published in Callaghan & Corbit 2015.]
Makes sense. Is there anything I can do to help Carys get better at understanding symbolic representation? Is her understanding of symbolic representation primarily driven by development (as Piaget might argue) or by external factors that I can influence?
It’s probably some combination of both. A colleague and I (Callaghan & Rankin, 2002) decided to test whether scaffolding [helping a child to learn a concept] between 2 to 3 years could accelerate symbolic functioning. We gave children a bag with a set of 12 toys. The child pulled out one toy at a time, and the experimenter quickly rendered a simple, unambiguous sketch of the toy (commenting “Let me draw that toy for you,” “Watch me,” etc.) on small cards. The experimenter ensured that the children were watching her draw and placed the toy on top of the picture she had drawn (commenting, “Let’s put the toy on its picture over here, etc.). After drawing and placing all of the objects once, the experimenter put them back in the bag and the same procedure was repeated. A control group received placebo training in which the children chose the toys, and the experimenter lined up the toys making comments to maintain the interest of the child (e.g. “That’s a neat toy, do you have one like that at home?”), but did not draw the toys.
The children were aged 28 months at the start of the study. Half of the children who were trained received this training for four months from the start of the study (the “Early Training” group), and the other half received the placebo training for four months followed by actual training for four weeks in the fifth month (the “Late Training” group).
We then tested the children’s symbolic comprehension by showing a picture of a ball to the child, who was instructed to choose the object portrayed in the picture from a choice of two similar balls. Previous studies found this task to be difficult for children aged under 3 years either because the children cannot distinguish the objects using a verbal label (because they are both balls) or because the children do not know the name of the object if unfamiliar objects are used.
We found that the Early Training group understood symbols about two months earlier and could make their own symbols about a month earlier than the Late Training group. The Late Training group caught up in one month after their own training occurred, and this improvement occurred only after training – no child showed evidence of understanding or producing graphic symbols before that time, which means that the improvement was likely due to the training rather than the child’s maturation over the course of the study. The Early Training group needed to be exposed to training for some time before we saw benefits, while the Late Trained children saw benefits after only a month of training.
We see that parents in Western culture show generic symbols in picture books to infants and toddlers and name the pictures, but the pictures are not presented as symbols of particular real-world objects – and it was this connection between a real-world object and a pictoral representation of that object that our study made explicit.
Using the three stages we discussed earlier, we see that we can’t train a child with presymbolic understanding to depict refined symbols, and older children (in the Late Training group) also needed less training to grasp symbolic comprehension than the younger (Early Training) children so there must be some developmental function at play. But the fact that we can help children move from one stage to the next implies that there is a social component to this development as well.
But is there a reason why I would want Carys to recognize symbols sooner? I know you did a study that compares Canadian children’s performance on the drawing and toy-switching test with that of children from Peru and India, who passed these tests on average 3-6 months later for the drawing task and 1-1.5 years later for the toy switching test (Callaghan et al. 2011). But I’m thinking about research showing that children who are in teaching-intensive kindergarten may learn to read early, but then they get burned out by second grade because they’re sick of having “learning” stuffed down their throats. Do you see any danger here in teaching children to recognize symbols before they pick it up for themselves, or even any advantage that the Indian or Peruvian children might have by learning this so much later?
We absolutely do not think the Indian or Peruvian children are at a disadvantage, or less capable, than Canadian children. We have to consider children within their respective cultural contexts – the Indian and Peruvian children thrive in their environments – but these environments do not have an expectation to produce pictoral or pretense symbols as part of early development of the mind. This is a very Western concept of child cognition. (As is the concept that a parent’s responsibility is to explicitly teach/scaffold/support their child’s cognitive development by facilitating these symbolic activities.) Language is likely the only universally supported (by parents) symbol system.
I can’t think of any reason why you would want your daughter to recognize symbols earlier. Our training study and the cross-cultural study had scientific goals – of demonstrating that social supports are necessary for symbol systems to develop – not applied goals of showing how to accelerate symbolic insight. I agree that too much parental pedagogy can stress a child (and the parent!) – it is better to enjoy the process, tune into what the child is capable of, and provide the supports when the child is ready.
I also don’t think the Indian or Peruvian children have any advantage by learning later – we must consider the culture and what is expected of children by parents, teachers, and other significant people in their worlds – in general, doing well at what matters in your culture puts you at an advantage.
Pennie Brownlee says that adults shouldn’t draw for a child:
“Imagine that a three-year-old asked you to draw a cat for her. You could comply and draw a cat for her, and it would probably look like this (photo of cat drawing from Brownlee’s Magic Places). The three-year-old has plenty of experience of cats. She knows about cats and she has built up her own concepts of cats. By drawing for her you have taken away her opportunity to show you just what sort of cat she had in mind. You have taken away her chance to sort out her ideas about cats and how to get them down in symbol form. Worse, she may think that is the only way to draw cats and model all hers on that particular one. This applies not only to drawing, but to all creative activities including blocks, construction sets, playdough, collage, carpentry, sewing…If you do it for them you stop them learning and growing, and they won’t think of themselves as someone who can give anything a go” (Magic Places p.16)
What do you think about this?
This is a possibility, but I think to some extent it depends on the child’s personality – I’m not aware of specific research that has been conducted on My older son seemed to be a person who held back if someone else did things better, but my younger son wasn’t like this. I think some balance of drawing together and encouraging independent drawing is not likely to stifle creativity.
One of the questions I went to Reggio Emilia with was to what extent my own creativity could impact Carys’ development. I’m learning to draw, I sew, knit, make presentation boxes, make books, fix stuff around the house…even if I’m not doing these things for her as part of her play, does exposing her to creativity help her to see its value or stifle it if she feels like I’m “better” at these things than she is?
Taking the big picture, I think your creativity will inspire your daughter, and enveloping her in a life filled with creativity may actually lead your daughter to think that these are worthwhile activities, since the adults she (and all children) so wants to emulate are doing them too – kind of like drinking coffee [Jen’s note: reminds me of my grandfather, who used to say that certain things – like asparagus, I think, and Brussels sprouts – were “adult tastes,” so my sister and I would desperately try to like them so we could have adult tastes too…]. There’s a lot of complexity to the development (cognitive, social, motivational, emotional…) that your daughter is going to go through in the coming years that will all impact the way she views creativity.
I’ve also had a conversation on some of these issues with Jeanette Prince, an Atelierista (Art Studio Teacher) at a school in Southern California, whom I met in Reggio Emilia. She neatly summarized for me: “As a parent, don’t hold back on your passion for creativity. Your child will fall in love with creativity too, but let them make their own choices.”
I acknowledge that I am a student of these ideas and also that I have not read all of Piaget’s copious works, or even all of the massive amount of literature available by many researchers on the topic of how symbolic representation develops. But from a review of some of his relevant writing and also summaries by others of his work, plus more recent published literature, it seems as though the Piagetian view of relatively young children (starting at around age 2) being capable of intentionally producing a drawing that symbolizes something else is probably out of date. Instead, this capacity seems to begin to develop between the ages of 3.5-4.
This is not to say that giving a child 18 months old or younger the opportunity to use paints is a pointless activity – far from it! I would think that the children get some enjoyment from spreading an unctuous material on paper, seeing the colors move and change shapes, and mixing colors together. They are likely forming an understanding of these properties that will help them to use these materials in a more sophisticated way when they do develop symbolic understanding, possibly leading to more “advanced” representations than children who have not had this exposure (anyone know of research on this?!). And while the claim that an 18 month-old child can symbolize an orange likely does the child no harm, I believe we are ignoring the preponderance of current research if we believe it to be true.
Ivana Soncini, a psychologist in the Reggio Schools, mentioned in a lecture during the Study Group that she and the other practitioners of Reggio Emilia are in an ongoing dialog with the field of neuroscientific research and implied that the practitioners are willing to and indeed do shift and have shifted their views on learning as our collective understanding of children’s development changes (although I will add that when queried as to which of the latest research we – the Study Group attendees – should read to get ourselves current on the latest thinking Soncini cited Margaret Mead, Piaget, Vygotsky, and Bruner – three of the four are dead and Bruner is 100 years old), yet I do not see that this shift has occurred with regard to symbolic representation.
As usual, I’m very receptive to a constructive discussion, particularly if you can point out something I’ve misunderstood which has lead me to an illogical conclusion.
I should also note that while this is the second recent post in which I’ve called something about the Reggio Emilia approach into question, I’m not down on it at all. I believe it’s a really solid way to think about early childhood education. But the interesting parts for me to explore are the pieces around the edges that don’t make as much sense to me; not the stuff in the middle that everyone agrees about.
Brownlee, P. Magic Places. (Currently out of print but may be back in print soon; check: http://penniebrownlee.weebly.com/books.html)
Callaghan, T.C., Rackozy, H., Behne, T., Moll, H, Lizkowski, U., Warneken, F., & Tomasello, (2011). Early social cognition in three cultural contexts. Monographs of the Society for Research in Child Development, 76(2), Serial Number 299. http://onlinelibrary.wiley.com/doi/10.1111/mono.2011.76.issue-2/issuetoc
Callaghan, T. & Corbit, J. (2015). The development of symbolic representation. In Vol. 2 (L. Liben & U. Muller, Vol. Eds.) of the 7th Edition (R. Lerner, Series Ed) of the Handbook of Child Psychology and Developmental Science (pp. 250-294). New York: Wiley.
Callaghan, T., & M. Rankin (2002). Emergence of graphic symbol functioning and the question of domain specificity: A longitudinal training study. Child Development, March/April 2002, 73:2, 359-376.
Callaghan, T., P. Rochat & J. Corbit (2012). Young children’s knowledge of the representational function of pictoral symbols: Development across the preschool years in three cultures. Journal of Cognition and Development, 13:3, 320-353. Available at: http://www.psychology.emory.edu/cognition/rochat/lab/CALLAGHAN,%20ROCHAT,%20&%20CORBIT,%202012.pdf
Jolley, R. P. & S. Rose (2008). The relationship between production and comprehension of representational drawing. In Children’s understanding and production of pictures, drawings, and art (C. Milbrath & H.M. Trautner (Eds)). Boston, MA, Hogrefe Publishing. Chapter available at: http://www.staffs.ac.uk/personal/sciences/rj2/publications/Jolley%20and%20Rose%20chapter.pdf
Rochat, P. & T. Callaghan (2005). What drives symbolic development? The case of pictoral comprehension and production. In L. Namy (Ed.) Symbol use and symbolic representation. Mahwah, NJ, Lawrence Erlbaum Assoc. Chapter available at: http://www.psychology.emory.edu/cognition/rochat/lab/WhatDrivesSymbolicDevelopment.pdf