How does sensory adaptation occur

Introduction to psychology: gateways to mind and behavior with concept maps. Wadsworth; Webster MA. Evolving concepts of sensory adaptation. F Biol Rep. Your Privacy Rights. To change or withdraw your consent choices for VerywellMind. At any time, you can update your settings through the "EU Privacy" link at the bottom of any page.

These choices will be signaled globally to our partners and will not affect browsing data. We and our partners process data to: Actively scan device characteristics for identification.

I Accept Show Purposes. Was this page helpful? Thanks for your feedback! Sign Up. From a modeling perspective, the perfect adaptation case is of particular interest, because it entails the presence of a particular form of regulation known in control theory as integral feedback Perfect step adaptation means that, regardless of the amplitude of the input step being applied, the system is able to recover exactly the nominal value it had before the stimulus. The upper and lower parts of the panel show the response to the two main input protocols described in the text, steps and multiple pulse pairs here a series of 4 pulse pairs in which the second pulse is progressively delayed with respect the first; the 4 double pulse responses are shown all simultaneously and the 4 first pulses of each pair are all identical and overlapping.

Notice that the step responses resembles those of Figure 5 of Koshland et al. In particular perfect step adaptation requires exact integral feedback i.

See Supplementary Figure S1 for blown-up plots of the various cases. If, as in our sensory systems, we are able to apply a richer class of input profiles than just steps, then more features than simply steady state responses can be studied. From a dynamical perspective, in fact, the stimulation with time-varying input protocols provides information which is nonredundant with the steady state responses.

Combining this with the possibility of monitoring precisely the entire time history of a response, then more fine-graded hypotheses on the regulatory mechanisms encoded in the pathways can be formulated theoretically and validated or falsified experimentally.

For instance, if in a system with integral feedback instead of steps we apply a pair of nonoverlapping pulses, no difference should emerge in the elicited responses as we progressively increase the delay of the second pulse with respect to the first.

This is not what happens in sensory systems: if for short delays between the pulses adaptation manifests itself in a reduced amplitude of the transient response to the second pulse, increasing the lag time adaptation progressively reduces, until the system recovers completely, i.

Exact integral feedback which corresponds to an infinite time constant cannot achieve this, but a dynamical feedback with a suitable memory decay can accomplish the task. However, replacing an exact integral feedback with a dynamical feedback having a memory decay implies that perfect step adaptation is no longer possible. Also this prediction is coherent with the experiments. In both sensory systems, in fact, the step responses reset themselves only partially, never completely.

While the gap is minimal in the olfactory neurons, it is consistent in phototransduction, see Figure 2A and Figure 3A. In blue the fit with the dynamical model S9 described in the Supplementary Information.

In blue the corresponding fits with the dynamical model S9. Experiments were performed on two isolated olfatory neurons from Ambystoma Tigrinum salamander one for panel A and one for panel B. The blue traces show the response of the dynamical model S11 described in the Supplementary Information. In blue the fit of the dynamical model S11 is shown. In Ref. The limiting case of perfect step adaptation corresponds to no recovery at all in multipulse adaptation.

In the present paper this trade-off is investigated more in detail from both a theoretical and an experimental perspective. In particular, we observe that both our sensory systems obey to the rules imposed by this trade-off and the fact can be neatly observed in the transient profiles of the electrophysiological recordings. We show that the trade-off is naturally present also in basic regulatory circuits and that the time constant of the dynamical feedback can be used to decide the relative amount of the two forms of adaptation.

These elementary circuits help us understanding the key ingredients needed to have both forms of adaptation and confuting potential alternative models. For example, while it is in principle possible to realize some form of recovery in multipulse adaptation also in presence of exact integral feedback, we show that this requires necessarily a transient response that undershoots its baseline level during the deactivation phase, something that is not observed experimentally in neither sensor.

However, if we manage to artificially shift the baseline level for example performing phototransduction experiments in dim background light rather than in dark then our simplified model predicts that nonnegligible undershoots in the deactivation phase should emerge.

We have indeed verified their presence in experiments. Several input protocols, i. We have applied the first two protocols to olfactory neurons and photoreceptors, obtaining electrical recordings like those shown in Figures 2 — 3. The double step is instead used only for phototransduction shown in Figure 4.

A Double step protocol applied to the model 1. While the step response never exhibits deactivation undershooting i. The deactivation transients of the inner step overshoots the steady state corresponding to the outer step. The fitting for the model S11 is shown in blue. The responses to these input protocols for the two systems exhibit several common features which are highlighted below:.

No significant overshoot is observable for deactivation of the outer step. It is remarkable that both sensors exhibit input-output responses which are qualitatively similar for what concerns both the types of adaptation mentioned in the introduction.

Detailed dynamical models for the two sensory systems can be found in Refs. The approach followed in this paper is different: rather that including into our models all the kinetic details available for the two signaling pathways, we would like to introduce an elementary model which, in spite of its extreme simplicity, is nevertheless able to qualitatively capture the salient features of the various responses listed in the previous section. This basic model is presented now in general terms.

Later on an interpretation in terms of the specific signaling mechanisms of the two pathways is provided. More specific models tailored to the two transduction processes are discussed in the Supplementary Information. Consider the 2-variable prototype regulatory system depicted in Figure 1A. It represents a system in which two molecular species y and x are linked by a negative feedback loop. The following minimal mathematical model describes the reactions in the scheme of Figure 1A :. The external stimulus u favors the production of y , which is instead inhibited by the negative feedback from x.

In turn, the synthesis of x is enhanced by y. The model 1 is the simplest elementary dynamical system having an input-output behavior resembling that of olfactory transduction. A straightforward algebraic manipulation allows to rewrite the system 1 in terms of z.

In this case the regulatory actions have the opposite sign: u decreases z while the feedback from x promotes the formation of z. This is the minimal model which will serve as reference for the input-output behavior of phototransduction.

By construction, the models 1 and 2 exhibit the same dynamical behavior up to a flipping symmetry in the y and z variables. An exegesis of these models, explaining the role of each of the terms and including other technical details such as shifted baseline levels, is presented in the Supplementary Information. In particular, possible alternative minimal models are formulated and their responses investigated in Supplementary Figure S2 and S3.

While several models exist able to capture perfect step adaptation 14 , 15 , 20 , 12 , 23 , 24 , there is one general principle to which most proposals are equivalent, namely that perfect step adaptation in order to be robust to parametric variations must be obtained by means of a negative regulation and that this regulation achieving perfect step adaptation must be of integral feedback type, see Refs.

In our minimal models 1 and 2 , an integral feedback is obtained when the degradation rate constant for x vanishes i. This corresponds to the second differential equation of 1 being formally solvable as the time-varying integral. Hence the second pulse response is attenuated with respect to the first. However, lack of degradation of x t implies that the behavior occurs regardless of the lag time between the two pulses, which contradicts the experimental results shown in Figure 2B and Figure 3B.

Hence a perfect adaptation model is inadequate for our sensory transduction pathways, because i it fails to reproduce the non-exact return to the prestimulus level observed in the step responses of Figure 2A and Figure 3A and ii it completely misses the recovery in the multipulse adaptation observed in Figure 2B and Figure 3B.

In a model like 1 or 2 , both types of adaptation are determined by the ratio between the characteristic time constants of the two kinetics, which are captured with good approximation by the first order kinetic terms i.

This behavior is similar to what happens in our experiments with the olfactory transduction system shown in Figure 2. This situation resembles our experiments with phototransduction shown in Figure 3. Upon termination of a step, a response deactivates, meaning in our model 1 that the observable variable y returns to its pre-stimulus level y o which for simplicity and without loss of generality we are assuming equal to 0.

The way it does so is informative of the dynamics of the system. In a system with exact integral feedback, if the activation profile overshoots the baseline and then approaches it again, then the deactivation time course must follow a pattern which is qualitatively similar but flipped with respect to the baseline, i.

The undershooting in the deactivation phase should however be observable experimentally, i. No experiment with the olfactory system shows undershooting of the basal current. Also in phototransduction experiments, for both pulse and step responses in dark, no overshooting above the noise level can be observed in the deactivation phase.

For both sensors, this behavior is confirmed by many more experiments available in the literature 33 , 34 , 35 , 7 , 29 , 9. As discussed more thoroughly in the Supplementary Information , the lack of deactivation undershooting is another element that can be used to rule out the presence of exact integral feedback regulation in our systems. This is coherent with the step deactivation recordings shown in Figs. Assume now that the input protocol consists of a double step as in Figure 4.

This is indeed what happens for the model 1 , see Figure 4A. Given the very strong adaptation in olfactory sensory neurons, the double step experiment has been performed only in photoreceptors: indeed the combination of near zero baseline and almost perfect adaptation implies that in olfactory sensory neurons the presence of an overshoot will be hardly detectable.

In photoreceptors, instead, the double step deactivation behavior of 1 is faithfully reproduced. In the input protocol, the broader step of smaller amplitude corresponds to a constant dim light on top of which a more intense light step is applied.

The current recording shown in Figure 4B indeed exhibits a consistent deactivation overshooting not observed in dark. In this paper we will not attempt to present comprehensive mathematical models of the two sensory pathways containing all the biochemical reactions known to be involved in the signaling transduction of the stimuli, but will limit ourselves to consider the section of the pathways involving the Cyclic Nucleotide-Gated CNG channels and a primary calcium-induced feedback regulation.

The taste buds in our mouth play a critical role during eating. Our tongues have approximately 2, to 8, taste buds divided into four basic tastes: sour, sweet, bitter and salty. When eating a specific food, the initial taste is very distinct and identified by the tongue's sensory neurons.

As you continue eating the food, the taste is not as strong and does not have the same impact, which is due to sensory adaptation. Susan Henrichon has more than 25 years of experience in education. She has taught special education and possesses administrative experience in the public school setting. How the Human Nose Works.

What Is Tactile Stimulation? What Are Perceptual Illusions? Visual adaptation and face perception. Clifford CWG. Perceptual adaptation: motion parallels orientation. Trends in Cognitive Sciences. Neural locus of color afterimages.

Curr Biol. Kohn A, Movshon JA. Neuronal adaptation to visual motion in area MT of the macaque. Adaptation across the cortical hierarchy: Low-level curve adaptation affects high-level facial-expression judgments. Journal of Neuroscience. Global shape aftereffects have a local substrate: A tilt aftereffect field. Gollisch T, Meister M. Eye smarter than scientists believed: neural computations in circuits of the retina. Retinal adaptation to object motion.

Dynamic predictive coding by the retina. Texture interactions determine perceived contrast. Afraz A, Cavanagh P. The gender-specific face aftereffect is based in retinotopic not spatiotopic coordinates across several natural image transformations. Pulling faces: An investigation of the face-distortion aftereffect.

Face Adaptation without a Face. Current Biology. A motion aftereffect from still photographs depicting motion. Psychol Sci. Dils AT, Boroditsky L. Visual motion aftereffect from understanding motion language.

Sex-contingent face aftereffects depend on perceptual category rather than structural encoding. Morphing Marilyn into Maggie dissociates physical and identity face representations in the brain. Face adaptation depends on seeing the face. Opposite modulation of high- and low-level visual aftereffects by perceptual grouping.

Carandini M, Heeger DJ. Normalization as a canonical neural computation. Nat Rev Neurosci. Valentine T, Endo M. Towards an exemplar model of face processing: the effects of race and distinctiveness. Quarterly Journal of Experimental Psychology A.

Blakemore C, Campbell FW. On the existence of neurones in the human visual system selectively sensitive to the orientation and size of retinal images. Journal of Physiology. Human colour perception and its adaptation. Network: Computation in Neural Systems.

Adaptation and face perception - how aftereffects implicate norm based coding of faces. Aftereffects for face attributes with different natural variability: Adapter position effects and neural models. Rhodes G, Jeffery L.

Adaptive norm-based coding of facial identity. Figural aftereffects in the perception of faces. Psychonomic Bulletin and Review. Similar neural adaptation mechanisms underlying face gender and tilt aftereffects. Vision Res. Not all face aftereffects are equal. Hegde J. How reliable is the pattern adaptation technique? A modeling study. J Neurophysiol. Barlow HB. A theory about the functional role and synaptic mechanism of visual aftereffects. In: Blakemore C, editor. Visual Coding and Efficiency.

Cambridge: Cambridge University Press; Response normalization and blur adaptation: Data and multi-scale model. Kohn A. Visual adaptation: physiology, mechanisms, and functional benefits. Chopin A, Mamassian P.

Predictive properties of visual adaptation. Contrast gain control in the cat visual cortex. Greenlee MW, Heitger F.