Is the Sun Alive and are the Clouds Made of Sugar?

Before they step into a classroom, young children have the daunting task of making  sense of the world without being aware of all scientific knowledge that already gives an explanation as to why things are the way they are. As adults, our previous knowledge biases the way we understand the world, especially when it comes to things that we have been told by others or things we learned to be true in school. But take a moment and try to remember the questions you had about the world when you were little. What are the things you used to think were true but found out later to be otherwise? Perhaps you have a funny story of your own. At a very young age, I used to think that if I lose something I need to start crying and it will show up somehow – of course this was because I noticed that very often I would find the missing thing just as I began crying  with frustration. Such erroneous connections are not uncommon. Fully understanding causality is one of the tricky reasoning tasks for young children. As well as the task of classifying natural kinds – how to know what is alive and what is not alive.  An easy route to understand the world is to generalize your own human experience to other phenomena you observe, and children haven been known to use a lot of anthropocentric thinking – but when is this helpful and when not? This article will focus on how children employ their reasoning to understand natural phenomena without having been taught scientific theories in a formal setting.

Many things children observe in their environment do not come with a visible connection between cause and effect. This is often the case with natural phenomena. Every child is very familiar with natural phenomena such as the rain, the sunrise and sunset or the growth of plants, but the mechanisms that lead to  these phenomena are hidden from the naked eye. Researchers use such phenomena to investigate children’s ideas about causality. For instance, Piaget observed three tendencies in children’s ideas about what clouds are made of: until six years old they believe that clouds are solids created by either humans or deities, from six to  nine years old they see them as masses of smoke and from ten years onwards they start to understand that the clouds are suspended water droplets (Piaget, 1973). Continuing this stream of research, other scientists identified different developmental stages of comprehension about what clouds are: “bags of water that open up to create rain” at  seven years old, creation of clouds by the sun and generation of rain by wind which moves the clouds at  nine years old, and adopting the complete scientific understanding between nine and fifteen years old (Bar, 1989, as cited in Fragkiadaki & Ravanis, 2015). Moreover, Fragkiadaki and Ravanis (2015) studied preschool children’s assumptions about how clouds are formed and found that children of four, five and six years of age most frequently provide explanations that include causal associations of different physical entities, for example, the wind mixes water into a small ball which is then turned into a cloud by cold. The second most frequent type of explanation was similar to a recipe involving technical factors, for instance, a factory that makes clouds from water and sugar and places them in the sky with a big ladder. It is interesting that, in this study, only one child provided a scientific description, albeit incorrect, by referring to clouds as being water in gas form. This shows how diverse children’s explanations are, but also reveals that the emergence of causal thinking is a gradual process and children’s intuitions are not always compatible with the scientific explanations. Before reading such studies, I expected that children’s overactive imagination might make them say that little dwarfs are knitting clouds and pressing buttons to start rain more often than they did – however, interestingly enough, their explanations were often quite plausible, given the information available to them.

Sometimes children invoke an agent behind the natural phenomenon, other times they solely use natural forces to explain it. Regarding the cause of natural phenomena, pre-school children most frequently give responses that combine the two, as shown by Christidou and Hatzinikita (2006) who asked children how plants grow. An example of such explanation, documented by them, would be that: “God gives water to the plants when he makes it rain, which feeds the plants so that they grow.” Less frequently, Christidou and Hatzinikita (2006) encountered purely naturalistic explanations (plants get water from soil which makes them bigger) or purely non-naturalistic explanations (plants grow if you talk nicely to them). 

Children appear to have a tendency towards assigning a purpose to objects found in nature. For example, mountains are to be climbed or giraffes are for children to see them in the zoo (Kelemen, 2003). Intriguingly enough though, Kelemen and DiYanni (2005) found that this is not the case regarding natural phenomena, with children of ages six and ten preferring to explain natural events in a physical-reductionist way – how parts of a system work together to create a phenomenon – rather than in terms of “the use” of the natural phenomenon. Additionally, in this study, children of six years old and older did not believe that natural phenomena are man-made. However, when it came to objects rather than phenomena, children relied on purpose to explain the existence of artifacts, animals and non-living natural entities. In this case, believing that something exists for a purpose was also positively correlated to the belief that it was created with intention, by different kinds of agents, like humans, God, the Earth or an unidentified “they”. An example of an explanation containing both agency and purpose was given by one of the children about birds: “ because…they needed more nature in the country because it probably was dirty in the olden days so they invented…invented birds come up” (Kelemen & DiYanni, 2005).  One wonders whether it is not the parents that had used such explanations and children assimilated them. Maybe parents say things like “the sun is here to make light for us”, “dogs are here to be humans’ friends” or “the sun goes down so that we can go to sleep”.  Research based on parents’ recollections and recordings of spontaneous speech between parents and children found that parents explanations are most often causal – “the milk turned brown because of the chocolate in the cereals”, then teleological – “the sun is hot to make us warm” (Kelemen, Callanan, Casler, & Pérez-Granados 2005).  

When children look around in nature, they see living beings, artifacts and non-living entities. Being able to classify them correctly is necessary if you want to reason about their origins or create expectations about it. Stepping on a rock won’t hurt it, but stepping on a cat’s tail would. Making the animate-inanimate distinction is fundamental for causal reasoning because living beings have self-generating actions, some can have goals or can actively engage with the environment, while non-living objects cannot move at their own will, cannot grow or reproduce. Notably, the problem of agency in causal reasoning is especially relevant to the animate-inanimate distinction. Starting from one and a half years old, infants show an emergent understanding that alive and not alive are different categories, as shown in studies which use sequential touching paradigms (the infant is presented with a set of objects and the order in which they touch the objects is recorded) (Rostad & Poulin-Dubois, 2012). Later on, four-year olds appear to use movement to distinguish living from non-living (Wright, Poulin‐Dubois, & Kelley, 2015) – but this leads to categorization errors because mushrooms are alive even though they do not move, while the sun moves on the sky even though it is not alive. That is why young children (before age six) have trouble categorizing plants as alive (Brule et al., 1993). An exception is when children are asked to classify non-living and living as a function of regrowth – in this situation, four-year olds show that they know that if damaged, a plant can regenerate by itself, while a chair or a car needs to be fixed by people if broken, and therefore we cannot expect artifacts to regrow and people cannot fix living things (as in putting back the leaves on a plant) ( Backscheider, Shatz,  & Gelman, 1993). The fact that children perform better in tasks that explicitly use a key distinction between alive and non-alive such as growth (asking them what things can grow and what things can’t) has made researchers wonder whether the way children interpret the word “alive” might steer them in a wrong direction. Maybe children associate “alive” with liveliness and movement and less with the idea of a “living thing”.  Leddon, Waxman and Medin (2009) found that when using the term alive, even ten years old children misclassify plants as not alive. However, when the authors repeated the same experiment using the term “living thing”, children from six years old onwards made far less mistakes in clarifying plants as living things. This shows the importance of picking the right words and checking how children interpret them when we try to teach them something. 

We can also look at pre-school children’s failure to classify plants as alive from a different perspective. Many of the studies that reached this conclusion compared plants with animals, and found that children include animals in the alive category way earlier (Inagaki & Hatano, 1996). What is different about animals? Children might find it easier to label animals as alive because they notice the commonalities between features and functioning of their own bodies and animal’s bodies. Like humans, most animals have eyes, noses, mouths and they show biological behaviors like eating and breathing. Plants have an appearance which does not resemble humans, and physiological processes such as breathing and assimilating nutrients are not observable from the outside. This idea is supported by studies that look at children’s anthropomorphic thinking. When we anthropomorphize, we attribute human properties to non-humans entities, for instance saying: the plants love the sun, the mosquitoes are mean or the computer is tired. Children anthropomorphize animals more often than plants (Tamir & Zohar, 1991). Until age seven, children understand biological functioning in terms of wants and beliefs because they transfer human properties to animals, while making few generalizations from animals to humans or across animals (Geerdts, 2016). This proves useful when learning about animals because it serves as a conceptual bridge. The fact that anthropomorphism is a way for filling a knowledge gap is supported by studies which show that urban children, who have less direct contact with animals, reason more anthropocentrically than rural children, and children who have pets are also less likely to anthropomorphize (Unsworth et al.,2012). Although anthropocentric thinking makes the concept of animals more accessible to children, it might make it more difficult for them to understand the nature of plants.

What all this research shows is that children’s preconceptions about nature tend to follow certain patterns. The often funny explanations preschool children give are proof that their mind is actively trying to engage with the world and use all bits of information it can gather to form a meaningful story about what is happening around them. Studying how children understand nature is interesting in itself, but also very relevant for education. I believe that in learning, addressing incorrect intuitions is more important than making sure children have access to the correct facts. This is because the sort of reasoning that feels right and has been in place for many years will stick to them more strongly than facts that contradict intuition, and, when not addressed, this can become a barrier to forming a good scientific foundation.