The review, published in BMC Psychiatry by Amir Jahanian Najafabadi and Alireza Rastegari of the Department of Cognitive Neuroscience at Bielefeld University in Germany, synthesizes 37 empirical studies published between 2010 and August 2025. The authors frame their question around what they call the active self: the embodied sense of self that arises from goal-directed action and interaction with the environment. Rather than treating the self as something the brain merely observes, this framework sees it as something the brain actively constructs, moment by moment, through sensorimotor loops that link intention, movement, and sensory feedback. When those loops operate differently, the reasoning goes, the resulting self-model may be differently calibrated, and the review set out to map exactly where in the literature that calibration appears to diverge in autism.

The authors organized the evidence into three themes. The first is multisensory integration, the process by which signals from different senses are combined into a unified percept. The second is self-bias and self-other differentiation, the tendency of the brain to prioritize information belonging to one’s own body and to distinguish it from information about other people’s bodies. The third is the motor underpinning of bodily awareness, covering how movement planning and execution feed the sense of agency, the feeling that you are the author of your own actions. Across these domains, the review identifies a pattern of atypical processing in autistic participants, but also striking inconsistencies between studies, which the authors treat as scientifically informative rather than simply as noise.

In the multisensory domain, the clearest signal involves timing and weighting. The brain’s rules for binding touch, vision, and proprioception depend on narrow temporal windows; stimuli arriving outside those windows are normally experienced as separate. The reviewed studies frequently found differences in autistic participants in these temporal integration windows and in how much weight each sense receives in the final percept. Concretely, autistic individuals often showed reduced or delayed susceptibility to some bodily ownership illusions, laboratory phenomena such as the rubber hand illusion, in which synchronized stroking makes people adopt a fake hand as their own. A weaker illusion response suggests that the neural machinery deciding whether something is part of your body may operate under different parameters in autism, integrating sensory evidence with different thresholds or different priorities.

Equally significant is the finding of greater reliance on proprioceptive or immediate sensory information. Under many laboratory conditions, the brain resolves conflicts between senses by letting vision dominate touch or proprioception. Several studies reviewed suggest that autistic participants may lean more heavily on internal, body-based signals, or on whichever information stream is most immediate and least ambiguous. This detail matters because it reframes sensory atypicalities in autism: instead of a simple story of over- or under-sensitivity, the data point toward altered cue weighting, where the brain’s calculus for combining noisy signals differs systematically. Interestingly, the review also reports relatively preserved performance under some unimodal or low-noise conditions, implying that differences may emerge most clearly precisely when the brain must arbitrate between competing sensory streams.

The second theme, self-other differentiation, produced evidence of differences on several fronts. One set of paradigms measured the flexibility of peripersonal space, the buffer zone around the body where external stimuli feel personal, and of interpersonal space, the comfortable distance maintained from other people. Studies reviewed by the authors reported differences in how flexibly these zones are represented in autism. Other work examined self-bias effects, including the self-prioritization effect, in which people identify stimuli associated with themselves faster and more accurately than stimuli associated with others, and related friend-prioritization effects. Differences in these biases, alongside differences in agency-related processing, suggest that the default weighting of self versus other information, normally tilted heavily toward the self, may be calibrated differently in autistic individuals.

The third theme connects these bodily representations to movement. Motor planning and execution differences have long been documented in autism, and the review links them to bodily awareness through the sense of agency. Experiments using measures such as intentional binding, the temporal compression people perceive between their voluntary actions and their consequences, revealed differences in agency-related processing in autistic participants. Because agency is one of the pillars of the active self, altered motor signals may propagate upward into a differently structured sense of self. The review also notes associations with developmental coordination difficulties, a domain that has traditionally been studied separately from self-representation research, and the authors argue that the two literatures should be read together.

What makes the review compelling is its candor about heterogeneity. Findings were not uniform across paradigms, developmental stages, or participant groups, and the authors explicitly caution that substantial methodological and sample diversity limits strong causal or clinical conclusions. Illusion paradigms, questionnaire-based measures of embodiment, behavioral tasks, and electrophysiological methods each capture a different slice of the phenomenon, and the 37 studies varied widely in sample sizes, age ranges, and how autism was characterized and confirmed. Terms such as high-functioning autism and Asperger syndrome appear across the older literature, complicating comparison. The authors therefore frame their conclusions as supporting an association, not a mechanism proven, between multisensory and sensorimotor processing and several components of bodily self-representation in autism.

The road forward, according to the authors, requires methodological discipline. Future research should prioritize larger and more representative samples, longitudinal designs that track bodily self-representation across development, standardized experimental paradigms that allow direct comparison between laboratories, and objective measures of embodiment rather than reliance on subjective report alone. Multimethod designs combining behavioral, psychophysical, and neural measures such as electroencephalography would help determine whether the observed differences in sensory weighting and self-bias reflect a single underlying computational difference or a family of distinct mechanisms. The authors also emphasize that representative sampling is essential, since much of the existing literature over-represents intellectually able autistic adults, leaving developmental trajectories and intellectual disability largely unexplored.

The implications extend beyond theory. If differences in multisensory integration and sensorimotor processing do underpin aspects of bodily self-representation in autism, they could ultimately inform individualized assessment and intervention, for example by identifying sensory profiles that predict which environments or tasks will be difficult, or by designing interventions that target the specific sensory computations involved. The authors are careful not to overpromise; whether these mechanisms can be translated into clinical practice remains an open empirical question. But the conceptual payoff is already substantial. By locating the sense of self in the body’s active dialogue with the world, and by showing that this dialogue is measurably different in autism, the review invites a shift in how autistic experience is understood, from a disorder of social cognition layered on top of an otherwise typical self, toward a differently constructed self, built from the first sensory exchange onward.

Subject of Research: Multisensory and sensorimotor foundations of bodily self-representation in autism spectrum disorder

Article Title: The active self in autism: multisensory and sensorimotor foundations of bodily self-representation

Article References: The active self in autism: multisensory and sensorimotor foundations of bodily self-representation. (n.d.). https://doi.org/10.1186/s12888-026-08756-w

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08756-w

Keywords: autism spectrum disorder, active self, bodily self-representation, multisensory integration, proprioception, sensorimotor processing, self-other differentiation, sense of agency, rubber hand illusion, embodiment, systematic review, BMC Psychiatry