Origins of Mind : 08
\section{Crossing the Gap}
\section{Crossing the Gap}
(Something like) core knowledge exists.
There is a gap between core knowledge and knowledge knowledge.
Crossing the gap involves social interactions, perhaps involving words.
*for bk: include \citep{meyer:2016_monitoring} on crawling infants’
(informative about relations between performance and observation)
Action: When?
\section{Action: When?}
\section{Action: When?}
Consider a much-replicated study by Woodward and colleagues.
Woodward et al 2001, figure 1
anatomy of a goal
A goal typically involves a type of action---kicking rather than smashing, say---as well as manners of
action---discretely, firmly, and precisely, for example---and perhaps more besides.
For a process to \emph{track} a target of an action is for how that process unfolds to nonaccidentally
depend in some way on whether that thing is indeed a target of the action.
The studies we have considered so far focus on infants’ abilities to track the targets of actions.
anatomy of a goal
anatomy of a goal
anatomy of a goal
anatomy of a goal
For a process to \emph{track} a target of an action is for how that process unfolds to nonaccidentally
depend in some way on whether that thing is indeed a target of the action.
The studies we have considered so far focus on infants’ abilities to track the targets of actions.
Given the liberal way I have defined tracking a goal, a process which tracks a target is thereby a process which tracks a goal. (The converse will often hold too, although there could be exceptions.)
It is conceivable that some types of processes \emph{merely} track targets: that is,
processes of this type only ever track goals in virtue of tracking targets.
The studies we have considered so far are consistent with the hypothesis that infants
are merely tracking targets in this sense.
To answer this question, we would ideally have pairs of scenarios in which the target of an action
is kept constant while the type of action varies.
To the extent that subjects respond appropriately to the difference in type of action, we can be
confident that they can distinguish actions not just by their targets but also by their types.
This is illustrated here, where you are habituated to a grasping event and the test events are
(a) grasping but with novel kinematics (from a different angle), or (b) a novel goal event (pushing).
Unfortunately, as far as I know this has not yet been done.
But there are some studies which, although not intended to get at exactly the issue of whether
infants track goals and not just targets of actions, do indirectly shed light on this issue.
These studies demonstrate competence in goal-tracking from nine months of age, and give us no reason
to doubt that, in simple enough cases, infants might show competence in goal tracking much earlier.
\citet{Behne:2005dw} created just such pairs of contrasting scenarios.
In one of their contrasts, an experimenter holds a ball out for an infant to grasp and then
either ‘accidentally’ drops it or teasingly pulls it back.
So in each case there is a goal-directed action involving the ball, but in one case the goal
of the action is to pass the ball to the infant whereas in the other case the goal is to
tease the infant.
\citet[Study 2]{Behne:2005dw} found that nine-month-olds (but not six-month-olds)
consistently and appropriately discriminated between these scenarios by, for example,
banging more when the ball was ‘accidentally’ dropped than when it was teasingly retracted.
This and other research
(e.g. \citealp{ambrosini:2013_looking}) suggests that, at least from nine months of age,
infants can indeed distinguish both the type and target of a goal-directed action.
% \citet{ambrosini:2013_looking} is important: shows that type of grasp and not just target is tracked, implying that anticipatory looking is not merely perceptual animacy.
The Teleological Stance
\section{The Teleological Stance}
\section{The Teleological Stance}
The Simple View
The Simple View is a view about the PROCESSES involved ...
\begin{quote}
Infants' (and adults') goal tracking depends on beliefs concerning relations which hold
quite generally between means and goals; and they identify particular goals by making
inferences from these beliefs plus their observations.
\end{quote}
Infants and adults engaged in goal-tracking reason about
to which outcome a means is the best available
in fundamentally the same way that you or I do when trying to work it out explicitly.
A Limit on Goal Tracking in the First Nine Months
\section{A Limit on Goal Tracking in the First Nine Months}
\section{A Limit on Goal Tracking in the First Nine Months}
Flanagan and Johansson, 2003 figure 1 (part)
To explain this I have to step back and show you something interesting about
adults when they perform, and when they observe actions.
Performing actions (e.g. stacking blocks): you don't look at your hand but at
the block it will pick up, or, when holding a block, at the location where it will place a block.
That is, in acting the eyes move just ahead of the action.
Kanakogi and Itakura, 2011 figure 1
‘Positive relative times of the arrival of gaze at the goal area indicate that gaze precedes the agent’s arrival (predictive);
negative values indicate gaze arrival after agent arrival (non-predictive). Each age group is n = 12.
Error bars are s.e.m.’
GH : grasping hand; BH : back of hand; MC : mechanical claw
Kanakogi and Itakura, 2011 figure 5 (part)
Kanakogi & Itakuar, 2011 show that abilities to grasp objects are correlated with
abilities to track the goal of a grasping action (as measured by proactive gaze).
x-axis is alpha, grasping angle. ‘An α angle value from 90 to 180° indicates that the infant
is engaged in a one-handed grasping action.’
‘The angle α is an index of the development of the onehanded grasping action and was
calculated by measuring the angle of a straight line de ned by the infant’s two hands (the
apex of the junction of the thumb and index nger) when crossed by an imaginary line
projecting frontally from the infant (Fig. 2b). If infants grasped for the objects with their
le hand, we reversed the red rightangled triangle from one side to the other side and
calculated the angle α in the same way. e angle α value of 90° corresponded to a perfect
alignment of the hands in a twohanded reach. erefore, the angle α value deviates from 90°
towards 180°, and bigger angle α value indicates more mature onehanded grasping. If the
angle α was over 90°, the infant was considered to be engaged in a onehanded grasping
action.’
Ambrosini et al, 2013 figure 1 (part)
Further support for a link between action performance and goal tracking
comes from a developmental study by Ambrosini et al which studied whether proactive
gaze in infants is influenced by pre-shaping of the hand, and, in particular,
whether it is influenced by precision grips.
Ambrosini et al, 2013 figure 1 (part)
Ambrosini et al, 2013 figure 1 (part)
By using no shaping (a fist), Ambrosini et al could treat sensitivity
to whole-hand grasp and precision grip separately.
Ambrosini et al, 2013 figure 3
‘infants’ ability to perform specific grasping actions with fewer fingers directly predicted the degree with which they took advantage of the availability of corresponding pre-shape motor information in shifting their gaze towards the goal of others’ actions’ \citep[p.~6]{ambrosini:2013_looking}.
Kanakogi and Itakura, 2011 figure 1C (part)
Further, changing from a bodily action to the operation of a mechanical claw
(say) undermines the goal tracking effect.
So Kanakogi & Itakuar, 2011 make two points:
(i) goal-tracking depends on action capabilities; and
(ii) only works for events involving biomechanically similar affectors
Kanakogi and Itakura, 2011 figure 5
Needham et al, 2002 / https://news.vanderbilt.edu/files/sticky-mittens.jpg
Needham et al, 2002 showed that putting ‘sticky mittens’ on 3-month-old infants
(for 10-14 play sessions of 10 minutes each) resulted in their spending
more time visually and manually exporing novel objects.
The observation was based on this study, which we saw earlier
Woodward et al 2001, figure 1
Sommerville, Woodward and Needham, 2005 figure 3
Sommerville, Woodward and Needham, 2005 figure 3
The results show that infants who played wearing the mittens first
were more attentive to the goal.
From at least three months of age, some of infants’ abilities to identify
the goals of actions they observe are linked to their abilities to perform
actions \citep{woodward:2009_infants}.
But one potential objection to this study concerns observation vs performance.
The infants who played wearing sticky mittens first had spent longer observing
actions by the time it came to the violation of expectations trial.
Could it be observation of action (including one’s own) rather than performance
that matters?
In adults, tying the hands impairs proactive gaze \citep{ambrosini:2012_tie}; in
infants, boosting grasping with ‘sticky mittens’ facilitates proactive gaze
(\citealp{sommerville:2005_action}; see also \citealp{sommerville:2008_experience},
\citealp{ambrosini:2013_looking}).
Sommerville et al 2008, figure 1
To address this issue, \citet{sommerville:2008_experience} did a study in
which one group had observation while the other group had performance.
The participants were 10-month-old infants this time.
The materials were a bit different: so that training vs observation could
be as similar as possible with respect to the causal structure exposed,
there was a hook to get an object.
Sommerville et al 2008, figure 2
The results show that infants with the training paid attention to the
distal goal (choice of toy) whereas those without paid attention to the
choice of cane.
... or IMPAIR INDIRECTLY ...
Bruderer et al, 2015 figures 1, 4
Experiment 1 : shows that 6-month-old infants can distinguish a phonetic contrast
they have never heard before (one that occurs in Hindi but not their linguistic
environments.) (The contrast used was the Hindi dental /d/̪ versus retroflex /ɖ/
distinction.)
These graphs show a difference in mean looking time between cases in which phonemes
are alternated and cases in which they are not. (Iff infants distinguish, they should
find the alternating phonemes more interesting.)
Experiment 2: but not when they have a tongue-controlling dummy in their mouths
Experiment 3: but yes when they have a dummy which leaves the tongue free.
The Motor Theory of Goal Tracking
\section{The Motor Theory of Goal Tracking}
\section{The Motor Theory of Goal Tracking}
Fogassi et al 2005, figure 5
Motor Representation
What follows rests on three background assumptions about the control of action.
First, I follow \citet{jeannerod_motor_2006} and others in rejecting the view that all motor representations specify only bodily configurations, joint displacements and end states.
Instead some motor representations specify outcomes to which actions are directed,
such as the grasping of a particular handle or the transporting of a given object.
Second, some motor processes involve computing means from ends and generating sensory expectations concerning the effects of actions \citep[e.g.][]{Wolpert:2003mg}.
Third, multiple means--ends computations can occur simultaneously, or at least rapidly enough for action preparation to involve selection on the basis of multiple means-ends computations
\citep[e.g.][]{wolpert_internal_1998}.
Suppose you are reaching for, grasping, transporting and then placing a pen. Performing even
relatively simple action sequences like this involves satisfying many constraints that cannot
normally be satisfied by explicit practical reasoning, especially if performance is to be rapid and
fluent. Rather, such performances require motor representations.
These representations are paradigmatically involved in preparing, executing and monitoring actions.%
\footnote{%
See \citet{wolpert:1995internal, miall:1996_forward, jeannerod:1998nbo, zhang:2007_planning}.
Note that motor representations sometimes occur in an agent who has prepared an action and is required (as it turns out) not to perform it: although she has prevented herself from acting, motor representations specifying the action persist, perhaps because they are necessary for monitoring whether prevention has succeeded \citep{bonini:2014_ventral}.
}
But they also live a double life. Motor representations concerning a particular type of action are
involved not only in performing an action of that type but also sometimes in observing one. That is,
if you were to observe Ayesha reach for, grasp, transport and then place a pen, motor representations
would occur in you much like those that would also occur in you if it were you---not Ayesha---who was
doing this.
Converging evidence for this assertion comes from a variety of methods and measures;
but I won’t mention any of that here.
Sinigalia & Butterfill 2015, figure 1
Goal-tracking is acting in reverse.
-- in action observation, possible outcomes of observed actions are represented
-- these representations trigger planning as if performing actions directed to the outcomes
-- such planning generates predictions
-- a triggering representation is weakened if its predictions fail
The result is that the only only outcomes to which the observed action is a means
are represented strongly.
There is evidence that a motor representation of an outcome can cause a determination of which movements are likely to be performed to achieve that outcome \citep[see, for instance,][]{kilner:2004_motor, urgesi:2010_simulating}. Further, the processes involved in determining how observed actions are likely to unfold given their outcomes are closely related, or identical, to processes involved in performing actions.
This is known in part thanks to studies of how observing actions can facilitate performing actions congruent with those observed, and can interfere with performing incongruent actions \citep{
brass:2000_compatibility,
craighero:2002_hand,
kilner:2003_interference,
costantini:2012_does}.
Planning-like processes in action observation have also been demonstrated by measuring observers' predictive gaze. If you were to observe just the early phases of a grasping movement, your eyes might jump to its likely target, ignoring nearby objects \citep{ambrosini:2011_grasping}. These proactive eye movements resemble those you would typically make if you were acting yourself \citep{Flanagan:2003lm}.
Importantly, the occurrence of such proactive eye movements in action observation depends on your
representing the outcome of an action motorically; even temporary interference in the observer's
motor abilities will interfere with the eye movements \citep{Costantini:2012fk}.
These proactive eye movements also depend on planning-like processes; requiring the observer to
perform actions incongruent with those she is observing can eliminate proactive eye movements
\citep{Costantini:2012uq}. This, then, is further evidence for planning-like motor processes in
action observation.
So observers represent outcomes motorically and these representations trigger planning-like processes
which generate expectations about how the observed actions will unfold and their sensory consequences.
Now the mere occurrence of these processes is not sufficient to explain why, in action observation,
an outcome represented motorically is likely to be an outcome to which the observed action is
directed.
To take a tiny step further, we conjecture that, in action observation, \textbf{motor representations of
outcomes are weakened to the extent that the expectations they generate are unmet}
\citep[compare][]{Fogassi:2005nf}.
A motor representation of an outcome to which an observed action is not directed is likely to
generate incorrect expectations about how this action will unfold, and failures of these
expectations to be met will weaken the representation.
This is what ensures that there is a correspondence between outcomes represented motorically in
observing actions and the goals of those actions.
‘proactive gaze’ and ‘explicit judgement’ are variables whose values represent
whether there is a proactive gaze or explicit judgement, and what is it to or about.
Likewise, ‘motor process’ is a variable whose values represent ...
The lines depict how the variables are causally related. I’ve used thick and thin
lines informally, to indicate weaker and stronger influences.
The dual-process theory of goal-tracking makes perfect sense of development.
It says that what we observe in six- and nine-month-olds is motor-based goal-tracking.
Presumably the more flexible, reasoning-based goal-tracking processes emerge some time
later in development.
A Puzzle About Goal Tracking
\section{A Puzzle About Goal Tracking}
\section{A Puzzle About Goal Tracking}
Kanakogi and Itakura, 2011 figure 1C (part)
Earlier I mentioned this experiment, which shows that infants
fail to track goals involving things which would appear to them to be
biomechanically non-agent-like events.
But ...
\#source 'research/teleological stance -- csibra and gergely.doc'
\#source 'lectures/mindreading and joint action - philosophical tools (ceu budapest 2012-autumn fall)/lecture05 actions intentions goals'
\#source 'lectures/mindreading and joint action - philosophical tools (ceu budapest 2012-autumn fall)/lecture06 goal ascription teleological motor'
When do human infants first track goal-directed actions and not just movements?
Here's a classic experiment from way back in 1995.
The subjects were 12 month old infants.
They were habituated to this sequence of events.
Gergely et al 1995, figure 1
For the test condition, infants were divided into two groups. One saw a new action, ...
... the other saw an old action.
Now if infants were considering the movements only and ignoring information about the goal, the 'new action' (movement in a straight line) should be more interesting because it is most different.
But if infants are taking goal-related information into action, the 'old action' might be unexpected and so might generate greater dishabituation.
Gergely et al 1995, figure 3
Gergely et al 1995, figure 5
‘by the end of the first year infants are indeed capable of taking the intentional stance (Dennett, 1987) in interpreting the goal- directed behavior of rational agents.’
\citep[p.\ 184]{Gergely:1995sq}
‘12-month-old babies could identify the agent’s goal and analyze its actions causally in relation to it’
\citep[p.\ 190]{Gergely:1995sq}
You might say, it's bizarre to have used balls in this study, that can't show us anything about infants' understanding of action.
But adult humans naturally interpret the movements of even very simple shapes in terms of goals.
So using even very simple stimuli doesn't undermine the interpretation of these results.
Heider and Simmel, figure 1
Consider a further experiment by \citet{Csibra:2003jv}, also with
12-month-olds. This is just like the first ball-jumping experiment
except that here infants see the action but not the circumstances in
which it occurs. Do they expect there to be an object in the way behind
that barrier?
Csibra et al 2003, figure 6
Daum et al, 2012 figure 1
Daum et al created a modified version of Woodward’s paradigm which allowed them to measure
both anticipatory looking and dishabituation.
where researchers have measured two different responses to a single
scenario, anticipatory looking and either dishabituation or pupil dilation.
Generally, they appear to find evidence for goal tracking in dishabituation
or pupil dilation but not in anticipatory looking.
(This is true of Daum et al, 2012; and Gredeback and Melinder, 2010.)
Why the discrepancy?
This is another question we can’t answer with the Motor Theory of Goal
Tracking.
Daum et al, 2012 figure 2
[skip -- just here in case need for discussion; shows that anticipatory looking
to cartoon fish takes years to develop]
Perceptual Animacy
\section{Perceptual Animacy}
\section{Perceptual Animacy}
Perceptual Animacy
\emph{perceptual animacy},
the detection by broadly perceptual processes of animate objects and their targets.
Gao et al, 2009 figure 2
Evidence for the existence of perceptual animacy comes from a variety of studies
where adults are given a simple visual task such as identifying which circle is the ‘wolf’
and which the ‘sheep’, or, in another experiment, moving the sheep around in order to
avoid being caught by the wolf.
It is also possible to examine how having a sheep and a wolf affects how attention is allocated by
measuring how well participants can detect probes placed on different shapes.
The literature on perceptual animacy mostly confounds it with goal tracking.
But there are two key differences.
First, perceptual animacy is a broadly
perceptual phenomenon.
Second,
the perceptual detection of animacy is clearly distinct from goal tracking
for it involves detecting relations between objects in motion rather than
outcomes to which actions are directed.
Relatedly, it does not involve sensitivity to the type of action.
And the perceptual detection of animacy appears to depend on simple cues
and heuristics and is unlikely to be correctly described by the
Teleological Stance.
So perceptual animacy does not involve tracking goals to which an action is
directed: it is a matter of tracking objects to which animate movements are
directed.
Gao et al, 2009 figure 3b
\emph{{perceptual animacy}},
the detection by broadly perceptual processes of animate objects and their targets.
To illustrate, consider a groundbreaking experiment by \citet[experiment 1]{gao:2009_psychophysics}.
Adults were shown a display which contained some moving circles.
In some cases the circles moved independently of each other, but in other cases there was a ‘wolf’
which chased a ‘sheep’ with varying degrees of subtlety.
The adults’ task was simply to detect the presence of a wolf.
\citeauthor{gao:2009_psychophysics} established that adults can do this providing the chasing is not too subtle.
In further experiments, they also showed that adults’ abilities to perceptually detect chasing
depend on several cues including whether the chaser ‘faces’ its target (‘directionality’) and how
directly the chaser approaches its target (‘subtlety’).
The detection of animacy appears to be a broadly perceptual phenomena since it depends on areas of
the brain associated with vision and influences how perceptual attention is allocated
\citep{scholl:2013_perceiving} irrespective of your beliefs and intentions
\citep{buren:2016_automaticity}.
Gao et al, 2009 figures 3b, 4a
To illustrate, we have to recognise that, in adults,
there are multiple processes
involved in tracking actions, which are differently affected by changes in
context, and which probably have different affects on different kinds of responses to
observed and anticipated actions.
Here is a rough guess about how things look IN ADULTS.
what’s already in infants ...
Gao et al, 2009 figure 3b; Csibra et al 2003, figure 6
further complication: associative learning for sequencing
Key thing is that we don’t have to postulate knowledge of goals;
earliest goal-tracking abilities can be invoked to explain the developmental emergence of knowledge
(of goals, and of other things).