Lecture 4 Imitation And Action Understanding Flashcards
How do we Map perceptually opaque actions
Requires neuro cognitive mechanism that relates the seen to the unfelt and the felt to the unseen
Define perceptually transparent actions
Actions where can see yourself doing it
What are perceptually opaque actions
Actions that aren’t a guided by sensory info about degree of similarity between the model and the observer - can’t see if doing well
Evolutionary reasoning for imitation
Innate mechanism that matches observed action to the one executed
Environmental reason for imitation
Learn to match our expressions
describe meltzoff and moore 1977 infant imitation
Presented infants with 3 facial expressions - mouth widening, lip pursing and tongue protrusion and control
found infants successfully imitated all actions with no confusing mapping actions between similar modalities
What is imitation
thought to be an innate mapping mechanism of Social behaviours of actions onto own body from another’s
the act of copying anothers behaviour
define mimicry
imitation across species
protective mechanism
difference between human imitation compared to other species
all animals imitate
BUT humans tend to over imitate - imitate both the behaviourally relevant and irrelevant actions that they see performed
describe hamer and whiten 2005 human over imitation
imitation in 3 y/o infants and chimps observing a human model
performing actions behind an opaqe or transparent box
opaque - cant see the inner workings
transparent - can see how actions impact behind box/outcome
chimps only imitate all actions when opaque - dont know which actions are necessary
BUT humans also imitate when transparent - even when obviously not relevant to the outcome
Meltzoff and moore 1997 original imitation model
facial imitationbased on active intermodal mapping
use match to target process using proprioceptive feedback to determine correct imitaition of expression
1- visual perception
2- supramodal representation (equivalence detector)
3- infant motor acts + proprioceptive info (feedback to equivalence detector)
detailed Melzoff and moore AIM model ‘active intermodal mapping’
1 - perceptual system - info from worlds (exteroceptive) and body (somatosensory field)
2- supramodal representational system - specifies equivalence of observed and performed action - organ relation between target and infant
3- if mismatch then to Action Systen which co ord movements, forms new end goal and execute new action - feedback to body for change in config
through repetitive body play ‘body babbling’, infants learn relationships between self-generated movement and the organ relations BUT the mechanism that maps movements onto self is mature from birth.
What does the AIM model overall suggest
infants have a good cognitive mechanism that allows them to compute imitative errors effectively and perform novel behaviours with a high level of accuracy
anisfeld 1996 critique of melzoff and moore
eval infant imitation studies
support only tongue protrusion imitation and no other actions
measures used not sensitive enough
+ m&m experimenter bias - readminister modelling period when trying to get attention means more likely to coincide similar actions
jones 1996 critique of m&m
tongue protrusions are not driven by an innate imitative mathcing mechanism but is driven by infant liking/interest expression which is usually expressed in the face of novel stimui
infants show tongue protrusion to wide variety of sitmuli ie lights as well as novel facial expressions
when learn reaching behaviours, tongue protrusions decline
heyes 2010 mesmerising mirror neurons
associative learning hypothesis of imitation
mirror neurons in moneys fire when observe and perform the same action
- mirror neurons develop from normal sensory and motor neurons when similar observed and performed actions coincide with one another
learns an association between the observed and performed action
catmur et al 2008 mirror neurons as learning and not innate
if the associations are learnt then should be able to reverse the activity
mirror neurons usually more responsive to hands > feet
BUT reverse pairing reverses this activity
sources of imitation learning
self observation (white et al 1964) mirror self observation (amsterdam 1972) caregiver imitation (uzigris et al 1986) schronious action (o'toole and dubin 1968)
self observation
white et al 1964
2-3m/o spend majority of time awake learning about own hands
builiding an understanding of motor movements and relevant sensations
mirror self observation
amsterdam 1972
95% of 6-12 m/o respond to mirror image as playmate
interact and play with it
recieve imitative info from reflection
likely to for mirror neurons - see action whilst performing as same time
caregiver imitation
uzgiris et al 1986
infants spend large proportion of time actively interacting with adults
synchronous action
o’toole and dubin 1968
55% spoonfeeding sequences involve caregiver infant imitation to get to open mouth
associate sensory representation with perception of action
heyes et al 2005 assoc learning and mapping
if learning then should be able to modify associations via experience
look at degree to hich hand responses are performed
- faster to respond if prev learnt assoc than non imitative
either recieve matching or non matching action sof hand opening and clasping
training alters mappings - faster to respond to assoc even if incompatible
ray and heyes 2011 wealth of the stimulus
imitation succes is dependent on the development of the infants environment
greater inclination to imitate matching over mismatching action sis because there is greater matching assoc within the environment
there is sufficient information in the individual’s environment, and particularly in their social interactions, to support ontogenetic development of the capacity for imitation without the guidance of a specialized, innate cognitive mechanism.`
ray and heyes 2011 poverty of the stimulus
the capacity to imitate depends on a complex innate imitative mechanism ie AIM model
If infants are able to imitate within hours of birth and their imitative capacity could not be based on learning; if they can imitate a range of behaviours, the innate endowment must be something more complex than a couple of reflexes.
problem with poverty of the stimulus argument
ray and heyes 2011
says nothing about the nature of the innate mechanism
unable to replicate the neonatal imitation findings
how does learning discriminate between similar and different actions
if similar then the probability of my seeing x while doing x must be higher than the probability of my seeing any other single action while doing x.
rhe associative mechanisms that make imitation possible via matching vertical associations do not encode or ‘know about’ similarity.
more inclined to imitate than to counter-imitate because our developmental environments have exposed us to more matching, x–x, than non-matching, x–y, sensori-motor relationships
problems with infantile imitation studies
infants difficult to manipulate
fail to attend
at a period of rapid learning when overwhelming with range of sensory info
- m&m repeat because not initially attend and must attend to be able to imitate action
brain disorders and imitation as biological
merlons et al 1997
imitative dyspraxia in patients with left frontal love lesions have good motor skill BUT fail to successfully imitate and tend to make many errors
suggests there may be a problem through their lesions which means they fail to successfully find equivalence between observed and performed motor action
problem in supramodal model? AIM
developmental disorders and imitation as biological
BC
BC believed theory of mind developed as a result of an innate mechanism that drove development
autism caused by ‘mindblindness’ - inability to share attention with others
sharing attention necessary for the recognition of outward behaviours and mapping onto own body - use imitaiton to recognise others internal states
smith and bryson autism and biological imitation
autistic children do not readily imitate the actions of others
may lead to problems in social functioning
autism and imitation as genetic
autism thought to have high heritability - such a high genetic aspect may infer a biological link between imitation ability
BUT problems imitation may simply be a result of another problem that makes imitation difficult - ie need high social functioning
hammes and langdell 1989 autism
autistic individual find imitating imaginary > real behaviours harder
make more errors when shown unconventional use of an object - more likely to not imitate or use for norm
may suggest that have problem in unconventional/novel imitatio but hold simple conventional system
define action understanding
activate motor representations of actions not only when hold the goal to imitate but when pasively watching
- motor activations serve as a way for us to interpret the actions of those around us
serve to interprest what we see anticipate how an action may unfold and predict future behaviours
gallese et al 1996 action understanding and mirror neurons
recorded electircal activity of motor neurons in vPMC of macaque monkeys
fire when monkey executes and when observes an action following association - recog what someone else doing as what self would so and reasons behind
southgate et al 2009 predicting behaviours via action understanding
9m/o recorded EEG while reaching or observing
subthreshold activity before an action is fully performed
suggest anticipate its occurance
southgate et al 2010 predicting behaviours via action understanding
action prediction paradigm
grasping hand (action) disappear behind occluder - predict outcome
1 - back of hand action not interpretted as action
2- grasping hand goal directed - elicit anticipatory motor activation
motor system active to grasping>back hand
when no motor activation for a goal directed behaviour then do not interpret as goal directed
