Consciousness Live! with Myrto Mylopoulos (S3 E22)

0:00 all right we are live so joining me today an extra special guest all the way from our northern friendly borders uh mirto milopolis um thank you for joining me today mr tom really looking forward to talking with you so i wonder if you could say a few words about who you are and what you do for those who might not know you yeah sure uh first i'll say thank you so much uh richard i think this is really cool uh what you're doing uh with this series and um yeah i'm excited to be part of it um uh about me i am

0:35 a professor of philosophy and positive science at tarleton university and um you know broadly speaking my research is in policy of mind cogside action and uh in particular i'm interested in the ways in which agency and consciousness uh intersect um and i think there are a lot of really interesting uh features of that of that relationship so some of it some of them have to do with uh the role of consciousness and facilitating or enabling certain types

1:07 of flexibility or action control um they have to do with the kinds of conscious experiences that are um characteristic of our exercises of effort or control or execution um there's questions about the relationship between uh agency and attributions of consciousness so do certain types of agency give us give us warrant for attributing uh creature or state consciousness in certain contexts so i'm really interested in all these kinds of questions and how to kind of in our pictures of our notions of

1:42 consciousness that we we can connect them to agency and in the ways that that are that are suitable yeah interesting i'm a big fan of your work i have enjoyed reading it quite a bit learned a real real a lot actually from your careful sorts of things but you also have stuff so agency seems to be like the main focus of a lot of your work i guess but you also dabble in other things too like you have a paper with tony rowe on synesthesia for example um something on the intentions conscious intentions uh sort of criticizing jesse um so are these other projects do you

2:14 see them as fitting into this the the story about agency or are these just sort of side projects that you got interested in the the project with tony was was sort of a side project that uh came out of a conversation that we were having at uh the assc in kyoto uh several years ago so we're he's done some amazing work on uh visual you know perception and consciousness and and uh had some interest in synesthesia and in particular and really we're just trying to um with that project understand uh the relationship between uh synesthesia

2:50 and and consciousness so the the extent to which um uh we might have a certain interesting case study uh in in synesthetic experience with respect to um you know do all synesthetic experiences have to be conscious can they be prime can they be triggered by subliminal perceptions can uh so just just kind of general questions exploring the the connections between synesthesia and you know theories of consciousness and what we might learn about it um so that was sort of a side project that

3:21 i came out of that conversation that's not something that i i take to be uh super um connected with with what i'm interested more and more these days uh with respect to the uh conscious intentions that was that was more so i mean i had general interest in theories of consciousness and uh um how those might go and and again i think the theory of consciousness should it should be able to account for uh facets of agency just as well as it does uh account for facets of perception or other other parts of cognition

3:54 and so um you know the run at air theory was was to raise a concern that perhaps conscious intentions are properly accounted for within that framework um but those are those are those those are sort of one-off projects that i've kind of left behind at this point yeah okay i think that's an interesting point that you just made though and i think i want to it's i want to kind of underscore it the idea that we want a theory of conscious male account a consciousness to be able to account for these very kind of conscious experiences and um the extensive agency agent phenomenology or whatever you want to

4:28 call it isn't something that really gets talked about a lot with relation to other theories of consciousness i think that's important um in fact i don't know so i was going to ask you about this anyway so because i guess when i first started reading a bunch of your stuff i was i guess i thought of myself as sort of a skeptic about um the existence of agenda phenomenology um and thought well maybe it's probably reducible to some other kind of phenomenology but after i mean after reading a lot of your stuff i came to be sort of i sort of reversed my course on that you convinced me that there is something that's maybe proprietary here

5:00 but but i know that i wasn't alone in this lots of there are other people out there who sort of have these same feelings so so can you say something about well first of all what a gentle phenomenology is and why someone might be skeptical about it oh sure um i mean i think we're thinking about it agenda phenomenology i i don't think it's this i don't think it's one thing really so i think there are varieties of agenda experience um i have a chapter in uh uri klugel's handbook on positive consciousness with with

5:32 joshua shepard on this and um in that chapter we sort of outlined different uh facets of agent experience and um i think one can be skeptical about some of them while endorsing others right so these range from feelings of effort or exertion they might include the sense of control over some action as it unfolds uh they might include a sense of authorship over over some action some folks want to include a

6:04 phenomenology of free will as as a kind of experience that might underlie say libertarian beliefs about free will um so there's a kind of range of experiences that i do think deserve our full attention and in terms of theorizing about the kinds of psychological mechanisms and representations that underlie them uh i think that probably most of the skeptical um concerns target this what i call legend of awareness or awareness of oneself as

6:36 as a source of action and um i think in part the skepticism might just be one of okay where what is this experience can you tell me more about the content how do we describe it when when is it absent when is it present and uh the best way that i've i've come across i guess to motivate that uh something like this this kind of agenda of experience of authorship might might exist um is when we look at cases uh where it really seems to just be um

7:13 impaired or absent and a lot of these cases come from uh these are these are clinical cases um cases of schizophrenic individuals with uh delusions of control where they might attribute their actions to external agents um cases of uh interdependent syndrome which involves you know lesions to parts of the motor cortex where uh you have the individual engaging in kind of seemingly goal directed actions right so they'll uh they'll they'll you know pick up an object and

7:45 manipulate tools and the right kinds of ways but uh all the while um disowning the object as one of their own authorship uh so those kinds of cases i mean of course there are there are issues with drawing influences from you know neural atypical cases to what's going on in the typical case um but i do think they point to a kind of a mechanism that might typically undergird our experience of authorship uh that can break down um in certain special cases and um

8:19 so uh so that's one way to kind of motivate uh the idea that that these experiences uh exist but i also think that you know that in some ways they're just it's kind of like a proprioceptive experience and they're just ever present right so we're we go about our day we we express and we exercise our own agency all day long uh both in the context of over bodily action and mental action as well um and so i think it's just it's such a ubiquitous part of our experience that it's it's hard to sort of pinpoint and isolate until you start really um attending to some of its

8:53 uh features yeah so and you left a cup i guess there's a couple of other ones in there too like there's the sense of mindness um which is uh part of the is that the same as the source of agency that's is that distinct yeah so you might you might think that um sense of authorship uh just comes down to a sense of ownership over in action uh so um there's a postdoc at ubc named aaron henry who's who characterizes things in that way and i think that's that's a good way to think about it so it's basically tagging an action as

9:27 as my own um some sense that i'm i'm the source of that that behavior and again that might be absent in these cases of um a breakdown uh and maybe in cases of compulsion as well so you might have um people with uh ocd repetitive patterns of behavior that in some sense uh they might not feel uh ownership over yeah i mean so it's it's a bit interesting that you're aiming to explain something which is like most noticeable when it's absent

9:59 yeah um and that people spend a lot of time kind of debating like is this sense of you know agency of um because i when i was reading a lot of this stuff i was paying a lot of attention to my movements within a way that i never had before and i was like am i does it feel like i'm doing something right now or am i you know because there's a kind of fluid effortlessness that comes with moving one's own body at least in most cases unless you're learning to do something or something but it's it's if it's there it's subtle it's hard to find in i mean would you agree that it's hard to find like introspectively the thing that that

10:32 you're trying to explain unless you've had it absent or something like that yeah i mean i do think it becomes the the experience comes becomes more salient with these these contrast cases and you know in typical cases when i think of simple reflexes you know you go to the doctor's office and they test your patellar reflex your knee flies up your leg your leg shoots out and you don't have a sense of authorship over i mean you can you can uh infer that it's yours based on

11:02 observation but um but you don't have that same sense of control or relationship that you do with your your typical actions uh i think the fact that introspectively it's not um an evident kind of experience or sailing experience might also offer clues to the kind of experience that it is right so um i'm i'm i'm not really sympathetic to the view that it's a that it's just the brand of sensory phenomenology um some people have have argued

11:33 uh i don't think it has a lot of the qualities of sensory phenomenology and we can talk about its vividness and its aliens as as as ways in which is this analogous um so uh so that so i don't i don't think it's i mean it might provide foods as to what kind of experience it is the fact that it doesn't share these qualities with other kinds of sensory phenomenology for instance right that's something you've pointed out i think is that we would expect there to be a sensory quality of some sort like some

12:04 some way some sense organized organ or something like that which was responsible for detecting these things we don't really find something like that yeah we i mean it doesn't seem as though whatever we're um aware of in these cases uh is anything like a perceptible property and and the regular way that our sense organs and our sensory systems operate to to track and um uh you know it doesn't there's hard to make sense of how for instance the quality space theory would get off the ground and trying to talk about agenda

12:36 qualities in the same kind of way um uh so that that might lead us to think that what we're looking for is a kind of non-sensory phenomenology and um i've tried to make a case for thinking of a vagina of awareness in terms of a type of uh cognitive phenomenology um so awareness of yourself by way of first-person thoughts that you're you're the agent of some action um which i think fits in perhaps with i mean that that's that's for some folks already an overly deflationary way of thinking about the the experience but um i think it

13:11 does kind of track uh the the some of the skeptical um concerns that you were you were articulating yeah so if it is something that's cognitive then i mean there's a whole debate about cognitive phenomenology on in its own right and so that's a kind of subtle thing in the background too i mean you hear a lot it's the same sort of response so how come people don't notice you know cognitive phenomenology people say well it's you know ubiquitous it's always there in the background so to speak that sort of what i heard you saying about why we may not notice uh agenda phenomenology in the same way um so i want to ask you about your code

13:43 because i know i know about your cognitive theory i think i think it's interesting but before we get to that so you mentioned briefly the free will stuff um and but that's not what your theory is aimed at explaining is it or is is this uh it's you know the sense of choice or poisonous or however people want to put it a sense of choosing that's that's a separate thing from what you're interested in i would say it's a separate thing i mean i think um

14:16 well i'm not sure i mean i i think it depends what one has in mind of course with i think i think there's one way of thinking about the phenomenology of free will which is such that um all it all it requires is the sense of uh woman called self as source and so um kerry horrigan wrote a bunch of stuff on this and i mean the idea is that uh if we kind of are reading a description of

14:50 uh the phenomenology of agency off of like a libertarian account of free will that involves some sort of agent causation so you don't have it's not it's not mental events that are causing bodily events it's it's you right you the agent um are causing uh some some action and and that's uh one way of securing this kind of again libertarian notion notion of free will then you then you might think that the phenomenal knowledge of free will in that guys is just self the source of some action but we might be talking about

15:24 about the same thing but then there there are other ways of thinking about the phenol phenomenology of free will that have more to do with um the sense of being an uncaused cause right so uh your your behavior not being um caused by uh or whatever else right so um if we're talking about those the those ways of pinning down the phenomenology of free will that i'm not sure that so i wouldn't say that my my um my uh my own account kind of

15:58 it speaks to to those questions i'm more interested in uh when is it that i have the sense that i'm the author i'm the person controlling some behavior right so and that might come apart from the sense in which you feel that you uh that nobody's making you do it or that and there's the other famous phrase you know that you could have done otherwise that you there's a sense in what you feel in the moment of choosing you felt like you could have made a different choice and so that's not something that you're worried about either it sounds like primarily what you're interested in is that it's that i feel like it's me that's performing the action

16:31 yeah so just just all right um and again if we think of fruit phenomology fruit as itself is source and then maybe we get some convergence there but uh yeah not in terms of the ability to do otherwise this might have to do with stimulating counter factual scenarios or what have you not not with respect to the the feeling that you're not compelled or constrained or you know caused in any uh those kinds of senses to to engage in your actions it's really just this um okay it was me that that performed the action

17:02 right so so on the way you present these kind of uh problem cases i don't know what to call them problem case or whatever but like anarchic hand syndrome um so the the people who have this happen to them they they say you know the arm is moving but they're not moving the arm or that they they're not responsible for it but they recognize that the arm belongs to them um and so they often attribute the movement to an external agent or to someone's so what i'm curious about is

17:34 how do they describe the experience um is it someone controlling their their arm externally or is it someone controlling their brain internally or like how do they what exactly do they report yeah good so i mean in the case of of anarchic hand syndrome typically these these are not um uh individuals who suffer from from delusions right so their sense of reality um is intact uh um the kinds of cases where you have these kind of external attributions or confabulations um have to do with are the ones i'm familiar with are from

18:07 schizophrenic individuals with western um symptoms but in the case of anarchic hand um it's interesting because they're i mean they're aware that it's their their body moving so they're they're um sometimes uh i guess forced into metaphors so they'll say you know it's behaving as if it has a will of its own and sometimes the anarchic land behaves at cross purposes with the agent so there's this famous case of an agent an interdecant patient trying to play checkers and

18:41 make a move with their healthy limb and the energy you know takes it back and um so they so they sometimes have a sense that they're kind of fighting with their uh their anarchic uh limb so they they do speak sometimes i think metaphorically and attributing a will or end of its own to the limb but they're they're very much aware that it's um uh it's it's their body that's a little advantage don't feel that it's that they themselves are offering the action i think in this case what's what's missing is a sense of control over the behavior of course so i mean part of

19:14 because this is sort of tangential to your main line of work and i'll get back to that second but since you know a lot about this i want to ask you um it's so if if a person has damage to like the motor air cortex and they undergo this which is that's where the damage typically is for anarch candidates like the parietal cortex or i guess in the frontal cortex supplementary motor area yeah so okay so what explains the the uh the the actions in this case i mean these aren't the sorts of things that they would be doing normally one takes it

19:45 so i mean in the case where there's antagonism between the hand and the person so to speak i don't know what the right way to describe it is but what is there any theory or explanation for why this these kinds of actions would manifest in the first place yeah i mean what they seem to be uh are often um just really highly stimulus driven responses right so um they're often uh cases where the relevant behavior is like highly highly familiar highly routinized automatized behavior right so um they're responding

20:18 to stimulating the environment in automatic ways and we have already evidence independently uh to think that um uh objects that we're familiar with interacting with so a cup uh maybe a checkerboard piece whatever uh these are things that um kind of automatically activate uh action routines that sometimes they then go on to performance sometimes not and this is you know this is why uh people have engaged in action slips right so you you go to the fridge you go to you're trying to get milk out uh you want to pour milk on your cereal

20:52 and you find yourself pouring orange juice on it right so it's just kind of these his automatic actions where if you're not paying enough that sometimes require attentional or inhibitory mechanisms uh in order to prevent the unfolding of um those those mechanisms are not in place at least the inhibition mechanisms are not in place and in the context of anaerobic candles is one um one one view on offer with respect to what's happening so uh so again the the although the the actions of the anarchy

21:25 glamour are they're goal oriented in a certain way um they're typically these very kind of familiar routine types of behavior yeah because i mean so the sense that they're antagonistic is kind of illusionary then really it seems like because i mean i've read cases where like a guy's buttoning a shirt and the hand comes and unbuttons the shirt or he reaches for the cup and the hand slaps the cup out of his out of his way so it does seem like you know more i mean they seem purposeful but they seem more like trying to prevent somebody from doing

21:57 something but i guess what you're saying is that the that the consensus that that's kind of an accidental feature of the actions i mean i think that's that's a kind of um uh it's a narrative right it's it's a kind of narrative way of understanding what's what's happening and it's i think natural for us to try to attribute when we see behavior that's complex it's natural for us to attribute agency to attribute some goal to make sense of it in some in some way um what seems more likely here that's happening is that it's just the same stimulus that's triggering uh

22:31 one in the very limited set of responses we have to very stimulate unbutton or surge or you button it up right those are the two things we do with certain buttons i mean it's just funny you never read cases about the helpful end on a an archived hand you know it's like uh the guy's looking for his glasses and the hand reaches everywhere yeah um yeah there's one so i mean this actually points to an interesting feature as well of of of the of this behavior which is that agents don't seem to have a

23:04 first-person access to it right so it's by observation that right uh they come to be aware that their inner cocaine is doing something so there's one famous uh this one woman reports that veronica can would would take stuff out of her purse and just like you know scattered about it took her very long time to realize what was happening because she was never aware that that her hand was grabbing these items and dropping them so um so in terms of the subjective access we have for our own agency right obviously it's very different

23:36 from our appearance that's very different from from the access we have to uh what others are doing and what whether our bodies are doing from the outside but in anarchy campus into they seem to lose that subjective first-person awareness right yeah that is interesting okay so i i think probably the what from my understanding the kind of dominant understanding of these um the phenomenological aspect of it is the comparator model um so first of all do you think that's right that's kind of the the dominant model um and secondly can you tell us a little

24:09 bit about it and why why you i mean i know you argue against it pretty convincingly in my view but can you take us through some of that yeah i mean i think um so it's it's been the comparator model i think has been the dominant model especially among uh neuroscientists and psychologists for for explaining uh the sense of the sense of agency or um pausing there because sense of agency is also used in different this is kind of an

24:39 umbrella term but um some people use it to mean different things um so some people use it to mean sense of control some people use it to mean a sense of authorship um but uh the the general idea and this is something that's been taken up in philosophy by people uh like uh tim bain and uh to a certain degree elizabeth pescheri the general idea is that uh every time we we perform an action uh the motor system outputs um a kind of prediction of what the sensory

25:13 consequences of that action will be and uh before the action unfolds some uh there's you know there's a comparison between his predicted sensory consequences and the agent's goal to to make sure that um there's a match there and then after the action up folds there unfolds there's a sensory feedback which is compared with the predicted sensory feedback and if there's a match there then um the action is as experienced as one's own

25:43 uh or there's a sense of control right depending on what you think the the result of that comparison is um uh and if there's a mismatch then you get a sense that something's gone wrong and some adjustment needs to be made uh so it's actually at first in the first instance the comparator model is is meant to be a model of motor control and has some uh some history and um computational engineering uh some work in the 90s but but it's been kind of

26:15 used as a model also of how certain inductive experiences uh come to arise um and so so can i just before you go on can i just ask so in the case of like the anarchic limb how is that the comparator model gonna give us an explanation of what's going on there is it the thief the feedback that's interrupted or the the error prediction that's and so which part of it is interrupted well my my worry is that um it is exactly that it doesn't really give us the right uh verdict no good um so it would seem that in that

26:50 case you know the motor system is still uh able to output motor commands that enable the agent to interact with parts of their environment so they can still pick up the the checker they can still unbutcher whatever um and there will be a match between the forward model prediction and the sensory feedback insofar as that's what's guiding the behavior over the course of the action as it unfolds um but there's there's something missing

27:23 which is maybe the intention to perform the action uh maybe uh attempts to inhibit the action don't i don't succeed they don't succeed directly so um so what that suggests to me is that uh the comparator model is not pitched as it is at the level of motor control fine-grained motor control is not the right level at which we want to account for magenta phenomenology um and sets of authorship more more

27:56 generally so we want to talk a bit more about um intentions and beliefs right well so how did so i mean yeah i guess why wouldn't there be uh a mismatch between the feed forward and feedback well yeah so i guess what you're saying is since they're able to do it then they're going to get the the match right and so they're very accurately reaching for the cup right and um they're they're not showing impairments in terms of uh the localization of the object

28:29 [Music] um the disconnect really comes at the level of course of their intentions and their ability to suppress these movements and that's a different set of psychological mechanisms and states than than those that are um the direct uh products of the the comparator so that's that's a concern that uh the comparative model might be useful for helping us understand motor control and maybe even certain uh forms of agenda experience um that accompany a company action

29:03 performance at that level but they're not going to be the right it's not going to be the right kind of model for helping us understand agenda experience or the experience of authorship or more broadly yeah okay so yeah i've been reading so much of your stuff that i guess i internalized that argument because i was like yeah how does it explain these uh but that was exactly your point [Laughter] so what did so what did the people who endorse the comparative model how and want to use it as an explanation for phenomenology people like bane and those guys how do they see it so as

29:35 ideally working yeah so um i think here it's it's worth thinking about some of the earlier work on the comparative model and why it became uh how it became linked to a sense of agency in the first place is as we're thinking about so that the um one of the the people who's who's whose work has has uh appealed to the model the most and has done some really incredible work

30:08 with with schizophrenic individuals with the kinds of delusions that i was describing earlier is is uh chris prith um at ucl and he uh he argues that uh so uh originally right the the comparator model was was meant to um explain what seems to be a feature of our voluntary uh movements which is that there's a kind of sensory attenuation when it comes to the sensory feedback uh we get from those movements versus our versus involuntary behavior so

30:43 uh the the kind of folk psychological observation that's meant to um support this or at least point to this phenomenon is is the fact that we can't tipple ourselves so you yeah no there's no surprise elements and there's obviously you're aware of what your body's going to do but if someone else tries to deal with you then successful um so so the thought is that because the thought is that these predictions that our motor system uh automatically outputs um when there's a comparison with

31:15 uh sensory feedback and there's a match between the predicted sensory consequences what actually comes back from movement uh there's a kind of dampening or attenuating effect right so because there's a match somehow the sensation is sort of cancelled out to some degree this is this is the kind of language that people use when they're describing this um and so uh the thought was did some experiments with um schizophrenic individuals with uh delusions of control in particular and they found that um they don't really get this attenuation

31:49 effect uh so when they're when they're rating the intensity of uh some tactile consequence or even auditory uh tone they don't uh they don't have this kind of dampening effect that neurotypical individuals do and so the thought was that in order to explain um the the judgments and that they're not the authors of their actions in order to explain the kinds of beliefs that they have uh there's some some subtle impairment at the level of uh agenda experience which has to do

32:24 with this sensory attenuation right so because um they don't get this attenuated uh response their actions feel different and because their actions feel different this is all at the level of appropriate reception uh most likely um because they feel different and then they have a reason to come up with alternate explanations of their behavior so that that was kind of the idea um and uh i think yeah the idea was sort of taken up and and and amongst some

32:57 philosophers as well as as being able to maybe explain even more than just sensory sensory attenuation or lack thereof but also a special phenomenal feature of action which is yeah it's mindness or it's it's being controlled by me or whatever else so does this suggest that schizophrenic people can't tickle themselves yeah that's yeah interesting okay so then the idea would be that

33:30 um there's a mismatch so when you know you're moving the match is correct um but the in the schizophrenic or these other kinds of patients they get this mismatch and therefore they don't sort of recognize the emotion as their being their own or there's a there's a less a sense of it being their own or something yeah and i mean i think there are different ways that i've been put forward to explain what exactly is is happening so one is that's just that there's no match at all because there's no

34:01 there's no proper forward model prediction outputted um so it could be at the level of uh it could be an issue with the prediction itself it could be a an issue with the with the matching process um so um different ways to think about uh what's what's different there and what's um yeah so i think um that's that's a general idea that they're they're missing some part of this process that is typically present and sustains our sense of agency and

34:35 explains why they lack it okay yeah interesting but then so it sounds like in these other more problematic cases like with the alien anarchic hand syndrome um that you don't you don't get this kind of mismatch have the people that are interested in this in this model of uh phenomenology have they addressed this worry at all because i haven't really seen anyone talking about so what seems like a like like something they should be talking about well i think so one one thing that can be said here is that um uh if you're a fan of the comparator

35:09 model you you needn't restrict it to the level of motor control right so that's that's i think um uh a popular way of thinking about it but there have been others like like elizabeth pachari who uh who view these kinds of comparisons as being um uh a feature of action control even at the level of intentions and beliefs and feedback right so if you're if you're interested in kind of extending the model and saying well look every time we have an intention

35:42 uh we can inform maybe unconscious predictions about what the outcomes of our actions are are going to be and we use those predictions uh to evaluate whether our actions on on the right course um or not and so you can think of uh what's happening with respect to the their anarchy canned patient as a case where you know there's feedback but there was no intention and no prediction and so there's this kind of uh error signal uh with respect to to the to the behavior so i think are there are ways of extending um the model but

36:14 uh for my purposes that's that's good enough because we're now at the level that i think of intentions and beliefs which is where you wanted to push them anyway um so it's you just said that's in the case of the anarchic engine that there's no intention to move the arm is that the way it's typically thought of or is it just that the person is not they have an unconscious intention it's a good question i think that um there are reasons to resist positing unconscious

36:49 intentions in these cases and i don't have problems with unconscious intentions more generally i think i think that's that's something i think they exist and in some cases it is appropriate to posit them um in the case of the anarchic hand uh patient or any case really i mean you can think of um intentions of having a sort of special property uh that's different from that that makes them differ from uh desires for instance and that um they uh they they imply some sort of commitment to a course of action right so um i think uh

37:25 in the case of anarcho-cancer we would have to we'd have to posit uh intentions two two separate intentions committed to opposing courses of action um and i think that seems sort of unlikely i think one open possibility is that there is maybe an unconscious desire to do to perform some action that's at odds with what the agent is intending to do that's that's possible um but that still wouldn't be the relevant kind of state

37:57 to kind of trigger i think uh this sort of predictive process and yeah but i guess the reason i was wondering that is because i mean this is sort of a secondary point but it seems like if there's no intention i wonder why we call it an action at all as opposed to a behavior but these things seem like they are purposeful um uh and so if their actions then well i don't know so are there actions that more generally speaking aren't don't have intentions besides these kinds of cases would you say i think so so i i tend to think of um

38:33 actions as uh mental or bodily events over which there is some control and i think uh in order to talk about control we have to talk about goal states and we have to talk about the satisfaction of those goal states um but goal states can be intentions they can be motor commands they can be i think there's there's there are even intermediate levels of action representation that fall in between low level motor commands and and intentions if we want to think of

39:05 those in the traditional kind of propositional attitude sort of way and so uh the behaviors of the automatic uh of the anarchic hand which which i'd be comfortable calling actions uh in the sense that there is some sort of motor command that's guiding behavior in a way that satisfies its content um in that sense there is a level of control you might want to call it automatic control uh we might not want to call it intentional control i think i think there there might be reasons to

39:38 to to use to to distinguish among different types of control that we can have on our behavior um but i don't think uh actions are solely in the in the domain of intentions and and their effects but are there any cases of like neurotypical act people who perform actions in this category well well think of um uh so even the kinds of action slips we're talking about before right so habitual behaviors right that are pretty strongly tied to

40:14 to stimulate but um maybe are initiated by an intention but uh um not not every aspect of the of the action is going to be intentional right so they're going to be some aspects so you intended to grab something but not the orange juice that sort of thing yeah or or you might just find yourself um you know walk by a light switch and there's there's such a strong uh habitual link with with flipping it you just find yourself flipping it at some point right so there's there's times when i mean i made it sound like i'm i'm just walking around

40:48 my house hopefully i'm not but um but there are cases where we have like really strongly stimulus driven kind of reactions uh and i think the those those might be cases where there is there again there are motor commands or certain goal states that are getting satisfied but we wouldn't call them uh different behaviors that are different by intention yeah but so yeah i guess i don't know why i have this feeling but uh i guess i do have it that those are like not really actions um

41:20 i don't think i really have an argument or a good reason for thinking this i guess that's just my gut reaction but i understand what you're saying there there's a motor command it satisfies you know it's there's some some goal that it's trying to satisfy so why not call them actions um yeah and you can think of non-human animals as well um if you're someone who who's not comfortable uh attributing intentions uh to non-human animals um but you think there's reason for for viewing them as agents yeah yeah

41:52 well let me put it let's so let me try this so what in on your view is an intention then um like how does it differ from these things that we're talking about yeah so i i mean i have a little bit of a uh when it comes to intentions i have a fairly traditional view so i do i do treat them as propositional attitude space or often the terminators of practical reasoning although they need to be um they uh yeah they interact with our beliefs and our desires um they interface with their motor

42:25 system in in some way uh so i i tend to think of them in what i think is a fairly uh traditional uh way but but i also think that i mean classical action theory um for a long time the story sort of stopped that intention so once we figured out how practical reasoning works and how we come to form intentions uh sometimes there's talk about intentions matching up with actions or their consequences but in terms of psychological representations uh uh not a lot there's a lot of silence and

42:59 you know let's pass that on to the neuroscientists or the psychologists um and one thing that i think has been really interesting about the the trajectory of action theory in the last 20 years or so um in part because of people like mark shawn rowe and elizabeth cheri um is that there's now more emphasis on action representations that that exist downstream of intention and guide behavior and i think really important to kind of figure out what the format of those i mean there's a whole

43:31 whole set of questions that opens up once we allow that these are um these are these are phenomena that are in the domain that you know that philosophers should care about too right if we want a full story of how action control works um it can't stop it at intention and there are there are folks that recognize this so harry frankfurt among others but the idea is that look we have these cases of causal deviance and an action theory right so cases where it seems as though some action matches the agent's intention or desire belief desire pair but it's not what we would want to call either an intentional action or

44:05 an action at all so um but but if we if we start to fill in the story a little bit um what's happening downstream of intention how does a tension as a mechanism does a lot of work on this area to um help fill in the story then we start to have uh there seem to be fewer cases where we we want to say that a causal theory can't explain what we're trying to explain right so so the motor commands the outputs of the motor command system then they're not propositional i would say that

44:39 um the that motor commands are a non-propositional uh state i think that um they have they have a certain uh um so the way i like to think about them is having a certain kind of uh structure right so they they're able to uh they they uh determine both the spatial and the temporal kind of parameters of some

45:10 action type and they have uh open parameters for things like force and and velocity and other qualities that the action might might have and i think that the story we tell for uh the content of motor representations is going to be different than the story that we tell for the content of intention then and one clue to that is they do seem to just play different much different functional roles um for the agent yeah so yeah that's interesting i mean so they are going to have some content but what you're saying is that they

45:42 i mean do they like is the content truth valuable uh i think it seems like the whole story about comparator stuff it kind of depends on it being something which could be accurate or inaccurate which sounds like already kind of semi-propositional if not fully propositional to me i think it's satisfiable but i don't think um i mean i'm not even sure it makes sense to to talk about intentions as estates that are uh accurate or inaccurate versus satisfiable

46:15 uh-huh yeah okay i mean okay so i mean i don't want to bring up the the famous l word but i guess i will so there's the libid experiments um which uh you know gets maligned and everyone has a has a pet theory about the lipid stuff but one way of like taking the livid stuff and i know you have a particular line on this so i want to get your response to this but one way of thinking about this is that what you know you have the readiness potential and you know all the details of the living stuff but that what you're basically finding out is that the un the unconscious and tension occurs prior to their awareness

46:49 of the attention um and so but it sounds like the way that you're thinking about this that would be kind of a mistake to take that to be the the results of the limit because the intention is going to be downstream psychologically speaking or something like that and this is you know more in line with the kind of motor commands uh sort of thing you're talking about um i'm not sure so i think uh one way to think about

47:21 one way to one kind of uh one consideration to use as a guide to determining when you want to be positive an intention versus a lower level state like a motor representation um one has to do with the complexity of the action right so uh i can have an you know an intention to cook pasta for dinner i'm not gonna have a motor representation that's capable of carrying that kind of content so yeah other representations are what representations are going to carry

47:54 content about um bodily parameters um and you know ways in which we can execute uh behavior um but uh another and so when it comes to the the limit task uh you might think well that that sounds like a mode of representation it's just very simple either if you're flexing the wrist or something like that um another way though is is to think about intentions as uh as as and now immediately talks talks in these

48:26 ways as uh you know um spontaneous acts of decision formation right so in the case of uncertainty i decide to do x rather than than what i say and uh in the case of the limit task because it's set up so that the agent has to make this decision right so they're supposed to spontaneously decide when to move um you might think that they're forming an intention to move now and the intention is going to trigger a motor representation this very simple representation of the action to be to be

49:00 performed um but you might still think that when when they they decide to act at the time that you decide to act it's an intention that's driving that behavior yes it's preceded by a prior intention to engage in the task to begin with so you can you can still posit that earlier uh intention um but there might be on those grounds some reason to think that the intention to move that the decision to move is itself an intention so so you said there's you could think of it in both those ways

49:32 um which way do you think of it uh i i think of it as as an intention okay you do yeah okay i mean i still think that there there might be reason to think of um the the redness potential is not yet the the early phases of the the activity that we connect with the readiness potential as not being an intention but being something like a desire to move before there's a commitment right i mean it's an interesting question how

50:06 so one because one of the reasons i was asking my earlier question was about you know the level the distinction between psychology on the one hand and neuroscience on the other hand and the the worry that we have about implementation so we want to posit a bunch of like psychological states and then you go and you look at what the brain is doing and there's a question about well are these just implementations of those psychological states or are these separate things so um yeah it's it's well anyway so i'm not i'm not really sure like what in the story about the comparator model

50:38 is really how if how it would interface with the intentional story um or how intentions as psychological states generally are going to be able to interface with motor commands representations given that one's proposition on the other is not so so how does how does it trigger it or um how does it generate the right kind of state i guess sort of similar to something like you know that wayne wu's talk about this uh the mini mini problem how does our our intentions connect up with the many kinds of motor representations that we have um do do you have any thoughts about that or is this

51:11 mega yeah on the so-called interface problem yeah elizabeth peshari uh ellen friedman's done some good work on this josh sheppard daniel bernstein there's lots of people who have sort of weighed in on on this on this debate and originally uh um steve butterfield synagogue uh kind of coined coined that that that that term for the problem um the name of the problem um but the idea is that um i mean really it's just a problem so

51:44 if some people kind of are unhappy with uh with the ways that the problem is sometimes formulated so um uh one way in which is important is to say well intentions are propositional attitudes and motor representations are not and so we we need some explanation of how these two interface because it looks like we're dealing with um you know these different formats and so in some way metaphorically you think of it as like they're speaking different languages so how is it that they

52:16 were meaning not non-accidentally um if if one doesn't like that way of putting it because it presupposes uh things about the format of intentions that maybe we don't want to presuppose and maybe the formative motor representations we can just think of it as a more general question of how how is there how is action control possible to begin with right so how do we get from intentions uh to behavior and what are the intervening representations if if any um and so i mean one way uh to

52:49 to respond is to um and this is this is the the way that elizabeth and pachari and i have argued is to is to posit these intermediate level representations that exist between intentions and uh low level low-level motor representations um that uh uh are known in motor control literature as motor schemas and motor schemas uh are more abstract than motor uh

53:21 commands um with low-level motor representations they they represent the uh invariant features of action types so um everything uh that all throwing the spatial and temporal features that all throwing movements for instance have in common um and they uh allow a hookup between um intentions and the motor system because um possession of a motor schema for some action type entails that you have

53:53 what we call executable action concept so right that means that you have a kind of representation that enables you to execute an action it's not the same as the kinds of action concepts that we might acquire uh on the basis of observing someone performing some action um it's it's uh a type of concept uh that can only be acquired on the basis of execution and so you might think that um intentions have as part of their content they have conceptual content and that that links up to these uh these are executable action concepts that are deployed in the

54:26 content of perception and that helps us that links up to the motor schema which then uh triggers the formation of appropriate motor commands so um we view these representations as kind of a building helping to build a bridge between this propositional level conceptually you know lay in states um and then these really low level motor representations that uh specify like detailed features of each angle trajectory and uh force and and and all these other parameters um and paying attention to this

54:59 intermediate level i think opens up uh ways of thinking about action control that yeah and i now just remember that uh i did read that paper with you uh and elizabeth cashford so i should have known that you had something to say about this but so i'm confused about the idea of a expedial executable action concept so this is the um this would be a concept in the intention or is it in the schema or where is it in the intermediate area or at the higher level i mean we leave it sort of sort of loose

55:32 and in the paper we just say that you know one condition on if you have the motor the motor schema then you possess the executable action concept we don't go so far to identify the motor schema with with the concept um but but we do suggest that in the content of a proximal intention uh executable action concepts are are deployed right so um you might think that perceptual judgments are special in terms of the kinds of perceptual concept that they that they involve in their content

56:04 um because they're so closely tied to the sensory system and we're positing a kind of analog and the case of intention and special kinds of action concepts so so to take like a concrete example if i want to reach from this pen over here i did it um so i the concept that's executable having that concept entails that i have the schema which allows me to generate the right motor command is that the right way of putting it yeah so if you if you have i mean the

56:38 this the acquisition of of of the maybe the schema direction secures your your possession of the concept okay and how so i guess what i what's confused so i'm worried i don't know i don't understand the interface deposited interface between the two levels here so how does the intermediary uh schema translate between the two languages so to speak between the intention and the motor command what so what would be the

57:10 content of the schema so the the content of the schema i mean it's um it's an action uh type and it uh it can the schema can possess a certain kind of uh structure in in in so far as it's specifying uh a spatial temporal sequence of steps that'll give rise to the action um and those steps have to go in a certain order and so there's there's it's important what the what the structure of the schema

57:43 looks like and it can go wrong in various cases and there's some evidence that skilled agents have um more complex structure to their and more accurate structure to their their motor schema than novices for instance um so uh the the thought is that um the diesel so that the schema leaves open various parameters so for instance i can reach for my for my cup and the reaching movement will have uh that's an action type um that can be specified by a motor schema but

58:18 there will be certain parameters left open with respect to how much force i i use to grip the cup handle so if it's a fragile cup versus a sturdier cup um how quickly i reach towards it and so on which might be um might change from a circumstance of circumstance now how am i going to determine as an agent uh or how is it going to be determined by the motor system how those parameters get filled in that's going to be determined on the basis of our perceptual link to our environment so you know not all

58:50 background knowledge i might have about the club um about whether how far away it is and and so on and also my attentional mechanisms so in this kind of coordinated way using the schema and using my intention as a starting point so this is similar to how when who talks about the intention attention nexus right so you have an intention to reach for the cup and um perceptual and intentional resources are deployed then on our view to fill in the motor schema which the intention also triggers um which is then going to

59:23 uh result in a detailed motor command of exactly the action to be to be performed that's that's the idea yeah and i mean you might you might think we're just creating more interface problems because yeah that's what i was going to say um but you know you don't think that so why wouldn't why shouldn't we think that well because i think that once you have uh what we now have is uh a matching control structure with an intention attentional and perceptual mechanisms an

59:56 intermediate representation and a low-level mode of representation and then it starts not to be as much of a mystery how how they can coordinate right so again if we if we if we kind of resist the language of well the metaphor i think they're talking different languages right so how can how can we translate between them um what we're basically saying is we don't need a translation process where what we need is uh structures that have been associated enough times in the past right to build up these robust connections among them

1:00:28 and again it's a whole control system it's not just an intention and just that it's intention plus attention plus perception plus uh intermediate action representations that we're calling schemas plus the low-level stuff and um once you have that coordination among all these different moving parts um then i mean our sense is that it starts to be less of a question of how how do we get from intention all the way to this low level stuff well you have this whole structure in place that's how yeah okay and so can can you trigger the action schema

1:01:02 without the intention yeah so that that might be these cases of um you know strongly driven behaviors right okay good yeah you might think if we wanted to to import this language to some of the stuff we're talking about earlier you might think that in the case of another can what are being activated are motor schemas which can be fairly complex right so it's not just low-level motor representations but motor schemas representing certain action types that guide the behavior um and it explains why it can still be goal-oriented and propulsive without the presence

1:01:36 if that's right of an intention and and is it you could also explain certain kind of agnosias right sort of like uh but also i guess like uncons like what it would like blind type type uh actions like in the case of df who's got you know got good um slot posting acts you know she can she can take the piece of paper and orient it in the right way but she's she can't like draw the slot angle or anything like that is it is this going to help us in those cases too i kind of want to say i don't know what your your your take is on on the case of df everyone has their

1:02:09 their favorite um kind of interpretation but i think i think she's operating on the basis of of intentions at least in some of the tasks that we're most familiar with right so in the um and the the mail slot task where she's posting uh a letter like object um she she intends to do that right that's that's the goal of her task that's set up to set up for her and then that intention again is going to trigger these downstream action representations um it's it's true i think that at the level of motor schemas we're probably dealing with dorsal stream vision uh so i talked about um you know

1:02:43 parameterizing those schemas filling in some of those details i think dorsal stream vision is going to play um a central role in that process rather than ventral stream vision which which is going to be probably what at the level of proximal intentions the agent uses to fix the target of their of their action and maybe even in general terms um an action type um but um so so i'm you know might help us understand what's going on in the case of of uh df only in so far as uh

1:03:18 she in terms of the visual contribution she seems to be it's just that the level of motor schemas that that that vision is going to be contributing right i mean yeah it is kind of interesting because yeah now that i'm hearing you say that yeah it makes sense because in a lot of blind site type cases they still have this sense of like source of self as source i guess they so it depends the blind side so that so in her case a visual form agnosi i don't i don't think there's a disruption of

1:03:49 her sense of agency that i've heard right but but when you're talking about the blind side cases um i'm thinking of that like the dramatic case like tn you know he walks down the hallway um he still seems like he's feels like he's the source of the action she just has a weird explanation for why he's doing what he's doing right yeah i think that's right and and in a way that it's that a sense of um i'm thinking of of the you know blind side in the

1:04:21 perceptual case right where there's a sense of guessing or lack of confidence that's an interesting question i yeah i've not heard reports that that that cases like tns there's there's any sort of degradation of the sense of agency evolved in the action and what about the uh the reverse kind of uh case where you know that you have the the case of the guy who's like looking at the light switch and he says i i know this thing turns on the light but i can't figure out how to do it uh like what are you supposed to do to it

1:04:54 um so in those kinds of cases what you're what you're saying is that he can't the sort of action scheme is not activatable or something like that yeah that's an interesting um i mean in that case i guess the the damage that if we're thinking like of optic ataxia so it's it's really that they can't uh one one story or one hypothesis would be and i don't i don't feel uh this is above my my pig right now but but the

1:05:26 one hypothesis would be to explore um it's just that the the motor scheme i can't get properly parameterized because because the dorsal vision isn't um there's no contribution from dorsals from vision right um but you're raising an interesting question about whether whether the fact that there's this robust damage to to that interaction between dorsal stream vision and the parameterization of schemas might mean that the executable action concepts if we're going to go with that story are also

1:05:58 somehow missing right so they're unable to determine what to do with an object um and then and and one prediction might be that they've literally lost that concept of uh turning off the light switch right yeah i guess you'd have to check in yes okay all right so i i noticed we're getting on in time here and i i don't i want to make sure we have time to talk about the cognitive theory that you have so um i think it's very interesting that uh so you know if

1:06:31 i found myself like i was saying pretty much convinced by by reading all of all of your stuff about the at least two things so one is that it doesn't make sense to try to explain the agenda phenomenology in terms of uh other kinds of phenomenology seems something um proprietary to it and then also the comparative model the paratora model sorry has at least you got to do a lot more work to make it compatible with these sorts of things um so that leaves us with these other high level things like we've been talking about intentions and beliefs and so on so can you tell us can you can you flesh

1:07:04 out the story a little bit more so in your view what gives rise to the sense of uh self is source yeah so the the story is embarrassingly simple it's um but it's uh um a first-person thought uh to the effect that uh i am aiming where aing is you know is a variable that stands in for an action type that's inherited from the proximal intention right so i think that um when we're performing actions for which

1:07:39 we have some positive sense of authorship or experience of authorship uh there's a proximal intention that um triggers one of these first-person thoughts and um here i sort of wheel in some some uh some work from higher order thought theory right so the idea is that um just i've heard of these things yeah uh just as we might have so you know a higher thought uh with the content i

1:08:14 see red gives rise to some phenomenology um module whatever anything uh whatever story you might have in mind on your uh horror view uh which is gonna be maybe slightly different but um uh the idea is that you have uh similarly you give a first person thought to the fact that you're performing some action right makes you aware of yourself as performing some action not as being in a mental state but as performing some action and that is on its own sufficient to

1:08:48 give rise to a certain kind of phenomenology um so yeah well go ahead finish it yeah so so the idea is really just that um in general terms we have this question of where where in our cognitive architecture to to locate digentive experiences um i want to say that they don't seem like they're the right kinds of things to be sensory experiences um you might think that they're features of experiential or phenomenal features of

1:09:20 the intention themselves uh i think that intentions don't quite have the right direction of fit um and the mental attitude just isn't isn't right for for capturing a master torque what i think is an esoteric form of awareness i'm aware that i'm performing some action i'm not um i'm not in a state uh and intentions are such that they um you know they they drive some action they there's a they don't assert that an action is being performed um and so if that's right um and if we have a kind of background

1:09:56 story which i think is is is is reasonable to to to buy into which is that um look every time we we perceive something or at least frequently we automatically form these perceptual judgments about her perceptual belief say about what's in our environment um you could say that every time we form an intention there's also kind of automatic uh thought to the effect that i'm performing um some action uh that's that's just triggered as a matter of of um of strong association built up over time

1:10:29 and uh and then if we also um i'm asking people to buy buy a lot of things here but if we also buy into the the idea of how how we get phenomenology from high order states in this case not a not a higher order state that's about another mental state but about an action um then then the story comes together so that's yeah i would say yeah i guess that's at the beginning you said this would help be part of the explanation for how thin the phenomenology was but you know the higher the theorist wants to say no we

1:11:02 get regular phenomenology too from this uh but that that's neither here another but one question i do have about this is in what sense the kind of thoughts that you're talking about are higher order um and you sort of said yourself they're not really thoughts about mental states um their thoughts about actions and those don't seem like that seems like a kind of first order thought to me not that it would really count against your view or anything but it does seem like it's a thought about uh yeah and and i've gone back and forth about whether to call them uh higher order they're certainly not however in the sense of

1:11:33 being directed at mental representation um there's some wiggle room here so you know if you think actions are like cyril um let's argue that their intentions plus uh bodily movements or mental events then then you might think that that having a thought directed towards an action is having a thought directed towards a mental state just not described as as such in the context of the thought um so in that sense they might be higher order um i think probably

1:12:07 it doesn't matter for for the for the view uh i mean what what it what if they really are first order thoughts it might just turn out to be the case that um first order thoughts directed towards actions uh give rise to phenomenology the same way that higher order talks directed towards mental states too right yeah no i didn't think it was really consequential for your theory i just think it's a terminological issue mostly um but but i do wonder about like in the going back to the antarctic hand phenomenon which we've been talking about

1:12:38 previously so it seems like on your you're committed to them not having these kind of thoughts or at least if they do they they've got to have it in the wrong way um because i mean you might think that when the hand reaches for the checker they are having a thought like oh i'm reaching for that checker uh but maybe they arrive at it by inference or something maybe part of the story i don't know so like how do you right so so it's not going to be a first-person awareness from the inside of the action they're performing they might have that for for the action that i mean that they're performing intentionally

1:13:10 right the the that the intact um so to speak uh limb uh performs um but uh if if my interpretation of what's going on maybe some of these cases is correct then they don't have a proximal intention so they're not going to have a first-person agent of attach to the effect that they're performing some action um from the inside or not seemingly non-inferentially right so uh it matters that they come uh they come about to these uh self-attributions um by way of observation rather than seemingly subjectively immediate

1:13:44 right so that's sort of similar to the way like on the hierarchy it's like these thoughts have to be arrived at from a seemingly non-inferential automatic and so forth and so on then you get the phenomenology the ideology matters in this case as well exactly and these thoughts are typically unconscious on your view typically unconscious um i think you know they're they can be conscious and that's when we we might come to have uh some awareness of ourselves as as as consciously

1:14:20 performing some action i think that things got a little bit murky here so i'm not entirely sure that in introspection we're aware of much more than maybe a thought to the effect that we're performing some action and then some sensory experience that accompanies the action itself um uh i think one odd prediction of the view and i have some things to say uh to make this sound less hot perhaps but um for automatic behaviors that might not be accompanied by so i've been describing all these other

1:14:53 right these intermediate levels of action representation that our downstream of intention i'm hooking up agent of thoughts to intentions now what about all the behaviors that aren't initiated by way of of an intention but that we don't we don't disown them right so it's not as though we have a sense of alienation from them um i do think it's true though that we don't have the same uh sense of of authorship we might have a sense of controlling the behavior in terms of the fine green details and the same way that

1:15:25 the anarchy can can be directed towards an object um but i don't think we have the same phenomenology for those automatic habitual behaviors so when you're reaching for the orange juice well that might be a different case though because you do have a you do have the intention to reach for something so you're still going to fail like you own it or something like that um but there can be cases what you're saying is there can be cases of automatic behaviors um you know where this happens to me all the time where

1:15:59 i'm like okay i want to get something out of this cupboard so i shouldn't shut the cupboard and then i just automatically shut the cupboard and i'm like oh why did i do that with my light switch case thank you yeah you're shoving clubbers all over the place uh yeah so so that kind of so you might have you might have some sense of authorship for you know getting getting the thing out of the cupboard but not for the um reaching out and shutting it automatically that that would be the idea i mean i think i mentioned a little bit uh earlier that there are other facets of

1:16:32 the experience of agency and i think that these other facets um exist at the level of motor control and that there's some really interesting work in metacognition of action that can tell us about um a sense of control in your behaviorism so it's kind of smooth execution of behavior that's different on my view than the sense of authorship that's a separate feature um so we might still have a sense of smoothly executing some behavior uh even if it's not even if we don't have that sense of authorship

1:17:07 and is it your is are you do you feel the pull to want to give a different account for those kinds of experiences so they wouldn't just be cognitive maybe those would be a metacognitive or sensory or something that's right so i i think of those as the outputs of a metacognitive evaluation so i think that you can think of and here's actually where uh this is this is quite friendly to a comparator model approach so the idea is that um we have

1:17:42 metacognitive monitoring mechanisms that evaluate the success of our of our ongoing actions and um one way they might do that is by way of this comparison between predicted and actual sensory consequences and um you might get uh metacognitive feelings of control that are the results of these comparisons and so that's going to be something much more low level and having more to do with action execution

1:18:14 whether the action is going well or whether it's going poorly so it has this kind of evaluative uh feature built in rather than action attribution which is what we're trying to explain in the other in the other case so um yeah so here's where uh um i i have uh i mean and there's something actually quite friendly for the higher order theories as well because these are these are not thoughts right and they're not perceptions they're metacognitive states they're they are meta representations though

1:18:47 that represent the success of some action outcome yeah okay so so for the like you guys list like six different things in in the paper with uh shepherd so maybe you're going to have a kind of different account for several of these maybe some of them can be grouped together maybe others need a different account or something like that yeah and i think i think it's important to be um specific about which which facet of the experience of agency one is interested in explaining because as you said different accounts are going to be

1:19:19 uh probably on offer for different uh features of the phenomena i don't think it's one kind of thing it's many people right so one one thing i was wondering about um and probably you've already addressed this somewhere uh is what they'll find what level of um i don't know how to put this what level of grain or uh coarseness the concepts and the kind of thoughts you posit are so i am aiming um okay so if i'm reaching for something is that that i'm reaching is it that i'm moving my arm is it that these are going to give different

1:19:53 experiences um if you have one but not the other kind of concept or so i'm just curious like and then secondly what the relation to the concepts and these kind of thoughts are to the executable ones we were talking about before yeah um these are good questions so i think that the way i've been thinking about them is is is fairly poor screening concepts relative to the contents that our you know our motorcycles have so i think um again they're gonna they're gonna borrow the content uh so to speak um from from the proximal intention

1:20:26 and um i think uh uh so that might be a description of the action in terms of the bodily movements that are required to perform it uh so you you might think in cases where you're thinking really hard about what you're what you're doing you want to be really careful to move your body in just the right way that the agenda phenomenology is going to pertain to the bodily movements themselves but you might think in other cases if you just have an intention to get the orange juice out of the fridge and

1:20:59 that's your proximal intention it gives rise to the action sequence that you that you then perform um your your agenda phenomenology is going to be quite um quite general quite loose in terms of the content um so uh it's gonna matter uh just what's in the content of the intention uh whether the agenda of thoughts themselves deploy executable action concepts that's uh that's a good question i mean i'd be inclined um i guess i'm sort of committed to the

1:21:31 view that they do because i'm talking about them as inheriting the content yeah the borrowing is what made me think of it right yeah um then i have to think about whether that's something i want to be committed to but yeah that's i think that would be that would be the story there yeah um so let's see because i i'm running i i hate to be rushing like this because it's a lot of things i want to ask but i have another commitment after this you already know but for anyone listening i i have to

1:22:04 normally i'd be torturing you for much longer in your toe but i have a lot of things that i find very interesting about your approach but so i wonder because one of what i'm thinking of is if you especially with the language about higher order thoughts so one of the things that comes from there is that the content of the hierarchy determines what it's like for you um so you know if i am seeing red uh there's the i am part and then what it's like for you is the seeing red you know i am hearing a sound there's the i am part which attributes it to you and then the seeing uh hearing a sound

1:22:38 or whatever but so is this is it on your view the case that that's also true so when you say i am reaching is the experience not just that you are the source of it but does it feel like reaching and if you were like i am you know grabbing that's a different kind of experience because of the different contents or is that not the way you think about this i think it's going to be quite quite thin right so you're not because you're not aware of yourself as as um being in the sensory state for the company or

1:23:12 actions that's not what i take to be the content of the if we're thinking of executable executable action concepts they're not yet they're they're not in sensory format that's just not the kind of state uh the kind of representation it is um so what i would say is that um in some ways the most the the most the way in which the the agent of thoughts come to shape our experiences um is is maybe primarily through modulating our

1:23:48 sensory experiences right by um by the sensory experience is now shaped by the indent of thought to the effect that i'm aiming versus being um so i don't think that um the agent of thought uh on its i mean i think the agenda of thought on its own is going to give rise to some phenomenology to the effect that you're that you're aiming but i think in terms of again the salience of the experience it's gonna it's gonna come from um the way that your sensory experience is shaped by way of having

1:24:21 that that soft yeah okay um and so if you you have these cases where you um you know you can stimulate the the parental cortex or the front are the um so much sensory cortex or the motor cortex and you get people saying things like you know i have this urge to move but they they don't move or they'll even if you stimulate it hard enough strong enough say i just moved um when they didn't move at all

1:24:52 so on on your view then is that a triggering of one of these kinds of thoughts or is this this because they do have the sense that there was a movement even when they like i mean you had to hide their arm from them but they clearly didn't move um so it does seem like that would be your account but also that you kind of have to predict that there can be these kinds of um like if the thought is triggered then you have the sense of moving even if there were no movement involved yeah so i think those kinds of cases i mean it's you know it is hard to interpret um

1:25:26 to go from just the the what we know about brain stimulation to exactly what kinds of representations have been have been triggered especially we're talking about psychological terms um but uh i think a promising way of interpreting those kinds of cases is is where you have agenda phenomenology but you don't have sensory um activation of the kind that you would normally have in the case of the movement um i also talk about uh libit always always comes up but you know there's um in his work in

1:26:01 addition to the the judgment of of when to move your patients or individuals participants i should say are asked to uh report the time in which they have moved um and patrick haggard has done some lots of follow-up work on this kind of in this kind of paradigm um but the the awareness of having moved uh typically precedes um the ex the uh first burst of muscle activity so so basically before there's any sensory feedback from the action um people seem to have this kind of

1:26:34 anticipatory awareness of that they have moved so uh that combined with maybe with with some of this um uh neurostimulation work we're see people seem to have just on the basis of stimulation and parts of the motor cortex experiences of having moved might seem to suggest that um we don't have we don't need sensory states right to explain the phenomenology of the experience of authorship so that's yeah but also so part of what i was wondering about was like where neurologically neural and not anatomically sorry

1:27:07 uh you're going to locate these thoughts um so it sounds like because you know thoughts i would associate i don't know we don't know enough about thoughts to say where any of them are but the prefrontal cortex is often like thought of as a possible candidate yeah um and but maybe the parietal cortex or the insulin i don't know so people will go back and forth but do you have any guesses about where these where where you could disrupt these sorts of thoughts if you wanted to um i don't know if i want to venture

1:27:40 venture i guess that's why location i think it's a good question that i don't think we we maybe know enough about yet in terms of um but i mean i would say in case of these uh these neurostimulation experiments um uh the story i would tell but what's going on there is that there is at least an intention being triggered and want to put intentions um somewhere in the motor cortex whether it's supplementary motor air whether it's you know earlier so again i don't want to say

1:28:12 i don't want to commit to a particular uh neural location but um that would be the idea that first there's a proximal intention triggered and then a thought to the effect that i'm performing some action yeah and then so it also seems like you're committed to this prediction of a kind of illusion um and i'm not sure i wonder if there are examples of this kind of illusion so where you are actually doing a but it seems to you like you're doing b um you have a kind of mismatch between

1:28:46 the agent of thoughts and what actual action you are performing so you know you may be reaching for the orange juice but have the thoughts i am jumping or i don't know that's an extreme case or something like that but that seems like you are committed to some that in principle at least being possible and i wonder if there are any cases that can be described that way um so i mean i think i'm trying to think of cases that don't involve this kind of brain stimulation um right i should say that uh

1:29:22 we wouldn't really expect in the wild um to to find many of these cases because uh typically um you know our proximal intentions are again tightly integrated with this interface story that um that i'm telling is accurate they're really tightly integrated with with our motor system and by way of this set of representations and attentional mechanisms so again there's kind of this coordinated system and um i i've i've said that uh agent of thoughts are kind of automatically triggered on the basis of

1:29:54 approximate intention as well so the proximal intention is both um leading to this cascade of downstream effects and the actual action execution and the agent of thoughts so we would expect that they would typically be typically aligned might match and so uh the kinds of cases where there's there's misalignment or these weird ones where you have you know brain stimulation so you don't get the um i mean there there have been experiments done that i have i have uh my views about um whether they

1:30:25 they actually offer us uh support for the things that they claim but um this will work in the past by uh the psychologist and dan wagner and the early 2000s really an influential work where uh he would he would set people up in various experimental situations where they would uh they would report they would have a sense of uh controlling some movement that they weren't in fact performing uh or of engaging in some some having some goal or some intention

1:30:58 to do something that they didn't in fact do again i think those are weird cases right so you know if you distort feedback or if you you set up a confederate and it's all sorts of um is this like the ouija board case is that one of the cases uh the the movement of a mouse so it's even it's even one step for move it's not it's not their bodily agency it's their control over a cursor on a screen right and if they're primed in certain

1:31:30 ways then they they rate on a likert scale um a sense of uh higher sense of intending uh to perform the action um to move the cursor to a certain location so yeah these are it's it's hard to know you know exactly what to make of those results so so maybe uh maybe what they're showing is um that we can attribute intentions to ourselves uh under certain circumstances that based on inferences that we make about the setting that we're in

1:32:03 um i'm not sure that they they tell us anything about whether a sense of agency or sense of authorship was actually um instantiated in in these kinds of cases uh typically the likert scale ratings tend to hover around the midway mark so it's not a super strong right of having intended to do something so um so i'm not sure what to make of those kinds of classic uh cases in terms of giving us um some evidence for a robust association between a sense of agency or a sense of authorship and then actually performing the action

1:32:37 um i would but i would yeah i mean again i would i would suspect very few cases where you actually have a sense of uh of of aim where you're where there's a a dramatic mismatch of course there are cases where our our actions just go wrong right so we just say can we find out through observation that something's gone wrong but that's that's yes uh us klutzy people are really familiar with uh this sense i have that happen to me all the time um

1:33:09 so finally i want to ask about the schizophrenia case so i want because so there's a case where they have a sense of someone else controlling their their movements or whatever um and so i was curious on your view is uh is is this just like okay so i guess there'd be two different ways of thinking about one is like they uh they don't have the agent of thought and so the the there's the motion and it doesn't feel like there so by inference then they think someone else is doing it um but then there's another way of thinking about it where they have the experience like

1:33:40 someone else is moving my arm not my by inference just like oh i see the arm moving i didn't move it so someone wants to do it but they sort of like experience the other like in a thought insertion sort of way they're like this this other person is literally controlling my arm um so i'm wondering if your view is committed to a certain interpretation of those kinds of cases like more towards the first one than the second one or if or what you think in general about uh yeah i mean i'm not entirely sure what to say but these cases i think there's a lot we don't we don't know about what's going on

1:34:12 um with these these delusions and i'm certainly no expert myself in terms of having worked with any of these individuals but i i think that um one thing to say is that uh i mean my view would seem to require that what's what's happening in these cases is that um there's uh either yeah so for some reason the the connection i'm seeing between the the proximal intention and the agent of

1:34:44 thought is um is severed and replaced with uh so so you might think okay here's a story you might tell you might you might you might say that um the mechanism by way of which we output these agents of thoughts requires some awareness of the of the intention uh so there's some awareness of the intention and then on the basis of so the idea would be we go through some sort of unconscious inference so i i intend to a um in the past every time i intend to a i a uh it's proximal intention therefore i'm aiming right something

1:35:16 like that um directly speaking might might describe the inferential process that we go through automatically unconsciously and you might think that um some step in that process is missing for the schizophrenic individual so they don't have they might not be drawing they may not have awareness of the the proximal intention uh there might be some reason why the inference is blocked to the effect that they're performing in action maybe the inference goes through but um uh the the thought is doesn't uh drive their verbal report behavior for for

1:35:50 some reason so there's lots of um there are lots of uh points in the story which you something might have gone wrong um yeah that's okay well i'm really out of time now so i do want to ask just finally though if there was anything what you were hoping would come up but hasn't yet come up and the answer can be no but i always like to check um i mean i think i think we've covered covered everything that i wanted i was hoping we would be good

1:36:22 great job you're an thank you i feel like i just bombard you with a bunch of naive questions i really have to say i admire your work i i think it's we need people like you at the interface of philosophy cognitive science thinking sophisticatedly about these sorts of problems i think that you're doing great work um i'll put links to your papers and stuff underneath the end of this or whatever but uh yeah i hope people go and check your work out um but thanks again mirto and uh just hang out for a second i'll say bye fair here thanks so much richard i really appreciate it okay all right so let me am i ending

1:36:55 this now how do i ah yes philosophy versus technology