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

 

Conclusion to Philosophical Issues in Behavioural Science

 

Lecture 09

Philosophical Issues in Behavioural Science

conclusion

In conclusion, ...
I am going to start with a 90 second version, and then repeat

To understand why people act, individually and jointly.

The aim was to understand why people act both individually and jointly.
We notice that there are independent sources that we can use from philosophy, psychology and also the formal parts of economics. And those answers all seem relevant to understanding why people act.

Philosophical, psychological and formal answers are—or appear—both mutually dependent and inconsistent.

But there is an interesting problem, which is that these answers appear, on the one hand, to be mutually dependent in the sense that they offer complementary pieces of a puzzle. And yet, on the other hand, the answers also appear to be inconsistent with each other. And that, of course, is a problem.

We cannot have a full understanding of why people act. If we've got three sources, all of them seeming to offer complementary bits of puzzle but also offering consistent theories.

This is an obstacle to full understanding,
but one that you can overcome.

The good thing is that this is a puzzle which you can overcome.

slower conclusion

challenge

Discover why people act,
individually and jointly.

When you act,

there are reasons why you act;

you know the reasons;

you act because you know the reasons; and

the reasons justify your action. make your action intelligible.

So I ask NN to why did you throw the chair out of the window?
And she says, well, Steve, you know, I knew that there was a fire and that the route through the door was blocked and that this was the best available way of smashing the window.
And suddenly an otherwise act of vandalism now appears in a completely different light.
I've seen it through NN's eyes, so it seems like a kind of sensible, rewarding picture.

How can we turn this into a theory? Is it true?

This is the basic picture that nearly all philosophers start from in attempting to meet the challenge.
How can we turn it into a theory? We attempted to do that in two ways.
[Way 1] One was by thinking about the philosophers, particularly Donald Davidson, who has the idea that we can solve the problem of action by appeal to an intention. So an intention is something that specifies a reason and is arguably also a mode of knowing. So if you intend to throw the chair through the window, then arguably you know that you are throwing the chair through the window in virtue of intending or that you were intending to do so.
[Way 2] You can also think, if you like, of decision theory as a way of capturing part of the simple story. The problem there is that with decision theory, we're cutting out the bit about knowledge. So it seems like we're maybe missing bits of the the story. And that's not surprising if we start off with a story and then we try to turn it into a theory, it's inevitable that bits are going to be gained and bits are going to be lost along the way.
Okay, so we had a little go at that. We didn't get very far.
[in outline: repeats the above ...]
Davidson’s view is a version: answer the Problem of Action by invoking intentions; to intend is to know a reason for your actions.
We can also draw on decision theory in turning the simple picture into a theory (decision theory and game theory are supposed to be elaborations of what it is to be rational). Of course, if we do this we lose the knowledge component.
One difficulty is that any such theory needs to be consistent with discoveries from various behavioural sciences, and that proves to be a large challenge.
The other question is whether the theory is true. And this is where the challenge of integrating the philosophical, the psychological and the formal has bite.
If you just looked at the philosophers, the philosophers are all about the simple story. So when we ask whether it's true, if we just want to look at philosophical writers, we would find that it has this kind of wide degree of credence, and you can make sense of it by appealing to Davidson or by appeal to Anscombe, or by appeal to Korsgarrd or Velleman or a variety of other philosophers of action, all in one way or another, offering ways to make the simple picture true.
But when we turn to psychology and neuroscience bits of economics, it's much harder to see how the theory could be true.
We can draw on them to show that the simple picture is either wrong or else only a very small part of the answer.
Just here we've got two choices, I think.
[Choice 1] One would be to reject this Simple Picture altogether (as, perhaps, Chater (2018), does in different words).
But it seems to me we haven't seen an argument for such a radical conclusion, because there's another possibility, which is that the simple picture is getting at some deep and interesting truth, and that truth is being missed by the other disciplines.
[Choice 2] Another way to go is to embrace the challenge: we need what we do not have, which is a way of making systematic sense of the role of reasons in explaining why people act.
This raises a number of questions. [1] Clearly reasons are not invariably involved. So what distinguishes cases where they are from where they are not? [2] Is the Simple Picture about processes or about something else? [3] Is the Simple Picture merely a gloss on existing scientific theories or does it point to the existence of phenomena that are not yet explained by existing scientific theories?

relevance to essay questions

But that is all big picture ... how to link it to essay questions?
So far I have been giving you the summary in a colourful way. But I don't expect that in your essays you would be writing about the simple picture. That would be an unusual way to go, just because the questions are sort of too big to be really manageable. And to my mind, it's better to focus on details where we can actually make progress and construct arguments.
All of the essay topics, apart from those about the interface problem, involve integration questions.

Integration Questions

Identify theories from two or more disciplines
(philosophical, psychological or formal)

which appear to target a single set of phenomena
while saying incompatible things about it ...

1. Are they actually inconsistent? ✔?

2. If so: how, if at all, should either or both theories be refined?

How many integration questions can you identify from the course?
[1] Standard Answer to the Problem of Action vs the dual-process theory of instrumental action [2] Standard Answer to the Problem of Action vs theories of motor control [3] Decision Theory vs the dual-process theory of instrumental action [4] Bratman’s theory of shared intention vs team reasoning [5] Bratman’s theory of shared intention vs motor representations of collective goals

Standard Solution to The Problem of Action vs the dual-process theory of instrumental action

Standard Solution to The Problem of Action vs theories of motor control

Decision Theory vs the dual-process theory of instrumental action

Bratman’s theory of shared intention vs team reasoning

Bratman’s theory of shared intention vs motor representations of collective goals

intention decision theory habitual vs goal-directed
We always have to know where we are in relation to the triange.

philosophy

What distinguishes actions from things that merely happen to you?

psychology

Which processes are involved in selecting the goals of actions?

Actions are those events which are appropriately related to intentions.

At least two: habitual and goal-directed.

apparent inconsistency

Nothing to say about processes.

Nothing to say about what action is.

apparent mutual dependence

Psychology relies on some understanding of what action is. Without this we have no idea about our subject matter.
Natural to think that philosophy will provide this.

1. Preferences shape intention:
pathological cases aside, if there are two outcomes and you prefer one outcome to the other, and there are no reasons to pursue the other outcome, then you will not intend an action that brings about the less preferred action.

2. Where habitual processes dominate, you sometimes pursue less preferred outcomes.

Therefore

3. Not everything you do involves intention.

philosophy

What distinguishes actions from things that merely happen to you?

psychology

Which processes are involved in selecting the goals of actions?

Actions are those events which are appropriately related to intentions.

At least two: habitual and goal-directed.

apparent inconsistency

Nothing to say about processes.

Nothing to say about what action is.

apparent mutual dependence

intention decision theory habitual vs goal-directed
Decision Theory vs the dual-process theory of instrumental action
Fewer people got this but I think only because I didn’t explain it clearly enough.
Decision theory is a model. A model says nothing about the world but can be construed in one way or another.
We can construe decision theory as a predictive device, as a normative device or as a an elucidation of the notions of preference and belief.
If we construe decision theory as an elucidation of the notions of preference and belief, then we can only apply it to agents whose actions invariably reflect their preferences and beliefs. (After all, the central idea is that you can extract preferences and beliefs from the choices people make.)
The problem is that the characteristic of habitual processes is that devaluation has no immediate effect on the actions they trigger.
[alt:] The problem is that the dual-process theory of instrumental action shows this: there are situations where (1) the goal-directed process would result in one action; (2) the habitual process would result in another, quite different action; and (3) which process dominates does not depend on your beliefs or desires but on your training history, or on how stressed you are, or on which situation you are in [popcorn: cinema vs work environment] ...
It follows that your actions will not invariably reflect your preferences and beliefs; and so that you have no preferences or beliefs at all.
But this is absurd; and so we cannot construe decision theory as an elucidation of the notions of preference and belief if we also accept the dual-process theory of instrumental action.

Integration Questions

Identify theories from two or more disciplines
(philosophical, psychological or formal)

which appear to target a single set of phenomena
while saying incompatible things about it ...

1. Are they actually inconsistent? ✔?

2. If so: how, if at all, should either or both theories be refined?

Over to you. (I genuinely do not know, although I will offer some clues at the end of the lecture.)

Integration Questions for Joint Action

shared intention game theory dyadic motor plans
This is partly a preview because I did not tell you what dyadic motor plans are yet—that is coming later in the lecture.
Is there both mutual dependence and inconsistent claims? I am unsure about this.
Yes, they are inconistent. On team reasoning there is no role for shared intention; on shared intention there is no role for team reasoning. And since each attempts to say how joint actions are selected, they cannot be combined.
But are they mutually dependent? (Maybe on Pacherie’s view: one is about fast, uncomplicated cases.)
But in the lectures we skipped over that because there is actually a substantial attempt to unify the two theories. And if unification succeeds, we do not need to worry about whether it was necessary or not.

philosophy

What distinguishes joint actions from things we do in parallel but merely individually?

economics

How can we model rational behaviour in social interactions?

shared intention

game theory team reasoning

Pacherie’s Reconciliation:
‘shared intention lite’

nothing to say about agents without planning abilities

no role for planning abilities

apparent mutual dependence

There appears to be mutual dependence here, because the philosophers have got nothing much to say about agents who don't have planning abilities at all. For example, children under the age of 5 or 6 seem not to be super good at planning and certainly are not good at sharing their plans with other people. They're not really plan coordinators.
On the other hand, when we look at the theory of team reasoning, in other cases, we see that there seems to be no role at all for planning abilities.
But planning abilities are a massive part of human's competence in acting, and their ability to share their plans seems to be a really important part of joint actions.
It's strange that there's something missing here.
So on the face of it. We've got mutual dependence. I'm not saying I've given you a compelling argument, but just on the face of it, it looks like these two theories ought to be somehow combined, able to give us a better view of the whole picture.
But at the same time, there's a clear incompatibility insofar as from the philosophical point of view, you only have the joint action when you have shared intention, whereas with the team reasoning case, there's no need for shared intention at all.
And conversely. This team reasoning is not generally a feature of joint action from a philosophical point of view.
So it looks now like, well, they're filling in different bits of the story, but they also offering theories that are incompatible with each other.
So we have a problem.
You might remember that Pachiere offered a reconciliation. She proposed that we can think of shared intention as involving two quite separate things.
One form of shared intention occurs when a piece of team reasoning results in a decision that is, in and of itself, a shared intention. So that was one of her ideas.
And her second idea was that there's also another form of shared intention, which is the kind characterised by Michael Bratman and other philosophers, and we need both kinds of shared intention.
So that's our way of trying to reconcile the two theories and put them together and avoid the incompatibility between them.
So this looks to be a case where the Integration Challenge was resolved.
Quick reminder what Pacherie’s theory says, in essence
individual reasoning team reasoning decision decision intention shared intention

Integration Questions

Identify theories from two or more disciplines
(philosophical, psychological or formal)

which appear to target a single set of phenomena
while saying incompatible things about it ...

1. Are they actually inconsistent? ✔?

2. If so: how, if at all, should either or both theories be refined?

Yes, they are inconistent. On team reasoning there is no role for shared intention; on shared intention there is no role for team reasoning. And since each attempts to say how joint actions are selected, they cannot be combined.
Actually if you remember there might have been an objection (the autonomy challenge). But still, at least we have a candidate here
So we want to be careful here. There is a solution. It's detailed. It's well worked out. It looks, on the face of it, credible. But there's at least one objection it needs to overcome. And that's probably going to be the same for you writing an essay.
If you do come up with a solution, it's going to be important to think about where the next objection might be coming from, because surely there will be one.

challenge

Discover why people act,
individually and jointly.

basic theories and discoveries
from three disciplines needed
to answer the question

I always wanted to pure philosophy ... in fact maths and philosophy because I felt stats and mechanics were too applied ... and for a long time I hoped that this would be enough but in talking to scientists I realised that it isn’t

to reach beyond
you need to look beyond

🔑

inconsistencies abound (maybe)

but integration is possible (definitely)

This is the key to the essay questions. In one way or another, you are addressing this question: find significant inconsistencies or consider how integration is possible. (Except for questions about interface problems)
In a sense, I'm not concluding at all because the essay that you write for me and that I then read and have a think about it as well, that is the conclusion of the course rather than the lectures.