9. Active exploration: how movement produces and renews sensory information

Knowledge Base – Expert Guide Sensory Flows

Blue Portance Knowledge Base · Non-commercial reproduction permitted with source citation

To begin with, the previous chapters described how several sensory sources are integrated (B7) and how their respective weight is adjusted according to their reliability (B8). In both cases, the nervous system processes information that is already available. However, this chapter changes perspective. It concerns cases in which the subject’s movement can itself produce, renew or refine sensory information, rather than merely receiving or reweighting it. Thus, the relationship between perception and action does not run in one direction only. The nervous system uses sensory information to guide a movement. That movement then in turn changes the conditions under which new information is produced.

By contrast, this chapter does not revisit the mechanics of multisensory integration (B7) or of reweighting (B8): it deals specifically with the exploratory function of movement, that is, the fact that certain movements can be directed toward obtaining information or can take part, through their sensory consequences, in updating the state of the body and its environment.

Movement is not only a means of acting on the environment: it can also take an active part in gathering sensory information. Exploring an object with the hand, adjusting one’s posture or shifting one’s gaze produces information that passive reception would not provide; motor control relies on continuously updated sensorimotor loops rather than on a linear chain running from a stimulus to a response, and the nervous system distinguishes, at least partially, the sensory consequences of its own movements from those caused by an external source (Gibson, 1962 ; Bizzi & Ajemian, 2020 ; Blakemore et al., 2000).

1. Voluntary movement as a producer of information

The distinction between active and passive perception is long-standing in sensory psychology. Gibson already contrasted active touch with passive touch. In the first case, the person directs their own exploratory movements. In the second, a stimulus is simply applied to the skin without any voluntary movement by the subject. Gibson also showed that active touch gives access to certain properties of an object that passive contact conveys less completely, because the subject can orient and adjust their exploratory movements. However, the advantage gained depends on the property sought, the task and the possibilities for movement (Gibson, 1962).

A simplified model of motor control could be written as: stimulus → sensation → motor command → movement. This model is useful for introducing the notion of a response. However, it wrongly assumes that movement has no effect on the information that follows it. A more complete description instead links the state of the body and environment, the resulting sensory signals, the estimate made from them, the motor command and the movement produced. It then integrates the change in the state of the body and environment. This change in turn generates new sensory signals.

Consequently, motor control cannot be described as a linear chain running from a stimulus to a response: it relies on converging, redundant and continuously updated sensorimotor loops (Bizzi & Ajemian, 2020). Thus, the motor response does not close the chain. On the contrary, it restarts it. This chapter deals specifically with this exploratory function of movement, distinct from its purely motor function.

Loop in which voluntary movement changes the body, contacts and support, produces new afferents and updates the sensory estimate.
Figure 1 — Movement restarts the sensory loop By transforming posture, contacts and support, voluntary movement renews the available sensory information and restarts the estimation of the body. © Blue Portance 2026.

2. Active touch: exploring in order to know

Next, the work of Lederman and Klatzky formalized this idea. They show that people use stereotyped, reproducible sequences of hand movements. These sequences serve to identify specific properties of an object. For example, people use lateral motion to explore texture, pressure to judge hardness, static contact for temperature, unsupported holding for weight, and enclosure for shape or overall volume. Finally, contour following makes it possible to refine shape. Each of these “exploratory procedures” is spontaneously selected when the information sought corresponds to the property it best allows one to assess (Lederman & Klatzky, 1987).

This result thus illustrates a general principle. Active exploration is not a random movement followed by passive reception of sensations. On the contrary, it is a strategy directed toward an informational goal. The movement itself is adapted to the nature of the information sought, which fundamentally distinguishes this mechanism from a simple stimulus received without voluntary movement.

Five procedures: lateral motion for texture, pressure for hardness, contour following for shape, unsupported holding for weight and static contact for temperature.
Figure 2 — Exploring actively to seek a property Exploratory movements are organized according to the property sought. © Blue Portance 2026.

3. Active exploration beyond touch

Furthermore, this principle of actively sampling the environment can be observed in several sensory systems. It does not concern manual touch alone. For example, gaze movements direct vision toward a region of interest. Likewise, movements of the head and trunk change access to certain information. An influential synthesis proposes that a large part of sensory processing is thus shaped by motor and attentional routines of active sampling, rather than by purely passive reception of the sensory flow (Schroeder et al., 2010).

However, this perspective does not claim that all perception results from voluntary movement. Indeed, a large part of sensory processing remains involuntary and continues even in the absence of exploratory movement, as the previous chapters have shown. It simply indicates that a significant part of the sensory information available at a given moment depends on movements produced, at least in part, for an exploratory purpose.

4. Converging and redundant loops

In addition, the same movement can be signaled by several simultaneous sources. These are proprioceptive, tactile, visual and vestibular information. Added to these are signals linked to the motor command itself and the mechanical consequences of the ongoing movement. These sources are partly redundant. Indeed, several afferents can provide information about the same bodily change, with different precision and timing (Bizzi & Ajemian, 2020).

On the one hand, this redundancy can improve the robustness of motor control. If one source becomes less available, others can continue to contribute. On the other hand, it can complicate interpretation when sources diverge, since the system must then process information that does not provide exactly the same estimate. There is therefore not necessarily a simple relationship between a given receptor and a given afferent activity. Nor is the relationship with a specific motor response simple. For example, the same postural correction can be triggered by several different sensory configurations. Conversely, the same sensory signal can contribute to different responses depending on the task and context.

5. Predicting the sensory consequences of one’s own movement

More precisely, exploring actively requires the nervous system to be able to distinguish two origins of sensory information. Some is produced by its own movement. Some is caused by an external source. This distinction relies on the internal models and efference copy already presented in B7. When a movement is planned, a copy of the motor command makes it possible to predict its expected sensory consequences (Franklin & Wolpert, 2011). A difference between the expected consequences and those actually produced can then provide distinct information. For example, a support may move more than expected. Likewise, support may unexpectedly become asymmetrical. This discrepancy allows the system to distinguish a predictable consequence from an unexpected event, without this necessarily implying conscious perception of the difference.

The tickle example and internal feedback

The best-known experimental example of this mechanism concerns tickling. A tactile stimulus self-produced by a voluntary hand movement is perceived as less intense. It is compared with an identical stimulus produced by another person. Indeed, its sensory consequences are anticipated by the motor system (Blakemore et al., 2000). More recent work proposes a computational mechanism consistent with this observation. Internal feedback signals would travel from motor regions to sensory regions. They would thus help compensate for the delays of the sensorimotor system. To do this, they would filter out part of the self-produced or predictable variations, which would allow unexpected information to be transmitted more quickly (Li et al., 2023).

However, this model remains computational. It proposes a mechanism consistent with the measured delays of the sensorimotor system. By contrast, it does not directly demonstrate how it works in the sitting position. Nor does it mean that predictable signals would be completely suppressed. They can continue to contribute to the estimate of body state, but with processing distinct from that of an unexpected event.

Comparison between the expected sensory consequence of a voluntary movement and the information actually received after it is executed.
Figure 3 — Predicting the consequences of one’s own movement When producing a movement, the subject can compare its expected sensory consequences with the actual feedback and detect agreement or any discrepancy. © Blue Portance 2026.

6. Active exploration and reweighting: two distinct mechanisms

It is therefore useful to explicitly distinguish active exploration from the sensory reweighting presented in B8. Reweighting describes how the nervous system adjusts the relative weight given to sensory sources that are already available, according to their contextual reliability. Active exploration describes a different mechanism: through their movement, the subject changes the conditions of sensory sampling and can thus produce, renew or refine certain information.

However, these two mechanisms are not mutually exclusive. An exploratory movement can produce, renew or refine information which, once available, will then be integrated and weighted according to the principles described in B7 and B8. Conflating them, however, would amount to presenting every adjustment of sensory weight as the result of voluntary movement, which neither B8 nor this chapter demonstrates: reweighting can occur without exploratory movement, and an exploratory movement does not on its own guarantee a change in sensory weight.

Comparison between the production of new information through action and the change in the weight of information already available.
Figure 4 — Active exploration and sensory reweighting: two distinct mechanisms Exploring changes sensory sampling; reweighting changes the relative contribution of sources. © Blue Portance 2026.

7. What active exploration does not allow us to claim

Statements to rule out
  • “Moving always reveals the truth of the body.”
  • “An exploratory movement automatically produces better perception.”
  • “The absence of movement prevents any reliable perception.”

Recommended formulation: a movement can change sensory conditions and make certain information available, more discriminable or differently organized. This possibility guarantees neither its accuracy, nor its actual use by the nervous system, nor a systematic improvement in bodily experience.

In sum, active exploration is an additional source of information. It is not a mechanism that would make every other source superfluous or every perception automatically more reliable. Indeed, the principles of integration (B7) and reliability-based weighting (B8) continue to apply to information produced by exploration, exactly as to any other sensory source. Likewise, a change in pressure measured at the interface is not enough to demonstrate sensorimotor exploration. Moreover, a visible movement is not enough to demonstrate a perceptual intention. Finally, a reduction in postural sway is not enough to demonstrate better bodily perception. Mechanical measurement, observed behavioral response and any subjective experience describe distinct levels, which must be related without being confused.

8. Application to sitting

In the sitting position, changing position, shifting the pelvis or adjusting support changes the contacts, tissue deformations and associated afferent activity. These sensory consequences can contribute to updating the bodily estimate. In some cases, a movement can also have an exploratory function, but the mere fact that a postural adjustment occurs does not establish that it was directed toward obtaining information. For example, a shift of the pelvis can be described as a loop. It begins with a motor adjustment, followed by a movement of the pelvis. Then come the change in contacts and body configuration, followed by the renewal of afferent activity. A possible update of the estimate and a new adjustment may follow. This loop can thus repeat at very different scales, from a brief adjustment to a gradual lateral shift.

Controlled pelvic adjustment, new contacts, new afferents, possible update and next adjustment.
Figure 5 — The adjustment loop in the sitting position In the sitting position, a controlled adjustment can renew the contacts and the information available for the next adjustment. © Blue Portance 2026.

A transposition that remains cautious

However, this possibility remains a cautious transposition hypothesis. Indeed, none of the references used in this chapter deals specifically with sitting or furniture. It therefore does not support the conclusion that a seat allowing more movement necessarily produces better bodily perception, nor that immobility prevents any reliable perception — only that a support-adjusting movement can, in some cases and among other functions, play an exploratory role. Moreover, the fact that a support allows changes of support does not demonstrate that the user will make use of them. Nor does it demonstrate that they will improve the user’s bodily perception. Indeed, this use depends on factors specific to the subject — motor abilities, sensory sensitivity, fatigue, task, attention, experience — which remain distinct from the mere mechanical possibility offered by the support.

9. Key points

  • Movement is not only a means of acting on the environment: it can also produce, renew or refine sensory information.
  • Indeed, motor control is not described as a linear stimulus → response chain, but as converging, redundant and continuously updated sensorimotor loops.
  • In addition, active touch gives access to certain properties of an object that passive contact conveys less completely, depending on the task and the movements possible; exploratory hand movements are adapted to the nature of the information sought.
  • This principle of active sampling extends to other sensory systems beyond touch.
  • Furthermore, the nervous system distinguishes, at least partially, the predictable sensory consequences of its own movement from those caused by an external source, through internal models and the efference copy already presented in B7; recent computational models propose a filtering mechanism for these predictable consequences, without directly demonstrating how it works in the sitting position.
  • Active exploration is a mechanism distinct from the sensory reweighting described in B8: one changes sampling conditions and can produce, renew or refine certain information; the other adjusts the relative contribution of the information available.
  • An exploratory movement guarantees neither better perception nor its actual use by the nervous system: mechanical measurement, observed behavior and subjective experience remain distinct levels that must not be confused.

In sum, active exploration complements the mechanisms described in this sub-guide. After multisensory integration (B7) and reliability-based reweighting (B8), it introduces the possibility that movement itself produces, renews or refines sensory information, rather than merely receiving it. Its specific contribution to prolonged holding and sensory variability is developed in B11.


Scientific references

Active touch and manual exploration

  1. Gibson, J. J. (1962). Observations on active touch. Psychological Review, 69(6), 477–491. PMID: 13947730.
  2. Lederman, S. J., & Klatzky, R. L. (1987). Hand movements: A window into haptic object recognition. Cognitive Psychology, 19(3), 342–368. PMID: 3608405.

Sensorimotor loops and active sensory sampling

  1. Bizzi, E., & Ajemian, R. (2020). From motor planning to execution: a sensorimotor loop perspective. Journal of Neurophysiology, 124(6), 1815–1823. PMID: 33052779.
  2. Schroeder, C. E., Wilson, D. A., Radman, T., Scharfman, H., & Lakatos, P. (2010). Dynamics of active sensing and perceptual selection. Current Opinion in Neurobiology, 20(2), 172–176. PMID: 20307966.

Prediction and internal feedback

  1. Blakemore, S.-J., Wolpert, D., & Frith, C. (2000). Why can’t you tickle yourself? NeuroReport, 11(11), R11–R16. PMID: 10943682.
  2. Li, J. S., Sarma, A. A., Sejnowski, T. J., & Doyle, J. C. (2023). Internal feedback in the cortical perception-action loop enables fast and accurate behavior. Proceedings of the National Academy of Sciences, 120(39), e2300445120. PMID: 37738297.
  3. Franklin, D. W., & Wolpert, D. M. (2011). Computational mechanisms of sensorimotor control. Neuron, 72(3), 425–442. PMID: 22078503.
Note : this content aims to explain mechanisms. It is neither a medical diagnosis nor a therapeutic prescription.