Morality Guest Lecture Flashcards
Theoretical perspective:
- Arguments that morality is linked to the evolution of cooperation (be it group hunting, caretaking, etc etc)
- to cooperate safely, humans need mechanisms to
- distinguish friends from foes (social evaluation)
- influence others’ social behaviours (inclusion/exclusion;rewar/punishment)
- because humans are the most cooperative species on earth (e.g., Tomasello, 2009) we should also be
the most evaluative and the most retributive
Basic Methodology: Helper/Hinderer Puppet Shows
hill (reach hilltop/fall steep)
ball(obtain object/someone elese has it)
box(obtain object/in box,cant lift lid
result
babies everytime prefered pro-social behaviour
Maybe low-level perceptual cues?
Infants prefer:
- prefer victims then harmer
- protectors to non-protectors
- fair to unfair distributors
Why? Aversion to antisocial, attraction toprosocial, or both? (Hill)
10 months
prefer helper vs harmer
prefer helper then nutral(not helping or harming) - very small difference
prefer nultral then harmer - very big difference
Infants’ Social Preferences…WHY?
- A third-party judgment: Infants do not stand to (immediately) gain or lose from these actions
- Still many possibilities for what babies
attending to: - lowest-level account: something is inherent to the stimuli, not a social judgment at all
- maybe infants just prefer pushing up to pushing down, ball-leaving to ball-taking, or box-opening to box-closing
Maybe low-level perceptual cues?
Infants don’t prefer agentive actors who perform similar acts on inanimate objects
Maybe any action that facilities an agent’s path of movement?
Or, about helpful/unhelpful nature of those acts?
Study 1: Prevent some babies from inferring goal, examine choice
1/2 babies: eyes fixed;
gazes uphill:
Clear, though
unfulfilled goal
1/2 babies: eyes unfixed;
gazes downhill:
unclear goal
Maybe any action that facilities an agent’s path of movement?
Or, about helpful/unhelpful nature of those acts?
Study 2: Do individual differences in goal inference predict choice?
- Under some circumstances, infants visually anticipate agents’ goals
from ~6 months - Thought to reflect robust goal understanding
- 5-month-olds, cartoon hill, brief familiarization (= babies young, &
we made it hard)
result: babies only like helpers when they understanding what they need help with (what their goal is)
Seems to be social, and about protagonist’s unfulfilled goals
But a social preference for helpers need not reflect anything like a moral preference for helpers
alterism or for selfish help for the future(the helper might help me)
One important facet of moral judgment:
Intentions v. Outcomes
Two parts of each action that babies could be evaluating in studies to date:
* The outcome for the protagonist
* The intention of the helper/hinderer
* Adults use both, but privilege intention
* Kids notoriously bad at privileging
intention for moral judgment (e.g., Piaget,
1929 & dozens more)
* But, infants appear to focus on intentions in
other situations…
- 3 cases: Failed Attempts to help/harm,
Accidental Help/Harm, Ignorant help/harm
Intention/outcome distinction case 2: Failed Attempts
3 comparison types:
intent differs, same outcomes
intent pitted against outcome
same intent, different outcome
result
outcomes same with different intentions:prefered intension (good)
outcomes same but different intention:prefered intention
failed help vs failed hinder : prefered intention
intention both good- prefer outcome
intention both bad - prefered outcome
accidental
they prefered
the intended helper vs the accidental helper
they also prefered the accidental harmer then the intended harmer
Is it always good to intentionally facilitate goals?
- Adults sometimes see intentional goal-blocking as good
- e.g., bad/dangerous goals
- e.g., punishment for bad behavior
- Punishment is a human universal (with X-cultural
differences); theorized to stabilize cooperative systems
- When adults see others as deserving punishment, we sometimes positively evaluate those who perform it
- What about infants? 2 act plays
- Infants see 4 distinct events
act 1
protagonist with unfilled goal
helped act - deserve reward
hindering act - deserve punishment
act 2
hinder is helped
helper is helped
results
19 month
like when people who are helper (has helped before) are helped
they dont like when people who are hinders (has hinder before) are helped
Will older babies reward/punish themselves?
Training: Give each puppy a “treat”
Puppet show: Box OR Ball
Reward OR Punish
result
- take treat from mean puppy
- give treat to good puppy
- less likely to take away treat from a victam
experiment
- Are younger infants motivated to punish? Let them punish with their eyes!
- Contingency training: When infant looks to a character, a stone crushes it (Exp 1)
- Infants see one puppet hit another
- Later, they selectively look to the hitter (replicated in Exp 5)
- Because they want the hitter to be crushed?
- They do not selectively look to hitter if trained that something soft falls on
character they look at (Exp 2), nor if their looking makes stones fall/crush
randomly (Exp 3), nor if nothing is an agent (Exp 4
summary
- Infants prefer prosocial to antisocial others
- early as 3 months
- negative info primary
- Based on acts’ helpful/unhelpful nature
- not physical cues
- Based on mental states
- Sensitive to context
- Influences own sociomoral behaviours in year 2 (or before!)
failed replication
But, failed replications!
- A few failed replications for basic preferences for prosocial over antisocial over the years
- Generally, there have been far fewer failures than successes
- Meta-analyses suggest preference robust
- (But there may be publication bias)
- Very recent large scale collaborative effort “ManyBabies” to replicate the
hill study (*Yuen, Lucca, et al., in press). - ~500 babies around the world.
- No systematic preferences
But, failed replications!
Were we wrong?
Were we wrong?
- Maybe yes:
- At the same time as the ManyBabies failure, we have been running a large longitudinal study trying to track moral development from birth - age 3, where we re-ran all puppet show studies
- No preferences for helpers in any of them
- Maybe no:
- Also, failed to replicate a bunch of other findings
- We replicated a visual preference for helpers in an online study, and several other labs have had successful online replications too
- in lad failed.at home suceeded(being in a weird enviroment?)
- Issue with in-lab testing? Recent failures were all post covid (babies spending less time with strangers, experimenters wearing masks, etc).
- COVID effects will take years to sort out
Were we wrong?
How much experience?
- Even when babies succeed, they have all had 3+ months of experience within the sociomoral world.
- Maybe they (quickly) learn it
- What about newborns?
Newborns
- Preferentially attend to more over less social stimuli
ex:mother voice vs stranger voice
ex:prefer faces that look at them vs looking away
Also,
- Goal-directed action
over non-goal-
directed (hand) action - Action that
approaches (them)
versus moves away
(from them)
Do newborns prefer prosocial actions?
- Avg age = 5 days (tested at first checkup)
- Dual-video preferential looking method
- 2 trials; videos on opposite sides
- Two experiments, each with different prosocial/antisocial acts and a matched set of non-social control videos
result
babies in the social condition - looked more to the helper
babies in the non-social condition; had no preference
even after replication
