Layers of the Skin: Three Perspectives on Tactile Sensation in Autism Open Access

Already from the first minutes and hours of an infant’s extrauterine life, the sense of touch is the primary medium of its experience. Tactile activities such as affectionate soothing, stroking, kissing, holding, rocking, playful tickling, and the everyday tactile tasks of grooming and changing consume almost the entirety of the infant’s sensory world. As such, the developmental significance of touch can hardly be overstated. The idea of touch as a first and primary medium for early social experience and development has recently prompted an interest in exploring alterations of tactile processing in autistic persons. Given both the developmental nature of autism, its connection to differences in social behavior, and the prevalence of social touch aversion, the idea that alterations of tactile processing could underlie autistic development is an intriguing hypothesis (McGlone et al., 2014, p. 748).

The recent upsurge of interest in tactile sensation in autism is incited by the recognition of sensory differences in autism that, despite historically having been downplayed in favor of cognitivist models of explanation (e.g., Baron-Cohen et al., 1985), are estimated to occur in 70–90% of autistic persons (Kirby et al., 2022; Leekam et al., 2007). Autistic persons have consistently reported profound differences in tactile sensation, including amplified sensitivity to light touch, preference for deep pressure, atypical sensitivity to pain, and both strong aversion to and preference for specific textures of different fabrics and foods. Tactile defensiveness and aversion to social touch are observable already during the first year of life, thus far preceding the social behaviors forming the basis of an autism diagnosis (Baranek, 1999). Such evidence has motivated some authors, especially in the field of somatosensory research, to propose atypical tactile processing as a developmental catalyst of autism (Musser, 2019).

Scientific studies of tactile sensation in autism are in part motivated by first-hand accounts derived from both qualitative studies and the growing body of autobiographical literature describing hypersensitivity to touch as a key factor in atypical social behaviors. For example, Temple Grandin described how being hugged “was like a great, all-engulfing tidal wave of stimulation” to which she reacted “like a wild animal” (Grandin, 1996, p. 58). Such vivid portrayals of the experience of touch are highly cited in contemporary somatosensory literature but merely as descriptive examples. Many of such studies are developed out of a wish to move beyond subjective report, often construed as anecdotal evidence, and instead aspire to measure tactile processing objectively through psychophysical assessment (Mikkelsen et al., 2018; Puts et al., 2014). However, as Cascio et al. (2008) note, such study designs can end up far removed from the affective dimension of touch as well as the everyday social settings in which touch is experienced as problematic by autistic persons (Cascio et al., 2008, p. 134).

One research tradition that has dealt systematically with touch as an arena for emotional, social, and psychological development is psychoanalysis. Building on the pioneering works of Bick (1968) and Klein (1930), a number of psychoanalytic thinkers have thematized the significance and function of the skin for the infant’s rudimentary sense of self and other (Brenner, 2020; Meltzer, 1975; Ogden, 1989; Tustin, 1986). Psychoanalysis remains a controversial framework in autism research and for good reason following its historical connection with the “refrigerator mother” theory. However, despite this problematic role that psychoanalysis has played in autism research and practice, this tradition has nonetheless made a rich contribution to the understanding of the role played by bodily sensation in autistic development that precedes and anticipates the recent focus on sensory differences in mainstream autism research.

Summing up, the experience of touch in autism has attracted attention from three different perspectives: One, a highly empirical body of somatosensory research focusing on differences in tactile processing, two, a predominantly clinical and theoretical tradition of psychoanalytic literature focusing on the significance and function of the skin for the experience of self and other, and three, a growing body of autobiographical works by autistic authors containing detailed descriptions of the subjective experience of touch. Building on the skin’s stratification into epidermal, dermal, and hypodermal layers, it could tentatively be suggested that these three perspectives represent layers of tactile sensation that elaborate different dimensions of the tactile opening toward the world.

Despite a shared interest in how alterations in tactile sensation shape the development and experience of autism, somatosensory, psychoanalytic, and autobiographical perspectives on the skin have rarely been in touch. Not only do they operate with radically differing understandings of tactile sensation, the relationship between them has also been antagonistic and conflictual. As mentioned, somatosensory research often relegate evidence from the autobiographical literature merely subjective and anecdotal status. Conversely, psychoanalysis has traditionally developed from clinical practice rather than empirical science, and, as Peter Hobson has emphasized, a rift exists between mainstream autism research and practice and the psychoanalytical approach, with animosity and prejudice on both sides (Hobson, 2011, p. 229). Finally, autistic authors often retain a critical attitude toward the tendency in much autism research to construct autistic modes of sensing as deviant or abnormal (Yergeau, 2018, p. 21).

In response to the rift between contrasting perspectives on touch and skin, this article aims to gather the diverse and conflictual field of tactile sensation in autism, focusing on its somatosensory, psychoanalytic, and autobiographical elaborations. While the aim is first of all to develop shared insights concerning autistic differences in the sense of touch, the article also opens up a conceptual space and offers resources for interdisciplinary analyses of both the concordances and discordances between neurobiological, psychological, and experiential “layers” of the skin. By disentangling underlying philosophical assumptions concerning touch and the body, the article presents a critical discussion of the compatibility between differing perspectives on tactile sensation in autism. Building on this analysis, the article suggests that the skin, as elaborated across neurobiological, psychological, and experiential levels of description, is intrinsically connected to self-other differentiation. In this context, autistic narratives may serve as an important constraint for an interdisciplinary account of the skin by elaborating experiential and contextual dimensions of touch neglected in both the somatosensory and psychoanalytic literature as well as resisting hierarchization between the skin as palpable flesh and as psychical boundary. In closing, the article discusses the normative tension between competing paradigms in autism studies and urges increased consideration of autistic voices in the sciences of the skin.

Let us start by considering the substantial and growing evidence for differences in tactile processing in autism in the context of the broad and multidisciplinary field of somatosensation. Representing an area under rapid development, somatosensory research has moved from primarily employing questionnaire measures to increasingly detailed psychophysical assessment in experimental settings targeting specific low-level sensitivity thresholds such as tactile detection, discrimination, adaption, as well as their neuronal underpinnings (Mikkelsen et al., 2018).

Traditionally, researchers have dealt with the sensory system as fulfilling a discriminative purpose, linking the sense of touch with motor control through detection, discrimination, and localization of pressure, vibration, temperature, itch, and pain on the level of skin receptors functioning as part of a somatic reflex arc enabling the rapid triggering of a motor response (McGlone et al., 2014). Studies on tactile processing in autism have until recently primarily covered this discriminative function of touch and investigated differences in detection and discrimination thresholds focusing on tactile stimuli on the glabrous surfaces of the palm and fingers. Espenhahn et al. (2023) found that young children with autism displayed poorer performance in the ability to discriminate between different amplitudes of vibrotactile stimuli delivered to glabrous parts of the participants’ index and middle finger as well as poorer ability to judge the temporal order of two successively delivered stimuli. In similar studies, Powell et al. (2022), Puts et al. (2014), Tavassoli et al. (2016), and He et al. (2021) concluded that autistic participants displayed lower performance on discriminating between vibrotactile stimuli, with the latter study interestingly demonstrating a correlation between poorer tactile discrimination and core autism symptomatology.

Studies on discriminative touch most often administer stimuli through a vibrotactile stimulation device delivering vibrations at varying frequencies to different areas of the participants’ skin. Interestingly, Cascio et al. (2008) found that autistic participants display a greater sensitivity to vibrotactile stimuli delivered to the forearm; an area often explored as a site for social touch processing. Similarly, Blakemore et al. (2006) demonstrated a significantly higher sensitivity to high-frequency, but not low-frequency vibrotactile stimulation in autistic participants compared to controls. Taken to indicate a hypersensitivity to the vibration and “tickliness” of touch, Blakemore et al. (2006) employed a second experiment where participants rated the perceived intensity and tickliness of a tactile stimulus delivered to the palm with a piece of soft foam. Here, autistic participants rated touch significantly more intense and tickly than did controls. The trend toward a more intense perception of different tactile stimuli is also reported by Cascio et al. (2008), where autistic participants displayed a tendency for higher pleasantness ratings for the sensation of three textured surfaces with varying degrees of roughness. Similarly, Haigh et al. (2016) investigated alterations of roughness perception in autism and found that autistic participants rated surfaces as generally rougher than controls.

Interesting as these results are, it is surprising that studies generally struggle to establish firm relations between the clinical features of autism and performance on discriminative touch (Mikkelsen et al., 2018, p. 147). However, it may be that different measures tap into different levels of tactile processing and “are likely to measure complimentary, but not always directly related, aspects of tactile function in ASD” (Mikkelsen et al., 2018, p. 147). As Mikkelsen et al. (2018) emphasize in their review on studies of tactile processing in autism, empirical results in this field are mixed and somewhat inconsistent, likely due in part to differences in tactile measures and descriptive language (see also He et al., 2023). Indeed, one difficulty arising from the ambition to develop ever more precise psychophysical assessments methods is grasping multidimensional nature of touch. As Espenhahn et al. (2023) aptly note, “perception assessed in a tightly controlled laboratory setting is unlikely to capture the multifaceted emotional, attentional and behavioral aspects of atypical tactile responses” (Espenhahn et al., 2023, p. 2900).

The affective dimension of touch has only recently been well recognized in somatosensory research and social neuroscience (McGlone et al., 2014). Responding optimally to low-force dynamic touch with the thermal characteristic of human body temperature, C-tactile afferents are thought to provide “the neurobiological substrate for the development and function of the social brain” as well as the “affective and rewarding properties of touch” (McGlone et al., 2014, p. 737).

Following the increased attention to sensory differences in autism, researchers have recently become interested in exploring possible alterations of the affective touch system in autism, including its underlying neurobiological mechanisms (McGlone et al., 2014). Especially neural responses to tactile stimuli meeting the optimal conditions of C-tactile afferent activation have been of interest, with results fairly consistently demonstrating a reduced response in the “social brain regions” in autistic participants. For example, Kaiser et al. (2016) found both a reduced neural activation from social touch as well as an enhanced activation to nonsocial touch compared to controls, indicating a shift in autism toward the processing of discriminative as opposed to affective touch (Kaiser et al., 2016). In a similar study, Voos et al. (2013) found that autistic traits correlated negatively with neural responses to social touch, and further, that participants with more autistic traits also rated social touch less pleasant than participants with less autistic traits.

A recent study by Capiotto et al. (2024) focused similarly on both automatic and hedonic responses to affective touch in autism, but through the measurement of skin conductance (i.e., the electrodermal activity in the skin). The authors found that autistic participants displayed a lower overall skin conductance in response to both types of touch as well as a decreased distinction in skin conductance between social and nonsocial touch, as opposed to the control group, where the automatic response to social touch was significantly higher than for the control condition. While some studies indicate that autistic persons experience social touch as less pleasant (Croy et al., 2016; Voos et al., 2013), Capiotto et al. (2024) employed both “unpleasantness” and “affectiveness” ratings for social touch. Surprisingly, the decrease in automatic skin response to social touch in autistic participants co-occurred with amplified hedonic ratings on both scales. That is, touch was experienced both as more unpleasant and more intense on an affective level.

Findings from studies of both hedonic and automatic responses to social and nonsocial touch have led some authors to emphasize a seemingly reduced distinction between social- and nonsocial touch in autism. As mentioned, Capiotto et al. (2024) demonstrated a lack of differentiation between automatic response to social- and nonsocial touch, and on the level of hedonic experience, Croy et al. (2016) found that autistic participants indicated both lower pleasantness ratings for affective touch and higher pleasantness ratings for non-affective touch, and Kaiser et al. (2016) emphasized the same lack of differentiation on the level of neural responses. Such results are often taken to demonstrate a decreased “social touch awareness” (Croy et al., 2016, p. 493) or hypo-reactivity to social touch (Kaiser et al., 2016, p. 2712). However, the discrepancy indicated by Capiotto et al. (2024) between automatic and hedonic responses to social touch and the subjective experience of its affectiveness should lead us to interpret such findings with care and not infer a weakened experience of the emotional dimension of touch from reduced automatic responses.

Considering the bigger picture, the skin is understood in the somatosensory literature as a physiological organ and studied through the complex relation between its different classes of receptors, nerve endings, and neural pathways transporting signals to be processed in the somatosensory cortex. Through the employment of psychophysical assessment and neuroimaging, touch is viewed as an objectively measurable and quantifiable phenomenon playing a direct role as a neurophysiological basis for autistic perception. However, differences in tactile processing may be the result of alterations at multiple stages of sensory processing (Blakemore et al., 2006, p. 6; Mikkelsen et al., 2018, p. 141). Perhaps in response to this complexity, there is a tendency in much of the somatosensory literature summarized above to locate tactile differences at the lowest possible level through controlled experimental studies.

Nonetheless, the aim for such studies is ultimately to understand better the role of tactile processing in the core symptomatology and development of autism, a complex developmental condition involving multiple dimensions of experience and functioning. Here, the tendency to atomize and isolate specific mechanisms of tactile processing also implies a deepening rift between the neurobiological dimension of touch and the significance of tactile experience for subjective and intersubjective experience, development, and behavior. This further emphasizes the necessity of clarifying the relationship between, on the one hand, processes transpiring on the level of skin receptors, nerve fibers, neuronal pathways, and the cortical and subcortical areas of the brain, and on the other hand, processes transpiring on the level of psychological and social development. On this question, as Mikkelsen et al. (2018) also emphasize, we have yet to see any convincing accounts in the somatosensory literature.

Generally, autism is described in the psychoanalytic literature as a condition involving profound alterations in early self-other differentiation (Mayes & Cohen, 1994). Bodily sensation plays an important role in these accounts. Especially for Donald Meltzer, Frances Tustin, and Thomas Ogden, autism “takes place” at a level of bodily experience that precedes what Melanie Klein identified as the earliest stage of infantile psychic life (Klein, 1930, 1975). Indeed, the psychoanalytic exploration of autism in this tradition became the exploration of the very beginnings of consciousness. In the words of Tustin, “in studying autism we find that we are studying the beginnings of perception” (Tustin, 1986, p. 84).

This move toward placing the skin and touch at the center stage for a psychoanalytic understanding of autism is in part motivated by the important contribution of Esther Bick. In her canonical 1968 paper, The experience of the skin in early object relations, Bick developed an account of the function of the skin as a primary container for previously unintegrated elements of the self. The experience of the skin and its sensations constitutes the earliest and most basic experience of a self being held together, and thus already from the earliest stages of infancy fulfills an indispensable proto-mental function (Bick, 1968, p. 139). In her work, Bick highlights the role of anxiety in the infant and its crucial role in psychopathological development. A compromised development of the skin function leaves the infant with “catastrophic anxieties” (Bick, 1968, p. 140) of disintegration, and consequently, a lack of the experience of an inner space. Bick’s analysis of the skin became highly influential to psychoanalytic studies of autism, e.g., Tustin’s notions of autistic shapes and objects (Tustin, 1986), and Ogden’s analysis of the autistic-contiguous position (Ogden, 1989). These accounts will be discussed in the following.

Tustin’s main idea is that autism should be understood in terms of sensation-based self-protective strategies employed to ward off anxieties toward the disintegration of one’s bodily surface (Tustin, 1986). Here, she makes the important distinction between autistic shapes (1984) and autistic objects (1980) in securing a feeling of being soothed and tranquilized in the case of the former, and a feeling of one’s boundaries being reinforced and rigidified as a hardened shell in the case of the latter. One of Tustin’s (and subsequently Ogden’s) central contributions is bringing to light the emergence of the infant’s earliest and most rudimentary experience of self and other. There is neither inside nor outside in this state of being, but rather the infant exists through the sense of interface or contact surface brought forth by tactile sensations on the level of the skin. As Ogden later emphasizes, “the sensory experience is the infant in this mode, and the abrupt disruption of shape, symmetry, rhythm, skin moldedness, and so on, marks the end of the infant” (Ogden, 1989, p. 35, italics in original).

Tustin describes how the infant’s emerging sense of its own boundaries is mediated by the experience of the malleable and soft shapes of its own bodily fluids and substances (Tustin, 1984). While occurring spontaneously in the beginning, the child will gradually learn to manipulate and self-induce such shapes, which come to function as forms of sensory tranquilization, or “a kind of autogenerated hypnosis which makes the child feel safe and comfortable” (Tustin, 1992, p. 20, quoted in Taipale, 2023). In normal development, there is a trajectory from “merely” experiencing the tactile feel, or tactile contour (Taipale, 2023, p. 7) of the shape toward experiencing a multisensory object existing in a shared space. Here, Tustin identifies the onset of autism with a turning in on itself of sensory experience so that the shape remains a two-dimensional tactile impression rather than the outline of a three-dimensional object (Tustin, 1984, p. 280).

As opposed to the softness and malleability of autistic shapes, autistic objects have hard and rigid surfaces that serve to reinforce a sense of one’s surface and boundary. According to Tustin, autistic children rely heavily on these to provide them with a sense of protection and impenetrability, where the felt hardness of the object’s surface is experienced as a hardness of one’s own bodily outline (Tustin, 1980, p. 27). For Tustin, this excessive reliance on the felt hardness of autistic objects to reinforce one’s own bodily definition essentially serves the purpose of forming a sense of “encapsulation” (Tustin, 1972), which inevitably ends up isolating the child from a world shared with others. Through her analysis of the autistic use of shapes and objects, Tustin paints a picture of autism as involving a mode of experience dominated by the sense of touch, ultimately presenting a world perceived through its felt impressions on the skin. In other words, this particular mode of experience, where sensation on the skin surface is all there is, becomes inflated and feeds into perceptual experience as such, so that also seeing, hearing, smelling, etc., become “touch-like” (Tustin, 1980, p. 28).

It was Ogden, however, who has described this tactile-dominated mode of experience most directly. Building on the works of Tustin, Bick, Meltzer, and Klein, Ogden aims to explore the earliest and most primitive organization of experience through what he terms the autistic-contiguous position. Ogden identifies the notion of a position with a particular mode of generating experience, that is, “a process through which perception is attributed meaning in a particular way” (Ogden, 1989, p. 11). This, he emphasizes, implies that neither the depressive, the paranoid-schizoid, nor the autistic-contiguous position are developmental phases that are merely passed through to reach more complex ways of experiencing. Rather, all three positions are conceptualized as “synchronic dimensions” (Ogden, 1989, p. 11) that form continuous organizing principles for our ongoing experiential life. In the case of autism, there is a collapse in the direction of the autistic-contiguous position, where “connections are established through the experience of sensory surfaces ‘touching’ one another” (Ogden, 1989, p. 31).

Ogden specifies the primary function of contiguity in establishing a “feeling of boundedness” of sensory experience, specified as a nonsymbolic and bodily “beginning of a sense of place in which experience is occurring” (Ogden, 1989, p. 35). Leaning on Tustin, Ogden emphasizes that all there is contained in the experience of contiguity is the feeling of softness, hardness, angularity, or shape of tactile impressions on the skin (Ogden, 1989, p. 36); impressions whose interconnection come to constitute bounded surfaces marking the origin of differentiation between self and other (Ogden, 1989, p. 49). Indeed, according to Ogden, this basic way of organizing experience comes to form a continuous background of sensory boundedness “of all subjective states” (Ogden, 1989, p. 50).

Ogden associates with autism a defensive reaction toward what he terms “autistic-contiguous anxiety” (Ogden, 1989, p. 67) concerning “the feeling of leaking, dissolving, disappearing, or falling into shapeless unbounded space” (Ogden, 1989, p. 68). According to Ogden, when such anxieties become extreme and overwhelming, a rigidification of defenses takes place in the attempt to restore a threatened or fragile sense of boundedness. In normal cases, such restoration may take the form of self-soothing practices, such as twirling one’s hair or tapping one’s foot to reinstitute a sense of bodily cohesion. In pathological cases, however, such defensive strategies become hypertrophied such that “the effort is to maintain a perfectly insulated closed system (in which sensory experience does not lead anywhere except back to itself)” (Ogden, 1989, p. 59).

Psychoanalytic theories view the skin as functioning both as a physiological and psychological condition of experience, where the development of the skin as a container and boundary through tactile stimulation marks the emerging distinction between the physical and psychological as well as the infants’ rudimentary sense of its own mental space as distinguished from the outside world. However, in following Bick’s account the “psychic skin,” it is not always clear to what extend psychoanalytic descriptions concern the skin as a physiological organ or as a proto-mental experience of the boundedness of the self. According to Leon Brenner, psychoanalysis has primarily dealt with the skin in a metaphorical sense, where references to the skin as a bodily surface are used figuratively to describe certain psychic functions and structures. Taking distance from this approach, he suggests that “it is best to steer clear as much as possible from the use of metaphorical language” and instead “conceptualizing them using a formalistic language that concerns abstract representations of boundaries that separate internal and external spaces” (Brenner, 2022, p. 201).

While this suggestion seems persuasive, it may risk overlooking an important question in relation to autism, namely that of the role played by differences in tactile sensation in the development and structuring of such boundaries. On the very first lines of Bick (1968) paper, she remarks that the “primal function” of the skin concerns “the most primitive binding together of parts of the personality not as yet differentiated from parts of the body” (Bick, 1968, p. 139). Indeed, when Tustin describes how autistic states require a therapeutic space “in which “fractures” can be “splinted,” “abscesses” can be “drained,” and the overlaid crusts of unhealing dead scar tissue can give place to healthy skin which is receptive as well as protective,” she adds that it is due to the lack of separation between physical and mental that “physiological images seem exact” (Tustin, 2016, p. 66).

Ogden defines contiguity as a process “in which raw sensory data are ordered by means of forming presymbolic connections between sensory impressions that come to constitute bounded surfaces” (Ogden, 1989, p. 49). Ogden’s use of the notion of sensory impression derives from Tustin’s famous example, in which she describes the early experience of the skin surface by asking us to imagine sitting on a chair:

On Ogden’s interpretation, “there is neither a chair nor one’s buttocks, simply a sensory ‘impression’ in the most literal sense of the word” (Ogden, 1989, pp. 35–36). Ogden’s position here seems to be that a sensory impression is the immediate feel of skin moldedness brought about by tactile impingement, and that such impressions order “raw sensory data” and come to constitute the experience of the skin as a bounded surface.

Ogden is quite clear that the sense of contiguity cannot be reduced to the biological level. He argues that although physiological reflex may have “a locus,” “periodicity,” and “a temporal and spatial beginning and end,” these are different from the “beginnings of a sense of place,” the “feeling of rhythm,” and of “boundedness,” which he associates with the sense of contiguity (Ogden, 1989, p. 35). Following Ogden’s more general formulation that the autistic-contiguous position describes a particular mode of generating experience, we should understand it as such that the sense of contiguity amounts to the very first manifestations of subjectivity or, in Ogden’s words, the “frontiers of consciousness” (Ogden, 1989, p. 3).

The following discussion focuses on three pieces of life writing by autistic authors (Gerland, 1997; Lawson, 2000; Williams, 1992), which have been selected based on the volume and level of detail of their descriptions of tactile experience. Themes have been highlighted to represent the different aspects of touch integral to these descriptions while taking care not to isolate them from their significance in everyday life experiences, emotional value, and connection to broader sensory experiences involving multiple modalities. Recognizing that there is no autistic experience of touch “in the singular,” autistic narratives point to the diversity of tactile experience and the manifold ways that touch is significant for the experience of self and other. Yet, such accounts also offer highly similar experiences that echo shared themes and modes of expression, which according to Irene Rose “exceed the individualisation of autobiography” by “asserting a communal response through the repetition of similar experiences represented across a variety of autistic life narratives” (Rose, 2008, p. 44). As such, they “defy the totalising constraints of discursive pathologisation of cognitive difference” and “demand the discursive space to explore and communicate different modes of being” (Rose, 2008, p. 52).

It is well known that the textured surfaces of objects, clothing, and foods represent a crucial arena for tactile sensitivity in autism, with some textures causing debilitating discomfort and unease, and others functioning as sources of comfort and enjoyment. Wendy Lawson describes that “I found cooked vegetables unpleasant on my tongue and the roof of my mouth – sometimes I felt like the food was choking me” (Lawson, 2000, p. 5). Contrary to this feeling of tactile resistance in chewing and swallowing food, Lawson describes pleasurable encounters with soft textures: “It’s hard to resist the temptation to run my fingers through my cat’s fur and lose myself in the feeling” (Lawson, 2000, p. 44). Lawson furthermore emphasizes the softness of fur as a feeling she can “lose herself” in. As Gunilla Gerland indicates, such experiences involve an intense emotional connectedness to specific features of the surfaces of objects. Recounting early experiences of playing in her room as a child, Gerland describes an intense fascination with curved surfaces:

Gerland here describes a fascination with curviness that only later develops into a need to hold and touch curved objects, thus emphasizing the complexity of sensory seeking behaviors, which often are understood as ways to compensate for tactile hyposensitivity. In another passage, Gerland echoes Lawson’s aversion to food textures, emphasizing intense bodily discomfort arising from the feel of certain consistencies:

Gerland’s description illustrates tactile discomfort with the experience of an unbearable electricity and needing to release this tension by biting down on something firm. This contrast between the electric tension associated with tactile discomfort and the stabilizing effect of biting down on something hard parallels a similar distinction between light and firm touch in social context, which is widely described across the autobiographical literature.

Light touch sensitivity is a frequently described phenomenon in autism, which can be experienced as unpleasant, overwhelming, and even painful. Gerland emphasizes how “light, soft touches tensed me, tightening the springs hard inside me and becoming unbearable” (Gerland, 1997, p. 15). She elaborates through an example of playing a specific game with a peer as a child:

For Gerland, the light, gentle, brushing movements associated with affective touch are described as an attack on her nervous system, almost as if every nerve fiber convulses from the sensation. Moreover, describing “even pressure” against a larger area of skin as favorable to brushing or tickling touch indicates a contrast between the skin as focalized (through light touch experienced intensely on a smaller area of skin) and the skin as a broader plane or surface (through firm touch experienced on a larger area of skin). Lawson alludes to a similar distinction in the context of the feel of the skin in the encounter with different textures:

Lawson describes how the feeling of sticky resistance of damp skin invites Lawson to find a way to separate skin from skin by introducing an alternative texture in the form of a bedsheet as an intervening membrane. Elaborating on the function of such strategies, Lawson writes:

These descriptions suggest that touch can both be discomforting to the point of dissociation and a fundamental means for reinstituting a sense of self and of being at home in the face of the “unbearable unbearableness” described by Gerland. As indicated by Gerland, this double quality of touch becomes particularly ambivalent in social contexts, where there is both a desire for closeness with others while at the same time a paralyzing discomfort from tactile contact. In the following, I will look at this ambivalence in more detail.

The experience of affective touch is a prominent theme in Donna Williams’ writings and is often described as painful and deeply terrifying. As Williams puts it, “I had always experienced being touched emotionally as the threat of death” (Williams, 1992, p. 151). Contrary to “safe physical contact,” which she defines as “that which does not threaten to hold or consume” (Williams, 1992, p. 190), Williams emphasizes that “otherwise, all touch was experienced either as pain or tolerated as though one was made of wood” (Williams, 1992, p. 190). Describing how she, as a child, felt comforted by touching her grandmother’s crocheted cardigan rather than the feeling of being physically close to her, Williams specifies what “death” signifies in such experiences:

For Williams, the physical touch of another person involves a terrifying loss of boundaries between herself and the other. To avoid this surrendering of one’s autonomous existence, Williams connects instead to the feel of her grandmother’s cardigan. Describing this emotional connectedness to crocheted and wooly fabrics in greater detail, Williams elaborates:

Williams’ emotional withdrawal from touch and intensified emotional connectedness to touching physical things suggests a somewhat idiosyncratic differentiation between socially salient and “merely physical” aspects of the sensory environment. In this case, her grandmother’s emotional comfort is manifest in the crocheted texture of her (clothed) bodily surface, which in Williams’ experience is principally detachable from her person in the form of woolen bits of fabric. This form of differentiation between the social and the sensory features of touch ultimately becomes a divisibility between parts of the bodily surface, where some areas of the skin become arenas for the frightening social touch, and others for enjoyable sensorial touch:

In the case of her grandmother’s cardigan, Williams secures a sense of being comforted and soothed by touch without the terrifying physical closeness with another person. And in the case of depersonalizing parts of the body, she renders the body “touchable” without feeling as if her existence is at stake. This split between the social and sensory features of touch appears motivated both by a need to protect oneself against the threat of social touch, and at the same time by an intense fascination and emotional appraisal of the soothing feel of specific tactile sensations.

Autistic life writing not only details the experience of touch as disruptive and disturbing, or the subjective dimension of a “deviant” way of sensing. Paraphrasing Melanie Yergeau, autistic stories are importantly also about “the beauty of shiny objects” (Yergeau, 2018, p. 21), in this context the “love for curved things” (Gerland, 1997, p. 11) or “the silkiness of inside coat pockets” (Lawson, 2000, p. 5). This should not be taken to undermine the emotional stress associated with experiences of touch and the traumatic quality that tactile experience often presents with in the autobiographical writings discussed above. Indeed, autistic autobiographies portray touch through the language of affect through experiences of vulnerability, unbearableness, panic, overwhelm, anxiety, threat, or even the sense of being eaten up, drowned, or choked.

Importantly, the affectivity of touch in these writings makes no distinction between its physical and psychological dimensions. For example, in the case of Williams, the unwelcomed touch of another person is experienced as the feeling of losing one’s sense of separateness from the other. And in the case of Lawson, the tactile sensation of soft materials is experienced as the feeling of being at home. This emphasis on touch as something that resists the boundary between physical and psychological shares some features with a psychoanalytic understanding of the infant’s experience of the skin. Yet, autobiographical accounts indicate how this irreducibility of touch characterizes tactile experience throughout the lifespan, stretching far beyond the specific developmental phases or functions described in the psychoanalytic literature. As such, autistic life writing communicates an almost phenomenological dimension of touch as a continuous structure of subjective experience, intimately connected with affective life and encounters with others.

Autistic narratives thus foreground the affectivity of touch, which is a recurring theme across all three perspectives on tactile sensation in autism. On a very general level, the skin is cast in a social and affective role, mediating the experience of others through sensations on the skin surface. Here, somatosensory studies tend to focus on the role of affective touch in the processing of social information, psychoanalysis construes the skin as the first and primary site for affective communication within the infant-caregiver dyad, and autistic life writing details the intense affective impact of different tactile encounters. While this shared focus on the affectivity of the skin is striking, this should not lead to the assumption that the meaning of affectivity converges across the three paradigms. For example, whereas the somatosensory literature seems to collapse the affectivity of touch into the neural activation of C-tactile afferents, autobiographical present affectivity from a felt and subjective perspective that does not distinguish between neurophysiology and experience. Indeed, as each perspective approaches the skin and its sensations with radically different epistemological and ontological assumptions, any attempt toward their synthesis must put such differences front and center.

The three perspectives described above each represent in-depth accounts of the centrality of tactile sensitivity for autistic experience and development but take as their point of departure profoundly diverging approaches to the meaning of the skin and its sensations. It is often emphasized that science deals with different levels of description, of explanation, and of reality, and that each branch of science deals with a level suitable for its own aims. Following List (2019), such levels may be understood to mark differences in scientific vocabulary and conceptual framework (as levels of description), or they may be understood as ontological levels into which reality itself is stratified. Regardless of whatever conception one is inclined toward, an urgent issue is the hierarchical structure between levels. Such hierarchization usually hinges on the idea of a fundamental level of reality, a “unique lowest level” in the words of Oppenheim and Putnam (1958, p. 9) on which everything else supervenes (for a critical discussion, see Dupré, 1993). From this perspective, it is easy to assume the problematic position that supervenience implies explanatory reducibility (List, 2019, p. 22). On this reductionistic view, the manifold experiences of touch as well as the complex developmental implications of the skin would in principle be explainable with reference to the lower level neurobiological processes described in the somatosensory literature.

Attempts at a levelled account of the skin thus inevitably carry with them an import of assumptions with profound implications for how the nature of autism is more fundamentally construed. Keeping in mind this normative dimension of a levelled account of the skin, the remainder of this discussion will focus on a specific point of intersection between the three perspectives on tactile sensation, namely the link between touch and self-other differentiation. By indicating more precisely the role of touch for the experience of self and other, this theoretical intersection between levels is fundamental both for an interdisciplinary account of the skin and its implications for human development as well as for a richer understanding of autism. Pursuing this line of inquiry is inevitably an optimistic project that seeks compromise between profound epistemological and ontological divergences – and a project which one may regard with some apprehension. For this reason, the following discussion proceeds with caution but nonetheless forward to tentatively map out directions for future studies of the skin and its sensations.

Self-other differentiation has until now remained an underlying, yet fundamental, function of the sense of touch. In the field of somatosensation, touch is related to self-other differentiation on the most general level through the skin’s double function as protecting the vulnerable inside of the organism, while at the same time allowing for the perception of the outside world and the distinct qualitative properties of objects impinging on the skin. Issues in the processing of tactile information and integration with other sensory modalities have been explicitly linked to alterations in self-other differentiation in this literature (Cascio et al., 2012; Ropar et al., 2018), with results taken to suggest a sharper and less flexible boundary between self and other in autism (Noel et al., 2017). Moreover, social touch is demonstrated to profoundly affect the internal physiological condition of the body, most prominently in infancy where the organism relies entirely on the caregiver (Zoltowski et al., 2022). This not only points to the crucial function of the skin to mediate both a sense of what goes on inside and what impinges from outside, and the ability to regulate the former in relation to the latter, but also suggests that self-other differentiation is a developmental process, where the organism through its gradual maturation becomes able to autonomously maintain allostasis.

Here, we arrive at an important junction with the psychoanalytic literature, where self-other differentiation is considered a developmental achievement in which the infant in the context of experiencing care forms a unit with the caregiver, who only gradually becomes differentiated into a separately existing entity. As emphasized in the work of Esther Bick, it is the experience of a “shared skin” with the caregiver that serves as a developmental foundation for the infant’s experience of its own bodily boundary. Here, it is worth emphasizing that despite this shared acknowledgment of the close connection between the sense of touch and self-other differentiation, somatosensory research and psychoanalysis deal with these processes on fundamentally different levels of description.

In the former case, tactile sensation represents a process internal to the organism, where the skin functions as a medium for encoding and transporting tactile information to be processed in the brain. In the latter case, the skin and its sensations are conceptualized within the interactive social context of infant care. From this point of view (further cemented with Donald Winnicott’s provocative claim that “there is no such thing as a baby”), it seems that somatosensory research, by focusing on the internal coding and processing of external stimuli, relies on a pre-established distinction between self and other. For example, studies of discriminative touch rely on the ability to evaluate and judge the qualitative and quantitative features of tactile stimuli – explicitly judging what, when, and how something external impinges on the skin. Importantly, such tasks presuppose a more fundamental experience of sensory boundedness and boundary explaining how the experienced differentiation arises in the first place, and this is what is taken up and interrogated directly in the psychoanalysis of the skin (Ogden, 1989, p. 35).

This contrast with respect to how tactile sensation is conceptualized has significant implications for conceptual commitments as to the nature and development of autism. Where authors in the somatosensory field typically commit to the widely accepted view that autism is a neurodevelopmental condition, psychoanalytic thinkers (despite acknowledging multifactorial explanation, see, e.g., Houzel & Rhode, 2018, p. xvi) remain interested in the significance of early interaction dynamics. For example, according to Tustin and Ogden, autism involves a terrifying experience of separateness connected to the absence of sufficiently soothing sensory experiences (provided in the interaction with a primary caregiver) confirming the infant in its sense of boundedness (Tustin, 2018, p. xii; Ogden, 1989, pp. 51–52). In the somatosensory literature, efforts seem to be more directed toward locating the source of tactile differences inside the autistic body through the study of, e.g., potential neurobiological alterations of the excitation/inhibition balance of the central nervous system (Mikkelsen et al., 2018).

One of the issues in arriving at a shared understanding of the nature, implications, and developmental underpinnings of autistic differences in tactile sensation is that researchers often seem to be talking about different things. This is true both within each perspective on touch (He et al., 2023; Mikkelsen et al., 2018) but certainly also between them. While efforts have been made to develop a shared nomenclature and hierarchical taxonomy for describing sensory features of autism (He et al., 2023), this project only seems to be taking place on an intra-level rather than an inter-level basis. That is, developing a shared language for describing tactile sensation is inevitably less thorny when the problem is indeed a terminological one and not one of deeper ontological and epistemological disaccord. The latter is exactly the issue for a more comprehensive layered account of tactile sensation in autism as each perspective on the skin imports diverging philosophical assumptions concerning both the nature of autism and the very differentiation between the somatic, the psychological, and the experiential.

As it were, the project of synthesizing the somatosensory, psychoanalytic, and autobiographical perspectives on tactile sensation in autism in the form of a “shared skin” stratified into three layers of description runs into significant obstacles. However, I want to suggest that, even given the profound divergences between the perspectives discussed above, each of them can nonetheless in a dialectic fashion provoke elaboration, development, or refinement in the other through mutual and critical inquiry. For example, a pertinent question for further study is the role played by alterations of tactile processing (as described in the somatosensory literature) in autistic experiences of the loss of sensory boundedness (as described in the psychoanalytic literature). And conversely, a critical concern for somatosensory studies is the task of integrating its neurobiological evidence within a theoretical framework capable of explicating its connection psychological development and subjective experience.

In the context of such critical engagement, a privileged role should be granted first-person descriptions of autistic experience as a potential corrective to scientific and theoretical elaborations of the skin and its sensations. Not only because such texts explicate the subjective complexity of tactile experience from the point of view of the autistic person but also because they flesh out the profound implications of tactile sensation for navigating social encounters and everyday contexts throughout the lifespan (see also MacLennan et al., 2022). As detailed in autistic life writing, these are crucial dimensions of tactile experience in autism but remain underemphasized in both the somatosensory and psychoanalytic literature. In the context of touch and self-other differentiation, autistic narratives point to theoretical intersections between differing perspectives on tactile sensation, and by presenting rich descriptions of autistic experiences of touch as they play out in everyday settings, may serve as an important constraint for interdisciplinary dialogue at such crucial junctions.

For example, in the writings of Donna Williams, Wendy Lawson, and Gunilla Gerland, the experience of touch presents with a traumatic quality and intense terror, not unlike what Tustin and Ogden describes as unthinkable anxieties of disintegration, spilling out, or falling into shapeless space. However, this traumatic quality is for Tustin and Ogden connected to the absence of sufficient experiences of “sensory holding,” whereas in the cases of Williams, Lawson, and Gerland, it is the qualities of certain tactile sensations that are themselves experienced as overwhelming and traumatic. Here, the important question arises of how neurobiological differences in tactile sensitivity and processing may contribute more directly to the harrowing emotional experiences detailed in psychoanalytic studies and fleshed out in autistic narratives. Considering the empirical evidence, it is likely that autistic persons from birth onwards are vulnerable to experiencing tactile impingement as overwhelming and struggle with binding, regulating, and adapting to such stimuli. Following this cue, autistic persons may from the outset be more exposed to traumatic experiences, where, in the words of Werner Bohleber, “the onslaught of quantities of excitation is too great to be mastered and physically bound” (Bohleber, 2010, p. 79). This does not mean, however, that autistic narratives provide an argument in favor of positing alterations of tactile processing as primary to this emotional dimension of touch. Rather, such narratives, by resisting the distinction between the skin as palpable matter and the skin as a psychical boundary, illustrate the futility of such hierarchization when dealing with autistic experiences.

Following these suggestions for developing a more integrated account of touch and its implications for autism, the important task remains of further detailing the role that autistic narratives can play in the sciences of the skin. The final part of this discussion will address this question through considering the normative and political tensions between the three perspectives on tactile sensation in autism.

In the context of literature on tactile sensation in autism, the relationship between competing understandings of autism is cashed out through the adoption of radically different “genres of sensory writing” (Fulton et al., 2020, p. 1). Where the somatosensory literature is replete with terms such as hyper- and hyposensitivity, altered thresholds, enhanced or reduced neural activity, deficit, dysfunction, or abnormality, the psychoanalytic literature is equally replete with notions such as breakdown, pathology, rigidification, isolation, encapsulation, and even descriptions of autism as a way of “not-being” (Ogden, 1989, p. 60, italics in original). In both cases, such ways of describing autism rely on the idea of an average or healthy mode of experiencing from which autistic persons diverge, thus constituting a “normative reflex” (Fulton et al., 2020, p. 3) pervading the vast majority of works on autism and touch.

This point should not be taken to indicate that autistic persons do not have difficult tactile experiences, but to further emphasize the underlying current of most autobiographical accounts to “work at a connective and emotional level to resist the pathologization of difference” (Rose, 2008, p. 46). As mentioned previously, autistic testimony is often cited in contemporary somatosensory studies on tactile processing, but rarely as more than illustrative example. Indeed, it has been assumed in autism research that autistic persons lack the sufficient self-awareness and reflective and linguistic capability for their perspectives on their own situation to be taken seriously in scientific studies of autism (Chapman & Carel, 2022; Fulton et al., 2020; McGeer, 2005). According to Fulton et al. (2020, p. 3), it is time “to create a discursive space in which autistic sensory experience might be pondered free from any normative tilt or filter.”

While such liberation from normative filter seems somewhat idealistic, making the attempt at least involves taking autistic testimony at face value. For empirical research, this could involve adopting a participatory research design where autistic persons are involved and consulted in designing and conducting research (Fletcher-Watson et al., 2019). For experimental somatosensory research specifically, involving autistic persons in developing the study design, experimental protocol, and crucial operationalizations could be fundamental for ensuring an ecologically valid account of touch aligned with autistic experience. For theoretical studies, e.g., in psychoanalysis, taking autistic testimony seriously could invite critical reflection on underlying normative assumptions, thus avoiding a rigidified application of theoretical concepts at risk of misconstruing autistic experience (Ekdahl & Boldsen, 2024).

In closing, this article has discussed findings and insights concerning tactile sensation in autism, focusing on its somatosensory, psychoanalytic, and autobiographical elaborations, and offered a conceptual space for critical analyses of the diverse features of autistic experiences of touch as well as the relation between neurobiological, psychological, and experiential “layers” of the skin. Such a project is ambitious and, importantly, not without its ambivalences, conflicts, and even contradictions. Yet, an important payoff is fertilizing the ground for interdisciplinary exchange between radically different approaches to the skin, its sensations, and their alteration in autistic experience, potentially invigorating broader dialogue between the conflicting paradigms in autism studies that such approaches represent. A promising avenue for developing a fuller and more integrated account of touch in autism is its intimate relationship with the experience of – and differentiation between – self and other. Here, autobiographical accounts may serve as an important constraint for such interdisciplinary exchange by resisting hierarchization between the skin as palpable flesh and as psychical boundary when dealing with the subjective and contextual complexity of tactile experience in autism. However, realizing the corrective potential of autistic narratives in scientific work requires an appreciation of the normative tension between competing discourses surrounding autism to counteract the tendency to neglect autistic voices at the crucial points of knowledge production.

A deepfelt thanks goes to my research group, Asymmetric Encounters: Intersubjectivity and the Sense of Boundaries, for their support and helpful discussion on an earlier version of this manuscript.

Ethical approval was not required for this study since no empirical data was collected.

The author has no conflicts of interest to declare.

This research was supported by the Research Council of Finland (Grant No. 348858). The funder had no role in the design and reporting of this study.

With reference to the Contributor Roles Taxonomy (CRediT), the author confirms sole responsibility for the conceptualization, investigation, methodology, and validation of this study, as well as writing, reviewing, and editing the original draft and final article.

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.