Source: Laboratories of Judith Danovitch and Nicholaus Noles—University of Louisville
Children are not the best artists. Sometimes it’s easy to pick out the characteristic triangular head, whiskers, and tail of a cat, but children often describe elaborate scenarios that they depict as a beautifully unrecognizable mess. Thus, given children’s questionable artistic talent, how do they know what their drawings, and the drawings of others, represent? One way children identify pictures is by relying on resemblance. If it looks like a cat, then it’s a cat. However, some pictures do not clearly resemble any real object. In this situation, children must use other means to figure out what the picture represents, including their understanding of what the person who created the picture intended it to represent.
By their first birthday, children are sensitive to the intentions of other people. They know that people’s actions are driven by their goals, and they can infer a person’s intentions even if the goal-directed action is not successful (e.g., they understand a person struggling to turn a lid intends to open a jar, even if they never see them succeed in opening it). By about age 3, children can use this understanding of intention to guide their interpretation of drawings and other pictorial representations. They apply this understanding both to identifying their own drawings, and to interpreting drawings created by another person.
This experiment demonstrates how to measure children’s use of intention to interpret otherwise ambiguous pictures based on the methods developed by Bloom and Markson.1
Recruit 3- and 4-year-old children. Participants should have no history of developmental disorders and have normal hearing and vision. For the purposes of this demonstration, only one child is tested. Larger sample sizes (as in the Bloom and Markson study1) are recommended when conducting any experiments.
1. Data collection
- Drawing task: Part 1
- Prepare a crayon and four sheets of blank paper. Invite the child into the room and have them sit at a table.
- Ask the child to draw a lollipop and a balloon on two separate sheets of paper.
- Ask the child to draw a picture of themselves and a picture of the experimenter on the remaining two sheets of paper.
- Alternate the order in which the drawings are created between subjects, so half the children draw the lollipop and balloon first, and half draw themselves and the experimenter first.
- Collect all of the child’s drawings and put them aside for approximately 15 min while the child completes the size task and the oddity task.
- Size task
- For this task, create four sets of hand-drawn pictures on separate sheets of paper in advance. Each picture should contain two of the same shapes of unequal sizes (e.g., a large circle and a small circle).
- At the start of the task, tell the child: "I am going to show you some pictures drawn by a boy or girl (gender should match the child’s gender) your age who has a broken arm. This child tried really hard to draw good pictures, but because of the broken arm, the pictures did not always come out looking the way they wanted."
- After placing the first picture in the child’s view, say: "They drew a picture of a spider and a tree. Would you like to see their picture of a tree and a spider?"
- Point to each object in the picture and ask: "What is this?"
- If the child is reluctant to respond, repeat the introduction and follow up by asking: "Can you point to a spider? Can you point to a tree?"
- Repeat the procedure for three additional item pairs: mouse and elephant, dog and house, and flower and bicycle.
- Counterbalance the order of the items in each pair across children and drawings.
- Record the child’s responses using audio or video.
- Oddity task
- Immediately follow the size task with the oddity task.
- For this task, create four sets of hand-drawn ovals on separate sheets of paper, where one oval has a different orientation than the rest. Place the dissimilar oval in either the second or third position in the series of four.
- Tell the child: "I am going to show you some more pictures drawn by the child your age who has a broken arm. They drew a picture of three shoes and one sock. Would you like to see their picture of three shoes and one sock?"
- Point to the dissimilar oval and ask: "What is this?"
- If the child is reluctant to respond, repeat the introduction and follow up by asking: "Can you point to a sock?"
- Repeat this procedure for three additional item pairs: cows and horses, pigs and chickens, and cookies and apples.
- Counterbalance the order of the items across children and drawings.
- Record the child’s responses using audio or video.
- Drawing task
- Following the size and oddity tasks, retrieve the drawings and say: 'I found the drawings you made earlier. I really enjoyed looking at them. What is this one?" Then present the drawings one by one in random order and record the child’s responses.
- Two independent raters code the child’s responses to each question to determine if they were accurate or not. Resolve disagreements through discussion.
- The dependent variable is the total number of trials (0-4) for each task in which the child described the entire drawing correctly.
- Analyze whether the child used the artist’s intention to correctly identify the objects in the drawing for each task. To do this, conduct t-tests against chance performance (a score of 2) for each task. Comparisons can also be made across age groups and tasks.
A child’s ability to interpret an illustration—identifying the object or person it represents—is based on the physical characteristics of the picture, in addition to social-cognitive processes.
On one hand, children can use a drawing’s actual resemblance to an object to make their interpretation. For instance, if a picture looks like a dog, then they’ll call it a dog.
However, pictures can be abstract, and don’t always closely resemble what they are meant to depict.
In such instances, psychologists Paul Bloom and Lori Markson have shown that as long as a child is told of the artist’s intention—for example, that they meant to draw a mouse—the child will be able to correctly interpret the drawing, even if it’s just a disorganized collection of shapes.
Employing the techniques of Bloom and Markson, this video investigates the criteria 3- and 4-year-olds use to identify what illustrations represent, describes the experimental design, how to collect and interpret data, and notes that similar methods can help elucidate how people understand symbols, abstract art, and physical objects.
In this experiment, 3- to 4-year-old children complete three tasks—size, oddity, and drawing—in which they must interpret hand-drawings based on intent and analogy rather than physical resemblance.
During the first part of the drawing task, children—generally thought of as unskilled artists—are asked to draw two sets of very similar items on separate sheets of paper: a lollipop and balloon, and a self-portrait and picture of the researcher involved. These pictures are put aside until the second phase of this task, which occurs at the end of the experiment.
In the intervening period, the size task is performed. This involves sequentially showing children four sets of researcher-created hand-drawings, each of which consists of two differently sized shapes—a small and large version of the same shape, side-by-side.
The trick here is that children are told the pictures were created by a gender-matched child who had trouble drawing with a broken arm. They’re further informed that the illustrations are meant to represent a small and large object, such as a mouse and an elephant.
For each given set, children are asked what each picture represents.
The idea is that children will use what they know about the injured artist’s intent—what they were aiming to draw—to interpret the small and large shapes as being the smaller and bigger animals, like a mouse and elephant, respectively.
In this task, the number of trials in which children correctly identify both drawings is the dependent variable.
This is followed by the oddity task, where children are again presented with four sets of drawings, which they are told were generated by the same broken-armed child.
The difference is that these illustration sets consist of four same-sized ovals, but one oval is in a different orientation than the other three.
Depending on the trial, children are informed that these pictures are meant to depict three of the same, and one dissimilar, object—like three cookies and an orange.
For each set, they are asked what the unique, rotated oval represents. Again, it’s expected that children will use the artist’s intent to identify the oval that’s unalike as being the singular object named
The number of trials where the atypical oval—and by extension the three similarly-positioned ovals—are correctly interpreted serves as the dependent variable for this task.
In the last phase of the experiment, children are presented with the pictures they initially drew, and asked what each represents.
Unlike the size and oddity tasks, here children must use memory of their previous intentions to interpret their own drawings.
In this case, the dependent variable is the number of trials where a child correctly identifies the pictures they drew. Children are expected to remember their original intent and perform more accurately here versus the other two analogy-based tasks.
Before the experiment begins, organize the drawing stimuli into two sets: one to be used in the size task, and the other for the oddity task. In addition, prepare crayons and four sheets of blank paper for the child to use when drawing.
Greet the child when they arrive, and present them with crayons and blank paper.
For the first part of the drawing task, ask the child to create—in a random order and each in a different color of crayon—a balloon and lollipop, as well as a self-portrait and picture of the researcher. Afterwards, gather the four drawings and place them off to the side.
Then, begin the size task by showing the child one of the drawing sets with a large and small shape, and tell them that these images are meant to represent a spider and a tree.
Gesture to each shape on the paper, and ask the child what they represent. If the child is reluctant to respond at first, encourage them by asking if they can point to the tree and to the spider.
Repeat this procedure for the remaining three sets of large and small shapes, and for each of these trials, tell the child that the drawings represent either a house and dog, elephant and mouse, or bicycle and flower. Record their answers using audio or video.
After all four trials of the size task have been completed, start the oddity task. For the first trial, tell the child that the person who created the pictures meant to draw a sock and three shoes.
Point to the dissimilar oval in the illustration, and ask the child what this shape represents. If the child hesitates in their response, inquire if they can point to the sock in this picture.
Repeat these steps for the remaining three oddity task trials, telling the child that the ovals represent different objects or animals in each. Again, record the child’s responses.
Approximately 15 min after the start of the experiment, perform the second phase of the drawing task by showing the child—in a random order—the four drawings they initially created. For each, ask the child what they drew and record their responses.
When the three tasks have been completed, have two independent raters code all of the responses, in order to determine their accuracy.
To analyze the data, average the number of trials in which children correctly interpreted the drawings by age and task.
For both age groups, conduct t-tests against chance performance for each of the three tasks.
Notice that the majority of pictures shown during the trials of the drawing and size tasks were correctly interpreted by 4-year-olds, reflecting significantly better than chance performance. Similar results were also obtained for these tasks in the 3-year-old group.
However, notice that although the 4-year-olds correctly interpreted drawings in the oddity task at a level significantly above chance, the 3-year-olds performed only as well as at this task as they would have by chance alone.
Collectively, these results suggest that children as young as age 3 can use what they know of an artist’s intent to interpret drawings, even if the pictures are themselves ambiguous. This ability is better developed at age 4 though.
Now that you’re familiar with how to design an experiment to explore the relationship between artists’ intentions and a child’s ability to interpret drawings, let’s look at other ways researchers assess how knowledge of intent can shape human perceptions.
In certain instances, drawings can serve as symbols, for example, to represent locations.
As a result, some researchers are trying to assess how children come to understand symbolic representations, such as maps.
In particular, psychologists have determined that what matters in such cases is not the appearance or shape of a symbol, but rather shared knowledge about what a symbol is meant to represent.
There is also interest in whether knowing an artist’s intent contributes to both children’s and adults’ ability to appreciate abstract art.
To this end, some psychologists are examining how the title of an art piece affects a person’s enjoyment of it.
Finally, although applied here to drawings, psychologists are also looking at how knowledge of a creator’s intent enables people of all ages to categorize and use physical objects.
For example, if an object was created and sold with the intent of being a vase—but it resembles a tall drinking glass—most people would still categorize it as a vase, and use it to hold flowers.
Many researchers are extending this work to compare the degrees to which knowledge of intent and morphological features of an object—like size, shape, and color—factor in this categorization process.
You’ve just watched JoVE’s video exploring how an illustrator’s intent influences a child’s ability to identify an illustration. By now, you should know how to present hand-drawings to children, and collect and assess data relating to their interpretations. You should also have an appreciation of how knowledge of the creator’s intent—whether for physical or symbolic objects—influences a person’s ability to use or understand an object.
Thanks for watching!
Researchers tested 24 4-year-olds and found that 4-year-old children correctly identified 87% of the pictures in the drawing task (Figure 1). They also identified 83% of the pictures correctly in the size task and 68% of the pictures in the oddity task. All of these results were significantly better than chance at the p < 0.05 level. The researchers also tested 24 3-year-olds and found that they identified 76% of the pictures in the drawing task and 69% of the pictures in the size task correctly, which also reflects better than chance performance. However, the 3-year-olds only identified 54% of the pictures correctly in the oddity task, which is no different from chance performance. This suggests that children as young as age 3 can name a representation of an object based on the creator’s intention, even when it does not have a strong resemblance to the object’s actual shape, although their ability to do so is more fragile at age 3 than at age 4.
Figure 1: Percentage of trials in which children from each age group identified drawings correctly in the drawing, size, and oddity tasks.
Applications and Summary
Pictures and drawings are symbols, and the ability to identify what a symbol represents is important for the development of a wide range of skills. As early as age 3, children realize that understanding the intentions of a drawing’s creator can allow them to identify a drawing that might otherwise be unidentifiable. Moreover, children as young as age 3 can do this even if the drawing bears no resemblance to the intended object. Although appearance and shape are certainly still valuable for identifying pictorial representations, this demonstration shows that children can use social-cognitive processes to infer what a drawing represents.
Understanding that an artifact’s identity is a function of its creator’s intent is also important for categorizing objects and knowing how to use them. For example, an object might look like a tall drinking glass, but if the person who created it intended it to be a vase, then people call it a vase and put flowers in it instead. This principle also applies to understanding language, including written words and other types of symbolic representations, such as maps. What matters in these cases is not the appearance or shape of the symbol, but shared knowledge about what a symbol is intended to represent.
Children’s early understanding that what a picture represents is determined by the artist’s intentions may also be the basis for appreciating abstract art. Adults can look at a painting that appears to be only blobs of paint and understand that it represents the night sky or a group of people. Not only can children do the same thing, but they can also appreciate that even when their own drawings do not turn out looking exactly like a cat or a birthday party, they still represent the objects they set out to draw.
- Bloom, P., & Markson, L. Intention and analogy in children's naming of pictorial representations. Psychological Science. 9(3), 200-204 (1998).