Week 4 RF-Object individuation

Question 1

Define Object individuation

Answer 1

A cognitive process which establishes and recognises 2+ distinct objects in an event.

Question 2

Go through the steps of the individuation process

Answer 2

1.You have to register an object (e.g. notice a bird at the feeder)
2.Maintain a mental representation of the object once it’s out of view (e.g. a bird that leaves the feeder)
3.Register a new object in the scene (e.g. a new bird that comes into view)
4.You then have to compare this new object to the other object
representations you have in your mind (are they the same?)
5.If it’s a new object you then have to form another object representation for the new object.

Question 3

Why is object individuation harder for infants?

Answer 3

lack of specific knowledge around objects

Question 4

Sources of information: Define spatiotemporal information

Answer 4

Info about the object’s location and motion
(An object can’t be in 2 places at once and can only travel on
unblocked paths)

Question 5

Sources of information: Define featural information

Answer 5

Info about the object’s size, shape and colour, surface pattern or
texture (an object can’t change these spontaneously)

Question 6

How did Xu and Carey (1996) use the Violation of Expectation paradigm to test if 10-month-old infants can individuate objects using spatiotemporal or property information?

Answer 6

(process same in both conditions)
-had a screen where 1 object would be brought out and returned and vice versa
-then screen is revealed where the expected outcome would show both objects whereas the unexpected outcome would show just one object (any reactions of surprise would indicate that they’ve distinguished the 2 objects as different)

Question 7

How can we measure infant-looking times when they’re currently unable to talk?

Answer 7

*Infants look longer at different stimulus, something surprising or
alarming
*Eye tracker and certain methodologies e.g., Violation of expectation paradigm

Question 8

What were the results/conclusions of Xu and Carey’s (1996) Violation of expectation paradigm?

Answer 8

*Expected as a baseline for infants to look at the 2 object outcome, as it’s more interesting.
*No difference between baseline and property/kind condition.
Conclusions:
-10 month old Infants were able to use spatiotemporal information BUT NOT property information to set up representations of objects.
*Follow up experiments show they only successfully use feature properties at 12 months.

Question 9

What were Wilcox and Baillargeon’s (1998) procedure which involved event monitoring?

Answer 9

*Tested 7.5 month olds and 9.5 month olds
-Ball-box wide screen condition (possible event=space for 2 objects behind the occluder)
-Ball-box narrow screen condition (impossible event=no space for 2 objects behind the occluder)
Both 7.5 month olds and 9.5
month olds looked longer at the Ball-box narrow screen condition

Question 10

Wilcox and Balleirgon (1998):
What is Event Monitoring?

Answer 10

*1 scene is shown with an object moving on a single trajectory.
*In order to succeed at the task,
infants must monitor a single
object

Question 11

Xu and Carey (1996): What’s Event Mapping?

Answer 11

*2 scenes being shown:initial
presentation of the object +
final reveal of the object
*In order to succeed at the task
infants must map one representation onto another.

Question 12

Why do infants prioritise spatiotemporal cues (‘where’) before using feature
cues (‘what’)?

Answer 12

*Spatiotemporal info causes infants to assign an ‘object index’
that sticks with the object following it through space and time. (like a mental pointer finger)
*This ‘object index’ doesn’t necessarily have to contain the featural info (the finger only tells you where something is not what something is)

Question 13

How do infants begin to use this feature ‘what’/featural info at 12 months? Leslie, Hall and Tremoulet (1998)

Answer 13

*Infants look longer at the
change in shape outcome showing they have encoded shape in their object ‘index’
-In terms of colour, infants look at both outcomes equally. They haven’t noticed the change in colour (shows 12 month olds don’t use colour to identify)

Question 14

Why does featural info do object individuation but not identification?

Answer 14

‘what’ and ‘where’ info is processed by different streams in the neural system
*Dorsal stream processes the where
*Ventral stream processes the what

Question 15

What’s the issue with object indexes in relation to memory?

Answer 15

-object index is held in STM so limits the number of objects that can be attended to at one time (more objects on conveyor belt=harder task)
*Adult object tracking studies show this limit to be 3
*Evidence of these limits provides support for the indexing theory

Question 16

How did Feigenson and Carey (2005) investigate whether there was a set limit on the no. of objects an infant can remember at one time

Answer 16

*Used a manual search task.
4 Different trials type:
* 1 object vs. 2 objects
* 2 objects vs. 3 objects
* 2 objects vs. 4 objects
-infant expected to search for ball after seeing it be put in box/retrieved by experimenter (expect it to be empty)

Question 17

Define the Crucial trial in Feigenson and Carey’s (2005) experiment

Answer 17

Did the infant expect to find
another object and therefore continue to look?

Question 18

What were the results + conclusions of Feigenson and Carey’s (2005) experiment?

Answer 18

*The infants showed increased
looking at the crucial trial
compared to the baseline trials,
showing they expected another
object to be there
*EXCEPT on the 2 vs. 4 trials
Conclusion: Infants have a limit of 3 object representations.
*Provides evidence in support of babies using object indexes to help
individuate.

Question 19

How do adults overcome memory limitations?

Answer 19

They use chunking=organising stimuli into groups based on perceptual, conceptual or spatial info e.g., alphabet, companies etc.

Question 20

How did Feigenson and Halberda (2008) investigate whether when given conceptual cues, infants can
overcome their limit on their memory via chunking

Answer 20

4 objects were hidden with 3 types of trial:
*4 were hidden only 2 could be retrieved
*4 were hidden all 4 could be retrieved
*2 were hidden and 2 could be retrieved
*Importantly, these 4 objects could be organised into ‘conceptual’
groups (also novel OR familiar)
spatially grouped=cat cat car car OR spatially mixed cat car cat car

Question 21

What were the results of Feigenson and Halberda’s (2008) experiment 1?

Answer 21

Increased searching (light blue bar) is a result of subtracting averaged
searching on the 2 trials when the box was empty (red bars) FROM searching when the box was expected to contain more objects (green bar)
*All infants showed an increase in searching for the missing objects.
*Provides evidence infants expected 4 objects and could exceed the limit of 3 object indexes
*EXCEPT for novel objects not spatially grouped together

Question 22

What were the object groups used in Feigenson and Halberda’s (2008) experiment 2?

Answer 22

*2 tokens, 2 types- 2 non-identical cats and 2 non-identical cars
* 4 tokens, 1 type- 4 non-identical cats
Aim was to see would infants perform the same with non-identical tokens from the same
category?

Question 23

What were the results of Feigenson and Halberda’s (2008) experiment 2?

Answer 23

*Infants showed increased searching for 2 tokens, 2 types.
*As in experiment 1, had grouped
the objects to remember there were 4 hidden.
*Didn’t show increased searching
for 4 tokens and 1 type, as
couldn’t group the objects to
overcome memory limits.

Question 24

What conclusions can be reached from Feigenson and Halberda’s (2008) experiment?

Answer 24

*Memory limitations mean infants struggle to individuate if shown
more than 3 objects
*BUT chunking can overcome this, if objects can be conceptually
grouped (categorised) or perceptually grouped (e.g., there were 3 cars and 3 cats)

Question 25

What’s the role of language in Xu and Carey’s (1996) study

Answer 25

-Found a correlation between
looking time and whether the children understood the object labels from the task.

*Suggests that maybe learning language can assist object
processing independent of memory and perception development.

Question 26

Explain Xu’s (2002) procedure

Answer 26

Similar procedure to Xu and
Carey (1996), except when
objects were brought out they
were presented with either:
*Two Object labels (“look it’s a X)
*One object label for both
*Silence (baseline)

Question 27

What was the result of Xu’s (2002) study?

Answer 27

*Infants in Baseline and 1 word
condition showed longer looking at the two object outcome.
*Infants in two-word condition showed longer looking at the one object (unexpected outcome)
*Provides evidence that infants in two-word condition had successfully individuated the objects and set up two distinct object representations.

Question 28

What was found in the follow up experiments from Xu (2002)?

Answer 28

*Objects presented with tones (experiment 2) and emotional
expression (experiment 4, e.g. ‘wow’) didn’t facilitate object
individuation
*However infants did succeed at individuating the objects when
presented with unfamiliar objects and nonsense words.

Question 29

What was Xu, Cote and Baker’s (2005) procedure?

Answer 29

*Also showed the same effect but
with a manual search task
*Also did a 2nd experiment with
emotional expressions e.g. yuck

Question 30

What were the result of Xu, Cote and Baker’s (2005) experiment?

Answer 30

Object labels: Infants searched in
the box for longer when researchers gave two object
labels
Emotional Expressions:No difference in the searching when researchers gave two words or one

Question 31

Why were the results seen in Xu, Cote and Baker’s (2005) experiment?

Answer 31

*Presence of object labels facilitated infants to individuate objects based on features alone, where at the same age in a similar task they had failed to do so (Xu and Carey, 1996).
*Proven not to be due to an increase in attention