Lecture 4: Conceptual Development

Question 1

How might children categorise objects?

Answer 1

 There are a number of different possibilities:
 Taxonomic categorisation (i.e. an ordered arrangement of groups or categories in a hierarchy)
 Perceptual categorisation (i.e. based on perceptual features - colour, shape)
 Thematic categorisation (i.e. based on themes – what would the object “go with”?
->A football might be associated with a t-shirt the child wears to play sports)

Question 2

Conceptual organisation

Answer 2

• concepts are organised in hierachically
• basic level: dog, cat, chair, cow, tree
• global (superordinate) category: furniture and animal
• > more difficult, ask furniture word think of basic level concept
• hierarchical organisation of concepts
• > global level then basic level

Question 3

Three hypotheses of conceptual development

Answer 3

Hypothesis 1
 Development involves first acquiring basic level categories, then global/superordinate categories can be constructed (Rosch, 1978)

Hypothesis 2
 Basic level and global/superordinate categories develop together

Hypothesis 3
 Global/superordinate categories develop before basic level categories

Which is more likely?
-we chose hypothesis 1

Question 4

How can we research young children?

Answer 4

 Infants and toddlers can be tested in different ways
depending on their age:
 Match to sample (around 2 years) (target and choose between options)
 Sequential touching (around 12 months)
 Tactual exploration and novelty preference (around 11
months)
 Preferential looking (2+ months)

Question 5

Match to sample task

Answer 5

 Bauer & Mandler (1989) were interested in how young children organised the concepts that they hold.
 They tested 1- and 2-year olds (using the match-to-sample task)
 The purpose of the experiment was to see whether children were able to match based on basic-level AND superordinate level (i.e. can they match a dog with a dog (basic level) and also a chair with a bed (superordinate)).

• basic level: see this one? can you find another one just like this one? (lion picture)
• global/superordinate match: see this one? can you find another one just like this one? (monkey, choice between banana (thematic) or bear (categorical)
• > choosing bear correct as categorical match

Results:

• not much difference between basic and global at 19 months, 25 months and 31 months
• wasn’t an age effect
• > suggest that children might forget variable instructions??? (look study can’t understand prof)

Question 6

Sequential touching as a measure of categorisation level

Answer 6

-Mandler and Bauer 1988
Sequential touching task:
 Given a set of small objects, infants tend to touch objects sequentially
 Infants touch objects from the same category in sequence more than chance
 Systematic touching in the first year is followed by sorting the in the second year
-basic view in literature that basic formed first then subordinate

• different toys
• basic: dogs and cars
• superordinate: animals and modes of transportation

 What did they find?
 12 & 15 months: Basic level
 Sequential touching for dogs v cars
 Random touching for vehicles v animals
 20 months: Global/Superordinate as well
 Sequential touching for both sets

Question 7

Do infants use knowledge to identify global/superordinate categories?

Answer 7

• Pauen 2002
• hypothesis: (look at study can’t understand prof)
• wanted to see whether perceptual categorisation or based on knowledge

Realistic looking animals and furniture
 11-month-old infants given sequence of objects to examine (either animals or furniture)
 Test object was novel and from either the same or a different category
 Infants examined the new-category object for longer than the familiar category object
 This could be evidence for superordinate categories, or it could be based on simple perceptual differences (e.g. materials)

artificial looking animals and furniture (all look more perceptually similar)
 Repeated with new sets made from artificial materials
 Perceptual differences between sets reduced
 If perceptual differences are important, infants should find this discrimination much more difficult
 Results were exactly the same – longer examining of object from new category
 Infants used pre-existing knowledge of categories to guide their examining

Question 8

What about very young infants?

Answer 8

• Quinn and Johnson 2000
• 2-month-old infants familiarised to pictures of 16 different mammals
• Significant preference (59.8%) for furniture picture (global category)
• wanted to test the three hypothesises (which first or at same time)
• familiarize with animals, then presents with picture from same category or different
• infants look longer at the chair
• > shows able to do global categorisation as chair novel
• > habituated to mammals

2-month-old infants familiarised to pictures of 16 different cats

• then showed dog and cat pictures
• No preference (47.6%) for dog picture (basic
  category) at 2 months, but sig. preference (63.6%) at 3 months
• global form before basic level in this experiment

Question 9

Protypes

-can infants learn prototypes?

Answer 9

• prototypes: good examples of category
• > have prototype in head and helps us to categorise
• Younger and Gotieb 1988
• habituation study
• dots forming shapes/lines
• prototype, then presented distortions (low to high)
• then presented another???

->seem to be familiar with prototype
(can’t understand prof look at article)

Question 10

Do category preferences change with age?

Answer 10

-Blanchet et al. 2001
-picture of a cat
-thematic match: milk
-basic match: another cat
-superordinate match: dog
 Early development of basic-level categories may promote deeper understanding of links between things
 Target picture: Dog 1
 “Can you find another picture that goes with this one?”
 Alternatives: Dog 2 Bone Pen
 Basic-level match: Dog 2 (same category)
 Thematic match: Bone

3-year olds classified more with basic-level categories
 4-year-olds preferred thematic relations (involves knowledge)
 What does this suggest?
-thematic relations involved application of knowledge
->could be younger don’t have as much knowledge
->or could be processing

Question 11

Biological Knowledge

example of distinction between animate and inanimate being fundamental to human experience

Answer 11

 A key distinction is between biological and non-biological (animate vs. inanimate)

 The man who mistook his wife for a hat had lost the ability to identify the animate

Question 12

Biological motion

Answer 12

 ‘Biological motion’ is the term used to describe the unique way that an animate object moves.
 Biological motion is commonly portrayed using ‘point light displays’, where white dots are placed at key
places on the body and recorded while an individual performs an action.

• infants able to identify biological motion but not when its inverted
• > couldn’t discriminate between inverted dots and random dots (bertenthal et al 1985)

Question 13

How well do infants understand intentions?

Answer 13

-Meltzoff 1995
 18-month-olds watched experimenter model 5 actions
 4 groups
 Demonstration (target): specific target acts (dumbell pulled apart)
 Demonstration (intention): target acts not demonstrated, experimenter failed
 Control (baseline): no demonstration
 Control (manipulation): no demonstration, but experimenter handled objects

big difference between demonstration and control (20ish % vs 80ish%)
-infants able to understand intentions of adult

Question 14

Do infants imitate actions for inanimate objects as well?

Answer 14

• Meltzoff 1995
• human demonstrator and mechanical device mimicking these movements (two different conditions)
• is this specific to humans?
• can they recognise intention with machine?

 Human attempt: 60% pull dumbbell apart
 Machine attempt: 10% pull dumbbell apart
 Infants were significantly more likely to pull the dumbbell apart when they watched the human than when they watched the machine (p

Question 15

Goal directed actions

Answer 15

• woodward et al. 2009
• bear and ladybug, hand grasping bear
• constantly inferring behaviour of others
• saw over and over bear and hand grasping
• then presented with new trial
• > new goal trial goes for lady bug
• > new side trial goes for bear but on different side
• 5 months: new goal more of a response?? surprised going for ball
• > means have to infer something about intention
• > with claw results same, only human arm infer

Question 16

Can 11 month old infants predict the outcome of human actions?

Answer 16

• Cannon and Woodward 2012
• eyetracking
• see which one child looks at more
• hand and claw
• ball or frog
• Where will the hand or claw go?
• familiarisation, swap and test

YES, INFANTS ARE MORE LIKELY TO PREDICT ACTIONS IN THE HUMAN CONDITION THAN IN THE CLAW CONDITION

• Eye tracker data showed more predictive looks to goal object in hand than in claw condition
• hand: looked at original goal object
• claw: looked at original place

Question 17

Knowledge about true nature of living things and objects

Answer 17

-Gelman and wellman 1991
 Children make judgements about the ‘insides’ and ‘outsides’ of objects
 ‘Insides’ are more important than ‘outsides’ for understanding the true nature of an object
 Compare a person and a waxwork
 ‘Insides’ question
 “Which has the same kinds of insides as x?”
 ‘Outsides question’
 “Which looks most like x?”

target: pig
- piggy bank or dog
- then ask about outside what looks most like x?
- then ask which has the same kinds of insides as x?

 Target Picture: Pig
 Same outsides: Piggy-Bank
 Same insides: Puppy
 Both 3- and 4-year-olds performed better than chance
 Older children were more accurate
 3-year-olds – 58% correct
 4-year-olds – 73% correct
 Equal number of both types of error
 Appearance: Pig & Piggy Bank have same insides
 Internal properties: Pig & Puppy look similar

 Young children have knowledge of some key differences between biological and non-biological things
 Ability to distinguish insides and outsides didn’t develop
 What developed was the ability to cope with conflict between insides and appearance
 Piggy-bank: looks like the pig, but does not have pig insides
 Suggests improved processing capacity and possibly executive function (inhibition) could explain these effects (production deficit)

Question 18

Knowledge about growth

-animate vs inanimate

Answer 18

-It is the case that biological, animate things grow with
time and may change their appearance.
-Inanimate things do not grow with time, but may
change in appearance (e.g. knocks, scuffs, renovation)

-Rosengren et al. 1991
 Beliefs about naturally occurring transformations were examined in children aged 3 to 6 years in four experiments.
 Children were presented with baby animals and brand new artefacts. For each target, they were given two pictures – one picture was always the same size, and the other was either smaller or larger.
 Children were asked “Which picture shows x after it’s been around for a very long time?”

1) artefacts ( stimulus set with same size and smaller and stimulus set with same size and bigger)
- asked which is same? new to old?

2) animals (stimulus set with same size and smaller and stimulus set with same size and bigger)
- baby to adult?

 If children know animals grow and artefacts do not
 Always choose same size artefact
 Never choose smaller animal

• 3 y/o vs 5y/o
• older children know animals grow bigger and artefacts don’t
• younger children know animals grow bigger (more difficultly when animals smaller) but unsure about artefacts (were just as likely to say grew as stay the same)

 Principle of growth is first understood in the biological domain
 Artefacts are less well understood at younger ages
 Another characteristic feature of artefacts is that they are designed
 This ‘design stance’ guides our judgements about category membership
 It also promotes tool use
 Young children do not have a concept of design

Question 19

Artefacts and the concept of design

-3 experiments

Answer 19

-German and Johnson 2002

 How do people identify and reason about object functions? How do they decide what objects such as cars, clocks, and cradles are for?
 Therefore, as an object was intentionally designed to serve a specific function, we would expect people to retain the original name and function.
 Test this idea by asking children about objects that change
 Name (category)
 Function
(split up instead having together)

Category: A long time ago an inventor made this thing and called it a ‘tog’, but now someone else owns it and he calls it a ‘fep’. What is it really, a tog or a fep?

Function: This is a ‘tog’. A long time ago an inventor made it for collecting leaves, but now it belongs to someone else. One day, he used it for catching fish.
What is it really for, collecting leaves, or catching fish?

-adults vs 5 y/o
- Children know that the original category (name) cannot change
 But they don’t think that objects have a designed function

• looking to choose original
• adults high score
• 5 y/o do well in categorisation but not function (no difference)

Defeyter et al 2007
Brick
Novel functions: door stop ect
Design functions: build house
-5 y/o generate more novel functions, 7 y/o generate more design funstions
Could be a simple explanation, i.e. that older children have more experience with these objects.

• repeat experiments with novel objects and novel functions
• > a wiz for making holes in paper, a bif for rolling out play-doh
• same results
• 7 y/o thought of twice as many design functions, 5 y/o more likely to develop novel functions

->something greater than knowledge

Question 20

Functional fixedness in Duncker’s candle problem

Answer 20

• matches, box of tacks and candle
• task to mount candle on the front of an wall, standing upright and illuminate room
• tack box to wall and put candle inside
• if presented pins outside box more likely to come up with a solution
• pins inside box, fixate function in head

Question 21

Novel objects

-baseline vs function demonstration

Answer 21

• Defeyer and German 2003
• baseline: no comment on purpose, solve task without
• function demonstration: stick for making light (new function)
• > if people came up with novel way to use stick then more likely to solve then if fixated on fact used to make light
• 5, 6 and 7 y/o
• slower solutions in older children, aka functional fixedness effect at older age

 Younger children do not have a concept of design and can think of novel uses for artefacts
 They are immune to functional fixedness, and easily find a novel solution
 Older children are constrained by their concept of design, and take longer to set aside the function and think of a novel one
 However, there are many advantages to a design concept:
 Simplifies use of artefacts
 Able to immediately identify the function
 Don’t have to select the function from the range of possibilities