THE BOUNDARIES OF AFFECTIVITY (AND A CODA ON EXTENDED CONSCIOUSNESS) 1

In this contribution I will offer a rough guide through “situated affectivity”, a small galaxy of theories, ideas, intuitions which are recently growing at the boundaries between philosophy of emotion, philosophy of mind, philosophy of cognitive science. In the last years, we are witnessing a real “affective turn” in the domain of analytic philosophy of mind and cognitive science, and situated affectivity is an excellent example of what is going on in the field. In the first part, I will present the main versions of situated affectivity, distinguishing between the ideas of situated or scaffolded emotions (Griffi-ths, Scarantino 2009; Colombetti, Krueger 2015) and properly extended emotions (Carter et al. 2016; Colombetti, Roberts 2015), and showing their link to the corresponding idea of the mind: scaffolded, extended or otherwise. In the second part, I will further explore the connection between situated affectivity and the domain of extended mind: I will argue that, even if situated affectivity could be in principle evaluated independently from any particular theory about the mind, dealing with situated affectivity is, beside its independent interest, also a good way to tackle an unsettled problem in the literature on extended mind, namely extended consciousness.

In this contribution, my aim is twofold. On one hand, I would like to propose a certain classification of the landscape of studies on situated affectivity, a recent and prolific development at the border between philosophy of emotion and situated cognition. On the other hand, I would like to highlight and offer some comments on the connection between extended affectivity and an unsettled topic in the extended mind debate: namely, extended consciousness. I think that the topic of extended affectivity -a particular version of the more general thesis of situated affectivity -should be considered an occasion to come back to this "hot" and unsettled topic.

From situated cognition to situated affectivity
In order to fully understand the proposal of "situated affectivity", we will need to introduce some background of philosophy of emotions and philosophy of cognitive science. In general, we can say that the theoretical proposals concerning the topic of situated affectivity aim to focus on the contribution of the external environment, both physical and social, to our affective life, from both a diachronic and synchronic point of view (cf. Griffiths, Scarantino 2009). From the diachronic point of view, it is said that environmental resources contribute fundamentally to the development of our affective repertoire, i.e. the set of dispositions and affective states that characterize us; from the synchronic point of view, on the other hand, it is noted that these environmental resources are involved in the management and development of individual emotional episodes.
To argue that affective phenomena are situated is to say that they are dependent (in the weaker version) or partly constituted (in the stronger version) by elements of the external environment, or by bodily interaction with them. The theses attributable to situated affectivity range from weaker proposals about the role that certain objects, for example "affective scaffolding", play in the development and management of our affective experiences, to more challenging theses concerning the alleged extended ontology of affective states, such as the so-called extended emotions.
It is important to emphasize how the different declinations of the general thesis of situated affectivity are related to the different views of the mind offered in the field of situated cognition, or 4E cognitive science (embodied, embedded, extended, enacted). It is therefore impossible to understand the situated affectivity thesis without referring to situated cognition. I will deal with this in the remainder of this paragraph. In paragraph 2 I will offer a general characterization of situated affectivity, both in its scaffolded version, inspired by the concept of affective scaffold and other similar notions (cf. Colombetti, Krueger 2015;Piredda 2020;Sterelny 2010; below §2.1) and extended versions: the first inspired by the extended mind thesis proposed by Clark and Chalmers (1998) and based on the principle of parity (cf. Carter, Gordon, Palermos 2015;Colombetti, Roberts 2015;below §2.2) and the second connected to the enactivist theses and based on integration, related both to the theme of extended consciousness and to issues of social and political interest (cf. Colombetti 2014Colombetti , 2017Krueger, Szanto 2016;León, Szanto, Zahavi 2019;Slaby 2008Slaby , 2014Slaby , 2016.
In this section, I would like to briefly introduce the elements necessary to understand the general idea of situated affectivity, starting with the reference to situated cognition first, and then introducing the philosophical characterization of affectivity. I will thus provide the minimum contextual tools to understand what is meant by the phrase "affective phenomena are situated".
According to a well-established interpretation, from the 1990s onwards there has been a shift from the so-called "classical cognitive science", built around the computational and representational conception of mind, to a new version of cognitive science, which has been called new, post-classical, embodied or, more recently, 4E cognitive science, referring to the labels embodied, embedded, extended, enacted (Chemero 2009;Newen et al. 2018;Robbins, Aydede 2009;Rupert 2009). Opinions differ on the radicality of this shift, and its distance from the computational and representational model, but not on whether it has changed the way we look at the contribution of brain, body, and environment in human cognitive and mental activities. To quote the title of a famous book that helped crystallize this new view of the mind -Being There. Putting brain, body and world together again, by Andy Clark (1997) -the attempt has been to "put brain, body and world together again", overcoming the neurocentrism associated with the representational and computational view of mind (see also Clark 2008).
Whereas in the so-called "sandwich" model (Hurley 1998), cognition was seen as the middle layer between perception and action, the new perspective emphasizes that these three layers are deeply interconnected. Visual perception and action are more intimately linked than previously thought -see Ballard's (1991) animated vision studies and O'Regan and Noë's (2001) sensorimotor conception of vision. It is possible to build robots that adapt their behavior to the environment and do not act only by following instructions given from above, as in Brooks' (1991) situated robotics.
In general, humans use the external environment as a mnemonic resource that compensates for their limited biological resources (Donald 1991), they personalize and "engineer" it to meet their needs, establishing a relationship of strong dependence with some environmental resources, which qualifies them as natural-born cyborgs (Clark 1997(Clark , 2003(Clark , 2008. The intimate connection between the individual and the environment allows us to adapt the concept of ecological niche to the plane of cognition: Sterelny (2010) thus introduces the idea that human beings, by customizing their environment according to their needs, build "epistemic niches". The external supports, specially designed or occasionally exploited for their own purposes, are called "cognitive artifacts" or "cognitive scaffolds", using, in the latter case, a concept originally introduced by Lev Vygotskij (1986) in his theory on the zone of proximal development. We thus speak of a scaffolded mind, i.e. supported by external scaffolding, a theory of the functioning of the human mind that does not necessarily claim to offer a new ontology of the mind, as the model of the extended mind proposes to do. According to the latter, we should rethink the boundaries of our minds, beginning with observations of the intimate connections between mind and environment in everyday cognitive routines (Clark, Chalmers 1998;Clark 2003Clark , 2008. The theory of the extended mind is based on the principle of parity, according to which, with respect to a given task, if an environmental resource plays the same cognitive role as it would if it were an internal resource, contained in the brain box, then we should extend the boundary of the cognitive system to include the former. In other words, the vehicles through which cognitive and mental activity takes place can extend beyond the biological organism, provided certain conditions are met (several criteria have been proposed, in addition to the aforementioned parity principle). As an alternative to the parity principle, it has been proposed to base the identification of extended or hybrid systems, composed of an intertwining of biological and technological, on considerations of the degree of integration or complementarity between two systems or parts of systems (Menary 2007(Menary , 2010Sutton 2010).
In general, both the cognitive scaffolding-inspired and the extended view offer us a look at the mind that recognizes the deep debt of human beings to the environment, and vice versa: an environment that is actively manipulated by the agent to achieve its ends, in a constant interaction. This is a very brief presentation of the central themes related to situated and embedded (embodied) cognition (cf. Robbins, Aydede 2009; Wilson, Foglia 2017), which will serve to understand the proposals developed in the field of affectivity. We can ask, more specifically, how situated cognition relates to reflection on affectivity.
In the Western philosophical tradition, emotions and passions have generally been characterized as opposed to reason and outside the control of the agent, linked to predominantly bodily and instinctual aspects. It is not that the centrality of emotions has not been recognized in the philosophical field, but the reassessment of the importance of affective states, understood as complex states composed of different components (bodily, subjective/experiential and cognitive), is to be considered an achievement of contemporary affective sciences, and has also had its impact in philosophical reflection on affectivity (for general introductions to this field, see Brady 2019; Deonna, Teroni 2012;Scarantino 2016). A possible way to broadly reconstruct the landscape of philosophical positions on affectivity is to individuate three different traditions: the cognitivist-evaluative tradition, according to which affective states are essentially cognitive states of evaluation of situations in which the agent is embedded (the process of perception/evaluation of the situation is commonly called appraisal); the motivational tradition, for which the essence of affective states is the drive towards action; the feeling tradition, according to which what most characterizes affective states are the sensations that accompany them, or that constitute them. We will see that the various attempts to provide a situated view of affectivity address all these components of affective states. Just as it is possible to say that "situated cognition is the genus, and embodied, enactive, environmentally embedded, and distributed cognition are the species" (Robbins, Aydede 2009, p. 3), similarly one can bring within the category of situated affectivity the different ways in which affective states, in their multiple components, make use of the external, physical, and social environment to manifest themselves (see §2.1) or extend (see §2.2- §2.3). In the next section I offer an exploration of the varieties of situated affectivity. Traditional philosophical theories on emotions, ascribable to the cognitivist or Jamesian tradition, abstract them from the physical and social environment in which they occur, and which influences their unfolding in time and space. According to Griffiths and Scarantino (2009), emotions would be primarily designated to play a social role, rather than contributing to cognition or individual decision making. In addition to conveying information to the individual's decision-making system about a given situation, they should therefore be thought of as acts of reconfiguring relationships generated by conveying a social signal. They would not be mediated by conceptual thinking and would develop thanks to the network of affective supports offered by the environment, both physical and social, which influences the evolution of emotional episodes.

Varieties of situated affectivity: scaffolded and extended affectivity
The situated perspective on emotions is not intended to oppose or reject the findings of previous perspectives, but to integrate them into a dyna- shower and singing in front of an audience, or the fact that we smile more when we know we are not alone. This reinterpretation of emotions in a strategic-social sense has important similarities with the shift from the individualistic cognition of classical cognitive science to the situated perspective on cognition, although it is correct to point out that the situated affectivity thesis can be considered independent from situated cognition, and can therefore be evaluated independently. Emotions (or rather, their manifestation) would be strategic moves in the context of a social negotiation; therefore, they would not simply be responses to the current situation, but also responses that attempt to maximize the probability of achieving a given goal. For example, manifesting anger or sadness might be a good way to gain a reward, just as sulking during an argument might be interpreted as a strategy to gain some concessions within the relationship.
Thus, a situated perspective on emotions emphasizes the role of social context in the production and management of affective states, and the reciprocal influence of emotions in the evolving social context. Behaviors traditionally interpreted as the involuntary expression of an individual's psychological state -smiling, crying, getting angry -are interpreted as signals intended to influence the behavior of other individuals, or strategic moves in ongoing transactions. For example, smiling is interpreted as an affiliative gesture more than it is a mere expression of hilarity (Griffiths, Scarantino 2009, p. 440). In this sense, the situated perspective departs from the Darwinian conception of emotions, according to which the expression of emotions is an involuntary event and their recognition is inscribed in our biology (an idea echoed by basic emotion theory, see Ekman 1972).
The emphasis on the active character of the agent in strategically manipulating the manifestation of their affective states for social purposes and the manipulation of the environment itself to their own advantage, through the creation of epistemic niches, recalls the view of the mind inspired by cognitive scaffolding proposed by Kim Sterelny (2010) and briefly introduced in §1.

Scaffolded affectivity: from affective scaffolds to affective artifacts
Bringing together these two perspectives -the theory of the scaffolded mind and the view of situated emotions -Giovanna Colombetti and Joel Krueger (2015) propose to introduce the concept of "affective scaffolding" to account for the way in which agents manipulate the environment around them also for affective (and not only cognitive) purposes. In this way, the concept of niche is also extended to the field of affectivity, where affective niches are defined as "self-styled environments providing the developmental conditions for affective states to take shape and thrive" (see also Colombetti, Krueger, Roberts 2018, p. 1).
Affective scaffolds are distinguished into material supports and interpersonal supports, and in both cases are characterized according to certain dimensions, identified by Sterelny (2010) in the context of cognitive scaffolds: the degree of trust and reliability, the personalization and consolidation of the bond with the individual, and finally, only in some cases, the degree of sharing with other agents. Some affective scaffolds are particularly reliable because we can count on their effect whenever we turn to them. For example, I can count on a certain song or singer to relive emotions related to a past episode, or to regain vitality after a difficult moment.
Something similar happens when we scroll through the photos stored in our smartphones, evoking memories and feelings related to them. The smartphone is undoubtedly an interesting example of affective scaffolding, for its characteristic of storing many contents that are important to us: from our favorite music to photos, to messages that we have decided not to delete. We often turn to the smartphone for affective reasons, even if only to fight boredom and search for small informative or social rewards.
There is no need, however, to invoke technology to find examples of affective scaffolding or support: other fitting examples may be clothes, which we choose according to our mood and sometimes to achieve a certain effect on it, paying attention to colors and textures, or accessories, such as bags, identified as real instruments of mood regulation (Kaufmann 2011).
Instead, among non-portable material objects we find cinemas, concert halls, and nature itself: these are all environments that we choose to visit (also) because of their effect on our affective states.
The case of the relationship between a musician and his musical instrument is a good example of a deep interconnection -material and affective -between an agent and an environmental resource. The musician expresses his or her affective states through the use of the instrument, and this almost seems to represent an extension of the musician's body, which in effect adapts to it over time. Something similar can be said of the kitchen for cooks or the tennis racket for tennis players. The topic of incorporation, related to the phenomenological tradition, is well systematized in Colombetti (2016). Colombetti and Krueger's (2015) proposal, however, is not limited only to material supports; affective scaffolding is also identified in the interpersonal realm, based on the relationships we choose to have and turn to in certain circumstances. We undoubtedly rely on relatives and friends to lift our moods or to let off steam, to spend time in peaceful company, because we know what effect seeing or hearing a certain person will have on us. Sometimes we modulate our affective states according to the company we are in, we could say according to the "affective tone" of the group. We adjust, for example, to the wit and humor that works best or is most appropriate in a certain group, thus achieving an effect on an emotional level (feeling accep-ted or instead judged for being inappropriate). Interpersonal supports can also be distinguished according to the dimensions of trust or familiarity and involvement/personalization, as in cases where we have a fixed appointment with a friend to go to the bar at a certain time or even just to phone each other.
The last case considered by Colombetti and Krueger (2015) is affective scaffolding that is shared by a group of people (see also von Scheve, Salmela 2014). This is the case of religious, political, geographic, sport, and occupational symbols and environments: affective supports to which entire communities turn, sharing deep emotional experiences, which can sometimes take the form of collective emotions.
In conclusion, according to Colombetti and Krueger, "in order to understand and explain affective phenomena we need to consider the ways in which agents 'engineer' their affective environments -i.e., create affective niches -and in doing so allow these environments to influence their affective states on an ongoing basis" (2015, p. 1160).
A further proposal that fits into this landscape, and which has the advantage of being able to rely on an extensive supporting literature regarding the notion of artifact and cognitive artifact (cf. Heersmink 2013, Hilpinen 2011, is to speak, in the case of some affective supports, of "affective artifacts" (Piredda 2020). Affective artifacts would be material or immaterial objects that, for their designated purpose or instead only for the development of an idiosyncratic function (say, by serendipity), are habitually used by a certain agent to regulate his or her affective life. Objects specifically designated for this purpose include photo albums, wedding rings, souvenirs, and children's stuffed animals. Any object, however, can become an affective artifact without being designated for that purpose: just think of the cases in which we become attached to an object that is completely insignificant from an economic point of view, and we keep it for a long time, just because we have attributed to it a high affective value linked to some memory. Of course, the most representative examples of affective artifacts are to be found among personal objects, and this is also why the importance of these objects in reconstructing one's autobiographical narrative, and in consolida-ting a sense of self, has been emphasized (Piredda 2020, §6; but also Candiotto, Piredda 2019; Heersmink 2017Heersmink , 2018Heersmink , 2020James 1890).
The main difference between scaffolds and affective artifacts lies in the more specific and limited nature of the notion of artifact compared to that of scaffold. This can be seen as an advantage or a disadvantage. On the one hand, talking about artifacts per se excludes the possibility of considering the support of other social agents, which is instead included in the analysis of affective scaffolds. On the other hand, referring to a more specific and theoretically defined notion allows to avoid some aspects of vagueness and indeterminacy that the notion of scaffold could raise: given its breadth, it is in fact difficult to give a precise ontological characterization of the concept of "scaffold" and in general of "support". The discussion on these points is still open and some reflections in this sense can be found in Colombetti (2020) and Saarinen (2020).
Not surprisingly, the discourse on affective scaffolding has also turned towards the consideration of broader environmental factors, such as institutional, or work contexts, which also contribute, to manipulate our affective practices by influencing them, both on the part of the agent (bottom-up) and on the part of the institutional contexts (a work environment, an exhibition, the school context, the environments offered by digital technologies, thus assuming a top-down direction). To account for these phenomena, the concepts of techno-social niches (Krueger, Osler 2020), mind-invasion (Slaby 2016), and affective arrangement (Slaby et al. 2019) have been proposed. In particular, the notion of affective arrangement aims to recover in the current debate a tradition at the border between cultural studies, phenomenology and existentialism, emphasizing the relational, dynamic and performative nature of the perspective of situated affectivity. The scaffolded mind model emphasizes the way in which the biological mind relies on the external environment to meet its needs. This view of the mind remains neutral with respect to the ontology of the mental, which is instead strongly revised in the theoretical proposals that pertain to the model of the extended mind (Clark, Chalmers 1998;Clark 2003Clark , 2008Menary 2010). In this context, the interaction between mind and certain environmental resources is interpreted as a "constitutive" phenomenon of the men-tal, and not as a mere causal interaction, however close. This leads to reconsider the spatial and ontological boundaries of the mental, which sometimes transcend the limits set by skull and skin, as in the canonical examples of "extended mind". In the famous mental experiment devised in support of the extended mind, two fictional characters -Otto and Inga -differ only in the location where their memories are stored: biological memory for Inga and a notebook for Otto, who suffers from a mild form of Alzheimer's disease. All other things being equal, this difference, according to Clark and Chalmers (1998), is not enough to exclude the information contained in Otto's notebook from the set of his dispositional beliefs. According to this interpretation, we can therefore say that some of our mental states -such as Otto's -are realized on non-biological substrates, but involve the material world in which we are immersed: fragments of notes, photographs, calculators, smartphones, and -in the particular social declination of the extended-mind thesis (see e.g. Gallagher 2013) -also other social agents or constructs. In the literature on the extended mind, this reasoning is applied to the case of cognitive states and processes, but what about affective states? Is it possible to interpret affective phenomena as extended processes as well, in a similar sense? The extension of affectivity has been proposed in various forms, as we will see in the next sections.

Extending the extended mind: extended affectivity
Regarding the possibility of "extending affectivity", one finds a hint in David Chalmers' preface to Supersizing the mind (Clark 2008), when he writes: It is natural to ask whether the extended mind thesis might itself be extended. What about extended desires, extended reasoning, extended perception, extended imagination, and extended emotions? I think there is something to be said for each of these. Perhaps the camera on my iPhone can serve as an extended perceptual mechanism. And perhaps one might have something akin to an extended mood, if not an extended emotion, when one's environment is always nudging one toward happiness or sadness (Chalmers 2008, p. xiv).
It is with later work, however, that the issue of extended affectivity is specifically addressed. Giovanna Colombetti and Tom Roberts (2015) are the first to systematically defend the possibility and indeed the necessity of applying Clark and Chalmers' (1998) arguments not only to cognitive but also to affective states. A specific version of extending emotions, with particular reference to their cognitive-evaluative component, is instead brought forward by Carter et al. (2016). Colombetti and Roberts's (2015) argument in favor of extended affectivity is as follows: if vehicular externalism -the idea, that is, that the material vehicles that realize the mind encompass not just brain activity, but also that of the body and the material environment (p. 1243) -is true for cognitive states, then it must also be true for affective states, and thus affectivity is extended. The authors argue that proponents of the extended mind "should also accept that the vehicles of emotions, moods, sentiments, temperaments, and character traits can extend beyond skull and skin." (p. 1243) by bringing two scenarios to the reader's attention: the first concerns occurrent affective states, the second affective dispositions. In sum, Colombetti and Roberts (2015) propose to extend the thesis of vehicular externalism to emotional and affective states, both dispositional and occurrent.
The case of dispositional affective states, such as emotional dispositions or tendencies, temperaments, feelings, moods, or character traits, is constructed in analogy to the thought experiment of Otto and his notebook (Clark and Chalmers 1998). Eve is a girl who resents her parents, and this resentment, the authors argue, is realized in part through the presence of her secret diary, where Eve records the events that generate this emotion ("my father is disinterested in my feelings" or "my mother is never satisfied with my accomplishments"), venting and making the diary a central element of her dispositional emotional state. Reinier's case, on the other hand, concerns the possibility that our temperaments, the tendency, that is, to have certain moods, are supported, or even made possible, by the presence of adequate material supports, such as a notebook where Reinier, a boy with a melancholic personality, collects useful material to motivate and inspire him, trying to keep his mood and self-esteem high. Without this notebook, Reinier would sink back into self-pity. Similar cases can be constructed to show how even feelings and character traits can depend in a narrow sense on the presence of adequate material supports.
The case of occurrent emotional states is more controversial, and comes closest to the proposals we will explore in §2.3. In the model of the extended mind, the extension is not only about dispositional states, but also about some occurrent states. Of course, the way to defend this thesis is necessarily different, and refers to those cases of "coupling" and self-stimulating loops very often illustrated by Andy Clark (2008). Returning to the affective domain, then, it is a matter of identifying scenarios in which "recognizably occurrent affective processes are driven, and boosted, by a subject's self-stimulating loops of interaction with worldly materials, such that any intuitive disqualification of the extra-neural components from the proper parts of the system seems to be grounded in internalist bio-prejudice" (Colombetti, Roberts 2015, p. 1248). In the case of occurrent states, it is ne-  Carter, Gordon, and Palermos (2016), who argues for the extension of emotional states -conceived in a cognitivist framework -through the application of the extended cognition thesis to these states. The result is the defense of a circumstantial thesis about the extension of emotions, according to which "some emotions partly supervene on elements of the world external to the organism" (p. 198-99). In this proposal, extension applies specifically to the notion of appraisal, that cognitive process that helps to cause a judgment or evaluation. If, therefore, a cognitive dimension is necessarily involved in emotions, and if this cognitive dimension is extended, then emotion will also be extended. in the examples of cognitive extension offered by cognitive prostheses). Already now, moreover, any artificial device paired with the organic body to support a specific function (regulate heart rate, release hormones) would count as an extension of the autonomic system (Colombetti, Roberts 2015, p. 1258Stephan et al. 2014, p. 11). In this regard, it is worth noting that, with regard to the embodied dimension, the domains of cognition and emotion differ substantially, as has been rightly noted by Krueger (2014). Indeed, while in the case of cognition the thesis of embodiment represents a thesis to be defended, opposing the traditional view of cognition as fundamentally localized in the brain, arguing that emotions are embodied is a much less challenging thesis, because the idea of this embodiment somehow appears to be part of our intuitive conception of emotions (Krueger 2014, pp. 534-535;Stephan et al. 2014, p. 6).
On the extension of qualitative experiences, or feelings, the question is certainly more complex, and calls into question that controversial branch of the extended mind debate that concerns the possibility (or not) of extending conscious phenomena, getting closer to enactivist and sensorimotor approaches (cf. Chalmers 2008Chalmers , 2019Clark 2009Clark , 2012Noë 2004;O'Regan, Noë 2001;Pepper 2014;Ward 2012). On this, Clark and Chalmers have consistently expressed conservative views, arguing for them in various ways, and an independent debate on extended consciousness has developed inspired by enactivist interpretations of mental extension, focusing on the inseparability of thought from action and the concept of integration (cf. Gallagher 2017; Kirchoff, Kiverstein 2019). On this point, Colombetti and Roberts (2015) also remain more cautious -the thesis that the basis of su-pervenience of our qualitative experiences is extended is highly controversial -and merely defend the idea that the external environment should at least be acknowledged to play a fundamental role in explaining complex phenomena such as the evolution of our sensations, for example, during a concert or a jazz improvisation session. Instead of defending an ontological thesis about the presumed extended nature of sensations, they prefer to opt for a weaker position, limited to the epistemological field.
Evaluating Colombetti and Roberts' (2015) proposal on the extension of affective phenomena, we can say that the most acceptable versions of extended emotion concern bodily aspects and cognitive and evaluative aspects. With regard to sensations (feelings), however, the authors' position is (understandably) much more cautious, and less revolutionary (but see below, §3, where Roberts (2015) is discussed). However, the theme of the extension of emotions is also taken up in other theoretical frameworks, which refer to the enactivist tradition, emphasize integration instead of parity as a guiding principle to evaluate cases of extension, or support some version of the thesis regarding the social extension of mental states, whether cognitive or affective. These other theoretical frameworks are addressed in the next section.

3. Enactivist, integrationist and social extension
This paragraph is devoted to the presentation of approaches that refer to the enactivist tradition represented, in the case of affectivity, by the works once again of Colombetti (in particular 2014 and and to an integrationist-based principle of affectivity extension, alternative to the principle of parity (see Slaby 2008Slaby , 2014Slaby , 2016. A very brief mention will be devoted to the option of a "social" extension of emotions which, echoing the idea of a socially extended mind (cf. Gallagher 2013), proposes to discuss the possibility of interindividual extensions, configuring collective or shared emotions (Krueger, Szanto 2016;León, Szanto, Zahavi 2019). Colombetti (2014) attempts to integrate the enactivist and phenomenological perspective with an analysis of affectivity inspired by dynamical systems theory, which -as in the case of cognition -emphasizes the tempo-ral nature of affective processes, along with the close integration between agent and environment. In this framework, affective episodes are defined as "self-organizing patterns of the organism" and are described as contextdependent and flexible; despite this, they show a certain stability. As we know, both enactivism and dynamical systems theory tend to blur the boundary between agent and environment, configuring cognition and affectivity as complex phenomena, describable by referring to state variables referred to the whole organism-environment system. Therefore, this view of affectivity represents a further possibility to extend affective states beyond cerebral and even biological realization, given that the enactment of mental states, cognitive or affective, also directly involves the interaction between the organism and the environment as its constitutive part (the latter thesis is explicitly defended in Colombetti 2017).
In general, the phenomenology-inspired enactivist perspective questions the very presence of a precise distinction between cognition and affectivity. This thesis is also supported by Jan Slaby (2014) Clark and Chalmers (1998), which excludes the pos-sibility of extending conscious phenomena. The intention is to approach an enactivist, phenomenological, neo-existentialist reading, within which it is possible to conceive, through the cases of synchronic extension based on integration, an interpenetration between agent and environment also with regard to qualitative states. In cases of "phenomenal coupling", "part of the dynamic process of emotion can both originate in and be dynamically sustained and guided by processes that take place in the environment of the person experiencing the emotion, as when one is moved by a sad movie or driven to anger or even aggression under the influence of a raging crowd" (Slaby 2014, p. 39). And again: Phenomenal coupling is the direct, online engagement of an agent's affectivity with an environmental structure or process that itself manifests affect-like, expressive qualities -be it in the form of an affective atmosphere (Anderson, 2009;Schmitz et al., 2011) or as a dynamic gestalt feature of a different kind, such as an expressive quality of a piece of music (Levinson, 2009). The most relevant range of examples is in the social-interactive domain: nothing is as emotionally engaging as the expressivity of fellow humans -individuals as well as groups can draw us into emotional experiences that we would not be able to experience on our own. Another key range of examples is found in contexts of art reception, as many of the emotions we experience in response to dynamic art-forms such as music, film, theater, or dance are likewise such that their full phenomenal quality cannot be characterized without recourse to the expressivity of the artworks themselves. (Slaby 2014, p. 41).
As can be seen, Slaby insists heavily on the dynamic and temporally extended, socially and culturally complex character of affective states. A particularly interesting note of Slaby's (2016)  it is possible for the individual mind to extend, but the idea of a socially shared mind has remained beyond their predictions and expectations. For this aspect, they have been criticized by Shaun Gallagher (2013Gallagher ( , 2017Gallagher ( , 2018, who introduced the proposal of a socially extended mind, which would transcend the limits of individual minds and be located, for example, at the level of institutions or the legal system as a whole. Now, setting aside the particularly controversial nature of this hypothesis, of which I have given only a hint, it is worth asking, in the case of affectivity, how this thesis might be declined. Krueger's work (2013, p. 863) gives us a clear overview of the different hypotheses in the field, which, starting from a general thesis, the extended emotions thesis -according to which "some emotions incorporate external resources and thus extend beyond the neurophysiological boundaries of organisms" -also comes to examine the case of "collectively extended emotions". Within the environmentally extended emotions, in fact, Krueger distinguishes between individually extended emotions, defended through the example of music as an affective scaffolding, and collectively extended emotions, which he does not defend, except in the specific case of the infant-caregiver relationship, characterized by affective integration and emotional convergence. On the other hand, a rigorous treatment of the issue of shared emotions cannot be separated from reference to the contribution that the phenomenological tradition has made to this issue (see León, Szan-to, Zahavi 2019).

Into the wild: situated affectivity meets extended consciousness
The last point I would like to discuss in this paper concerns a question that has been just touched in the original paper on the extended mind by Clark and Chalmers (1998) and then has been handled again in several paper by Andy Clark (2009; and David Chalmers (2008;: the question of extended consciousness. I think that the topic of situated affectivity, and particularly the case of extended affectivity, represents a good occasion to recover the question of extended consciousness, for reasons that I will try to make explicit in this last paragraph. Let me begin from the outset, and present the problem of extended consciousness, which in itself is not always crystal-clear. Let me begin by saying that according to Clark and Chalmers (1998), extended cognition and mind do not affect conscious states. Consciousness remains internal to the boundaries of the individual, for reasons of bandwidth, direct access and in general for lack of compelling arguments (Clark 2009(Clark , 2012Chalmers 2008Chalmers , 2019. In the last years, this resistance to extend consciousness has been criticized by many authors from different points of view: from sensorimotor accounts of cognition (Hurley and Noë 2003;Noë 2004;Pepper 2014;Ward 2012); using parity considerations (Vold 2015); from the point of view of predictive processing (Kirchoff and Kiverstein 2019); and also from an internalist point of view (Farkas 2019).
The discussion in the literature has been focused on two main points: 0 Is consciousness extended? The first three cases are attempts to defend a form of extended consciousness.  (2015), Extending Emotional Consciousness, in which the author analyzes the connection between the sensorimotor account of perception (Hurley and Noë 2003;Noë 2004) and the hypothesis according to which emotional consciousness, that is the consciousness/awareness associated with an emotional episode, may be considered as extended. After having analyzed in detail the kind of relationalism that the sensorimotor account represents in the domain of the theories of perception, Roberts' conclusion is actually negative on the possibility that this could represent a good way to argue for extended consciousness. With two interesting exceptions, which recall Slaby's concept of phenomenal coupling, and even Colombetti and Roberts' consideration of the possibility of extending phenomenal consciousness.
The first exception -that is, a positive example of how emotional consciousness can be said to extend -regards some cases of emotional regulation. Emotion regulation is a process in which it may happen that "part of the world is modified, confronted, attended to, or manipulated in order that it have a certain effect upon one's emotional consciousness" (Roberts 2015, p. 7 of the penultimate draft). In most cases, the interaction with the world is purely causal and unidirectional: a change in the world generates a change in the subject's experience. Other cases are nonetheless more interesting for our purposes: these are scenarios in which objects outside in the world play a more significant role in explaining the temporal and phenomenological development of the emotional experience.
Here, there is a pattern of dynamic, temporally extended activity in which subject and object are densely and tightly woven together in the creation and expression of an emotional experience, such that there are loops of bi-directional causal influence spanning brain, body, and environment affordances are -according to Gibson -not representational devices, rather directly perceived, it will be the case that the external object responsible of the new perceived affordance will be part of the supervenience basis of the affordance itself. And in this sense, it will be part of the supervenience basis of the emotional experience, thus extending it beyond the organismic boundary. To accept this second case, one has to accept the notion of affordance interpreted in Gibsonian terms, and also to accept that a change in the space of affordances is part of the phenomenal emotional experience. For this reason, I find this second example slightly more problematic than the first one, which is a more classical example of extension (it is built on the same drives that are used even in cognitive cases of extension by Clark and by the defenders of the dynamical system theory applied to cognition). Nevertheless, I think that Roberts is on the right track in order to explore in a rigorous manner the problematic question of extended consciousness applied to the domain of affective states and that this line of inquiry would merit further reflections.

Conclusion
In this paper I offered a possible systematization of the recent fra- In the last paragraph of the paper I presented some reflections on the possible interactions between the idea of extended affectivity and of extended consciousness: I believe that the case of emotions and affective states can be a good test for the controversial hypothesis of extended consciousness.