1. The reliability of a sensory source is not fixed
To begin with, the reliability of a sensory source depends on the conditions under which it is used. Vision becomes less reliable in the dark or when the visual environment moves. Likewise, proprioceptionProprioceptionProprioception is the body’s ability to perceive the position, movement and effort of its various parts without needing to look at them.Read the full entry : Proprioception can be altered by fatigue, vibration or muscle disturbance. Finally, tactile information can be hard to interpret when contact slips or varies unpredictably, as B5 already pointed out regarding the reliability of contact.
Indeed, reliability is not an absolute property of a receptor or a sensory modality. Here it refers to the ability of a source to provide, in a given situation, sufficiently precise and consistent information. In a given situation, this information must be able to contribute to the postural estimate. Moreover, its reliability depends notably on the stability of the environment, the variability of the signal, the state of the subject, the task and its consistency with the other sources available.
Thus, the same modality can be highly informative under some conditions and become ambiguous under others. A stable visual field provides useful orientation cues. By contrast, a moving visual environment can attribute to the body a displacement that actually comes from the environment. Likewise, somatosensory information from a support surface becomes harder to interpret when that surface tilts or deforms independently of the body.
However, this variability is not an anomaly. It reflects the fact that each sensory source has a range of conditions in which it provides precise information, and other conditions in which its precision degrades. The nervous system must take this into account so as not to give a degraded source the same weight as under favorable conditions.
2. The principle of sensory reweighting
More precisely, weighting refers to the relative contribution of the different sources at a given moment. Reweighting refers to the process by which this contribution changes when conditions change. It therefore implies a temporal dimension. Indeed, the system does not merely combine available signals. On the contrary, it gradually adjusts how they take part in the postural estimate.
For example, standing postural control served as the model for establishing this principle. The experiments perturb the visual field and the support surface. Faced with these perturbations, the nervous system dynamically adjusts the relative contribution of visual, vestibular and proprioceptive information. These then contribute to the estimate of postural sway in a combination that is not fixed (Peterka, 2002).
In particular, this principle can be illustrated by a few typical cases studied in this experimental framework. When the support surface becomes unstable, the information coming from it tends to lose weight. Its contribution to the postural estimate then decreases. If vision is removed or made unreliable, vestibular and proprioceptive information tends to gain weight. Finally, when proprioceptive information is disturbed, the relative contribution of vision and the vestibular system may increase (Peterka, 2002). These examples illustrate a general principle rather than a strict rule: the magnitude and exact direction of these adjustments depend on the precise conditions of each situation.
This principle of dynamic reweighting complements the principle of reliability-based weighting already set out in B7. It is no longer just a matter of assigning a weight proportional to a source’s instantaneous reliability. This weight must also be gradually adjusted as sensory conditions change.
3. Reweighting and changing support conditions
Furthermore, the dynamics of this reweighting have been clarified by work on the combination of visual and proprioceptive information in response to tilts of the support surface. The weight assigned to each source changes gradually over several stimulation cycles. It therefore does not switch instantly from one source to another (Assländer & Peterka, 2014).
However, this adaptation does not necessarily follow the same dynamics in all directions. Assländer and Peterka observed slower adjustments after certain transitions. This is notably the case when moving from a large-amplitude perturbation to a smaller one. They also observed differences between reweighting within the same modality and reweighting across modalities. There is therefore no single, universal time course of reweighting.
Reweighting can thus be described at two levels. Intramodal reweighting changes the response to different signals or contexts within the same modality. By contrast, intermodal reweighting changes the relative contribution of different modalities, for example visual and proprioceptive information. The dynamics observed are not necessarily identical in these two situations (Assländer & Peterka, 2014).
Finally, this work concerns experimentally perturbed standing. The perturbation results from platform tilt and visual-field motion. By contrast, it does not address the sitting position or pelvic support. The principle it establishes is one of gradual, non-binary reweighting. It can provide a framework of analysis for other postural contexts. However, its transposition must be verified experimentally in each of these contexts.
4. Reweighting as a regulatory mechanism
For example, models of postural control have sought to represent reweighting as an automatic mechanism that reduces the contribution of a source. This occurs when the deviations it produces become incompatible with the overall estimate of movement. A model of this type was thus proposed to reproduce certain adjustments observed during visual and support-surface perturbations (Mahboobin et al., 2009).
This type of model thus shows that a reweighting mechanism can account for measured postural behavior. However, it does not demonstrate that the nervous system literally performs the mathematical calculation used by the model, nor that the same mechanism applies unchanged to the sitting position.
5. Sensory conflicts
In contrast, a sensory conflict arises when two sources provide clearly incompatible information. It is then no longer a matter of estimates that are simply more or less precise than each other. For example, a support surface tilts while a visual field signals movement in a different direction (Peterka, 2002). This case differs from the arbitration discussed in B7 for moderately discordant sources: here, the discordance is deliberately large and prolonged in the experimental protocols that study it.
Consequently, when faced with discordant information or information that has become less usable, the postural system can change its relative contribution to the estimate of movement. Experimental work notably shows reduced reliance on certain visual or proprioceptive information when the conditions that produce it are perturbed. This adaptation does not necessarily require explicit identification of a “false” source or its complete exclusion.
Several types of response remain compatible with this cautious description. On the one hand, the system can reduce the contribution of one source or increase that of another. On the other, it can rely more on the predictive models presented in B7, or slow down or reduce movement. These different responses are not mutually exclusive, and there is no indication that any one of them systematically dominates.
Finally, a source can be unavailable, like vision when the eyes are closed. It can also be degraded when its precision decreases. Conversely, it can be discordant when it indicates a state that is hard to reconcile with the other sources. These situations are not equivalent and do not necessarily trigger the same adaptation. Reweighting is therefore not a uniform response to every sensory change.
6. Not confusing reweighting with total substitution
Indeed, it would be excessive to describe sensory reweighting as the complete replacement of one source by another. The studies cited show a gradual and partial change in relative weights, not a wholesale switch from one source to another. Thus, even a degraded source generally continues to contribute, to a reduced extent, to the postural estimate.
This nuance matters to avoid an overly mechanical reading of this chapter: losing or degrading a sensory source does not mean that another source automatically and completely takes over.
7. Consequences for postural control
In addition, this reweighting is not merely perceptual. It has direct motor consequences, since the postural estimate adjusted in this way feeds the sensorimotor control loop described in B1. A change in the relative reliability of sources can therefore produce a measurable change. This concerns the postural response. Yet no equivalent external mechanical change has necessarily taken place. Multisensory models of postural control describe this integration as a central component of stability, studied mainly in standing (Chiba et al., 2016).
Moreover, the motor consequences of reweighting also depend on the task and on the subject’s response options. A change in sensory weight can alter the amplitude, timing or organization of postural corrections. It does not, however, determine motor strategy on its own: task goals, the subject’s biomechanical capacities, expectations and experience also shape the response.
This link between sensory reweighting and motor response remains established mainly for the control of standing. By contrast, its detailed transposition to postural adjustments in the sitting position is not demonstrated by the studies cited in this chapter.
8. What this chapter does not cover
In contrast, this chapter describes how the weight of sensory sources readjusts in response to changes in reliability or to conflicts. It does not address how the subject can actively produce new sensory information through voluntary movement, notably by changing position themselves — a question developed in B9.
9. Application to sitting
In the sitting position, support conditions can vary. For example, there may be a change in perceived firmness depending on posture or a change in load distribution. Likewise, contact with the backrest or armrests can vary. The principle of sensory reweighting established for standing could transpose, at least partially, to these situations. In that case, one can hypothesize that a change in support conditions gradually changes the weight the nervous system gives to the various tactile and proprioceptive sources available.
In addition, the predictability of a change of support could also affect how usable the resulting information is. However, the references used here do not make it possible to establish a distinction in the sitting position. The nervous system may encounter a change chosen by the subject or a predictable response of the support. It may also encounter an imposed or unpredictable displacement. This question calls for specific research.
Nevertheless, this transposition remains a working hypothesis. It is not a conclusion established by the studies cited. Indeed, none of them concerns the sitting position. The speed or magnitude of any reweighting in this context therefore remains to be documented. It would thus be premature to claim that a dynamic seat demonstrably triggers beneficial sensory reweighting.
10. Key points
- The reliability of a sensory source varies with conditions; it is not a fixed property but depends on the environment, the task and its consistency with the other sources.
- In addition, weighting describes the contribution of sources at a given instant; reweighting describes how this contribution changes over time, gradually and sometimes asymmetrically.
- Finally, reweighting can be described at two levels — intramodal and intermodal — and automatic models can reproduce certain measured postural adjustments, without demonstrating the exact neural mechanism.
- A discordance or degradation of sensory conditions can lead to a change in the relative contribution of sources, without automatically excluding any of them.
- This reweighting has direct motor consequences for postural control, modulated by the task and the subject’s capacities.
- Its transposition to the sitting position remains a working hypothesis, not demonstrated by the available studies, all of which concern standing.
In sum, the sensory reweighting described here complements the multisensory integration set out in B7. It explains how the weight of sources changes over time, not just how they are combined at a given instant. How the subject’s voluntary movement itself produces new information, rather than passively waiting for a change in conditions, is the subject of B9.
Scientific references
Sensory reweighting and postural control
- Peterka, R. J. (2002). Sensorimotor integration in human postural control. Journal of Neurophysiology, 88(3), 1097–1118. https://doi.org/10.1152/jn.2002.88.3.1097
- Assländer, L., & Peterka, R. J. (2014). Sensory reweighting dynamics in human postural control. Journal of Neurophysiology, 111(9), 1852–1864. https://doi.org/10.1152/jn.00669.2013
- Mahboobin, A., Loughlin, P. J., Atkeson, C. G., & Redfern, M. S. (2009). A mechanism for sensory re-weighting in postural control. Medical & Biological Engineering & Computing, 47(9), 921–929. https://doi.org/10.1007/s11517-009-0477-5
Multisensory models of postural control
- Chiba, R., Takakusaki, K., Ota, J., Yozu, A., & Haga, N. (2016). Human upright posture control models based on multisensory inputs; in fast and slow dynamics. Neuroscience Research, 104, 96–104. https://doi.org/10.1016/j.neures.2015.12.002
