Glossary of child dev.

Start Studying! Add Cards ↓

*approach does good job capturing HOW dev. might happen.....
-gradual changed in synaptic weights can lead to stage-like changes in behavior!

*Approach tries to provide links to neuroscience
-exciting links to how the brain might work

-not neurally realistic
-often requires many trials to learn
-use of feed back not always realistic
-kids spontaneous learn w/o feedback
-comp told right or wrong to learn
info processing:
gen. char.
- main focus:cog dev.
-computer metaphor (way most think)
- models of cog. processing and how info processing systems change over dev.

input=> stuff=> output
info processing:
1.perceptual input, sensory register, processing
- attn, working memeory, longterm memory, categorization, decision
Shiffrin & Atkinson's store model
-remember more stuff short term memory
-time diff long term memory/get rid of stuff/lot more stuff there when adult than when 2 yrs. old
-response generation kids get better older know how to act couple yrs. later
-from kinder. to 3rd grade controlled
-IONIC MEMORY- image still there when all lights go off still see image
info processing:
how does dev happen?
-size of sensory register increases
-size fo memory changes
-attn. resources get redirected
1.domain gen. vs. 2.domain specific
1.spill over other areas thinking

2.doesnt spill over
Does cog processing get faster over dev? (domain gen. view)

5 tasks
-visual search (find letter among #'s)
-mental rotation (letter h backwards)
-mental add (problem 1+2)
-tapping (how many in a min. can do)
-name retrieval
increase in processing speed:
-same rate of speed up across tasks
-dev. looks domain gen.
(other thinking: relate mylenation of brain to faster processing)
Do expert children follow the domain gen pattern?
3 tests for adults & expert chess kids:
1.digit span (read list # and say back)
2.chess test
3.chess digit span (pics chess board/ reproduce them/recognize board one thing)

-2&3 domain specific view
Results & conclusions
-digit span=> adults better
-chess test=> kids better
-chess digit span=> kids better

-chid knowledge more richly connected
-more chunked in domain of chess
=>dev. looks domain specific
Info processing:
how does dev. happen
-domain gen. changes in speed of processing, memory space, etc.

-changes in specific info child knows within a particular domain
-more info stored in LTM
-more rocust retrieval cues
-info. organized more efficiently
info processing:
how do these things happen
-changes in brain (mylination)
-creation of new insights (variability) and selection of what "works"
view of cog dev inspired by the brain's sturcture

-neural networks (connectionist networks) comp. models inspired how brain works

-info. in network is distributed across many, many neurons (all mult tasks)

-neurons are connected to synapses

-"weighting" given to signals at each synapse determine what neurons "know"
connectionism = learning
Changes in synaptic strength:Hebbian learning

*neurons that fire together,wire togethers

*when two neurons "turn on" at same time,increase the synaptic weight connection them

*this makes the neurons more likely to turn on together in the future

*a neural basis for forming associations (camels and horses look similar; balls are round)

-Edleman video strengthened neural connection
Variability & Selection:
-In challenging situations, child generate a varitey of problem solving strategies

-with repreated experience, strategies that work best "survive" and vice versa
Siegler and Jenkins (1989):
How do children learn the "min" strategy
(min. take time solve problem!)
-start counting from larger of two numbers

2 + 4=?
the connectionsit approach to dev.
-try to build models that minic how child learn in particular situations (behave)

-examine the performance of the neural network model and the child to see if they might be using the same processing "rules"
Connectionist Model Ex:
Model Piaget Balance Beam Problems
-kids go completely with weight regardless how far away from folcrum

-2nd phase distance in special circumstances

-3rd phase mult. dimensions
Connectionist Model Example
Model is presented w/ hundrends of balance-scale problems

-weight problems more frequently (as in "real world")

Examined the performance of neural network model-matches child's performance pretty well!

-model initially showed strong "weight"

-gradually model started to atend to distnace (strong "distance" connections) when weight =

-finally model able to attend to distance even when weight unequal
Siegler and Jenkins:
-children use a variety of strategies:
fingers, retrieval, min

-tons variability
-variable use each strategy
across sissions

-but also w/ in sessions.... used
diff. strategirs on SAME problem
in same session
overlapping waves model
-any 1pt. in dev. not single strategy

-always try out diff strategy but dont lose old ones!
Tests of dual representation hypothesis
Shrinking machine
*doesnt require think from one room from the other

-nonsymbolic task
=>child assume only 1 doll and 1 room
-2 1/2 succeed in task
-only one doll/ one room
-perceptual salience of symbol affects performance

-young childern's symbol use in task specific....symbolic thought not an all-or-none dev. change (sound familiar??)
The Relative Weighting Perspective:
Newcombe and Huttenlocher have provided a framework for thinking about these findings
-4 global spatial systems

-2 are egocentric:
-RESPONSE LEARNING-reach to right
-DEAD-RECKONING-i am currently at this place in the world (if you keep track of all your movements, you should always know exactly where you are! Desert ants and sailors do this really well)

-2 are allocentric:
-CUE LEARNING-the toy is in the toy box
-PLACE LEARNING- toy is 10in. from left edge of sandbox
The Relative Weighting Perspective:
the View
-spatial performance generally requires the combination of egocentric and allocentric info

-people "weight" info. differently in diff. situation

-over dev. child learn more and more effective ways to weight spatial info.
The Relative Weighting Perspective:
egocentric to allocentric shift revisited
-by this view, this shift in dev. does not reflect stages as Piaget proposed

-rather, the shift reflects a change in which info child have learned to rely on

-world-centered spatial info tends to be very reliable (becasue toys generally stay where we leave them!)good
A more sophisticated understanding of the world
given the gen trend toward a more world-centered understanding of space, where can ask how sophisticaed do kids reall get....
-do they, for instance, understand and use spatial symbols

-ex: maps
-need to understand the relative layout of stuff on the map (eg, distances)

-need to understand that the map refers to or stands for a real place in the world (symbol)
spacial symbols and dev
Using a map requires "dual representation"

-understanding detalils of map
-understand how themap and real space relate to one another

-DeLoache: young child have diff w/ dual representation
-2 rooms (1 life size, 1 scale)
-hide toy (snoopy) in scale model
-child searches in life-sized room

Child understand that scale model is a symbold for life-sized room?
-2 1/2 no!
-3 yes!
-not problem forget location but cant go back and forth btwn big/little
initial findings
-2 yrs. have trouble finding snoopy

-3 yrs. sucessful
why dont 2.5 yrs. succeed?
dual representation hypoth:
-scale model have both concrete and abstract nature

-must simultaneously represent both

-young child drawn to concrete
tests of dual representation hypoth
-less interesting (less concrete)
-2 1/2 do better
-3 yr.=great
-model more interesting (more concrete)
-3 yr. old do worse less likely do more abstract thing

2 1/2 = worse
beyond Piaget:
continuous view of spatial cog.
-sandbox data suggest that the "egocentric to allocentric" shift is not an all-or-none change over dev

-this makes sense since people must always coordinate egocentric all allocentric info

-it doesnt do me much good to know that my keys are on the desk in my office if i dont know how to get there!

-instead, it looks like there's a shift in emphasis over dev.- child rely more and more on world-centered info.
Spencer et al. (2001):
A-not-B Sandbox
-hide a toy 6 times at an A location

-then hide a toy 3 times at a nearby B location

-2 yr. old searches on the B trials were biased toward A

-so even though children are more allocentric, egocentric representations of space still influence what they do
Example of Continuity:
A-not-B Sandbox
-classic example of an egocentric spatial error: A-not-B error
-Smith and Thelen: the habitual response of reaching to A overrides the memory of the toy at B

-although 2 yr. olds show impressive world-centred memory abilities in the sandbox task, they still make this error!
Conclusions about Language:
Interactionist perspective
-child use basic perceptual and learning
(shapes vs. materials (functions))

-parent AND child actively involved

-context (especially social aspects) help
-matters joint-focus attn

-child learn to learn words

so gen processes become specific to the task of learning language
Var. isnt just present
approach provides insights into how dev. might happen
-var. and selection

microgenic method is becoming mroe prevelent
-provides insight into the details of how child learn new things

-overlapping wave model is discriptive only
-need mroe detailed models to understand mechanisms that produce variability and selection
Informaiton Processing
-has re-framed my classic questions
(domain gen v. specific)

-has lead to deeper understanding of HOW cog change happpens (connectionism)
Vygotsky's sociocultural theory
*-every psych process is social in orgin

1. 1st btwn individuals

2. later within the child and "process" is internalized

-memory is in social context
-ex: create memory with family
-went to zoo- random walk things come into their head (1st grader)
-parents help sturcture converstaion/ memory for them
Vygotsky's sociocultural theory:
EX: pointing
-early interacitve gesture
-initially: unsucessful attempt to grasp something
-parent aids child
-through interaciton, pointing becomes gesture for others/ incontext
Vygotsky's sociocultural theory:
-activity occurs over and over
-assoc. with interpersonal activity
-later, child does on own
-activity goes "underground" (becomes mental activity)

*fully internalized
Private Speech:

-preschoolers talk to themselves
-doesnt sound like conversation (no sense)
-declines over time (doesnt disappear)
Private Speech:
-private speech reflection of pre op. EGOCENTRISM

-child dont realize other people cant understand them

-thought went away after out pre op. period
Private Speech:
=>more private speech in tough situations

-form of self-direction

-initially from parents

-child adopt parents behavior turn it in words (prompts from parents)

-child talks to themselves as an adult providing cues might
Zone of Proximal Dev.(ZPD)
-can do some things without adult's help (independent problem solving)

-can do other things only with maximal help from adult

-ZPD is this range of ability
Scaffolding (kids learn)
adults create supported situations within the ZPD to help extend child's current skills

-recruit interest
-reduce steps to solving problem (1st do this then next do this)
-maintain pursuit of goal
-indicate divergence from ideal solution (bad)
-control frustration
-demonstrate how to do activity

*promote learning w/ in ZPD
Limitations of Vygotsky's theory
-no real explanation of how internalization occurs

-too much emphasis on interaction
-discounts discovery learning by the individual in socially isolated situations
Vygotsky's Major Insights
-cog dev. occurs in contexts that include social partners

-this social context critical to emergence of new behaviors
Components of Language
-PHONOLGY: speech sounds

-SYNTAX: arrangement of words into sentences
-MORPHOLOGY: use of markers to indicate meaning (eg: "s", "ed" past tense)

-SEMANTICS: meaning
Amazing language learning
-at 6 mo. babbling
-say 1st word around 1 yr.
-combine words around 2nd yr.
-understnad and procedure most grammatical constructions by 4-5 (ed) extensions
-between 12m -5 yrs learn on ave. 18 new words A DAY!
Theories of Language Development:
Nativist Argument
"Instead fo proceeding by trial and error -- unsure of whether "doggie" refers to apart of a dog, to one dog in particular or to anything with four legs-- children start with a set of innate biases"
Theories of Language Development:
Behaviorist arguement
-language learned by stnadard processing of operent and classical conditioning

Skinner- correct usage is reinforced, incorrect usage is not

Theories of Language Development:
Interactionist Arguement
-language is a product of interaction of genes, envior, and experience

BATES: basic perceptual and learning abilities are molded by culture and society to become language specific

*starts domain gen dev. set abilities domain specific
Spatial Cog. Dev:
what do child know about space?
-EX: how do child remember where things are? (socks, toys)

-PIAGET:child understanding of space changes through a sequence of stages

-OVERALL, there's an egocentric to allocentric shift
-early egocentric view fo space
(3 mountian task)
-later: formal understanding of space not linked to self
(euclidean Geometry- any pt. in space by x y z cordinates)
Semantic Development:
yr. old vs. words
-learning meanings of words

*1 yr. 1word
*2 yr. 300 words
*3 yr. 1000 words
*4 yr. many thousands
*5 yr. 10,000 words
Quine's Problem
how do children know the referent of a new word given an infinite number of possible referents(means)

Quine's Problem:
What parents do to help estb.
the referent (scaffolding)
-point(to rabbit)
-use voice inflection (focus right thing)
-use routines (get kid point to it)
-refer to previous experiences (look w/ if you have per w/ if at home)
-joint focus of attention (JFA)
Baldwin (1991):
-2 novel toys, kids pay with them
-one toy in view, one in bucket
-E waits for child to focus on visible toy and then labels:
-visible toy (follow-in condition)toma
-bucket toy (discrepant condition) child looks visiable to label one in bucket
-novel label (toma or peri)
-repeat 4 times
-observed child's behavior in both conditions and then asked child to "get the toma"
Baldwin (1991):
-follow-in: easliy learned words
-discrepant: more labeling error
-child look at E more indiscrepant condition

=>mis-match btwn child's focus of attention and direction of E's voice

=>child sensitive to JFA
Quine's Problem:
Things children do to help figure out the referent (what it is)
-follow parent's direction of gaze

-follow parent's point

-pick up objects and ask

-point to obj.

-look at obj.
Quine's Problem:
Not so hard
-parents and children use JFA

-social context helps

-salience helps
Quine's Problem:
-OVEREXTENSIONS: use "moon" to refere to anything that is cresent shaped
ex: cresent sandwich= moon

-UNDEREXTENSIONS: use "moon" only to reflex to pick in specific book
-not in sky

errors demonstrate how child is mapping words onto categories of things
Why do children make these errors?
Clark and Nelson
-CLARK: semantic feature story
-children's early word learning (and early errors) are perceptually based
moon = moon

-NELSON:children's early word learning (and early errors) are functionally based
-mislabel things has on serve sim. functions
Spatial Cog. Dev:
*is child early understanding of space purely egocentric? or does ti look like even young child understand somehting about htings like (Euclidean) distance?
sandbox task...
-hide toy
-smooth sand
-count to E or say something to mom
-find toy in sand

Gentner (1987):
pit form v. function w/ novel obj
Kids go with this when asked what Jiggy is?
Gentner (1987):
FORM: function determines early- learned names

NAMES: word- referent mapping based on form (shape!)
Samuelson & Smith (1999):
shape- bias
-58, 17-33 months old

Two independent variables:
-productive noun vocab- survey 600 words
-artificial noun learning- novel obj.
Samuelson & Smith (1999):
solid obj.
use shape when picking
Samuelson & Smith (1999):
non-solid substances
Samuelson & Smith (1999):
Results and Conclusions
Shape bias learned kick in after know bunch words!

-learned words- go w/ shape
-150+ words

-not shape

-looks like shape bias dev. as child learns 1st words
Samuelson (2000):
-if word learning biases (shape-bias) are learned by learning words, we should be able to change these biases by changing the words children know

-teach young child nouns & look for dev. of biases
Samuelson (2000):
only kids dont know many words

-taught 15 to 21 mths. old diff types of nouns
-SHAPE nouns group (bucket, pear)
-MATERIAL nouns group (lotion, jell-0)
screw up
-measure 1: artifical noun learning
-measure 2: parent report of vocab.
Samuelson (2000):
Newcomb et al. (1998):
Sandbox task
-but ther ws also a dev shift:
at 22 mths, child accuracy improved when external landmarks were visable

-so child do seem to rely on more "world-centered" or "allocentric" spatial information as they dev
Samuelson (2000):
results & conclusions
-teach shape nouns=> dev. early precocious shape bias
-teach material nouns=>no material bias!

-shape bias = product of word learning
-but, may also be influnced by prelingusitic abilities children bring to the language learning task (perceptual abilities that help them attend to shape)
*shape bias not enough learn words
*something else going on (prelinquistic)

-substance = hard to determine!!
Conclusions about Language
-some aspects of language are clearly learn-able (Quine's problem, generalization problem)

-but children may also have prelinguistic biases that help them start learning language
*cant conclude all learning
Spatial Cog Dev:
Recent research emphasizes continuity over dev (vs. Piagetian stages)

-child gradually come to rely on more world-centered spatial info

-child gradually learn that space can be thought of in multiple ways (eg, dual representation idea)

Add Cards

You must Login or Register to add cards