Tactile hair, or vibrissae, are a mammalian characteristic found on many mammals (Ahl, 1986). Vibrissae differ from ordinary (pelagic) hair by being longer and thicker, having large follicles containing blood-filled sinus tissues, and by having an identifiable representation in the somatosensory cortex. Vibrissae are found on various parts of the body, but those most frequently studied are the facial or mystacial vibrissae, also called whiskers. Long facial whiskers, or macrovibrissae, are found in many mammalian species, projecting outwards and forwards from the snout of the animal to form a tactile sensory array that surrounds the head. For example, in rats, the macrovibrissae form a two-dimensional grid of five rows on each side of the snout, each row containing between five and nine whiskers ranging between ~15 and ~50 mm in length (see Figure 1 for illustration). The macrovibrissae of many rodents and some other species can move back and forth at high-speed thus explaining the term "vibrissa" which derives from the Latin "vibrio" meaning to vibrate.
The study of any behavior involves identifying the circumstances under which it arises, and then characterizing its nature, as precisely as possible, in all of the relevant contexts. However, in addition to describing what animals do, behavioral science also seeks to understand the function of behavior, both 'proximally' in terms of its immediate consequences for the animal, and 'ultimately', in terms of its adaptive significance and contribution to the evolutionary fitness of the species. In the context of the vibrissal system, we are only just beginning to piece together descriptions of how, and in what contexts, animals use their whiskers. Even less is known about the function of vibrissae, beyond the obvious intuition that whiskers are 'for touch' just as the eyes are 'for sight'. We would like to understand much more about the specific contribution of the vibrissae to the life of the animal, both in order to explain the emergence, through natural selection, of this important mammalian sense, and also to be able to frame better functional hypotheses for physiologists investigating its biological and neural substrates. Here, we provide a brief comparative and ethological overview of vibrissal behavior and function.
Contents |
Comparative studies of vibrissal sensing are important as they indicate the various roles of vibrissal touch in different environmental settings (e.g. in air or under water) and can reveal evolutionary convergences and divergences that allow us to draw conclusions about the generality of observed solutions. Vibrissal sensing is most often studied in rodents, aquatic mammals, and most recently in shrews (insectivores). Brecht (1997) has provided a morphological analysis of vibrissae systems in ten mammalian species including examples from marsupials, rodents, insectivores, pinnipeds, and primates, concluding that the presence of multiple rows of macrovibrissae increasing in length along the rostrocaudal axis is a shared feature of mammalian vibrissal sensing systems. Ahl (1986, 1987) has also reviewed comparative data from a wide range of mammalian groups, concluding that there is great variation between species but low variation within species, and arguing that these differences in vibrissal morphology could provide useful clues to function. For instance, a study of Old World field mice (genus Apodemus) found that a smaller facial vibrissal field was associated with a burrowing lifestyle and a larger field with a more arboreal one. Vibrissal function in a number of tactile sensing specialists is explored in the articles on seals (Hanke and Dehnhardt, 2015), manatees (Reep and Sarko, 2009), Etruscan shrews (Roth-Alpermann and Brecht, 2009). These studies provide islands of illumination, in a predominantly dark comparative landscape; much further research is needed to characterize the natural variation in vibrissal sensing systems across the different mammalian orders. By far the largest amount of research has concerned the facial vibrissae of mice and rats, which therefore form the main focus of the remainder of this article.
In rats and mice the facial whiskers are repetitively and rapidly swept back and forth, during many behaviors including locomotion and exploration. These "whisking" movements generally occur in bouts of variable duration, and at rates between 3 to about twenty-five “whisks” per second (with rates generally being somewhat faster in mice than in rats). Movements of the whiskers are also closely coordinated with those of the head and body allowing the animal to locate interesting stimuli through whisker contact, then investigate them further using both the macrovibrissae and an array of shorter, non-actuated microvibrissae on the chin and lips (see Figure 2). Movement of the vibrissae and its measurement is discussed at length in Zeigler and Keller (2009) and Knutsen (2015). A video illustrating whisking behavior in rats is available here.
Whisking is observed in only a sub-set of animals possessing prominent macrovibrissae. In addition to some species of rats and mice, it has also been reported in flying squirrels, gerbils, chinchillas, hamsters, shrews, the African porcupine, and in two species of marsupial—the Brazilian short-tailed opossum Monodelphis domestica, and the Australian brush-tailed possum (Rice, 1995; Welker, 1964; Wineski, 1985; Woolsey, Welker, & Schwartz, 1975; Mitchinson et al., 2011). Although whisking appears to be most prominent in rodents, there are several rodent genera, such as capybara and gophers, that do not appear to whisk, and others, such as guinea pigs, that display only irregular and relatively short whisking bouts (Jin et al., 2004). Whisking behavior has not been observed in carnivores (e.g. cats, dogs, raccoons, pandas), although some species, such as pinnipeds have well-developed sinus muscles making the whiskers highly mobile (Ahl, 1986). In animals such as rats and mice, that are capable of whisking at high frequencies, the whisking musculature contains a high proportion of type 2B muscle fibers that can support faster contractions than normal skeletal muscles (Jin et al. 2004). A comparative study of whisker movement in rats, mice, and the marsupial M. domestica found similar patterns of whisker movement, and of active whisker control (see below), in all three species, with mouse whisking having the most complex movement patterns. The presence of whisking in both rodents and marsupials implies that early mammals may also have exhibited this behavior, with evidence of similarities between the whisking musculature of rats and the marsupial opossum adding further support for the idea of a whisking common ancestor for modern mammals (Grant et al., 2013) (see also Haidarliu, 2015).
Since rapid movement of the vibrissae consumes energy, and has required the evolution of specialised musculature, it can be assumed that whisking must convey some sensory advantages to the animal. Likely benefits are that it provides more degrees of freedom for sensor positioning, that it allows the animal to sample a larger volume of space with a given density of whiskers, and that it allows control over the velocity with which the whiskers contact surfaces. In addition, the ability to employ alternative whisking strategies in different contexts, may constitute an important gain. In other words, vibrissal specialists may whisk for the same reason that humans carefully and repeatedly adjust the position of their fingertips whilst exploring objects with their hands, and adopt different exploratory strategies depending on the type of tactile judgement they are seeking to make (see Klatzky and Reed, 2009). In both vibrissal and fingertip touch, better quality and more appropriate tactile information may be obtained by exerting precise control over how the sensory apparatus interacts with the world.
Evidence to support this "active sensing" view (see Prescott, Diamond et al. 2011; Lepora, 2015) of whisking behavior comes from a number of sources. First, conditioning studies have shown that rats can be trained to vary some of the key parameters of whisking, such as amplitude and frequency in a stimulus-dependent manner (Bermejo et al., 1996; Gao et al., 2003). Second, recordings of whisker movements in freely-moving animals, show that this often diverges from the regular, bilaterally-symmetric and synchronous motor pattern that is usually observed in head-restrained animals. In the freely-moving animal, asymmetries (see, e.g. Figure 3), asynchronies, and changes in whisker protraction angle and timing have been noted some of which seem likely to boost the amount of useful sensory information obtained by the animal. For instance, it has been shown that head movements tend to direct the whiskers in the direction in which the animal is turning (Towal & Hartmann, 2006), suggesting increased exploration of space in the direction in which the animal is moving. Unilateral contact with a nearby surface tends to reduce whisker protraction on the side of the snout ipsilateral to that contact and increase protraction on the contralateral side (Mitchinson et al., 2007). Such a strategy would tend to increase the number of contacts made with a surface of interest whilst ensuring that the whiskers do not press hard against the contacted object. Other work has shown that the frequency of whisking, the starting position of the whiskers (the minimum protraction angle), or the angular spread between the whiskers, could each be controlled in a context- or behavior-specific way (Berg and Kleinfeld 2003; Carvell and Simons 1990; Sachdev et al. 2003; Sellien et al. 2005; Grant et al., 2009). Perhaps the strongest evidence that whisking control is purposive, in an active perception sense, comes from a study of sightless rats trained to run up a corridor for food (Arkley et al., 2014). Animals that were trained to expect obstacles at unpredictable locations in the corridor tended to run more slowly, and push their whiskers further forward, than animals trained to expect an empty corridor. This evidence suggests that the configuration of the whiskers is dependent on the animal's expectations about what it may encounter in the environment. The growing evidence for active control of whisker movement also implies that the vibrissal system can provide an accessible model for studying purposive behavior in mammals.
Experiments in adult rats and shrews involving whisker clipping, cauterization of the whisker follicle, section of the peripheral nerves, or lesion of critical structures in the vibrissal pathway, using appropriate controls for other sensory modalities such as vision and olfaction, have found significant deficits in exploration, thigmotaxis, locomotion, maintenance of equilibrium, maze learning, swimming, locating food pellets, and fighting (for reviews see Ahl, 1982; Gustafson & Felbain-Keramidas, 1977). Adults in which whiskers have been removed also show alterations of posture that have the effect of bringing the snout (and whisker stubs) into contact with surfaces of interest (Meyer & Meyer, 1992; Vincent, 1912), whilst animals in which whiskers are plucked shortly after birth and repeatedly thereafter exhibit pronounced behavioral compensations as adults (Gustafson & Felbain-Keramidas, 1977; Volgyi, Farkas, & Toldi, 1993), and show altered behavior even when whiskers are allowed to regrow (Carvell & Simons, 1996). Whisker clipping in infants disrupts early (post-natal day 3-5) nipple attachment and huddling (Sullivan et al., 2003). One of the most interesting and demanding uses of the vibrissal sense is in predation. For instance, pygmy shrews are known to prey on insects such as crickets that are themselves highly agile and almost as large as the shrew itself. Anjum et al. (2006) have demonstrated that tactile shape cues are both necessary and sufficient for evoking and controlling these attacks, whereas visual and olfactory cues are not needed (see Roth-Alpermann and Brecht, 2009). Rats are also efficient predators that can detect, track, and immobilise prey animals without vision (see, e.g. Gregoire & Smith, 1975).
Disruptions of predation, following whisker removal or damage, or of other behaviors such as maze learning, food finding, gap crossing, and fighting, are presumably the consequence of the loss of fundamental tactile sensory abilities that use vibrissal signals to (i) localise, orient to, and track objects/surfaces in space, and (ii) discriminate between objects/surfaces based on their tactile properties. These vibrissal perceptual functions are considered in more detail next.
Object localization and distance, orientation detection. Rats have been shown to use vibrissal information in the following tasks: gap measurement (Krupa, Matell, Brisben, Oliveira, & Nicolelis, 2001), gap jumping (Hutson & Masterton, 1986; Jenkinson & Glickstein, 2000; Richardson, 1909), measuring angular position along the sweep of the whisker (Knutsen, Pietr, & Ahissar, 2006; Mehta, Whitmer, Figueroa, Williams, & Kleinfeld, 2007), and distinguishing horizontal versus vertical orientation of bars (Polley, Rickert, & Frostig, 2005). Schiffman et al. (1970) demonstrated that rats confronted with a visual cliff do not show cliff-avoidance behavior unless their whiskers have been trimmed. This finding illustrates the primacy of vibrissal tactile sensing over vision in these animals with regard to depth perception. Ahissar & Knutsen (2008) have proposed that whisker identity, activation intensity, and timing of contact relative to the whisk cycle, together provide sufficient information to localise the position of a contact in 3-dimensional space (elevation, azimuth, and radial distance, respectively) (See Ahissar and Knutsen, 2011).
Orienting. Contact of the macrovibrissae with a surface often brings about orientation towards that surface which is then further explored using both the macro- and micro- vibrissae (Brecht, 1997; Hartmann, 2001). Grant et al. (2012b) showed that rats reliably orient to the nearest macrovibrissal contact on an unexpected object, progressively homing in on the nearest contact point on the object in each subsequent whisk. When exploring an object, the touching or brushing of the microvibrissae against the surface is largely synchronised with maximal macrovibrissal protraction and with a special pattern or rapid inhalation termed “sniffing” (Welker, 1964; Hartmann, 2001). Long-term bilateral removal of the whiskers has been shown to reduce orienting towards a tactile stimulus on the snout (Symons & Tees, 1990). When the various modulations of active whisking control are considered together it is possible to consider them as part of a general orienting system that both moves the tactile "fovea", at the tip of the snout, towards points of interest, and, at the same time modifies the positions of the whiskers in the wider array in a way that should increase the number of whisker contacts within an attentional zone (Mitchinson and Prescott, 2013; Mitchinson, 2015).
Detection of movement. The importance of vibrissae to prey-capture in shrews (Anjum et al., 2006), and in mouse-killing in rats (Gregoire & Smith, 1975) suggests that these animals may be able to estimate some of movement parameters of their targets using vibrissal information. In the case of the Etruscan shrew, effective use of the whiskers in prey-capture appears to be experience-dependent (Anjum and Brecht, 2012).
Texture discrimination. Initial studies of vibrissal texture discrimination (Carvell & Simons, 1990; Guic-robles et al., 1989; Guic-robles et al., 1992) showed that rats can be trained to reliably discriminate multiple textures using only their macrovibrissae, and can make fine distinctions with similar sensory thresholds to humans. Movement of an individual whisker across a texture has been shown to give rise to a characteristic whisker vibration or “kinetic signature” which is thought to form the basis on which sensory discriminations are made (Diamond et al., 2008; Hipp et al., 2006; Wolfe et al., 2008; Arabzadeh et al.,2009).
Air flow/water currents/vibrations. Air, or water, currents are effective stimuli for generating responses to vibrissal signals (e.g. in rats, Hutson & Masterton, 1986; in seals, Dehnhardt et al., 2001). Rat vibrissae should also be capable of discriminating vibrations of appropriate frequencies either applied directly to the vibrissal shaft or transmitted through the air as sound waves (Shatz & Christensen, 2008).
Shape discrimination. Compared to texture (microgeometry), shape and other macrogeometric properties are relatively poorly studied in rats. Brecht et al. (1997) have shown that rats can distinguish between triangular and square cookies (each 6mm per side) using their whiskers, although the evidence suggests that the microvibrissae are used in this task alongside the macrovibrissae. Experiments in shrews, using artificial prey replica, suggest that these animals may respond to some macrogeometric properties of prey animals (Anjum et al., 2006). Dehnhardt and co-workers have shown that seals can discriminate size and shape with their vibrissae (Dehnhardt & Ducker, 1996). Experiments with artificial vibrissal systems show that 3D shape of complex surfaces can, in principle, be recovered from vibrissal data (Kim & Moller, 2007; Solomon & Hartmann, 2006).
Numerical discrimination. Davis et al. (1989) have demonstrated that rats can be trained to discriminate between 2, 3, or 4 anterior-posterior strokes on their vibrissae, but not on their flanks. This result is indicative of vibrissal short-term memory and of some enumerative capacity in the vibrissal system.
The emergence of active whisker control alongside locomotion during development (Grant et al., 2012a) suggests a potentially close relationship between vibrissal sensing and the control of locomotion. Niederschuh et al. (2015) studied rhythmic whisker movement alongside locomotor gait in rats running on a treadmill with or without intact vibrissae. Whilst arguing against a close coupling between whisking and running, they suggest a likely role for the mystacial vibrissae in foot placement, in addition, they found that the carpal (wrist) vibrissae may assist the animal in monitoring its speed of movement. This hypothesis is supported by comparative evidence showing that, in eleven species of small mammals, forepaw placement reliably falls within the ground contact zone of the whisker field (Grant et al. 2018) and that the size of the vibrissal field is consistently wider than the forepaw. However, it appears that the vibrissal sense may be used in different modes in support of locomotion depending on speed of travel. Specifically, in rats, walking appears to be accompanied by broad whisker sweeps, whereas running is accompanied by the protraction of whiskers in front of the animal with much less forward-and-back oscillatory movement (Arkley et al., 2014). This pattern suggests a switch from using the whiskers to explore the floor surface, perhaps to find good locations for footfalls, to one where the whiskers are primarily used for obstacle detection and collision avoidance. In rats, a three-whisker array on the underside of the chin, called the "trident" whiskers, has been shown to drag along the ground during exploratory locomotion (Thé et al., 2013), suggesting that these whiskers could provide information to the animal about its velocity and heading direction. The value of whiskers in complex locomotor tasks, such as climbing, is indicated by evidence that small arboreal mammals, particularly nocturnal ones, have longer macrovibrissae than similar ground-dwelling species (Ahl, 1986; Sokolov and Kulikov, 1987). Experiments with movement on more complex surfaces, and in the absence of light, will be needed to better understand the importance of the vibrissae to locomotion control in terrestrial mammals.
In rodents, the vibrissae are also used during social behaviors. Typically, during encounters between resident and intruder rats, physical contact is often initiated by whisker-to-whisker contact, followed by potential aggressive behavior. Such nose-off's, along with other behaviors such as teeth-chattering and threat postures, may be used to convey social signals such as submissiveness or dominance. A study of whisker-to-whisker interaction in rats found that the social context of the interaction modulates whisker-related activity in primary somatosensory cortex (Bobrov et al., 2014).
Vibrissal system development has been studied most extensively in rats. Rat pups are born with an intact whisker field constituted of very fine, immobile vibrissae, that are invisible to the naked eye. The whiskers grow to their adult size in the first month of life, however, rats sustaining denervation of the whisker pad grow whiskers that are thinner and smaller than those of normal adults (Landers, Hairdarliu, & Ziegler, 2006). Whisking begins around postnatal day 11 to 13 (Welker, 1964; Landers & Ziegler, 2006; Grant et al., 2012a), prior to the opening of the eyes, and achieves adult amplitudes and characteristics by the end of the third post-natal week. The onset of active whisking control emerges a few days after the initial appearance of synchronised bilateral whisking (Grant et al., 2012a). Prior to the onset of whisking, neonatal rats show behavioral activation in response to whisker stimulation, and tactile learning in a classical conditioning avoidance paradigm (Landers & Sullivan, 1999; Sullivan et al., 2003). Micro-movements of the vibrissae in the first week of life have also been observed (Grant et al., 2012a). Rat pups are able to orient to contacts with nearby conspecifics before their eyes open implying an important role for the macrovibrissae in maintaining contact with conspecifics (Grant et al., 2012b). The emergence of vibrissal tactile sensing in rats may parallel the gradually increasing motor capacities of these animals that allow adult upright locomotion to occur around the same time as whisking onset. Tactile experience may be very different for neonatal rats than for adults both because of the small size and relative immobility of the whiskers, and because rat pups spend much of their time in "huddles" with littermates which is likely to produce stimulation of the whiskers from many different directions. Vibrissal behavior in rats, aged between 2 and 21 days old, is illustrated in this video.
Interest in the rodent vibrissal system has often stemmed from its accessibility as a model of mammalian sensory processing, rather than from the perspective of trying to understand its role in the life of the animal. For this reason, an accurate characterization of the contribution of vibrissal touch to rat or mouse behavior is some way off. We contend that such an understanding will be important for understanding the processing of vibrissal signals throughout the brain; after all to understand how a system works it should certainly help to know how it is used.
Active Touch Laboratory @ Sheffield (ATL@S). Link to the authors' laboratory.
The ATL@S YouTube Channel has videos illustrating vibrissal behavior in animals and biomimetic robots.
Whiskers! A feel for the dark. A short article on vibrissal function written for the magazine California Wild by Kathleen Wong.