Third-Person Mind Reading and the False Belief Task:
Moving from belief-desire to embodied interactionism
Dan Sperber
Luca Surian
An external systematisation that philosophers impose upon commonsense practices?
(Stich and Ravenscroft, 1996, Ratcliffe, 2006)
"However, such statements do not start with a phenomenology of social life and show how FP is incorporated into it. They just describe all instances of interpersonal understanding in FP terms and assume that such descriptions adequately characterise what is going on." Ratcliffe, 2006
Everyday Social Phenomenology?
Folk Psychology
Remembering the Reading: False Belief Task
Why are we doing this again?
Why Children?
Why False Belief?
...Why Folk Psychology?
Briefly...
Attribution of Beliefs by 13-Month-Old Infants
Stefania Caldi
T. P. German
Paul Bloom
Two reasons to abandon the false belief task as a
test of theory of mind
Matthew Ratcliffe
Saun Gallagher
Theory of Mind or Interaction between people?
Uta Frith
Simon Baron-Cohen
Does the autistic child have a “theory of mind”?
Alan Leslie
Why should we constrict cognition to the "skin bag"?
Andy Clark
Examining our own experiences of awkward interaction
“Could my mental states be partially constituted by the states of other thinkers? We see no reason why not, in principle”
(Clark and Chalmers, 1998, p. 17)
The Practice of Mind:
Theory, Simulation, or Interaction?
What happens when someone does not engage properly in conversation?
One watches the increasingly exhausted climber reach in desperation for something to hold on to, knowing full well that a fatal fall is imminent. Each time his hand slips, he tries again, knowing that his predicament is growing increasingly dire.

"When we empathise with the distressed climber we do not merely hold a series of propositions about his mental life. We personally experience states very much like his. Our situation is not at all like that of a scientist. " (Ravenscroft, 1998, p. 172)
Baron-Cohen (1995, Chapter 4) postulates three devices that facilitate a prepropositional awareness of others:
(a) A perceptual ‘intentionality detector’.
(b) An ‘eye direction detector’ that, amongst other things, triggers arousal
and affective response when somebody else is looking at you.
(c) A ‘shared attention mechanism’ that enables an appreciation that one is
looking at the same object as somebody else.
Third-person mind reading (assuming FP and propositional ToM) to understand ToM and FP, which is most often second-person? Should second-person/interactionism be used to understand third-person?
False Belief to dig in and "find" WHEN children are able to ascribe internal mental states- but are the assumptions of belief and desire attribution justified?
The logic behind the false belief task was outlined by several commentators on the paper by Premack and Woodruff (1978) entitled `Does the chimpanzee have a theory of mind?' Suppose you want to know whether a chimpanzee can reason about the mental states of others. As Dennett (1978) and others pointed out, it is not enough to
demonstrate that individual A can predict the actions of individual B. In many cases, A can do so without an understanding of B's mental states, but by simply observing
the actual state of the world. (Suppose A knows the chocolate is in the basket and observes B searching for food. A might expect B to look in the basket, not because A is attributing a belief to B, but because the chocolate actually is in the basket.) A more robust test involves predicting the behavior of another animal based on an inferred mental state that differs from reality ± a false belief. This would show that
the individual understands that it is the mental state, rather than the state of the world, that causes the action.
(Target Article)
Part of a larger project of understanding social cognition?
Target Article
Related Papers
1:23-2:00
Differing Performance on False Belief Task
"Four-year-olds tend to succeed at this task, correctly attributing a false belief to Sally, saying that she will look for the object in the basket, while younger children tend to fail (see Wellman, Cross & Watson, in press, for review). This has led many scholars to conclude that children undergo a radical shift in their understanding of the mind. For instance, Gopnik (1993, p. 1) claims that at about age 4, there is an important developmental shift to a representational model of the mind
Reason 1: There is more to passing the false belief task than theory of mind
"The more serious problem is that false belief tasks are inherently difficult. This is because any false belief task requires, at minimum, that the child reasons about a belief that is false. As Leslie, among others, has pointed out, beliefs are supposed to be true. This is what they are for (Leslie, 1994).Hence, even for a child who clearly understands that beliefs can be false, getting the right answer places non-trivial processing demands."
"Furthermore, there is evidence that standard false belief tasks are difficult for children even independent of the requirement to reason about false belief. In the `false photograph' task, a picture is taken of a scene, the scene is changed, and the child is asked what the picture depicts. Here there are multiple representations and there is the realist pull of the way the world really is, but there are no false beliefs. If children pass the false photograph task and fail the false belief task, it would be reasonable to infer that their problem with the false belief task really does have to do with beliefs. But in fact, 3-year-olds fail both the false photograph task and the false belief task (Zaitchik, 1990; see also Leslie, 2000; Slaughter, 1998).
False Photograph Task
Inherent Difficulty of FB
Summary of Target Article
It gets worse, as there is evidence that standard false belief tasks are difficult for children, independent of the requirement to reason about any sort of representation. For instance, Riggs, Peterson, Robinson and Mitchell (1998) constructed a task in which children are required to set aside an actual state of affairs (the fact that an object has been placed in one location) and generate an answer based on an alternative counterfactual state of affairs (where it would have been placed had some other event not happened). Younger children are poor at this task, even though there is no representational content. In a similar vein, Roth and Leslie (1998) developed the `screens task'. Children are presented with a box and a basket, and a marble is placed in the box. These objects are then concealed by a screen. Then a replica box and basket are placed in front of the screen. A marble is placed in the replica box, and then moved to the replica basket. Children are simply asked to report where the marble is behind the screen. This task requires children to set aside a currently available, salient situation (in which the marble is in the basket) and generate an answer based on a similar, less salient situation (in which the marble is in the box). Once again, 3-year-olds fail this task.
Screens Task
Reason 2: There is more to theory of mind than passing the false belief task
If younger children do not show signs that they have an appreciation of false belief, how do you know that they are reasoning about mental states at all? Fortunately, there are other methods that can be used to determine whether or not children are attributing mental states. For instance, in an elegant study by O'Neill (1996), 2- year-olds observed as an attractive toy was put on a high shelf. As this happened, the child's parent was either present or absent. When later asking for help in retrieving the toy, the children were more likely to name the toy and gesture to the location when their parent had not been present to witness the placement of the toy than if their parent had been present. This suggests that they modify their behavior according to the knowledge states of other people (i.e. whether or not their parent possesses a given belief), and that they have a tacit appreciation of the circumstances under which beliefs are formed.
Furthermore, there are plenty of signs that even before their second birthday,
children have some appreciation of the workings of other minds. They are capable
of initiating pretend play and of understanding the pretence of others (Leslie, 1994).
They can attribute goals to other agents (Csibra, Gergely, BiroÂ, KooÂs & Brockbank,
1999; Gergely et al., 1995; Woodward, 1998). They can imitate the intended, as well
as completed actions of other agents (Carpenter, Akhtar & Tomasello, 1998; Meltzoff,
1995). They can use eye gaze as a cue to what someone is attending to when they
use a new word (Baldwin, 1991) and orient to follow the gaze of an inanimate object
if it displays evidence of being an intentional agent (Johnson, Slaughter & Carey,
1998). In fact, young children's learning of the meanings of words can be seen in a
large part as the direct consequence of their ability to infer the referential intentions
of other people (Bloom, 2000). It might be that developmental psychologists are so
obsessed with the false beliefs task just because it is the one measure of theory of
mind that children are not very good at.
ToM at 2 years old
ToM before 2
Autistic ≠3 years old
Consider finally the fact that older autistic children fail the false belief task.
Wimmer and Weichbold (1994) suggest that this speak[s] for the validity of the
false belief task, since social and communicative impairments are among the defining
characteristics of autism. But it also illustrates a weakness in the task. Normal
3-year-olds and older children with autism both fail the false belief task, but, in all
interesting regards, normal 3-year-olds are nothing like older children with autism
(e.g. see Happe, 1996). Normal 3-year-olds are far superior with regard to communicative
and linguistic skills, the ability to pretend and understand the pretence of
others, and the ability to engage in, understand and manipulate the actions of others.
This is a severe problem for any theory that lumps the two groups together as
individuals who lack theory of mind.
The proper role of the false belief task
It is an ingenious, but very difficult task that taps one aspect of people's understanding of the minds of others. Nothing more, nothing less.
A step further back...
The Extended Mind &
Supersizing the Mind
How can notions of second-person mind reading and embodied interactionism inform ideas about third-person mind reading?
Interactionists:
Rethinking Commonsense
Psychology: A Critique of Folk Psychology, Theory of Mind, and Simulation
Evidence for FP is drawn largely from variants of the false belief task (Wimmer and Perner, 1983). However, the very name ‘false belief task’ suggests that an FP interpretation is already written into this experimental paradigm. It is simply assumed that the abilities measured by such experiments should be interpreted in terms of an ability to assign true or false beliefs.
Back to Third-Person
Types of 3rd Person Stances
False Belief
Perception of Third-Person from Interactive Stance
Given the differences between children’s everyday abilities and an ability to assign beliefs in an experimental context, Bruner and Feldman (1993, p. 269) suggest that “to equate grasping other minds with getting a False Belief Diploma on Graduation Day is to oversimplify its form and function.” Although the test measures some kind of ability, this is not sufficient to imply that FP constitutes the best description of everyday interpersonal understanding.
Autism and Interactivism
different circumstances, different motivation, same mentalizing?
sensory-motor, affect, language
Why Are We the Way We Are?
Readiness to Act
Daniel Hutto
Without doubt the so-called friends of folk psychology have overstated and misunderstood its role in social cognition; typically they see it as (i) more basic and (ii) far more pervasive than it actually is. With respect to the question of primacy, folk psychology is not fundamental to social engagements; not even exclusively human ones. We have many other, more basic non-folk psychological means of engaging with one another socially and coordinating our interactions – these involve only end directed intentional and not propositional attitudes. Our primary embodied modes of responding, by my lights, do not involve the manipulation of representations by inferential operations (let alone representations of propositional attitudes). Nor do such engagements result in predictions or explanations, understood as couched in subpersonal propositions. In such cases we get by with script-like patterns of recognitionresponse (some more flexible and complex than others): these are initiated and guided indexically and iconically by the expressive behaviour of others. These sorts of abilities – and not a capacity for ‘mindreading’ – best explain the embodied expectations of non-verbal creatures.
Folk Psychology without Theory or Simulation 2007in Folk Psychology Reassessed. Hutto, D. and Ratcliffe, M. (eds). Doredrecht: Springer. 115–135
I would again argue that the evolved capacities are not for true theories or for “understanding minds” but that our understandings of the world are about readiness for action, and, in this case, interaction, and it would then be hypothesized that interactive abilities are the underlying core that allow complex things like false belief understanding. Thus, I subscribe to the interactive stance.
The basic claim that I will defend is that in most intersubjective situations we have a direct, pragmatic understanding of another person's intentions because their intentions are explicitly expressed in their embodied actions. For the most part this understanding does not require us to postulate some belief or desire that is hidden away in the other person's mind, since what we might reflectively or abstractly call their belief or desire is expressed directly in their behavior.
Gallagher, S. 2001a. In: Thompson (ed), The Practice of Mind: Theory, Simulation, or Interaction? (pp. 83–108).
The false belief task clearly measures something and autistic people clearly have some kind of social impairment. However, experimental results only constitute a more specific case for the existence of FP if they are already interpreted through the lens of FP.
-Ratcliffe
Uta Frith
Alan Leslie
Does the autistic child have a “theory of mind”?
Simon Baron-Cohen
Attribution of Beliefs by 13-Month-Old Infants
Luca Surian
Dan Sperber
Stefania Caldi
A ‘third-person’ understanding is not an ‘objective’ understanding, in the sense of regarding somebody as a mere thing, to be viewed with indifference (Goldie 2000, pp. 181–182). In fact, the ‘third-person perspective’ arguably conflates many different stances. One might adopt a third-person stance towards somebody out of dislike, as a refusal to engage with them as ‘you’. One might do so because communication is physically impossible. One may retain a distance out of respect, fear or shyness. One can also regard another person with the aim of manipulating them, one may sympathise with them, feel pity for them or adopt a stance of curiosity towards their behaviour. The ‘third-person’ grouping fails to do justice to the numerous different stances or attitudes that one might adopt towards another. In none of these cases is the other simply experienced as a kind of ‘object’.
an appreciation of intentional states does not underlie an ability to engage in interactive narrative construction but emerges through it. Normal two- to three-year olds have an understanding of intentional states, which is exhibited in their ability to construct stories. This understanding cannot be extricated from the narrative context that ‘scaffolds’ it. Even in younger children, the activity of play has a narrative structure (Bruner and Feldman1993 pp. 273–274).
Reasoning about Representations
Counterfactuals and primacy of currently available, salient situation
Folk Psychology
Re-Assessed
the violation of our expectations about behavior feels uncomfortable. We are alert to this in a way that highlights our phenomenal lack of attention to mental states in normal interaction.
regular people don't think of our  in terms of understanding mental states. Even when prompted, Ratcliffe's class did not come up with that sort of talk
So what do the "folk" themselves think about "social understanding?"