Keyboard Shortcuts?f

×
  • →Next step
  • ←Previous step
  • ↓Skip this slide
  • ↑Previous slide
  • mShow slide thumbnails
  • nShow notes
  • hShow handout latex source
  • NShow talk notes latex source

Click here and press the right key for the next slide.

(This may not work on mobile or ipad. You can try using chrome or firefox, but even that may fail. Sorry.)

also ...

Press the left key to go backwards (or swipe right)

Press n to toggle whether notes are shown (or add '?notes' to the url before the #)

Press m or double tap to slide thumbnails (menu)

Press ? at any time to show the keyboard shortcuts

 

Notes and Slides

‘The words and phrases of natural languages comprise a treacherous basis for identifying valid psychological constructs, as I illustrate in emotion research.’

‘the emotion labels in natural languages do not have definite, stable, mutually transparent meanings.’

‘it is pernicious to use one language’s dictionary as the source of psychological constructs.’

Then: Bart and I were wondering whether this problem is specific to labels for emotion research or whether it generalises to research in what is sometimes called folk psychology, theory of mind or mindreading.

Fiske (2020)

(Howell, 1984)
(Borg, Hansen, & Salomons, 2019) (Liu, 2022)
Steve interjects: ‘But surely we can’t avoid folk psychological terms altogether’?
Steve takes over here!
(Heider, 1958) (Westra, 2021)
We seem to have shifted quite quickly from (i) talk involving folk psychological terms to (ii) concepts of mental states and on to postulating a (iii) capacity to predict and explain.
It appears that folk psychologizing has turned in on itself.
What could go wrong? Have we overlooked the diversity, polysemy, indeterminacy and context-dependence of folk psychological terms and therefore ended up lacking any way of properly identifying the concepts and capacities being postulated?
Recently there has been some striking successes in demonstrating that nonhuman primates can pass false belief tasks.

‘the present evidence may constitute an implicit understanding of belief’ (Krupenye, Kano, Hirata, Call, & Tomasello, 2016)

‘Apes [...] treated the actor [...] as having a false belief’ (Kano, Krupenye, Hirata, Tomonaga, & Call, 2019, p. 20907)

As Krupenye et al. (2016) write, this is widely regarded as evidence that they understand belief.

passes false belief task ∴ understands belief

Why is this a correct inference? Two possibilities ...

How is the step from success on false belief task to understanding false beliefs justified?
In asking this question, I am not attempting to challenge these authors’ conclusion. I chose them because they are particularly careful and subtle.
In asking this question, I am not focussed on these authors’ explicit reasoning. They quite reasonably follow the orthodox view that passing a false belief task provides evidence understanding false beliefs.
Instead I want to know what kinds of consideration might support this orthodox view. Consider two possible lines of justification ...

Justification 1

‘Comparative psychologists test for mindreading in non-human animals by determining whether they detect the presence and absence of particular cognitive states.’

(Halina, 2015, p. 487)

On Halina’s view, the justification for attributing understand belief is exactly like the justification for attributing understanding mass.
This first line of justification cannot be used to attribute false belief understanding to chimps. Why not? If we are right, terms like ‘belief’ do not pick out particular cognitive states.
An alternative is to rely on an operationalisation with robust statistical properties.

Justification 2

1. [fact] False belief tasks appear to test for a single underlying competence in adult humans

(Flynn, 2006, p. 650; Wellman, Cross, & Watson, 2001)

2. [assumption] False belief tasks test for the same underlying competence in chimps (and infants).

3. [stipulation] ‘understanding belief’ is whatever that competence is.

If you agree with our view on folk psychologies, then this is one available way to justify Kano et al’s view.
We think it is actually a reasonable justification, and we are also not aware of an alternative way to justify it.
But note that if you rely on this line of justification, you face various consequences.
One is that this line of justifications straight-out fails for infants.
You cannot justify attributing ‘understanding belief’ to infants using this line because infant false belief tasks do not appear to test for a single underlying competence.
Further consequence of relying on this line of justification: it does not work for Theory of Mind more generally.
Why not? Two reasons ...
First, we do not really know much about the structure of Theory of Mind, and different researchers use different taxonomies (Happé, Cook, & Bird, 2017; Beaudoin, Leblanc, Gagner, & Beauchamp, 2020).[^tom-construct-problem]
Second, while there is some evidence that a wide range of false belief tasks appear to test for a single underlying competence (Flynn, 2006, p. 650; Wellman et al., 2001), when we turn to theory of mind tasks more broadly we find that different theory of mind tasks appear to test for different things in the sense that an exploratory factor analysis fails to find that they load on a single factor (Warnell & Redcay, 2019).[^why-problem]
In particular, there is no way to apply this second line of justification to thinking about communicative intentions ...
[^tom-construct-problem]: See Beaudoin et al. (2020, p. 15): ‘The lack of theoretical structure and shared taxonomy in ToM definitions and its underlying composition impedes our ability to fully integrate ToM in a coherent and comprehensive framework linking it to various socio-cognitive abilities, a pervasive issue observed across the domain of social cognition.’ [^why-problem]: It is important to be clear about why this is a problem. It is not a problem that Theory of Mind may involve a variety of different processes and models, so that no single factor will explain performance across a sufficiently diverse set of tasks. But if you want to say, independently of answering the question about models, that we have a solid operationalization of Theory of Mind, then you need statistics to show that your operationalization has some kind of internal coherence. And that is what appears to be missing.
Talk about intentionality pervades work on communication. But what does it all mean?
Chimps, ravens and grouper fish.

‘Of particular interest [...] is the question of what chimpanzees intend their gestures to mean’

(Hobaiter & Byrne, 2014, p. abstract)

‘These [ravens’] signals thus qualify [...] as intentionally produced communicative gestures and are understood by receivers’

(Pika & Bugnyar, 2011, p. 4)

‘The grouper’s signalling has hallmarks of intentionality [...]: persistence until the goal is reached, elaboration of the communication if it does not initially achieve its goal, and means-ends dissociation by using two signals for the same goal’

(Vail, Manica, & Bshary, 2013, p. 5)

Our view is that we cannot anchor notions of intentionality in folk psychological practices. We are therefore confronted with the question, what are these things (persistence, elaboration and means-end dissociation) hallmarks of?
You might think that philosophers would be able to help us out here, but in fact we run into a general problem about operationalisation ...
I think the challenge we face with respect to operationalisation comes into view more clearly if we focus on how planning relates to intention ...

operationalisations

Planning is a mark of intention

 

Some spiders (portids) can plan detours

(Jackson & Cross, 2011)

Therefore:

Spiders have intentions

Spiders do not have intentions

(Frankfurt, 1978, p. 162)

Some spiders (portids) can plan detours

(Jackson & Cross, 2011)

Therefore:

Planning is not a mark of intention

Consider the idea that we can operationalise intention in terms of behaviours that indicate planning.
How do we know whether the operationalisation is correct?
To know which inference to select, we would have to have a sense of what the operationalisation is supposed to capture.
But where does our understanding of intention come from? Our suggestion is that it cannot come from folk psychological practices because of diversity, polysemy, indeterminacy and context dependence. [Not sure I think normativity is a problem; still thinking about that.]
Maybe it could come from philosophical theories? ...
INTENTIONyesno
Is it a mental state?Davidson (1978)Thompson (2008)
Is it a belief?Velleman (1989)Bratman (1987)
Entails belief?Harman (1976)Levy (2018)
Linked to planning?(Bratman, 1985)
Incompatible with habitual? Kalis & Ometto (2021)
Entails non-observational knowledgeAnscombe (1957)
Comes in two kinds?Searle (1983)Brozzo (2021)
Conjecture: there is a consistent theory for any of the 128 patterns of answers to these seven questions.
This is not the kind of problem that philosophers can help with. In fact philosophers are partly responsible for the problem.
If we are right, there is a challenge for approaches to understanding communication in different species in terms of intentional communication.
The challenge is to say what intentional communication is, and to do so in a way that enables us to critically evaulate how good the usual operationalisation is.

Eschew Academic Folk Psychologizing

Challenge: What do operationalisations of ‘intentional communication’* operationalise?

*(persistence, elaboration and and means-end dissociation)

If we are wrong, there is still a challenge but it is a bit different. The challenge

Continue with Academic Folk Psychologizing

Challenge: Do fish have communicative intentions? Can spiders intend? Can fish experience betrayal?

This is a challenge because the question seems to be unanswerable.

‘species capable of theory of mind [...] would understand that their caregiver made a conscious decision to deceive them’ (Oldfield, 2022)

‘it is quite possible that [...] you can betray a fish’(Balcombe, 2022, pp. 60--1)

This is what I think we should do.

conclusions and recommendations

In philosophy of mind and action, in philosophy of language, and in epistemology, we have been been relying on academic folk psychologizing to ground theories.
The diversity, polysemy, indeterminacy, and context dependence of folk psychological terms like ‘intention’ and ‘knowledge’ suggests to me that this is not a sensible way to proceed.
If that’s correct, I think there are two bits of work we need to do.
First, we need an approach to characterising phenomena like communication which avoid explicitly or covertly importing folk psychological notions like knowledge and intention.
To this end, I suggest we take attempts to operationalise communicative intention as a starting point and ask the question what they operationalise. The challenge, though, is to understand whether, in humans, there is a single thing that different criteria all operationalise. We also need to know whether proposed criteria such persistence, elaboration and means-end dissociation operationalise the same capacities in different species.
The second bit of work we need concerns theory of mind. I used to think that there was a need to characterise minimal forms of theory of mind but that we already had a clear idea about what full-blown adult human theory of mind looks like. After all, I thought, it involves reasoning about beliefs, desires intentions and the rest.
Now I think this amounts to trying to characterise an ability by applying our own folk psychological practices. So that won’t work. We need a more rigorous way to characterise full-blown adult human theory of mind.

appendix

‘epistemic case intuitions are generated by [...] folk psychology’

Nagel (2012, p. 510)

In case someone asks for an example of academic folk psychologizing. (NB we are not saying that everything philosophers do is raw academic folk psychologizing, just that this is a key input, as here where it provides intuitions.)