Autism as gradual sensorimotor difference

Autism research is increasingly moving to a view centred around sensorimotor atypicalities instead of traditional, ethically problematical, views predicated on social-cognitive deficits. We explore how an enactivist approach to autism illuminates how social differences, stereotypically associated with autism, arise from such sensorimotor atypicalities. Indeed, in a state space description, this can be taken as a skewing of sensorimotor variables that influences social interaction and so also enculturation and habituation. We argue that this construal leads to autism being treated on a par with other sensorimotor atypicalities such as blindness or atypical height. This leads to our conclusion that, insofar there is an ethical call to inclusion in our public sphere regardless of contingent bodily difference, an enactivist take on autism naturally leads to extending such inclusion to autism. Moreover, our analysis suggests a concrete way forward to achieve inclusion of autistics: by being more attentive to autistic sensorimotor specifics.

Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Autism as gradual sensorimotor difference From enactivism to ethical inclusion Authors: Thomas van Es1*, Jo Bervoets2,3* Affiliation: 1 Centre of Philosophical Psychology, University of Antwerp, Belgium 2 NeuroEpigenEthics, Department of Philosophy, University of Antwerp, Belgium 3 Leuven Autism Research Unit, Department of Health Sciences, Katholieke Universiteit Leuven, Belgium * Corresponding authors (both authors contributed equally to this project) Abstract Autism research is increasingly moving to a view centred around sensorimotor atypicalities instead of traditional, ethically problematical, views predicated on social-cognitive deficits. We explore how an enactivist approach to autism illuminates how social differences, stereotypically associated with autism, arise from such sensorimotor atypicalities. Indeed, in a state space description, this can be taken as a skewing of sensorimotor variables that influences social interaction and so also enculturation and habituation. We argue that this construal leads to autism being treated on a par with other sensorimotor atypicalities such as blindness or atypical height. This leads to our conclusion that, insofar there is an ethical call to inclusion in our public sphere regardless of contingent bodily difference, an enactivist take on autism naturally leads to extending such inclusion to autism. Moreover, our analysis suggests a concrete way forward to achieve inclusion of autistics: by being more attentive to autistic sensorimotor specifics. Key words: enactivism, autism, normativity, inclusion, sensorimotor, neurodiversity 1 Introduction Autism has been intensively studied in the cognitive sciences. Initially, autism has therein been primarily conceived of as a social-cognitive deficit. The most prominent suggestion is the (lack of) Theory of Mind hypothesis (Baron-Cohen 2000), but other cognitive theories such as Weak Central Coherence (Happé and Frith 2006) and (deficits of) Executive Functioning (Frith 1996) are part of such deficit perspectives. Theories positing a social-cognitive deficit in autism have 1 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. been criticized for unwarrantedly positing cognitive modules to perform representational calculus in the brain (Hipolito et al. 2020), as well as for their ethical implications (McGeer 2007; Bervoets and Hens 2020). Severe criticism of autism as a deficit to social-cognitive functioning also comes from the neurodiversity movement. Scholars like Milton (2017) and Chapman (2020) argue that autism can be seen as a positive practical identity and should not be seen as a flaw in autistic individuals1 but rather as a mismatch between autistic individuals and their environment. In this context, recent autism research increasingly views autism as, fundamentally, a sensorimotor atypicality. The ‘High and Inflexible Precision of Prediction Error’ (HIPPEA) account (Van de Cruys et al. 2014) is a prominent example of this, but other proposals in this vein are made by Mottron and colleagues (2006), by Markram and Markram (2010) and by Murray (2018). All these theories are in the wide sense theories of sensorimotor atypicality and therefore, as theories of a different autistic embodiment, amenable to enactive treatment. Enactivism, broadly understood, is a strand in cognitive science that stresses the importance of embodiment, autonomy and sensorimotor interactivity over representations and brain states to understand cognition and behavior (Varela et al. 1991; Thompson 2007; Di Paolo et al. 2017; Hutto and Myin 2017). These theories of autism are not wedded to a homogeneous autistic phenotype consisting of social-cognitive deficits and instead naturally align with the addition of ‘hypersensitivity to auditory, visual or tactile sensations’ in the DSM5 (APA 2013: clause B4).2 The addition of hypersensitivity is in line with early characterizations of autism as well as with autistic self-advocacy forming the initial impetus for the neurodiversity movement (Pellicano 2013). Examples are violent reactions to sudden noise and fixation on specific visual patterns. Sensorimotor theories of autistic embodiment embrace the heterogeneity of the autistic phenotype by trying to account in a statistical way for how social-cognitive issues may emerge given sensorimotor atypicalities and societal and personal contexts. Constant et al., for instance, (2020) develop such a notion of emergence based on HIPPEA. The crucial concept linking the sensorimotor and societal contexts is scaffolding (Krueger and Maiese 2018).3 A scaffold, in its literal meaning, is a structure of platforms typically used for the construction or renovation 1 In line with autistic preferences, cfr. Kenny et al. (2016), we use ‘identity-first terms’ throughout this paper along related terms like ‘autism’ and ‘autistic’ that avoid one-to-one connection with terms like disorder or condition. 2 The DSM-5 also mentions hypo-responsivity but according to sensorimotor theories this can be seen as connected to, and derivative of, hypersensitivity on sensorimotor accounts, see for instance Van de Cruys et al. (2019). 3 Van de Cruys et al. (2014) and McGeer (2019) both independently also connect autism research with the notion of scaffolding. 2 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. of buildings, that allows construction workers to work at different heights. Used figuratively in the interpersonal context, it refers to the practice of using material and social environmental tools as the building blocks of novel forms of interaction: think of being taught, or of the digital devices we use on a daily basis (Colombetti and Krueger 2015; Sterelny 2010). By focusing on scaffolding, any specific trajectory of an autistic is never wholly reducible to autism as such, but includes sensorimotor and social interactions totally specific to any autistic individual. The sensorimotor, or perception/action-driven, approach of enactivism is more open to taking into account autistic lived experience and - by that token - able to overcome the ethical issues, as mentioned above, in an interdisciplinary and productive way (Bervoets and Hens 2020). There is a lot of recent work (Hipólito et al 2020; van Grunsven 2020) addressing how enactivism can lead to seeing autism in line with the neurodiversity tenets. That is, it does not reduce issues associated with autism to an inherently problematic individual neurology. Instead, these issues are seen as, at least in large part, the results of contingent social practices. Think of shaking hands, making eye contact or meetings in noisy rooms. However, there is, as of yet, no detailed bottom-up account of how specifically the enactive construal of autism as the type of sensorimotor atypicality here considered entails an ethical and inclusionary approach. Even if autism and enactivism have been linked in much depth via the key concept of ‘participatory sense-making’ (Di Paolo & De Jaegher 2012; De Jaegher 2013; Fuchs & De Jaegher 2009), a constructive account linking autistic embodiment via enactivist principles to a better inclusion of neurodiversity has not been concretely given. In fact, an enactivist treatment of autism and of inclusion of autistics and their lived experience is anything but selfevident, see for instance van Grunsven (2020) who refers, inter alia, to Hutto (2003). Our account shows instead how relaxing contingent social norms such as making eye contact can concretely lead to including autistics by avoiding unnecessary stress and anxiety related to their sensorimotor atypicality and this much in the same way as intercultural inclusion is promoted by accepting diversity of greeting practices. In this paper we shall explore how an autistic sensorimotor difference can lead to social atypicalities that autism is stereotypically associated with. We will do this on the one hand by employing a sensorimotor atypicality view on autism to show how skewing of certain variables associated with sensorimotor interaction influences autistic (developmental) trajectories.4 We will on the other hand rely on a critical reading of Linguistic Bodies (Di Paolo et al. 2018) to 4 Our treatment below relies on the HIPPEA account yet is not limited to this specific account. Its key premises are limited to sensorimotor atypicality on the one hand and the enactivist framework on the other. This also means (see section 4) that our argument is as such not limited to autism but extends to any sensorimotor difference. 3 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. see how in such enactivist framework social and sensorimotor habits are mutually constrained, leading to diverse tensions between autistic and neurotypical people. Taking both elements together, we argue autism to be one of many gradually distributed sensorimotor differences (such as height) which we overcome in each and every social encounter. Insofar the normative ethical claim consists in furthering inclusion in our public sphere of people with contingent bodily differences, an enactivist take on autism seamlessly leads to extending such inclusion to autism. Given that our analysis is based on a precise understanding of autistic embodiment, it also suggests a concrete way forward to achieve autistic inclusion through recognizing the specific autistic sensorimotor atypicalities. The paper is straightforwardly organized in three sections corresponding to the key elements in the above paragraph. The next section deals with how autistic sensorimotor atypicalities, as seen through the lens of HIPPEA, translate into a specific skewing of social trajectories. Section 3 links this specific skewing to sensorimotor and social habit-forming, and the effect this can have on intersubjective social autonomous interactions. In Section 4 we build on this to develop our ethical claim. The argumentative structure is then as follows: If, as we argue, 1) autism is fundamentally a gradual sensorimotor atypicality that, as sensorimotor habits do, potentially branches out to social atypicalities, and, 2) broadly following Di Paolo et al. (2018), the overcoming of sensorimotor differences is inherent to any social interaction, then, we infer, 3) we should strive to include autistic ways of being and overcome stereotypes as are typically associated with autism, furthering autistic inclusion in the public sphere in the process. That is, we first reiterate the HIPPEA literature that advocates for a sensorimotor definition of autism. Our contribution here is to connect this with the enactive perspective and explore how the sensorimotor perspective connects to the social one in the enactivist framework. Together, we reach the statement (1). Statement (2) reiterates the enactivist view on how social interaction presupposes an overcoming of sensorimotor difference. Our conclusion (3) connects the dots: insofar social interaction presupposes overcoming sensorimotor difference, we ought to strive for also overcoming autistic sensorimotor difference. What we do not argue for here is that the relevant atypicality associated with autism is of a sensorimotor kind: this is already a position in the literature that we build upon in HIPPEA for example. Similarly, we do not argue here that a cognitive deficit view is incompatible with inclusion: this is already a position in the 4 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. disability studies literature that we build upon, such as for example Barnes (2016). Furthermore, we do not argue that the enactive perspective is necessarily better at reaching inclusion than a cognitivist framework: this is not a comparative effort. Nor do we argue that enactivism necessarily leads to inclusion as the literature suggests otherwise, see Van Grunsven (2020). Given the structure of our argument, we suggest in our conclusion that this principled enactive treatment of overcoming gradual differences - or differences in degree - can be leveraged to also argue for better inclusion of people diagnosed with other cognitive or neurodevelopmental ‘disorders’. 2 Autism as atypical sensorimotor embodiment The recent trend in autism research is to try to capture autistic atypicality as flowing from an underlying sensorimotor difference. This aligns with common experiences reported in selfadvocacy and qualitative research focusing on sensory sensitivity (Hens & Langenberg 2018; Pellicano 2013) and is already enshrined in the diagnostic criteria of DSM-5 (APA 2013). Accepting atypical sensorimotor embodiment as a core commonality underlying autistic heterogeneity allows a) to avoid ethical issues equating autism with higher cognitive (dys)function (Bervoets and Hens 2020) and b) to link autism research to enactivist views that – instead of relying on higher cognitive brain modules – rely on the notion of ‘scaffolding’ (Krueger and Maiese 2018). Scaffolding as a term in cognitive science has been popularized by Sterelny (2010) and has been taken up in the enactive literature by Colombetti and Krueger (2015). Essentially, it refers to the ways in which our material environment is used as a toolkit that allows us to create novel forms of interaction. Metaphorically, one may visualize this as using the metal pipes and wooden boards to create a platform to reach ‘higher’ places like actual scaffolds in construction work. In reality, material tools are essential scaffolds in elaborating our practices, but socially we also provide scaffolds for one another in teaching environments (Vygotsky 1978). Because of our approach, the benefits of our account are in principle compatible with any sensorimotor difference postulated as underlying autism. However, we rely for the detailed exposition of how sensorimotor atypicality concretely influences developmental trajectories on the HIPPEA account (Van de Cruys et al. 2014). We do this because it allows us to make this connection as precise as possible for the reader. Also, autistic scaffolding accounts based on HIPPEA already have been proposed by, for instance, Constant and colleagues (2020), the underlying Predictive Coding view has been theoretically linked to dynamic misattunement 5 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. (Bolis et al. 2017) and such accounts are being intensively researched empirically, see Palmer et al. (2015) or Zaidel et al. (2015). In essence, what we want to show is that viewing autism as atypical sensorimotor embodiment leads, in a state space description of the agent, to skewing certain variables associated with the sensorimotor interaction that are crucial for social normativity (see section 3). Given such skewing we can then specify how autistic developmental trajectories are such that they give rise to the type of social misattunement that has been so often mistaken for the social-cognitive essence of autism5. Before we go into the specifics let us first illustrate this with a familiar paradigm case of sensorimotor difference: blindness. Nobody would say that it is impossible for blind people to get to certain states of social interaction. What is different between blind and non-blind people is how they get to the physical place in which the social interaction takes place. They clearly will not arrive there with the same environmental (accessibility) conditions. The blind person will rely on, for example, a touching stick or echolocation or acoustic guidance. The non-blind person will, typically, rely on visual cues and would - failing those - feel quite lost. The sensorimotor difference then is generally accepted to impact only the how of navigating physical space and not that it is possible to navigate to a certain physical place. What we will argue below is that, in an enactive view of autism as a sensorimotor atypicality, what is basically in question is the how of achieving certain states in a physical (or more abstractly: state) space. Using blindness or deafness as an analogy exemplifies how our approach aligns with that of the social model of disability where ‘minority bodies’ can never be just seen as deficient bodies (Barnes 2016). Indeed, whilst there are clearly obstacles to certain contingent ways of social interaction, such obstacles can always be met (and, consequently, social interaction as such can always be achieved) in suitably adapting the environment. One can take sign language interpreters or audio described movies as examples.6 HIPPEA, or in full “High and Inflexible Precision of Prediction Error in Autism”, is a conception of autism within the predictive coding theory. Predictive coding in cognitive science, roughly, is a theory on perception according to which the brain is essentially a prediction machine that aims to minimize the divergence between its internally generated prediction of the external world and the actual input (Hohwy 2013). In HIPPEA it is In other words: we specify how this mistake stems from a “double empathy problem” (Milton 2017). Two anonymous reviewers pointed out there might be specific social interactions inaccessible to some people and we agree. Our ethical point is limited to the fact that what is considered essential to social wellbeing can be made to become accessible regardless of sensorimotor atypicalities by assuming sufficient effort by all parties involved. In fact, in line with our enactive treatment, we believe all individuals are to some extent particular from sensorimotor point of view such that any successful social interaction presupposes overcoming such difference (see section 3). 5 6 6 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. hypothesized that autism consists in a specific way of dealing with perceptual uncertainty. Following HIPPEA autism is an atypical way to process a difference (the “Prediction Error”) between perceptual input and prior expectations. This atypicality then specifically consists in atypically giving a “High and Inflexible” weight to the difference between both (in other words, it consists in giving an atypical importance to “Precision”). Note that precision weighting is a general feature of perception under Predictive Coding and that all individuals display variance in both the weighting processes and the flexibility thereof. In other words: everyone engages in precision weighting and displays a certain range of flexibility in this. HIPPEA thus by definition denominates a difference in degree between non-autistic and autistic people.7 Predictive Coding is phrased as a perceptual theory but this - certainly on an enactivist reading, see further below - implies action as well. Indeed, perception has been for a long time recognized as an active process (Gibson 1986) of mutual engagement with one’s environment. In this case of atypical importance of precision, we would expect to see an atypical sensorimotor interaction with one’s environment; an interaction more attuned to the actual specifics of the environment than to abstracting away from those specifics in order to flexibly switch from one abstract form of interaction to another. This then leads to a learning style that is at odds with the requirements of a typical environment. Given the aforementioned scaffolding, autistic developmental trajectories will, but only statistically speaking (Bervoets and Hens 2020), lead to emergence of social-cognitive elements of the DSM-5 (APA 2013). Let us make this more precise with a specific example. When in a meeting, the autistic subject might be attuned to a certain flicker of light or background noise, thereby giving less prominence to somebody new entering in the room. At that time, the autistic subject might seem self-absorbed compared to neurotypical others who are more rapidly cued to the new social situation, more quickly meeting the new entrant by greeting them. This may at first sight seem just an innocuous 7 An anonymous reviewer pointed out that there may be different ethical implications of viewing autism as a sensorimotor atypicality or as a deficit in cognitive function. As mentioned above we restrict ourselves in this paper to the positive point related to ethical inclusion when seeing autism in an enactive way as sensorimotor atypicality and do not pursue a negative point of seeing cognitive deficit as incompatible with ethical inclusion. Relatedly, they pointed out that HIPPEA might be seen as a deficit in perceptual processing. Although we cannot pursue this in the present paper, we call attention to the fact that, in predictive coding approaches, precision is a general feature of all perception and that, therefore, differences in precision can only be seen as gradual differences in line with the main argument in this paper (where HIPPEA is just used as an illustration). The inflexibility that arises is, as we argue in Section 4, related to contingent social norms of switching between contexts, and it can, therefore, not be interpreted as a contextless deficit within an individual. The upshot of this is that there can indeed be no a priori bounds on inclusion based on such a gradual sensorimotor difference if social norms can never – in line with what we argue below – be enforced rigidly. This does not mean that there will not be practical difficulties or difficulties with respect to specific ways of social interaction. It just means, see section 4, that such tensions can be overcome in allowing substantive social exchange. 7 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. difference but in the context of learning it may be an element in a large series of missed opportunities (or scaffolds) to attend to what is typically thought to be salient for development. In fact, as worked out by Constant et al. (2020), the outcome of such an atypical learning process may account for many atypicalities associated with the autistic self. Let’s now take a closer, enactive look at how atypical precision may skew variables which are associated with sensorimotor interaction. Specifically, in enactivism we should be wary of reifying an atypicality related to precision as part of a computational module in the brain. Instead, this atypicality is to be seen as capturing the specific way autistic bodies tend to covary with their environment8. The difference between typical and atypical precision is a difference in such covariation. Autistic bodies tend to covary in a more precise way with their environments, picking up regularities in their environment in a more precise way. By that token they will tend to be less sensitive to social habits which rely on abstracting from the concrete perturbations in the environment (which, said in passing, can be seen as an enactive reinterpretation of the terms “High” and “Inflexible” underlying the HIPPEA acronym). Coming back to the analogy we made above with respect to blindness, this atypicality is not a static feature of a ‘modeler’ or of ‘a model’ restricted to be in certain atypical states. Rather, it concerns the more nuanced, atypical manner of attuning to one’s environment, i.e., an atypical way of navigating states afforded by one’s environment. Autistic atypicality is, then, maybe less ‘visible’ to external observers (noting that even blindness may be compensated for by for instance echolocation and therefore remain rather invisible, see Servick (2019)) but in itself very akin to classical sensorimotor differences such as visual impairment. Autistic people will tend to rely a lot more on current tangibles in their environment and much less on carrying out previously internalized habits aligned with social conventions (for the enactive development of such a scaffolding, see: Krueger and Maiese 2018).9 In a physical space metaphor: they rather tend to “feel” their way around in the here and now rather than “reenacting” behavior based on some abstract cues. This does not mean 8 For a complete argument to this effect see van Es (2020a; 2020b). There is a bit of complexity here. DSM-5 (APA 2013) also specifies Restricted and Repetitive Behaviors. This implies that autistic often display an insistence on sameness holding on to specific habits or repetitive patterns of behavior. This might at first glance seem at odds with our reading of HIPPEA which is focused on the immediate tangibility of the environment. However, in our view this can be seen as a developed coping strategy of autistics in making their environment more predictable so as to be less overburdened by spurious changes. We cannot develop this point further in the scope of this paper but see it as a crucial aspect of an enactivist reading of HIPPEA in which such behaviors can be explained without reference to inflexibility of internal representations. We thank Sander Van de Cruys for pointing this out to us. Whilst an important outstanding issue in autism research, we do not believe its resolution is critical for the argument made in this paper. We also refer to footnote 3 on hyper- and hyposensitivity for a similar connection between two apparently contradictory phenomena. 9 8 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. they can get to less (or indeed more) places. It just means they tend to get there in a different way (see also Hipólito and Hutto 2020). Take the example mentioned above: it is not because the autistic subject seems self-absorbed that they mean any disrespect to the one entering the meeting room. They are just more attentive to the ‘here and now’ of their sensory environment and therefore prone to miss the sensorily less pronounced cues associated to what is typically seen as socially salient. This will have no effect on the efficacy of the meeting as long as both parties can agree this momentary desynchronization is of no consequence to the content of the interaction. This means that when we compare autistic and typical bodies, they navigate similar territory or state spaces in different ways. An enactive reading of HIPPEA, and by extension of the other differences in ‘sensorimotor apparatus’, leads to higher sensitivity to environmental perturbation. Seen from a neurotypical eye, autistic embodiment is correlated to ‘overfitting’ sensory data but from an enactive point of view it just is another, atypical, way of navigating one’s immediate environment. Given autistic bodies are more sensitive to (deal more ‘precisely’ with) their immediate environment, their actions depend more on the ‘tangible’ here and now than on abstracting out of such tangibles based on ‘intangible’ cues that have become associated with certain expectations of default interaction patterns. Phrased like this one can see the link to other sensorimotor differences such as blindness: the difference with ‘typical’ agents lies in how certain states are reached and not in a principled restriction of what states can, ultimately, be reached. This fits with the atypicality view of autism, opposed to the deficiency view. The skewing of how certain states are reached by an autistic body clearly has an impact on sensorimotor interactions with its physical environment. It will be more sensitive to triggers in such an environment, tending to covary with them in a more ‘precise’ way. This way it picks up on regularities and patterns a ‘typical’ body will simply not pick up on. At this junction there is neither value nor disvalue in this difference; it just is. However, if assessed by the standards of habits formed by typical bodies or from the interaction patterns naturally sustained between typical bodies, autistic differences will tend to be perceived as flouting the conventions, or the norms, of ‘typical’ interaction and hence as some intrinsic ‘deficiency’.10 The relations between 10 That such a (dis)valuation of autistic difference as deficiency is compatible with enactive theorizing has been shown by Janna van Grunsven (2020). The view from ‘typical bodies’ can be compatible with the enactive views as developed by Gallagher (2004) and Hutto (2003) on autistic differences. An alternative enactive view based on autonomous sustainability of interaction patterns is left open by the ‘participatory sense-making’ approach by Di Paolo, Cuffari and De Jaegher (2018). In the next section we critically build on the latter approach in order to capitalize on its strengths, at the same time avoiding this weakness. 9 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. the individual differences, the emergence of shared norms and valuation are complex as can be gauged from the above: an increased sensitivity to one’s physical environment may quickly be perceived as an insensitivity to one’s social environment. In order to work out these relations and their link to interactional differences between autistic and neurotypical people, we will in the next section critically build on the enactivist framework of Linguistic Bodies (Di Paolo et al. 2018) to see how social and sensorimotor habits are mutually constrained. This can then serve as the springboard for our ethical argument to further the project of inclusion. 3 From sensorimotor atypicality to social normativity In the enactive literature, Di Paolo et al. (2018) distinguish three distinct, yet intimately intertwined levels of autonomous organization relevant to an organism: the biological and sensorimotor levels of organization are tied to each individual, whereas the social level is intersubjective, and unique to each social encounter. These levels are hierarchically layered so that the biological level of organization lies at the foundation and enables the formation of any other form of autonomy. Autonomy is defined as precarious operational closure (Di Paolo and Thompson 2014; Maturana and Varela 1980). This means that an autonomous system consists of a self-enabling process network, which captures operational closure, that is naturally inclined to disintegration, which implies precariousness. A process network is self-enabling, if each process in it enables or is enabled by at least one other process in the network. A process enables another process when its continuation aids the continuation of the other process; and is enabled by another process when the other process’s continuation aids its own continuation. In this sense, the self-enabling nature of the network is what allows it to, albeit temporarily, resist the natural incline towards disintegration. Sensorimotor autonomy does not concern the physical integration of the system, but instead its sensorimotor interactivity. Here too, it is thought, do we find clusters of activities or processes that are self-enabling, and display a similar autonomous structure as we find in biological autonomy. These autonomous sensorimotor activity clusters are also called habits (Di Paolo et al. 2017; Barandiaran 2017). In the remainder of this paper, we will refer to these clusters as interactional (sensorimotor) regularities. Crucially, sensorimotor autonomy is irreducible to biological autonomy, and thus does not need to be in service of it either. Think of the many routines that populate our daily activities, many of them irrelevant to biological sustenance. Nonetheless, we do experience a sort of urge to maintain the activity clusters we are used to, such as a need to see some light after having been in a dark room for an extended 10 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. period of time. It is in this way that the habits that constitute an organism’s sensorimotor autonomy can be thought of as autonomous, independent of an organism’s biological autonomy (Di Paolo et al. 2017). Of course, the restriction of sensorimotor autonomy can have biological effects; by way of stress, for example. Social autonomy is intersubjective, and unique to each social encounter. This, following Di Paolo et al. (2018), does not explicitly concern each individual’s autonomy, but instead the autonomy of the social interaction itself. In this sense, each social interaction is characterized as a self-enabling network (the ‘self’ here referring to the interaction, not to the ‘selves’ that are interacting) that, unless actively maintained, naturally inclines towards disintegration. What helps sustain an activity (and what does not) depends on the participants and the environment. Arguably, not all three levels are actually associated with each organism. Some, the argument goes, seem to only have biological autonomy so that their sensorimotor interactivity can be reduced to, and is merely in service of, their biological autonomy, whereas others engage at least in some sensorimotor activities for the sake of it: think of a pigeon rolling in the snow, for example. 3.1 Levelled sensitivity An autonomous system cannot be defined without at least implicitly defining the environment it is embedded in. This is because in continuously enabling its own structural integrity – i.e. maintaining its autonomy – a system necessarily interacts with its environment. This continuous interaction can bias an autonomous system to the interactional regularities that are encountered. Put differently, the interactional history of a system allows it to become attuned to the unfolding interactions: to the particular ways that certain movements influence sensory signals (also called sensorimotor contingencies, see O’Regan and Noë 2001; Di Paolo et al. 2017). A sensitivity, then, is nothing more than a system’s structural bias towards certain interactional patterns or regularities borne out of an interactional history, which includes biological, sensorimotor and social interactivity. This structural bias could be unpacked in terms of reliable covariance between the system and its environment (Bruineberg and Rietveld 2014; Hutto and Myin 2017; van Es 2020).11 For two systems to covary, a change in one brings about a change 11 Di Paolo et al. (2018) and Di Paolo et al. (2017) take this a step further by taking a system’s precarious autonomy to imply a normativity relative to each level of organization. More than sensitivity, this normativity is intended to account for a supposed ‘judging,’ or ‘evaluating’ capacity. Each system’s biological autonomy thus implies an ‘inherent normativity’ which offers a standard of evaluation by which to judge the possibilities of interaction in the environment as either harmful, positive, or neutral. There is a debate in the enactivist literature with regards to 11 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. in the other. Reliable covariance is then a covariation relation that does not have a one-to-one mapping but implies covariance to a certain degree. There is no definite threshold for this but, in our case, minimally one may expect that the system’s structural integrity is to be maintained for anything to count as reliable. The sorts of interactional regularities that a system becomes sensitive to, depend of course on the interactions it has been able to engage in: its interactional history. This means that when considering sensitivities, we need to consider the system’s embodiment, which determines the possible modes of interaction (Varela et al. 1991). After all, a system cannot become sensitive to environmental changes that its senses cannot pick up on one way or another. People, for example, cannot become sensitive to the ways particular winds affect one’s flying capabilities, whereas birds are unlikely to become sensitive to a standardized pen grip. Furthermore, interaction only happens in an environment. The sorts of interactions available are thus also dependent on the environment one is in. Much like birds in a world that does contain pens, humans only a few centuries ago could not have become attuned to pen-writing activities either. Indeed, any activity is necessarily performed by a system, in an environment, thus the possibilities for interactions are determined by the system and its environment, and so the interactional history out of which interactional sensitivities are borne is, also, determined by both the system and its historical trajectory through an environment. A sensitivity is a system’s attunement to the particular way that certain movements influence the system in one way or another. The different autonomous levels of organization are influenced rather differently by one and the same motoric movement. Consider how, biologically, eating food provides energy for continued sustenance, whereas sensorimotorically the foods’ way of resisting the chewing pressure entails a certain texture, the influences on the tongue as it swirls and moves the food about entail a certain flavor (Noë 2004). Socially, thus intersubjectively, eating the food may prolong the interaction — not eating may be rude — while simultaneously imply a potential break — perhaps the interaction is over once the food is finished. The different ways particular interactions influence the respective autonomous levels implies that the developed sensitivities are also different. Whereas biologically the system may become attuned to, say, energy intake regularities such as the frequency or the whether this individualistic account of normativity betrays a hidden representationalist commitment (Hutto and Myin 2013, 2017). Our present argument does not hinge upon this debate. We can speak of interactional sensitivities obtained on an individual level, and introduce normativity on a social, intersubjective level, not perfectly in line with, but amenable to both takes in enactivism (Di Paolo et al. 2018; Di Paolo et al. 2017; Hutto and Myin 2013, 2017). 12 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. amounts, sensorimotorically the system may become attuned to textures and flavours, the spaces food is eaten in, the tableware it is eaten out of or the tools it is eaten with. The individual system may also become attuned to social regularities particular to food-eating practices, such as the presence or absence of certain people, the appropriate manner of using the cutlery, and whether or not to engage in, say, conversation.12 Though conceptually we can divide the three levels, in practice they are intimately intertwined. We don’t, e.g., experience multiple levels of organization, nor do different levels of embodiment refer to multiple physical bodies. There are however ways we can distinguish them in practice. Consider that someone may sensorimotorically be a bespectacled person, so that their spectacles are an inherent part of their ‘sensorimotor self’, whereas, biologically, the spectacles do not figure in the self-enabling network. The same could arguably be said of shoewear or even of clothes. Moreover, we sometimes find ourselves in conflict, when the different interactional biases conflict; such as becoming peckish during an academic presentation since, biologically, one may be attuned to energy intake at that time whereas, socially, eating may be detrimental to a continued academic interaction. Conversely, each and every activity is always associated with all three levels of autonomy, to the extent the respective levels are associated with the system (Di Paolo et al. 2018). Every human activity minimally takes energy, is some sort of sensorimotor interaction in a world that is inherently social in nature. Though distinct, no level of autonomy ever ‘acts on its own’. It is in this particular sense, that the interactional regularities that any system becomes attuned to are dependent on its levels of autonomous embodiment as well as the environment it figures in. 3.2 How autism levels up Now that the basic framework is laid out, we can start looking at how the sensorimotor atypicalities common to autistic people fit in. Briefly rehearsing, the interactional regularities one becomes sensitive to (interactional sensitivities) are dependent on a system’s interactional history. Organismic interaction depends on organismic embodiment as well as environmental 12 There is a slight disparity here between the discussed levels of organization and their respective sensitivities. According to Di Paolo et al. (2018), biological and sensorimotor autonomy are both individual, yet social autonomy is inherently intersubjective. Following the in-text logic, this implies that the autonomy of the interaction, instead of the participating individuals, becomes biased towards certain interactional patterns. There may be something to this, as Di Paolo et al. (2018) show. Consider recurring fights with a friend that neither of you actually wants to have. It may be, indeed, helpful to think of this in terms of a bias of the social encounter itself, instead of one of the individual participants. However, in this paper, we shall also be concerned with the individual interactional sensitivities pertaining to social encounters. 13 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. constitution. This means that the interaction changes whenever the organism or the environment changes; the organism comes to covary reliably with its environment. The particular ways in which the organism has come to covary with its environment, has become sensitive to its environment, has historically interacted with its environment, can be described as a state space trajectory. Imagine a multi-dimensional state space, with one axis for each variable associated with the system.13 The state of the system is then a multidimensional coordinate, with a particular value for each variable. Whenever a variable changes, the state of the system changes, pushing it to a different multi-dimensional coordinate. As an organismic system is constantly in flux, over time we will see the changing coordinates of the system as traversing a particular trajectory through the multi-dimensional state space. Autism, we argue, is best taken to consist fundamentally in particular sensorimotor atypicalities. Within the state space formalism, this implies that certain sensorimotor-related variables in the state space may incline, or be skewed, to display particular values or dynamics.14 This skewing is consistent throughout their developmental trajectory, and thus consistently influences their modes of environmental interaction. Consequently, by impacting interactions, it will over time impact interactional history, and thus interactional sensitivities. Sensorimotorically, an autistic will thus display an atypical state space trajectory, meaning an autistic becomes sensitive to different interactions and forms different habits. This may explain certain common basic sensorimotor tendencies among autistics (Constant et al 2020). Yet it underdetermines the exact influence it will have on any specific autistic individual. Each trajectory is of course not only dependent on a particular skewing of certain variables, but it is a holistic pathway: each and every aspect of the individual, as well as each and every aspect of their environment affects the trajectory in their own ways. As such, not only is each trajectory unique, but each autistic trajectory is also uniquely atypical. In this way, our model supports the large heterogeneity in the atypicalities usually associated with autism. Now we can see how this can ‘scale up’ to social interactivity (but see Gallagher 2017; Zahnoun 2019 for criticism on the need to ‘scale up’). 13 What exactly it means for a variable to be ‘associated’ with a system is an interest-dependent matter. If we are to model a person’s kinetic dynamics when going down a slide on a playground, variables pertaining to their cycling habits are not considered ‘associated’ with the system for current purposes. However, when we are interested in modeling this person’s risks for cardiovascular diseases, we may want to consider those. In this sense, a state space description of a real-life system need not be complete to be scientifically valuable. 14 Whether the inclination is visible in particular variables or dynamics depends of course on which variables were chosen to figure in the state space description. A system that accelerates strongly either skews to high values in the ‘acceleration’ variable or quickly rising dynamics in the ‘movement speed’ variable. It depends on how we set up the formalism. 14 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Per our explanation in Section 3.1, anything happening on the sensorimotor level of embodiment impacts the system’s social interactional sensitivities. Sensorimotor atypicalities entail a mode of interaction with the world different from that conventionalized by interaction between typical people. As such, the social interactivity will be influenced by the autistic’s sensorimotor atypicalities just as their picking up of objects is impacted by them. This starts at the very fundament of social development. Consider that joint attention schemes, for example, play out differently when one’s relevant sensorimotor habits are mutually attuned than when they are not. Over time, this may result in atypical social habits, such as particular manners of speaking, body language or a preference for certain environments to engage in social interactions in. As such, consistent sensorimotor atypicalities will influence the social interactional regularities that the agent becomes sensitive to. Put differently, sensorimotor atypicalities entail atypical pattern-forming on a social level, which in itself implies a skewing in the interactional regularities that the person becomes attuned to on a social level. Take practices of joint attention - like looking each other in the eye - that are crucial for turn-taking in the social realm. If, as Constant et al. (2020) argue, such practices are more difficult because of sensorimotor atypicalities, they can lead to different trajectories with respect to social preferences. At this juncture, it is important to understand that autism is just one example of a possible skewing of a sensorimotor trajectory. Height, weight, or visual or acoustic acuity more broadly may just as well be atypical and are accompanied by their own atypical habits both sensorimotorically and socially. As such, placing the atypicality of autistics at a sensorimotor level allows for the wide variety of noted atypicalities in autistic social interaction (Jaswal and Akhtar 2019). One person’s trajectory may have resulted in habits that eschew fantastical imaginings (see, for example, DSM-5 (APA 2013) or Curry and Ravencroft (2002)), whereas another’s interactional history may have resulted in habitual interaction in fantastical worlds (see, for example, Lyons and Fitzgerald (2012)). However, they may, coincidentally, share a habit of avoiding eye contact (Valiyamattam et al. 2020). They also may not (Jones et al. 2017). This all depends on the particular state space trajectory of each individual, as it does for every person, autistic or not. Moreover, this trajectory is a sum of the organism’s state space and its environment’s, which means that the same person will develop different habits if it were to have grown up in a different environment. This is to say that from that point on, their state space trajectories will come to vary wildly, and they will grow up to be different persons altogether. 15 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. 3.3 From individual sensitivity to social normativity Individually, any organism has its own embodiment accompanied by its own set of sensitivities on a biological, a sensorimotor and a social level. However, according to Di Paolo et al. (2018), each social interaction also has its own autonomy. Following the definition of autonomy above, this implies that each social interaction is composed of a self-enabling network that naturally inclines towards disintegration. This means that, if the processes in the network were to stop, the entire interaction would break. Only by continued processual activity can the whole network be kept active: a conversation cannot be sustained without continued interactivity. Moreover, each social encounter is unique and entails its own state space description. The potential modes of interaction that could either sustain or break the intersubjective autonomous organization are thus also unique to each social encounter. Like the organismic case, what sustains or breaks a social interaction is dependent on the processes that make up the network itself, as well as the environment it is embedded in. Naturally there is some variability in participants’ sensitivities: some people may be used to speaking at a particular volume that may seem too loud or too soft to others. Yet some environments also may afford speaking loudly better than others. The range of speaking volume which best sustains or breaks a particular social interaction is thus dependent on both the participants’ sensitivities as well as the environment the interaction figures in. Indeed, contrary to the individual organismic case, the interactional processes that make up this network are composed of those of all participants, and thus depend, but cannot be reduced to, all the individual participants’ own interactional sensitivities (Di Paolo et al. 2018). This makes the interactional state space description an intersection of the individual interactional regularities and sensitivities of at least two participants, their environment, and their individual relation to the shared environment. An atypically tall person for example may well have adjusted to looking slightly downward in social interactions, which means they may not be well attuned to social interactions with someone even taller. As it is for the atypically tall person, so it is for the autistic. This is to say that an autistic’s own interactional history will in part determine what will sustain and what will break a social encounter they are engaged in, just as much as their interactional partner’s interactional history does. This illustrates how a gradual sensorimotor difference can over time reverberate into a qualitatively different social experience, dependent on the interplay of how pronounced such difference is as well as on how much one’s environment is willing to accommodate such a difference. Whilst tallness is not immediately seen as consequential, consider shortness: one may not reach certain items in a 16 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. supermarket, may not be able to use public restroom facilities, or, indeed, may not be able to look interlocutors in the eye. Indeed, although this concerns a difference in degree in one and the same quantitative dimension, it is immediately apparent how with current social norms this will translate easily into a disabling condition affecting social interaction in a lasting way. Furthermore, the analogy between atypical length and the case of autism with respect to eye contact can be taken literally. In both cases there are sensorimotor reasons complicating eye contact: differential tallness in the one case, and hypersensitivity to the other’s gaze in the other. Yet whether or not this develops into a lasting preference to avoid eye contact depends to a large extent on the reactions of others for whom there are no such complicating factors. If they insist on it, this may well make any meaningful social interaction problematic. If this problem is one encountered by individuals from very early on in development — as seems to be the case for many autistic people — then this may result in a lasting tension with respect to making eye contact. This tension in the social encounter needs to be overcome if the social encounter is to be sustained, but it is not different in kind to any other tension specific to each and every social encounter between two uniquely different individuals embedded in a unique environment (Di Paolo et al. 2018). In fact, when the environment actively makes the difference inconsequential, it is less likely to become such a focus of lasting, problematic tension. It is this theme of overcoming tensions and differences that we take up in the next section where we link it to the ethical theme of inclusivity of diversity. What we have argued for in section 2 and 3 respectively is that 1) autism is fundamentally a gradual sensorimotor atypicality which branches out to social atypicalities, and that 2) overcoming of sensorimotor differences is, in general, inherent to any social interaction. Section 4 then infers that we should, on the same grounds as accepted for promoting the inclusion of other diversities, also overcome the social tensions flowing from sensorimotor differences associated with neurodiversity. 4 The ethical value of overcoming differences in degree It is hardly controversial that differences in visual ability or height should not lead to exclusion of those on the extremes of our human spectrum from our community. However, we can also arrive at a principle of inclusion via classical ethical literature. Indeed, grounding of our moral stance in our social communicative practices is, for instance, given by Strawson (2008/1962) 17 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. via his notion of ‘reactive attitudes’.15 Crucial in this leading ethical framework is negotiation of ethical interaction across individuals characterized by individual differences. Indeed, Wallace (2019) develops a call to cosmopolitanism using reactive attitudes where we should maximally broaden our ethical community irrespective of contingent social and cultural practices and merely based on the possibility of entering into a mutuality of reactive attitudes. In another Strawsonian interpretation, McGeer (2019) works out an inclusionary account based on the ability to scaffold towards social communicative practices. She contrasts this to approaches where certain specific social capacities are required as preconditions of being full members of the moral community. Her approach avoids setting minimal entry conditions based on contingent social practices such as eye contact as prerequisite for reactive attitudes. Instead, she focuses, in line with our above argument, on the ability to scaffold regardless of specific initial sensorimotor differences. Obviously, this requires both parties to the interaction to put in an effort in order to overcome such differences. This is also in line with disability literature in general where inclusion is central, and this independent of meta-ethical debates. This is clear from, for instance, Barnes’ take on The Minority Body (2016), and neurodiversity literature specifically. As a particularly acute example we can take Milton’s (2017) expression of the double empathy problem. In the latter, Milton argues that understanding autism as a deficit in social-cognitive ability actually is tantamount to a lack of empathy from the point of view of a neurotypical agent. In the former, Barnes argues that attributing issues to individual disabled bodies is, equivalently, an inability to see from the perspective of typical, able, bodies how those issues actually stem, at least for the largest part, from a lack of societal concern for accessibility of minority bodies. Our argument can be seen as integrating these into a Minority Brain view (Bervoets and Hens 2020). Indeed, overcoming visible sensorimotor differences requires the non-disabled others to put in effort to overcome related accessibility issues. In the same way we argue that prima facie invisible neurological differences give rise to an analogical 15 The ‘reactive attitudes’ are attitudes of blame, resentment, forgiveness, gratitude etc. which we discern in the other and the other in us. They allow us to build practices in which we can hold each other responsible. There are two remarks to be made about our using this framework as an illustration of our argument. First, there is a lot of controversy on whether such an account does not risk excluding others based on the ability to recognize and show the reactive attitudes. It has been used as a way to rationalize either (partial) exclusion of autistics (Shoemaker 2015) or to positively argue for a ‘normalization’ of autistics to recognize conventional social cues (Richman and Bidshahri 2018). However, we use this theory for illustration precisely because of its ability to abstract away from sensorimotor (a)typicalities and its consequent bringing into focus of a communicative interaction as such. Second, we use this theory, just like we used HIPPEA above, as a concrete illustration of our argument without it being a necessary premise in what we contend. Any ethical framework that can develop a need for inclusivity over and above individual differences will do to support our conclusion. 18 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. societal responsibility: adapting neurotypical social practices in order to allow for better social accessibility to the neurodivergent (autism being our prototypical example of neurodiversity). Now, as we have seen in section 3, the sustenance of social interactions is in a crucial way independent of specific underlying sensorimotor preferences. We do not have to react in a specific way to be able to signal, for instance, blame or gratitude. Exactly like blind and seeing people navigate (see our discussion in section 2) the same physical space differently, tall people and people of average height navigate differently the interactional state space so as to recognize reactions of blame or gratitude. The ethical call to inclusion is to separate the how we achieve an ethical interaction from the that, the very possibility, of reaching such interactions. Here we need to clearly separate the unique trajectories individuals traverse in order to reach some point of social and ethical interaction and the fact that such points can be reached. In enactive terms: whilst everybody lays down a different path while walking (Thompson 2007), there is no reason why our paths may not intersect in participatory sense-making (Di Paolo et al 2018, specifically the example worked out with respect to autism in Chapter 8). Let us make this more precise with respect to autism and autistic embodiment. Where autistic people are sensorimotorically skewed to attune more precisely to the here and now of environmental perturbations, this is not the case for the majority of (neuro)typical people. It is in this way understandable that the latter mistake their specific way of coming to reactive attitudes for minimal, or essential, preconditions of achieving them. Whether it is by shaking hands, recognizing specific facial expressions or in other ways abstracting from environmental cues deemed inessential, typical people will tend to mistake a sensorimotor preference for certain habits in social interaction for the essence of that social interaction. In the above we have given the example of eye contact and other contingent social practices that may be selfevident for neurotypical people but, for sensorimotor reasons, hard for those deviating from the statistical sensorimotor norm. That these practices are contingent to the social interaction itself can easily be seen from the possibility to interact socially without eye contact, and they should thus not be seen as essential barriers to social entry. Autistic people will nevertheless from childhood on have difficulty attuning to such habits precisely because of a statistical atypicality in the way in which they tend to covary with environmental cues. We have shown that this skewing of the variables associated with navigating the multi-dimensional state space should not be reified into stereotypes of social interaction, because it is but one of a great many historical factors that contribute to how unique autistic individuals wind up trying to engage in social interactions optimized for a sensorimotor embodiment that is not theirs. In a society more forgiving of the preconditions of expressing reactive attitudes, they may more easily develop 19 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. shared habits than in one that is less forgiving. Obviously also, the longer it takes to be recognized as one who is able to communicate, the more this part of one’s history will further skew one’s potential to bridge the gap between one’s own sensorimotor preferences and the social interaction which is deemed foundational for inclusion in our moral community. A specifically acute example of overcoming such gaps is given in the context of early language development by De Jaegher (Di Paolo et al. 2018: Chapter 8). None of this is to say that these social, ethical, interactions are disembodied. Far from it, in line with Di Paolo et al. (2018), each human interaction is necessarily embodied. But, because of the relative autonomy of sensorimotor and social interactions we are not limited to a specific way, a specific habit, of sensorimotor interaction to achieve a specific, ethical interaction. Rather, by prescribing specific sensorimotor interactions as basic for ethical inclusion we violate the independence of the social and ethical field and – see our argument in the previous paragraph – risk to actively contribute to exclusion through ‘ableist’ bias. Indeed, there is a sense in which all of disability, at least the social model aspects of disability, can be seen as the result of too prescriptive a notion of sensorimotor interaction. Such too prescriptive notions literally exclude people from ethical interaction by making such interaction inaccessible for those who are sensorimotorically different. As is clear from the above, the overcoming of differences in degree (whether in visual acuity, height, culture, ...) is something we do daily among people. We recognize just societies by their capacity to overcome the superficial tensions that may arise from such differences, i.e., their tendency to embrace cosmopolitanism, per Wallace (2019). What we argued for is that the enactive construal of autism as an atypical embodiment leading to a specific skewing of sensorimotor preferences leads to recognizing it as one such difference in degree. Looking at how such difference is overcome in actual lived practice helps us explain the precise nature of this underlying difference and - consequently - may help us to create more inclusive conditions by relaxing sensorimotor requirements which underlie social interaction of (neuro)typical people. We have mostly given relatively straightforward examples which might seem relatively easy to overcome, such as whether to maintain eye contact. However, our argument (as per fn. 7) does not admit of a priori bounds on conditions to be relaxed. To give a concrete example of this: think of non-verbal autistics. One may be tempted to consider verbality a minimal condition for social interaction, but this is challenged, amongst others by disability scholars in an intriguing case study (Van Goidsenhoven and De Schauwer 2020). In fact, the practice of living with autistic people shows that what was once seen as an insurmountable barrier can be 20 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. overcome and scaffolding to rich communication can be achieved in a variety of ways, see also De Jaegher (2013). Clearly, there is a cost to cosmopolitanism on the side of typical people who need to adapt to new habits (or unlearn old habits), but we believe that not doing so (and consequently leaving all the cost of adaptation to the atypical) is a case of bigotry. This does not mean that the final result will not be a compromise between a cost incurred on both sides - see the double empathy problem of Milton (2017) - but it does mean there is an ethical (as well as scientific) call to at least figure out how to come to a better compromise, even if this is a path we can only lay down in walking. 5 Conclusion In conclusion, it may be relevant to revisit our argumentative structure. By connecting HIPPEA with enactivism, we have argued that (1) autism is fundamentally a gradual sensorimotor atypicality that, as sensorimotor habits do, potentially branches out to social atypicalities. Then, (2) broadly following Di Paolo et al. (2018), we argued that the overcoming of sensorimotor differences is inherent to any social interaction. Indeed, everyone’s interactional regularities, or their developmental trajectory more broadly, influences their concrete social interactions. Differences in the way one approaches social interactions can cause tensions to be overcome in any social interaction. According to our analysis of autism, it is just one of the many ways in which one can be sensorimotorically atypical. As such, we conclude with (3) we should strive to include autistic ways of being, furthering in the process more inclusivity in the public sphere (McGeer 2019). We have argued that our enactivist construal of autism entails an inclusionary approach in line with the neurodiversity tenets specifically and cosmopolitarian ethics generally.16 That this is not self-evident is clear from other enactivist approaches to autistic embodiment, see Van Grunsven (2020), which start either from the neurotypical interactional habits (Hutto 2003; Gallagher 2003) or from the mere autonomy of specific sensorimotor interactions (Di Paolo et al. 2018). Our approach instead starts from a relative - and ethical - independence of social interaction and so leaves room for an inclusionary approach to autistic embodiment seen in its sensorimotor specificity. Basically, we show that there are many ways to peel an orange and in our daily practice we overcome sensorimotor differences that unavoidably arise from different 16 This does not imply that a cognitivist construal cannot arrive at inclusion in a different way. 21 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. life histories. Autistic embodiment is just one of such bodily differences. Understanding autistic sensorimotor difference through a good empirical theory of autism, such as atypical precision, allows us to direct our search for practical ways in which we can increase mutual understanding. After all, we are all continuously changing bodies in need of being cut slack in order to maintain social, ethical, interaction. In the process, we have also engaged with the enactivist literature. We have expanded enactivism to provide a space for speaking of individual social sensitivities, over and above the biological and sensorimotor individual sensitivities. This has allowed us to expand the merely descriptive enactivist analysis into the normative debate. This provides new tools for the ethical project of inclusion, by specifying particular social tensions that we should work to overcome, as we do on a daily basis elsewhere. With this, we avoid slipping into enactivist developments of autistic embodiment that, ultimately, point back to basic, or essential, social capabilities and therefore deficiencies related to autism (Van Grunsven 2020). Furthermore, we connected the enactivist notion of scaffolding with its use in McGeer’s work, allowing for an ethical shift of focus from the overt capacities typically used in social interactivity to the minimally - shared ability allowing social interactivity at all (McGeer 2019; Krueger and Maiese 2018). Crucially, the toolkit provided here can be put to use in other disability studies as well, such as Tourette Syndrome or ADHD, and may serve as a general framework from which to analyse and help resolve social tensions. In general, this can bring ethics and empirical science more closely together via enactivism. However, we do not think a philosophical framework is the solution to inclusion. If this were the case the struggle for better inclusion of ‘physical’ disabilities would long be over. The solution lies in individuals ‘actively’ trying to overcome difference, thereby providing the evidence that this overcoming is possible and, hence, that the difference is not a difference in human quality but merely one of physical degree. Still, this framework can be, in our opinion, a good support for such activism in showing that inclusion should not be a mere targeting of individual difference but requires an effort on both sides of the interaction. Funding This work is partly funded (JB) by the NeuroEpigenEthics project that received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement 804881), and (TvE) by the Research Foundation Flanders (FWO), grant number 1124818N. 22 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Acknowledgements We want to acknowledge the invaluable comments of Kristien Hens, Inês Hipólito, Sander Van de Cruys and Damian Milton for helping formulate our argument as presented in this paper. Bibliography American Psychiatric Association (2013). Diagnostic and Statistical Manual of Mental Disorders (DSM-5§). Washington, DC: American Psychiatric Association Publishing. doi: 10.1176/appi.books.97808904 25596 Barandiaran, X. E. (2017). Autonomy and enactivism: towards a theory of sensorimotor autonomous agency. Topoi 36, 409–430. doi:10.1007/ s11245- 016- 9365- 4. Barnes, Elizabeth. (2016). The Minority Body: A Theory of Disability. Oxford University Press. Baron-Cohen, S. (2000). Theory of mind and autism: a review. Int. Rev. Res. Ment. Retard. 23, 169–184. doi: 10.1016/s0074-7750(00)80010-5 Bervoets, Jo, and Kristien Hens. (2020). Going Beyond the Catch-22 of Autism Diagnosis and Research. The Moral Implications of (Not) Asking ‘What Is Autism? Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2020.529193. Bolis, D., Balsters, J., Wenderoth, N., Becchio, C., and Schilbach, L. (2017). Beyond autism: introducing the dialectical misattunement hypothesis and a Bayesian account of intersubjectivity. Psychopathology 50, 355–372. doi: 10. 1159/000484353 Bruineberg, J., & Rietveld, E. (2014). Self-organization, free energy minimization, and optimal grip on a field of affordances. Frontiers in Human Neuroscience, 8, Article 599. Chapman, R. (2020). The reality of autism: on the metaphysics of disorder and diversity. Philos. Psychol. 33, 799–819. doi: 10.1080/09515089.2020.17 51103 Colombetti, G. & Krueger, J. (2015) Scaffoldings of the affective mind, Philosophical Psychology, 28:8, 1157-1176, DOI: 10.1080/09515089.2014.976334 Constant, A., Bervoets, J. Hens, K., and Van de Cruys, S. (2020). Precise Worlds for Certain Minds: An Ecological Perspective on the Relational Self in Autism. Topoi. https://doi.org/10.1007/s11245-018-9546-4. Curry, G. and Ravenscroft, I. (2002). Recreative Minds. Oxford University Press, Oxford. doi:10.1093/acprof:oso/9780198238089.001.0001 23 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. De Jaegher, H. (2013). Embodiment and sense-making in autism. Front. Integr. Neurosci. 7:15. doi: 10.3389/fnint.2013.00015 Di Paolo, E., & Thompson, E. (2014). The enactive approach. In L. Shapiro (Ed.), Routledge handbooks in philosophy. The Routledge handbook of embodied cognition (p. 68– 78). Routledge/Taylor & Francis Group Di Paolo, E., and De Jaegher, H. (2012). The interactive brain hypothesis. Front. Hum. Neurosci. 6:163. doi: 10.3389/fnhum.2012.00163 Di Paolo, E., Buhrmann, T., Barandiaran, X., (2017) Sensorimotor Life: An Enactive Proposal. Oxford University Press Di Paolo, E., Cuffari, E. C., & De Jaegher, H. (2018). Linguistic Bodies: The Continuity Between Life and Language. Cambridge, MA, USA: MIT Press. Frith, U. (1996). Cognitive explanations of autism. Acta Paediatr. 416, 63–68. doi: 10.1111/j.1651-2227.1996.tb14280.x Fuchs, T., and De Jaegher, H. (2009). Enactive intersubjectivity: participatory sensemaking and mutual incorporation. Phenomenol. Cogn. Sci. 8, 465–486. doi: 10.1007/s11097009-9136-4 Gallagher, S. (2004). Understanding Interpersonal Problems in Autism. Philosophy, Psychiatry, and Psychology 11 (3):199-217. Press. Gallagher, S. (2017). Enactivist interventions: Rethinking the mind. Oxford University Gibson, J. J. (1986). The Ecological Approach to Visual Perception. Psychology Press , London. Happé, F., and Frith, U. (2006). The weak coherence account: detail-focused cognitive style in autism spectrum disorders. J. Autism Dev. Disord. 36, 5–25. doi: 10.1007/s10803-0050039-0 Hens, K., and Langenberg, R. (2018). Experiences of Adults Following an Autism Diagnosis. Cham: Springer. doi: 10.1007/978-3-319-97973-1 Hipólito, I, Hutto, D, & Chown, N. (2020). Understanding autistic individuals. Cognitive diversity not theoretical deficit. In Rosqvist, H. B., Chown, N., and Stenning, A (eds), Neurodiversity Studies. A New Critical Paradigm. Routledge, London. https://doi.org/10.4324/9780429322297 Hohwy, J. (2013). The predictive mind. Oxford University Press. 24 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Hutto, D. D., & Myin, E. (2013). Radicalizing enactivism: Basic minds without content. Mit Press. press. Hutto, D. D., & Myin, E. (2017). Evolving enactivism: Basic minds meet content. MIT Hutto, Daniel D. (2003). Folk psychological narratives and the case of autism. _Philosophical Papers_ 32 (3):345-361. Jaswal, V., & Akhtar, N. (2019). Being versus appearing socially uninterested: Challenging assumptions about social motivation in autism. Behavioral and Brain Sciences, 42, E82. doi:10.1017/S0140525X18001826 Jones, R. M., Southerland, A., Hamo, A., Carberry, C., Bridges, C., Nay, S., Stubbs, E., Komarow, E., Washington, C., Rehg, J. M., Lord, C., & Rozga, A. (2017). Increased Eye Contact During Conversation Compared to Play in Children With Autism. Journal of Autism and Developmental Disorders, 47(3), 607. https://doi.org/10.1007/s10803-016-2981-4 Kenny, L., Hattersley, C., Molins, B., Buckley, C., Povey, C., and Pellicano, E. (2016). Which terms should be used to describe autism? Perspectives from the UK Autism Community. Autism 20, 442–462. doi: 10.1177/13623613155 88200 Krueger, Joel & Maiese, Michelle (2018). Mental institutions, habits of mind, and an extended approach to autism. Thaumàzein 6:10-41. Lyons V., and Fitzgerald M., (2012). Critical Evaluation of the Concept of Autistic Creativity, In Fitzgerald, M. (ed) Recent Advances in Autism Spectrum Disorders - Volume I, IntechOpen. doi: 10.5772/54465 Markram, K., and Markram, H. (2010). The intense world theory – a unifying theory of the neurobiology of autism. Front. Hum. Neurosci. 4:224. doi: 10.3389/ fnhum.2010.00224 Maturana, H. R., & Varela, F. J. (1980). Autopoiesis and Cognition—The Realization of the Living, ser. Boston Studies on the Philosophy of Science. Dordrecht, the Netherlands. McGeer, V. (2007). Why neuroscience matters to cognitive neuropsychology. Synthese 159, 347–371. doi: 10.1007/s11229-007-9234-1 McGeer, V. (2019). Scaffolding agency: A proleptic account of the reactive attitudes. Eur J Philos.; 27: 301– 323. https://doi.org/10.1111/ejop.12408 Milton, D. (2017). A Mismatch of Salience: Explorations of the Nature of Autism from Theory to Practice. West Sussex: Pavillion. 25 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Mottron, L., Dawson, M., Soulières, I., Hubert, B., and Burack, J. (2006). Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. J. Autism Dev. Disord. 36, 27–43. doi: 10.1007/s10803-005- 0040-7 Murray, D. (2018) Monotropism: An interest-based account of autism, Encyclopedia of autism spectrum disorders, Springer New York, 978–981 Noë, A. (2004). Action in perception. MIT press. O’Regan, J. K., & Noe, A. (2001). A sensorimotor account of vision and visual consciousness. Behavioral & Brain Sciences, 24(5), 939. Palmer, C. J., Paton, B., Kirkovski, M., Enticott, P. G., & Hohwy, J. (2015). Context sensitivity in action decreases along the autism spectrum: a predictive processing perspective. Proceedings of the Royal Society B: Biological Sciences, 282(1802), 20141557 Pellicano, E. (2013). Sensory symptoms in autism: a blooming, buzzing confusion? Child Dev. Perspect. 7, 143–148. doi: 10.1111/cdep.12031 Richman, KA, Bidshahri, R. Autism, theory of mind, and the reactive attitudes. Bioethics. 32: 43– 49. https://doi.org/10.1111/bioe.12370 Van de Cruys, S., Kelsey, P. & Hohwy, J., (2019): Explaining hyper-sensitivity and hypo-responsivity in autism with a common predictive coding-based mechanism, Cognitive Neuroscience, DOI: 10.1080/17588928.2019.1594746 Servick, K., (2019, October 1). Echolocation in blind people reveals the brain’s adaptive powers. Science Mag (retrieved 13th of June 2020), https://www.sciencemag.org/news/2019/10/echolocation-blind-people-reveals-brain-sadaptive-powers Shoemaker, D. (2015). Responsibility from the Margins, Oxford University Press Sterelny, K. (2010). Minds: Extended or scaffolded? Phenomenology and the Cognitive Sciences, 9, 465–481. Strawson, P. F. (2008). Freedom and Resentment and Other Essays. Routledge. https://doi.org/10.4324/9780203882566. Thompson, E. (2007). Mind in life. Harvard University Press. Valiyamattam, G. J., Katti, H., Chaganti, V. K., O’Haire, M. E., & Sachdeva, V. (2020). Do Animals Engage Greater Social Attention in Autism? An Eye Tracking Analysis. Frontiers in Psychology, 11, 1. 26 Penultimate draft. Accepted for publication in Topoi. Please cite the published version. Van de Cruys, S., Evers, K., Van der Hallen, R., Van Eylen, L., Boets, B., de-Wit, L., et al. (2014). Precise minds in uncertain worlds: predictive coding in autism. Psychol. Rev. 121, 649– 675. doi: 10.1037/a0037665 van Es, T. (2020a). Living models or life modelled? On the use of models in the free energy principle. Adaptive Behavior. https://doi.org/10.1177/1059712320918678 van Es, T. (2020b) Minimizing prediction errors in predictive processing: from inconsistency to non-representationalism. Phenom Cogn Sci 19, 997–1017. https://doi.org/10.1007/s11097-019-09649-y Van Goidsenhoven, L. and De Schauwer E. (2020), Listening beyond words: swinging together, Scandinavian journal of disability research, p. 330-339 van Grunsven, J. (2020). Perceiving 'Other' Minds: Autism, 4E Cognition, and the Idea of Neurodiversity. Journal of Consciousness Studies 27 (7-8):115-143 Varela, F. J., Thompson, E., and Rosch, E. (1991). The embodied mind: cognitive science and human experience. Cambridge, MA: MIT Press. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Wallace, R. (2019). The Moral Nexus. PRINCETON; OXFORD: Princeton University Press. doi:10.2307/j.ctv3znwhn Zahnoun, F. (2019) On representation hungry cognition (and why we should stop feeding it). Synthese. https://doi.org/10.1007/s11229-019-02277-8 Zaidel, A., Goin-Kochel, R. P., & Angelaki, D. E. (2015). Self-motion perception in autism is compromised by visual noise but integrated optimally across multiple senses. Proceedings of the National Academy of Sciences, 112(20), 6461-6466 27

RELATED PAPERS

RELATED TOPICS

Log In

Autism as gradual sensorimotor difference

Autism as gradual sensorimotor difference

Related Papers

RELATED PAPERS

RELATED TOPICS