Object representation

Naive Physics and Naive Psychology

Picture of the Tubes apparatus

Main researchers: Prof. Bruce M. Hood

Collaborators: Prof. Alan M. Leslie (Rutgers), Dr. Sara Baker(University of Cambridge)

A major area of interest in children's reasoning is related to perseveration. Children often make errors of returning to search at locations where objects were previously found. Previously we have demonstrated that intuitive notions about how objects will fall interfere with the ability to search for fallen objects that are no longer visible. This is demonstrated by the tubes task where toddlers erroneously search in an aligned box for a fallen ball, even after they have seen a tube deviating the ball's path. The fact that typically developing children repeat this error over many trials suggests that their behaviour is governed by a rigid concept of gravity that is difficult to overturn. With age most children solve the tubes task and we believe they achieve this by developing a general ability to inhibit impulsive behaviour. The ability to inhibit impulses leaves room for more flexible, context-appropriate behaviour, such as searching for the ball in a non-aligned box.

Recently we were awarded a 3-year ESRC grant (RES 062-23-0506, “Conceptual Change as the Combination of Domain-Specific and Domain-General Mechanisms”) to investigate this hypothesis by looking at a variety of children's na´ve and persistent notions of the world, and asking how these notions may develop alongside general inhibitory control.

In a current project with primary school children, we are studying a classic problem in na´ve physics: the relation between an object's weight and its velocity. In our "dropometer" task, children witness a race between two objects down a vertical chute. Both objects reach the ground at the same time, but when children are aware of the differing weights of the two objects, they tend to report that the heavier one fell first. The na´ve, and persistent, notion that heavier objects fall faster can also be found in adults under certain conditions. We are currently investigating how to help children modify their na´ve view of falling objects so that they are more objective observers of the world around them.

Another ongoing project conducted in a local preschool will give us insight into individual children's developing ideas of the world around them. Dr. Baker visits the same children every three weeks to chart changes in their na´ve physics (e.g., the tubes task, described above) and na´ve psychology (e.g., perspective-taking abilities). We'd like to know whether children's knowledge in one of these domains can influence their knowledge in the other one, or whether children's knowledge is "specialized" in physics or psychology from a very young age. Following the same young children throughout a period when their na´ve notions are developing rapidly will help to answer questions such as: "is there such a thing as general intelligence?", "do all children develop in the same way or are there many different individual profiles?", and "how flexible are children's ideas about the physical and social worlds around them?".

We are always looking for willing participants in our studies. If your child is interested and is between the ages of 24 months and 10 years, please contact us on (0117) 954 6613 or bcdc-lab@bristol.ac.uk.

Object Solidity

Main researchers: Dr. Nathalia Gjersoe, Prof. Bruce M. Hood

Collaborators: Prof. Alan M. Leslie (Rutgers)

Infants look longer at events that appear to violate the rule that one solid object cannot pass through another solid object. This finding has been taken to suggest that infants have an understanding of the solidity principle that they use to guide their expectations about events in the world. And yet, when we ask children at 2 and a half years of age to use this understanding to search for a toy that has rolled behind a screen and been stopped by a partially visible wall, they consistently fail. Do children lose the knowledge that is apparent in infancy or is the dissociation something to do with the nature of the task?

Picture of the Four Door ApparatusWe conducted a study with toddlers that showed that they frequently looked at the correct door yet searched at the wrong door. This indicates that they do have knowledge of the solidity principle and can use it to make predictions about events in the world but that there is something about the nature of getting up to search for the object that disrupts this understanding. We suggest that it may be competition from other doors, such as the one that was previously correct or one that the child has chosen as a favourite, that is getting in the way of correct search on this task.This interpretation has been supported by exploring the types of errors that young children make at different ages and by subsequent studies showing inhibition biases to the previously correct location on an approach and avoid paradigm.

We have also found that allowing the children to choose the location of the wall improved their ability to search correctly, suggesting that participation is important in figuring out these types of problems. However, if they returned a week later they were unlikely to have learnt from this, indicating that they still have a relatively weak grasp of the problem.

Tunnel Task

Main researcher: Dr. Nathalia Gjersoe

Collaborators: Prof. Bruce M. Hood (University of Bristol), Prof. Laurie Santos (Yale)

How do we come to understand that an object is the same individual even when it disappears and reappears from behind things? We explore this question about the development of object individuation in toddlers by making use of the “tunnel effect” – an illusion compelling us to see featurally different objects as one magically changing individual if they appear to be travelling along the same path. In this study a variety of objects that differed accross a number of featural dimensions were used to determine how susceptible young children are to the illusion.

We looked to see whether 3-5 year olds would assume that there where one or two objects involved in the display. Overall, there was a clear developmental change with nearly all 3-year-olds, half the 4-year-olds and none of the 5-year-olds succumbing to the illusion as evidenced by their real-world search. Control studies established that all of the children understand the instructions of the task and respond appropriately when there are no confounds in the display. Indeed, even when they are shown exactly how the trick works, 3-year-olds (but not 4-year-olds) still revert to the error, suggesting that this is a strong enough visuo-spatial bias to guide real-world behaviour in young children even when they know better.

This pattern held over apparent changes in colour, orientation, shape and combinations of surface features. Having the objects differ along ontological categories (animals and plants) increased the rate of 3-year-olds overcoming the illusion but to a rate no better than chance. Indeed, the only manipulation that resulted in young children robustly overcoming the illusion was giving the objects individual labels before test trials commenced. Ongoing studies are exploring the extent to which labeling acts as such a strong cue because it implies deeper categorical differences than surface properties alone.