Which term refers to the type of stimulus or sensation produced when a sensory receptor is activated multiple choice question?

Sensory pathway

A sensory unit is defined as a single nerve axon and all the sensory receptors which transmit information to it. Each sensory unit, therefore, has a receptive field: in the case of touch or temperature receptors on the body surface, it is the area of skin wherein are situated all the receptors connected to the single sensory axon. For visual sensation the receptive field of a sensory unit is the solid angle monitored by the receptors linked to the nerve axon of the unit. Discrimination is the ability to perceive two or more stimuli as separate. A classical test of touch discrimination is to apply the points of a pair of dividers to the skin surface and determine the least distance between the points at which they are perceived as separate. The lips and the tongue have a high discrimination for touch – the dividers can be detected as two points when they are barely separated – but on the trunk touch discrimination is low. Visually, discrimination represents the ability to see two points as separate. High discrimination implies a low ratio of receptors to nerve fibres: the sensory unit is small and the receptive field is small. High discrimination, however, usually carries the penalty of low sensitivity unless the receptor density is very high. Sensitivity is the ability to measure small changes in stimulus intensity: for this purpose a high ratio of receptors to nerve fibres is preferable. The intensity of stimulus is coded by single receptors but the more receptors there are involved the more effectively changes in intensity can be detected. The bigger the stimulus, the more receptors will be stimulated. Size of receptor field and density of receptor distribution are both important factors in sensitivity of a sensory unit.

The sensory nerve fibre represents the final common pathway for all its receptor areas. Impulses passing to the brain may, however, be modified at any synapse en route, and the final sensation perceived is a result of the original stimulus together with all the interactions that may have occurred at spinal cord, reticular formation, thalamic or cortical levels. It is the product not of discrete sensory impulses but of the pattern of activity in a number of nerves.

1.12.2.2 Structure

The sensory unit of the mechanosensory lateral line is the neuromast, an onion-shaped cluster of cells (typically tens or hundreds of micrometers in diameter) resting on the basement membrane of the epidermis (see Webb, 2014 for review). Neuromasts contain apically located receptor cells called hair cells intermingled with support cells of multiple types. Both the support cells and the sensory cells derive from the dorsolateral placodes (Ghysen and Dambly-Chaudiere, 2007; Northcutt et al., 1994; Piotrowski and Baker, 2014). Each hair cell projects a single apical kinocilium, flanked on one side by a bundle of microvilli (stereocilia) that decrease in height with distance from the kinocilium. This apical polarity defines the functional response of the hair cell, because deflections of the cilia toward the stereocilia hyperpolarize the receptor potential, whereas deflections away from the stereocilia are excitatory (Flock, 1971). Typically, each neuromast contains hair cells with a common single axis of sensitivity, often with pairs of opposing polarity, giving the neuromast a single axis of best sensitivity and two populations of responses according to direction of stimulation (Chagnaud and Coombs, 2014). The apical ciliary bundle of each neuromasts projects outward into a gelatinous cupula that is friction-coupled to the surrounding medium. In gnathostomes, there is a set of efferent neurons in the hindbrain that is thought to modulate neuromast function during locomotion and in response to ambient conditions (Chagnaud et al., 2015; Tricas and Highstein, 1991).

Neuromasts are found in a variety of configurations and integumental positions, generally divided into superficial neuromasts that rest on the surface of the epidermis and canal neuromasts, found in subdermal canals that have invaginated from the epidermis during development (Coombs et al., 1988). This division into canal and superficial organs is functionally very important, as the integumental structures dictate the hydrodynamic environment of the neuromast (Van Netten and Mchenry, 2014). Further, the canal orientation and pore distribution act as spatial filters and can provide great signal-to-noise averaging at some cost in spatial resolution (Klein and Bleckmann, 2015). Canal morphology, particularly its diameter and structural properties of the canal wall, alters the performance of the canal as a high-pass filter. Narrower, stiffer canals create more viscous resistance and reduced fluid flow within the canal at low frequencies. Wider canals allow more low-frequency flow within the canal (Denton and Gray, 1988, 1989) and are often seen as most useful in low-flow environments, where low-frequency stimuli are not swamped by ambient noise (Gray and Best, 1989; Janssen, 1997; Schleuter and Eckmann, 2006; Montgomery et al., 1994). Pits, trenches, and other integumental structures also influence the hydrodynamics of signal detection by channeling or altering flow past the neuromast.

The friction coupling of the cupula to the surrounding medium means that displacement of the ciliary bundle (the proximate stimulus) is proportionate to the velocity of fluid flow in the vector of the neuromast's best sensitivity (Denton and Gray, 1988; Kalmijn, 1988a). For neuromasts directly on the body surface, the cupula extends out past the boundary layer of the fish and is displaced in direct proportion to the velocity of water flowing past the cupula. When embedded in a canal, the neuromast's axis of sensitivity is aligned with the canal. The cupula of canal neuromasts also moves in proportion to the velocity of fluid flow within the canal, but flow within the canal is proportionate to the pressure difference at the two pores on either side of each neuromast, ie, the acceleration across the body surface. Thus the neuromast is a displacement-sensitive organ that detects a mixture of fluid velocity and acceleration (depending on the surrounding integumental structures) at a specific point (and in a specific vector of flow) on the body surface (Van Netten and Mchenry, 2014; Kalmijn, 1988a). The lateral line as a whole provides the central nervous system with a complex pattern of inputs that represents a time-varying three-dimensional pattern of fluid flow across the body surface (Chagnaud and Coombs, 2014; Coombs and Braun, 2003; Coombs et al., 1996, 2000). The contrasting properties of canal and superficial neuromast responses (and their distribution on the body) suggest that canal-based systems can provide good information about punctate sources (object localization), while clusters of superficial neuromasts may provide better information about more global flow patterns (Braun and Sand, 2014; Bleckmann and Mogdans, 2014).

The configuration of a neuromast, either superficially or within a canal, is extremely important for its biomechanical function, but it is also a very evolutionary-labile trait (Webb, 2014; Coombs et al., 1988). In many taxa, paedomorphic changes result in delayed or altogether reduced canal development. In these cases, neuromasts that were ancestrally housed in a canal are superficially located in some derived taxa. Thus homologous groups of neuromasts (eg, the supraorbital line) may be housed in a canal in one species but in a superficial groove in a closely related species. These secondarily superficial neuromasts have been called replacement neuromasts (Coombs et al., 1988). It is reasonable to assume that lateral line function plays an important role in the selective forces behind such evolutionary shifts, but the dermal canals are also so embedded in the overall development of the skull that many forces must interact to determine the evolutionary trajectory of the canals in any given species (Webb, 2014). In some taxa (especially in teleosts), large numbers of superficial neuromasts are also found in new locations on the body surface, with no evolutionary precedents (see Section 1.12.2.3.5) in nonteleosts. To properly understand the evolution of the lateral line system, it is important to recognize neuromasts not on the basis of their position in the skin and biophysical properties, but rather in the context of their history as specific components of a multichannel sensory system that arose with the origin of vertebrates and has been modified and augmented in descendent lineages.

The model and its behavior

In this model, two sensory units (analogous to B and Y in Figure 3A) represented the two decks of cards and provided weighted input to unit P. The output of P, δ(t), was used to decide whether the current deck was selected. The model made choices by making random transitions between Deck A and Deck B, thus inducing fluctuations in δ(t) as before. The probability that the current deck was selected was:

(8)Ps=11+expmδt+b

where r(t) = 0 before a card is actually chosen. The model randomly chose one deck as a starting point and “looked back and forth” between decks; the fluctuations in δ(t) assigned a value to the transitions between choices; and Ps determined the probability that a given deck was selected after a transition (see Figure 5A). Analogous to the bee example, the sensory weights (wA and wB) determined the sign and magnitude of fluctuations in δ(t), and thus influenced the choices between the decks. Once a deck was selected, the weight for the selected deck was updated according to a Rescorla-Wagner rule.

(9)Δwsfn=λrs fn−1−wsfn−1

fn is the fraction of choices from Deck A at iteration n, ws(fn-1) is the weight associated with the selected deck at iteration n-1, rs(fn-1) is the reward associated with the deck selected, and X is a learning rate (or equivalently a forgetting rate). For the networks, the fraction of choices from Deck A converged to the zone 0.31-0.41 (Figure 5B). This range includes the crossing point of the reward functions. The slope of the linear portion of the decision function (m) was varied from (-0.1,-5.0) and b was varied from (0.0, 15.0). In preliminary experiments, shown in Figure 6A, human subjects tended to fluctuate near the crossing point of the reward functions or remain near the optimal fraction of choice from Deck A (Egelman et al, 1995). These experimental results are compared to results with the model using various initial conditions, learning rates, and parameters for the decision function.

The humans and the networks tend to fluctuate around the crossing point of the reward functions, so that the average rate of return from the two choices is approximately equal. In experiments where an animal is given multiple behavioral alternatives, each of which yields rewards of various sizes or strengths, the animal tends to adjust its sampling of alternatives so as to match the relative rewards obtained from each alternative. In view of the importance of certain diffuse systems (e.g., dopamine) for reward-dependent behavior, our use of the diffuse-system output to constrain action choices provides one bottom-up explanation of how diffuse systems may establish constraints that favor matching.

Publisher Summary

This chapter discusses the analysis and interpretation of phasic potentials from the tooth. The recording of potentials from sensory units always poses problems as to the interpretation of these phasic potentials because intracellular recording under direct observation is rarely possible and would certainly invalidate any quantitative analysis of transducer action even if it could be done. The evidence for the validity of the recorded potentials arises in their consistency with the conditions of stimulus and environment. Phasic potentials can be recorded from electrodes in dentinal cavities and evidence has been presented to support the belief that these arise from the excitation of sensory units. The chapter discusses that these potentials are of two types based on their differential thermal excitability and other characteristics. It presents experiments which suggest that at least one of these two potential types may originate from structures in dentin because phasic potentials are seldom recorded from teeth having large pulps.

Publisher Summary

This chapter presents experiments to investigate the functional properties of nerve endings in teeth and, in particular, to record their responses to stimuli that produce pain in man. To study the properties of the primary afferent sensory units, the resolution of the recording system must be adequate to permit the discharge of a single nerve cell body or fiber to be identified and separated from that of adjacent cells or fibers. This can be done in one of three ways: (1) dentine recording with an electrode in contact with coronal dentine in a cat's canine tooth, (2) peripheral nerve recording using conventional techniques, and (3) cell body recording with microelectrodes in the trigeminal ganglion. This method has only recently been applied successfully to recording from pulpal afferents. It is hoped that it will enable one to record from a larger proportion of the nerves with small diameter axons than is possible with the other methods. In study, it is found that there is a little correlation between the properties of the nerve endings and the properties attributed to the nerve endings responsible for pain in man.

1.12.3.1 Generation of Sensilla Including Olfactory Receptor Neuron Neurogenesis

In all holometabolous insects, antennae derive from imaginal discs, which exist in the larval stage as flattened pouches of undifferentiated epithelial cells set aside for later use. Here the focus is on the generation of the individual sensory units, the sensilla, from the initially unpatterned disc epithelium; for further information (see also Chapter 1.11 on sensillar development). The formation of external sensory organs (of which olfactory sensilla are an example), and most particularly mechanosensory bristles, has been well studied in a large variety of insects. The general mode of development is the selection from cells in the epithelial field of sensory organ precursors (SOPs). Each SOP is usually selected out from one of the many proneural cell clusters in the epithelium – typically one cell within each cluster adopts the SOP fate and delaminates from the epithelium. The SOP then undergoes a small number of stereotyped asymmetric divisions to generate the different cell types within each sensillum (Jan and Jan, 1993).

Comparative studies have suggested that these lineages are rather conserved between similar sensory organs in different species (Bate, 1978), and that basic patterns of division may be reiterated with small variations in the generation of different kinds of sensory organs (Bellaiche and Schweisguth, 2001). This lineage-based generation of sensilla is in marked contrast to the generation of the photoreceptors of the compound eye, which proceeds by the sequential recruitment of neighboring cells via cell–cell interactions initiated by a founder cell (Dickson and Hafen, 1993).

There are several general themes in the developmental biology of external sensory organs. (1) How are the number, spacing, and position of SOPs controlled? Key mechanisms are the existence of prepatterns of proneural clusters (defined by the expression of bHLH transcription factors) in some epidermal fields (reviews: Simpson, 1997; Simpson et al., 1999), and the employment of lateral inhibition via the Notch pathway. (2) What is the relative importance of lineage versus recruitment? (3) How are different subtypes of sensillum generated, particularly different neuronal types (for a general review of proneural genes and subtype specification, see Bertrand et al., 2002)? (4) How does the SOP divide asymmetrically to generate different cell fates (review: Jan and Jan, 2001)? Some of these themes are addressed specifically in the generation of olfactory sensilla in Drosophila.

In Drosophila, the antennal imaginal discs are quite separate from the larval sensory structures, and already present as a group of 7–9 progenitor cells set aside in the embryo 2 h after egg laying (Postlethwait and Schneiderman, 1971). No further division takes place until 24 h after larval hatching (Madhavan and Schneiderman, 1977), when exponential growth proceeds through the remaining 4 days of larval life, so that there are about 2500 cells per disc at pupation (Postlethwait and Schneiderman, 1971). Around this time, differentiation of the epidermal field of the antennal disc begins and there is a halt in proliferation.

The concentric organization and development of the antennal imaginal disc has been described by Haynie and Bryant (1986). Precursor cells that will generate the third antennal segment (3AS) occupy a central location. These cells are surrounded by precursors of the second antennal segment (containing, in Drosophila, the auditory Johnston's organ) and, finally, at the periphery, by the first antennal segment precursors. During pupal metamorphosis, which lasts approximately 100 h in Drosophila, the antennal disc, like other imaginal discs, will telescope out so that the central regions become the most distal.

SOP specification and delamination occurs in three major waves (Figure 6) (Ray and Rodrigues, 1995; zur Lage et al., 2003), which can be visualized by expression of the senseless (sens) gene (Nolo et al., 2000). The first wave appears a few hours before puparium formation and consists of a well-spaced semicircle of SOPs at the periphery of the 3AS zone of the antennal disc. The second wave appears by 4 h after puparium formation (APF). It is more numerous, more centrally located, and consists of a stereotyped pattern of cells including two semicircles. Both of these waves of SOPs express the proneural gene atonal (ato), which is also required for the formation of chordotonal organs (Jarman et al., 1993) and photoreceptors (Jarman et al., 1994), and appear to be derived from ato expressing proneural clusters in the overlying epithelium (Jhaveri et al., 2000a; zur Lage et al., 2003). Both are absent in atonal mutant antennae and functionally will generate sensilla coeloconica of the sacculus (wave 1) and the main antennal surface (wave 2) (Figure 6) (zur Lage et al., 2003). In contrast to these first two waves, amos expressing SOPs derived from overlying Amos positive proneural clusters, appear in a third wave first evident at 8 h APF. These SOPs initially intercalate between the wave 2 ato-dependent SOPs. However, by 16 h APF these amos SOPs are much more numerous and less obviously spatially patterned.

zur Lage et al. (2003) were able to identify an amos enhancer whose expression was maintained in SOP progeny, which turned out to be all cells in the sensilla basiconica and trichodea; this is satisfyingly consistent with the amos loss-of-function phenotype, which is a complete loss of sensilla basiconica and trichodea (zur Lage et al., 2003).

In short, the generation of the different SOPs responsible for the three classes of olfactory sensilla has now been established with some precision and conforms, in general, to the expected norms of sensory organ development. However, the next step, which would normally be the stereotyped division of the SOPs, remains somewhat less clear. Mitotic cells are not detected in the 3AS zone of the disc until about 12 h APF and, even then, they are in the periphery of the lobe, corresponding to wave 1 cells, which delaminated at least 12 h earlier (Sen et al., 2003). In the meantime, it seems that the SOPs (or, more appropriately, founder cells) recruit additional neighboring cells to form a presensillum cluster (PSC) of 2–3 cells, which can be labeled by the A101 (neuralised; neu) enhancer trap (Reddy et al., 1997). This conclusion was supported both by BrdU labeling, with no division detected between the generation of SOPs and the formation of PSCs (Ray and Rodrigues, 1995), and by clonal analysis (Reddy et al., 1997). Clones were generated at mid first larval instar, i.e., very early in antennal disc formation, and then analyzed at 48 h APF, i.e., mid-pupal stage, in principle after the final division of sensillar progenitors (see below). In all 29 sensillar clusters analyzed, clonal boundaries separated cells of the same sensillum. This unusual mode of PSC development is similar to the recruitment of neighboring cells by R8 photoreceptors, which, incidentally, are also ato dependent like waves 1 and 2 olfactory SOPs. Reddy et al. (1997) confined their studies to those SOPs generated within the first 10 h of larval development. It will be very interesting to determine if the more numerous amos dependent SOPs of the third wave show a similar behavior.

After the formation of the PSC, these cells finally divide at least once to generate the sensillar components (Figure 7). This period of division is reported by Sen et al. (2003) to last from 12 to 22 h APF. Sen et al. (2003) have identified at least three different division patterns by analysis of clones generated at 10 h APF, i.e., before division starts, and analyzed at 36 h APF, i.e., after division ends (Figure 7b). Two of these patterns are directly analogous to the division pattern of the adult sensory bristle (Figure 7a versus b) and are named accordingly. The pIIa precursor generates the socket and shaft accessory cells, while pIIb generates a migratory glial cell and another intermediate, pIIIb, which gives rise to a neuron and the sheath (the most glial-like of the sensillar accessory cells). In fact, only the ato-dependent coeloconic sensilla show the full pIIb lineage, since they are gliogenic (Jhaveri et al., 2000b). The corresponding PSC cells in the trichoid and basiconic lineages, which are nongliogenic, may proceed directly to the pIIIb division, or the equivalent of the glial cell could undergo apoptosis. Finally, there is a division pattern labeled pIIc by Sen et al., (2003), which generates a pair of neurons.

In short, it is proposed that the generation of olfactory sensilla follows a hybrid pattern, in which the newly selected SOP recruits neighbors, rather than undergoing an initial division, but this cluster of cells then divides in a manner typical of other sensory organ lineages. It is possible that such a system allows more flexibility in the generation of sensilla with different numbers of ORNs, perhaps by the recruitment of 0, 1, or 2 cells, which will undergo the neurogenic pIIc division pattern.

Possible Functional Significance of the Sympathetic Vasomotor Control in the Pulp and the Periodontal Tissues

The present studies have indicated that activation of the sympathetic nerves to the oral tissues results in a pronounced reduction of blood flow in the pulp and the periodontal tissues. It is obvious that due to the small total mass of the jaws in relation to total body mass, the redistribution of cardiac output produced by maximal sympathetic nerve activation to the oral tissues is insignificant for the circulatory homeostatis. Instead, the influence of sympathetic nerve activation on the vascular beds in the pulp and the periodontal tissues may be of local significance, e.g. for the function of the jaws.

In order to study in some detail the possible local importance of the sympathetic vaso-constrictor control in the jaws, a series of experiments was performed in which the sympathetic influence on the excitability of intradental neural elements was investigated (Edwall and Scott, 1971).

In the canine tooth of the cat tracer disappearance measurements from a depot in a deep dentinal cavity as described above was combined with recording of intradental sensory nerve impulse frequency. A typical experiment is shown in Fig. 5. Sympathetic stimulation with 3 imp/sec (Fig. 5:1) reduced the tracer disappearance rate and induced an initial increase in impulse frequency of the intradental sensory unit. This increase was followed by a marked decrease. Rapid recovery with an overshoot following the stimulation was seen for both parameters. In contrast to these marked effects, stimulation with 1.5 imp/sec (Fig. 5:2) showed a small decrease in k-value during the period of stimulation. There was, however, an initial increase in impulse frequency of more than 50% during the first half of the stimulation period. During the third period of stimulation with 6 imp/sec (Fig. 5:3) the resultant changes in both impulse frequency and k-value showed a marked similarity to the pattern observed when stimulation at 3 imp/sec was applied. This experiment indicates that the excitability of sensory terminals may be modulated by a change in pulpal blood flow induced by the sympathetic nerve activation.

Rapid increases in impulse frequency with temperature have been known from previous studies (Scott, 1966) but it was not known to what extent this was the property of the sensory unit and to what extent this was an indication of the effect of vascular reactions. Figure 6 shows the isolated influence of a thermal heat pulse. As can be seen a rise in temperature of 6°C resulted in an increase in k-value and an explosive rise in impulse frequency from the intradental sensory unit.

Since sympathetic stimulation and heating produced opposite effects on circulation it was important to examine their relative effectiveness in a combined procedure.

For this purpose (Fig. 7) sympathetic stimulation was first applied and continued for 6 min at 6 imp/sec. Three minutes after the start of this stimulation the temperature was increased by 6°C and this was continued for 5 min. During the initial 3 min period the k-value fell to zero and the impulse frequency underwent a rapid increase followed by an equally abrupt fall to near zero value. The subsequent period of combined stimulation showed a slow rise in k-value and impulse frequency which does not resemble either the further depression seen with continued sympathetic stimulation (Fig. 5:1) or the explosive rise in frequency observed at the initiation of warmth alone (Fig. 6). In the final two min when warmth stimulation alone was applied both k-value and impulse frequency rose rapidly to supernormal values as though released from a restrictive constraint.

Thus, thermal stimulation superimposed during a maximal sympathetic stimulation, when k-value was depressed, was inadequate to evoke an appreciable increase in sensory nerve impulse frequency. These results indicate that the excitability of intradental sensory units, and consequently the afferent flow of impulses from these units, depends on the integrity of the pulpal microcirculation and is strongly modulated by stimulation of sympathetic vasoconstrictor fibres.

Sympathetic nerve stimulation has been shown to modulate the excitability of different types of mechanoreceptor, for example the muscle stretch receptor and cutaneous receptors (cf. Paintal, 1964), the pattern of response being similar to that observed in the present study.

Therefore, it seems possible that the sympathetic vasomotor control to the oral tissues may serve to modulate the afferent flow of impulses from receptors localized in other parts of the jaws, for example the periodontal membrane.

Predictive coding

We have now established an objective function that is maximized to enable inference and learning in E- and M-steps respectively. Here, we consider how that maximization might be implemented. In particular, we will look at predictive coding, which is based on minimizing prediction error (Rao and Ballard, 1998). Prediction error is the difference between the data observed and that predicted by the inferred causes. We will see that minimizing the free energy is equivalent to minimizing prediction error. Consider any static non-linear generative model under Gaussian assumptions:

36.6u=g(v,θ)+ɛ(1)v=η+ɛ(2)

where Cov(ɛ(1)) = Σ(1)) is the covariance of random fluctuations in the sensory data. Priors on the causes are specified in terms of their expectation η and covariance Cov(ɛ(2)) = Σ(2)). This form will be useful in the next section when we generalize to hierarchical models. For simplicity, we will approximate the recognition density with a point mass. From Eqn. 36.4:

36.7F=−12ξ(1)Tξ(1)−12ξ(2)Tξ(2)−12ln|∑(1)|−12ln|∑(2)|ξ(1)=∑(1)−1/2(u−g(μ,θ))ξ(2)=∑(2)−1/2(μ−η)

The first term in Eqn. 37.7 is the prediction error that is minimized in predictive coding. The second corresponds to a prior term that constrains or regularizes conditional estimates of the causes. The need for this term stems from the ill-posed nature of recognition discussed above and is a ubiquitous component of inverse solutions.

Predictive coding schemes can be seen in the context of forward and inverse models adopted in machine vision (Ballard et al., 1983; Kawato et al., 1993). Forward models generate data from causes (cf. generative models), whereas inverse models approximate the reverse transformation of data to causes (cf. recognition models). This distinction embraces the ill-posed nature of inverse problems. As with all underdetermined inverse problems, the role of constraints is central. In the inverse literature, a priori constraints usually enter in terms of regularized solutions. For example: ‘Descriptions of physical properties of visible surfaces, such as their distance and the presence of edges, must be recovered from the primary image inputs. Computational vision aims to understand how such descriptions can be obtained from inherently ambiguous and noisy inputs. A recent development in this field sees early vision as a set of ill-posed problems, which can be solved by the use of regularization methods’ (Poggio et al., 1985). The architectures that emerge from these schemes suggest that: ‘Feedforward connections from the lower visual cortical area to the higher visual cortical area provide an approximated inverse model of the imaging process (optics)’. Conversely: ‘while the back-projection connection from the higher area to the lower area provides a forward model of the optics’ (Kawato et al., 1993). This perspective highlights the importance of backward connections and the role of priors in enabling predictive coding schemes.

Mechanoreception

The lateral line functions to detect vibrations and water movement and allows fish to orientate themselves in a water current (rheotaxis), gain information about their spatial environment, and also plays a vital role in schooling (see also HEARING AND LATERAL LINE | Lateral Line Structure). The sensory cells within the lateral line are known as hair cells and are also present in the ear. In the lateral line, hair cells are contained in sensory units known as neuromasts. Toxicants that interfere with hair-cell function, therefore, have the potential to disrupt behaviors reliant on hearing and the lateral line. Ototoxins are contaminants known to specifically affect hair-cell function and include the pharmaceuticals such as gentamicin sulfate, streptomycin, and amiloride.

Trace metals may also interfere with lateral line function. For example, banded kokopu (Galaxius fasciatus) exposed to waterborne cadmium show a reduced ability to orientate in a water current. Waterborne copper exposure in zebrafish larvae reduces the number of functional neuromasts in the lateral line. In control zebrafish larvae, functional neuromasts can be visualized by staining with the fluorescent dye, 2-(4-dimethylaminostyryl)-N-ethylpyridinium iodide) (DASPEI). Figure 2(a) shows control zebrafish larvae with functional neuromasts running along either side of the body in the lateral line stained with DASPEI. Figure 2(b) shows zebrafish larvae exposed to waterborne copper before staining with DASPEI; a clear reduction in functional neuromasts can be seen. In consequence, zebrafish larvae exposed to waterborne copper during development have a reduced ability to orientate and maintain their position within a water current.

Receptors

Sensation begins most distally with the transduction of mechanical stimulation by mechanoreceptors in the dermis, epidermis, muscles, and joints. To stimulate the axon associated with a mechanoreceptor, the stimulus must first pass through intervening tissues. This process is referred to as stimulus accession. This is followed by stimulus transduction, in which the stimulus energy is transformed into electrical energy by depolarization of the axon terminal in proportion to the amount of mechanical energy applied. This depolarization is termed a generator potential and is due to influx of sodium ions through sodium channels that open in response to deformation of the mechanoreceptor. The greater the stimulus, the more sodium channels open and the larger the generator potential. If a threshold value is reached, an action potential occurs and nerve impulses travel along the sensory neuron. The amplitude of the mechanical stimulus determines the frequency at which the action potentials are initiated.8 This temporal summation of afferent impulses is preserved throughout the afferent course and is decoded centrally as subjective stimulus sensation magnitude. Increasing the intensity of a mechanical stimulus not only increases the firing frequency of the discharging mechanoreceptor but also recruits more sensory units. The intrinsic properties of the mechanoreceptors, however, determine the duration of the impulse activity when stimulated.9

Mechanoreceptors may be divided into rapidly adapting and slowly adapting based on their response to sustained stimuli (Fig. 19‐1). Rapidly adapting (RA) mechanoreceptors provide information about changes in skin contact. They quickly cease firing in response to a constant stimulus. Slowly adapting (SA) receptors provide information about long‐term contact with skin and do not stop firing with sustained stimulation. Mechanoreceptors may be further subdivided into type I and type II receptors. Type I receptors are located superficially in the skin and have small receptive fields. Type II receptors are deeply located and have broad, diffuse receptive fields.10

Four subtypes of mechanoreceptors have been identified: (1) Meissner's end organ (RA, type I) responds best to light touch; (2) Pacinian corpuscles (RA, type II) responds best to vibration; (3) Merkel's end organ (SA, type I) responds best to pressure; and (4) Ruffini's end organs (SA, type II) respond best to pressure.11 Any given mechanoreceptor may respond to various types of mechanical energy; however, most are particularly sensitive to one form of stimulation, as noted previously. This selectivity tends to be greatest at or near threshold levels of stimulation—that is, activation of the smallest number of sensory units necessary for stimulus perception.12 It has been demonstrated from intraneural microstimulation of a single myelinated fiber associated with a mechanoreceptor that elementary sensations can be perceived and sensory quality, magnitude, and localization can be resolved at a cognitive level.5 Microstimulation of a single RA unit produces a sensation of intermittent tapping or fluttering. These receptors are related to texture discrimination in glabrous skin such as the fingertips.13 Microstimulation of a Pacinian corpuscles produces a sensation of vibration or tickle. Microstimulation of an SA, type I receptor produces a sensation of pressure,5 and these receptors appear to be critical in resolving the spatial structures of objects or surfaces.13,14 Microstimulation of SA, type II receptors evokes no sensation when stimulated in isolation; however, they are excited by skin stretching and joint movement.15

The proprioceptive system arises from mechanoreceptors located in muscle spindles, Golgi tendon organs, and joint receptors. The Ia afferents that represent the primary endings of muscle spindles also produce no sensation when stimulated in isolation.16