sensing in a noisy world: lessons from auditory ... › content › jexbio › 220 › 24 ›...

REVIEW

Sensing in a noisy world lessons from auditory specialistsecholocating batsAaron J Corcoran1 and Cynthia F Moss2

ABSTRACTAll animals face the essential task of extracting biologicallymeaningful sensory information from the lsquonoisyrsquo backdrop of theirenvironments Here we examine mechanisms used by echolocatingbats to localize objects track small prey and communicate in complexand noisy acoustic environments Bats actively control and coordinateboth the emission and reception of sound stimuli through integratedsensory and motor mechanisms that have evolved together over tensof millions of years We discuss how bats behave in differentecological scenarios including detecting and discriminating targetechoes from background objects minimizing acoustic interferencefrom competing conspecifics and overcoming insect noise Batstackle these problems by deploying a remarkable array of auditorybehaviors sometimes in combination with the use of other sensesBehavioral strategies such as ceasing sonar call production andactive jamming of the signals of competitors provide further insightinto the capabilities and limitations of echolocation We relate thesefindings to the broader topic of how animals extract relevant sensoryinformation in noisy environments While bats have highly refinedabilities for operating under noisy conditions they face the samechallenges encountered by many other species We propose that thespecialized sensory mechanisms identified in bats are likely to occurin analogous systems across the animal kingdom

KEY WORDS Active sensing Acoustic interference Animalcommunication Jamming Noise

IntroductionTo find mates and food sources or evade predators animals must useone or more of their senses to detect localize and discriminatesignals from the noisy backdrop of their natural habitats (Stevens2013 Bradbury and Vehrencamp 2011) Sensory systems haveevolved to detect and discriminate stimuli that are important to theorganismrsquos survival and reproduction (Capranica and Moffat 1983Wehner 1987) However animals routinely encounter situationswhere there is an abundance of stimuli (abiotic or biotic) that havethe capacity to interfere with biologically relevant signals Thesestimuli are commonly referred to as noise (see Glossary) Naturalenvironments are composed of numerous competing stimuli whichhave the potential to serve as signal or noise depending on ananimalrsquos current behavioral goal (Fig 1)Animals have evolved a variety of mechanisms for sensing in

noisy environments Signalers are well known for making

behavioral adjustments to compensate for noise such as shiftingtheir calling frequencies to avoid spectral overlap with competingbackground signals (Shannon et al 2016) However receivers alsomake behavioral adjustments in noise for example moving to amore favorable location or increasing active scanning of theenvironment (Brumm and Slabbekoorn 2005) Animals also tendto rely more on multi-modal sensing (see Glossary) in noisyenvironments processing congruent stimuli acquired throughcomplementary senses andor placing greater weight onmodalities that are less subject to interference (Munoz andBlumstein 2012) For instance animals may rely on visual andacoustic stimuli from a common source (Taylor and Ryan 2013) ordepend more on hearing in darkness (Danilovich et al 2015)

Echolocation and electrolocation are considered lsquoactiversquo sensorymodalities because they operate through the animalrsquos productionof signals (sound and electricity respectively) into the environmentto generate sensory information which then guides behaviors(Nelson and MacIver 2006) Active sensing (see Glossary) alsorefers to an organismrsquos movements that influence sensory signalreception (Schroeder et al 2010) and the contribution ofmovement to perception is widespread throughout the animalkingdom Primates for example are well known for using rapideye movements or saccades to sequentially direct their foveae atobjects in a scene (Land 2006) Active gaze control (ie head oreye movements that control sampling of visual information) isimportant to animals as diverse as jumping spiders stalk-eyed fliesand zebra finches (Tarsitano and Andrew 1999 Ribak et al 2009Eckmeier et al 2008) Many animals actively control sniffing toimprove olfactory sampling (eg Catania 2006) explore objectsthrough touch by whisking (Ganguly and Kleinfeld 2004 Towaland Hartmann 2006) and localize sound sources by movingthe head and ears (pinnae) (Populin and Yin 1998) Activecontrol of sensing results in influxes of sensory informationwhich are used to inform further actions (Schroeder and Lakatos2009) This leads to actionndashperception feedback loops whereavailable sensory information drives the acquisition of futureinformation Under noisy conditions active sensing improvesstimulus detection and discrimination in support of diversesurvival tasks

In this review we consider echolocating bats as model organismsto gain a broader view of the mechanisms by which animals copewith noisy sensory environments To use echolocation or biologicalsonar an animal produces acoustic signals and then compares theirtemporal and spectral features with those of returning echoes(Griffin 1958 Surlykke et al 2014) This process allowsecholocating animals to reconstruct acoustic scenes Biologicalsonar can also be considered a form of lsquoauto-communicationrsquo ndash thatis a system in which signaler and receiver are the same individualMany of the factors that improve communication under conditionsof noise (Brumm and Slabbekoorn 2005) are well developed in batsfor auto-communication under low light levels

1Department of Biology Wake Forest University Box 7325 Reynolda StationWinston-Salem NC 27109 USA 2Department of Psychological and BrainSciences Johns Hopkins University 3400 N Charles Street Baltimore MD 21218USA

Author for correspondence (Aaronjcorcorangmailcom)

AJC 0000-0003-1457-3689

4554

copy 2017 Published by The Company of Biologists Ltd | Journal of Experimental Biology (2017) 220 4554-4566 doi101242jeb163063

Journal

ofEx

perim

entalB

iology

Echolocation itself is inherently subject to noise interferencetarget echoes are often several orders of magnitude weaker thanecholocation signals and sonar calls and echoes are separated bytens of milliseconds or less (Moss and Surlykke 2010) Targetechoes can easily be masked by the batrsquos own emission or bylsquoclutterrsquo echoes (see Glossary) returning from other objects in theenvironment (Fig 1) Bats also must contend with the rustling ofwind and noise from flowing water conspecific calls and chorusinginsects that produce sounds in the ultrasound range Despite thesechallenges bats perform natural echolocation behaviors withapparent ease For example bats exit cave roosts sometimesamongst thousands of echolocating conspecifics (Gillam et al2010) capture evasive insects in a fraction of a second (Kalko1995) and navigate in darkness through dense foliage while flyingat high speed (Kong et al 2016) Tight coordination and adaptive

control of both signal emission and echo reception are central to thesuccess of bats By studying these sensory specialists we hope togain broader and deeper insight into how animals in general copewith noisy sensory environments

We begin this review by briefly summarizing the mechanismsbats use to localize target echoes and separate them frombackground clutter echoes (see also Moss and Surlykke 2010)Next we consider how bats adapt to ecological scenarios involvingnoise including interactions with conspecifics and prey We alsohighlight the importance of multi-modal sensing to the success ofbats We finish by relating these findings to common sensoryproblems faced by a variety of animals

Fundamentals of echolocation common problems andsolutionsMore than 1100 bat species use echolocation for a wide variety oftasks and in diverse habitats (Denzinger and Schnitzler 2013Fenton and Simmons 2015) Bat auditory systems are composed ofthe same basic neural architecture and pathways as those of othermammals (Popper and Fay 1995) Like other animals bats comparethe amplitude and arrival time of sounds at their two ears todetermine the azimuth of sound sources (Wohlgemuth et al2016a) Bat ears typically feature a large tragus ndash a spear-likeprojection that modifies the spectral profile of echoes arriving withrespect to sound source elevation (Muumlller 2004 Wotton et al1995) Bats can use elevation-dependent spectral cues to determinethe vertical direction of a sound source (Wotton and Simmons2000) A distinguishing feature of spatial localization byecholocation is the ability to determine target distance (rangingsee Glossary) with high precision Bats accomplish this byestimating the time delay between outgoing pulses and returningechoes (Simmons 1973) Sound travels at approximately343 m sminus1 Big brown bats can discriminate echo arrival timedifferences of about 58 micros supporting range discriminationperformance of approximately 1 cm (summarized in Moss andSchnitzler 1995 Wohlgemuth et al 2016a) Some studies haveeven found evidence that bats can detect echo arrival timing changes( jitter) of less than 1 μs (Simmons 1979 Simmons et al 1990

GlossaryActive sensingUse of active processes to influence the influx of sensory informationAttentionSelectively processing stimuli that are relevant to the current behavioraltaskCall directionThe bat callrsquos aim as indicated by the axis of greatest acoustic energyCall directionalityA measure of the width of the sonar beamClutterEchoes returning from objects that are not the focus of an echolocatinganimalrsquos attentionFrequency modulatedA call that changes in frequency over timeMaskingA situation where one sensory stimulus influences the perception of asecond stimulusLow-pass filterChanging a sound so that lower frequencies are enhanced relative tohigher frequenciesMulti-modal sensingThe use of multiple sensory modalities for sensory perceptionNoiseEnvironmental stimuli that have the potential to interfere with thesensation of biologically meaningful stimuliNarrowbandA call whose acoustic energy is concentrated within a relatively shortrange of frequenciesRangingDetermining the distance to a target objectReceptive fieldThe specific range of stimuli that elicit a response from a sensory neuronsinFMSinusoidally frequency-modulated calls specialized signals used byMexican free-tailed bats to jam the echolocation of conspecificsSNRSignal to noise ratio a measure of the ratio between the strength of thesignal that is being attended to and the competing background noiseSpatial releaseThe process of using binaural hearing cues to reduce acoustic maskingof sounds that are coming from different directionsSPLSound pressure level a measure of the amplitude of an acoustic signalTymbal organA sound-producing structure found in some insects including tiger moths(Family Noctuidae subfamily Arctiinae)WaggleRapid head movements that change the orientation of the ears and areproposed to amplify auditory cues used for target localization

Fig 1 Cartoon illustration of a noisy acoustic environment of a batA focalbat (left) emits an echolocation signal (black) Echoes (gray) returnsimultaneously from a potential prey item and a tree obstacle A second insectemits an acoustic signal (red) and a conspecific echolocates nearby (green)Note that each of these acoustic stimuli could serve as signal or as noisedepending on the focal animalrsquos current behavioral goals

4555

REVIEW Journal of Experimental Biology (2017) 220 4554-4566 doi101242jeb163063

Journal

ofEx

perim

entalB

iology

Moss and Schnitzler 1989) corresponding to jitter along the rangeaxis of less than 1 mmMost bats produce short (lt20 ms) frequency-modulated (FM see

Glossary) calls that are followed by long silent intervals duringwhich time bats listen for echo returns from objects (Griffin 1958Schnitzler and Henson 1980 Moss and Schnitzler 1995) FM callshave a low duty cycle ie a low proportion of time filled withsound Approximately 200 bat species produce calls that start or endwith FM components but also include a constant frequency (CF)component (approximately 8ndash100 ms) Long CF calls (gt50 ms)have a high duty cycle ie a large percentage of the time is filledwith sound High duty cycle bats can hear echoes at the same timethat they are producing much more intense calls This is possiblebecause the relative motion of the flying bat with respect to its targetintroduces Doppler shifts in sonar returns which separates thesound frequencies of calls and echoes into different listeningchannels High duty cycle bats have highly specialized auditorysystems (Neuweiler et al 1980) that enable fine frequencydiscrimination (eg Long and Schnitzler 1975) Greaterhorseshoe bats species that use long CF signals for echolocationcan discriminate and recognize fluttering insect prey by listening tospectral and amplitude modulations in echoes (von der Emde andMenne 1989 von der Emde and Schnitzler 1990) This specializedform of echolocation also serves to reject noise outside a narrowfrequency band used for target echo detection and discriminationwhich for the greater horseshoe bat is about 83 kHz Unlessotherwise noted the discussion that follows applies to FM or lowduty cycle bats which rely heavily on the timing of calls and echoesto extract spatial information from the environment (Moss andSchnitzler 1995) For further discussion of bats that use high dutycycle calls we refer the reader to recent reviews (Fenton et al 2012Schnitzler and Denzinger 2011)

Assignment of calls and echoes through dynamic control of signalduration timing and frequencyTo accurately localize and sort object echoes in the environment abat must solve several problems target echo detection localizationrecognition and tracking in the midst of acoustic noise This isachieved in large part by the echolocating batrsquos dynamic andadaptive control of biosonar emissions bats modulate the durationintensity frequency directionality (see Glossary) direction of thesonar beam (see Glossary) and timing of emissions to optimize echoreception from targets while minimizing acoustic interference(Moss and Surlykke 2010) Bats actively adapt the features ofecholocation calls they use to localize and track insect prey (Fig 2)which can be categorized into three phases (Griffin et al 1960)During the lsquosearch phasersquo insectivorous FM bats produce sonarcalls at a rate of 3ndash20 Hz depending on the species (Holderied andvon Helversen 2003) After detecting a target echo the bat entersthe lsquoapproach phasersquo in which call rate steadily increases and callduration decreases The sweep of the FM sonar signals alsobecomes progressively steeper (Fig 2A) In the lsquoterminal buzzphasersquo the pulse rate approaches its maximum of 150ndash200 Hz andintensity decreases (Jakobsen et al 2013) Bats have specializedlsquosuperfastrsquo muscles in the larynx that allow such extraordinary callrates (Elemans et al 2011)By re-aligning the echolocation sequence so that the time from

the beginning of each echolocation call until the end of the next callis stacked in a sequence (to illustrate a lsquosonar streamrsquo Fig 1B)certain features of the echolocation sequence stand out Fig 2Ashows an acoustic recording of a Mexican free-tailed bat (Tadaridabrasiliensis) hunting prey under natural field conditions Call

emissions are shown in black Simulated target echoes have beenadded to the figure (in red) based on the physics of soundpropagation in air and measured 3D positions of the bat and preyAssumptions about echo arrival times have been verified in severalstudies where echoes were recorded from microphones placeddirectly behind the bat (eg Hiryu et al 2007) Fig 2B shows thatbats progressively shorten both pulse duration and pulse interval(time between successive pulses) such that the echo returns after theend of the outgoing pulse but before the beginning of the next pulseBy shortening pulse duration bats avoid temporal overlap betweenpulses and echoes that could lead to masking (see Glossary) orinterference of echo detection from the sonar call emission (Kalkoand Schnitzler 1993) Pulse structure also changes over thesequence The bat uses relatively long and narrowband calls (seeGlossary) in the search phase (pulse 4 in bottom row of Fig 2A)These calls concentrate acoustic energy within limited frequencybands which facilitates echo detection (Griffin et al 1960Surlykke and Moss 2000) Short broadband calls used late in theattack (eg pulse 28 in Fig 2A) allow the bat to integrate ranginginformation across many auditory neurons tuned to differentfrequencies (Simmons and Kick 1984) and therefore they arewell suited for distance measurement (Simmons and Stein 1980Surlykke 1992) This allows precise distance measurement for thefinal prey interception maneuver

The echo stream in Fig 2B also illustrates that despite thedramatic reduction in pulse intervals late in the sequence batstypically allow sufficient time for target echoes to return prior to thenext emission This allows bats to avoid ambiguity in assigningechoes to the correct pulse a requirement for accurate rangeestimation Another strategy for ensuring correct assignment ofpulses and echoes is to produce two or more calls in groups (ielsquosound groupsrsquo) flanked by sonar signals at longer intervals(Kothari et al 2014) which serves to link calls and echoes throughdistinct temporal patterning Bats may integrate echo informationwithin sound groups to increase sonar resolution (Moss et al 2006)Pulses within a group may also have distinct timendashfrequencyprofiles (Jung et al 2007 Ratcliffe et al 2011 Hiryu et al 2010)which could aid further in callndashecho assignment Fig 2D shows thebat Cormura brevirostris producing triplets of calls that increase insound frequency This may allow the bat to match calls and echoesnot only by their temporal patterning but also by their frequency Asimilar strategy has been reported in the big brown bat operating in ahighly cluttered environment (Hiryu et al 2010) This suggests thatbats can simultaneously store multiple timendashfrequency call profilesagainst which echo returns are compared

Echo feature recognitionEcholocating bats face the fundamental task of recognizing echoesfrom their own sonar emissions and distinguishing them from othersounds in the environment Psychophysical experiments have beenconducted that measure the echolocating batrsquos ranging performancein playback experiments that electronically delay the arrival ofsimulated sonar echoes These experiments support the idea thatbats compare the timendashfrequency structure of the outgoing call withthe echo return Bats that use FM signals suffer reduced rangingability when listening to echoes that are manipulated in timendashfrequency structure ie sweeping from low to high frequencies(Masters and Jacobs 1989) when natural FM signals are replacedby noise bursts (Surlykke 1992) have altered sweep shape (Mastersand Raver 2000) or have the timendashfrequency structure from anotherindividual bat (Masters and Raver 1996) Bats can also learn todifferentiate calls from different individuals based on subtle

4556


Journal

ofEx

perim

entalB

iology

differences in the distribution of energy across frequencies (Yovelet al 2009) If given sufficient training a bat can learn todiscriminate with high accuracy the delay between its own call andechoes having a different timendashfrequency structure (Masters andRaver 1996) In one study bat sonar ranging performance wasdisrupted by broadband insect clicks that arrived within a short timewindow of echoes (Miller 1991) Collectively these studies showthat bats utilize the distinct timendashfrequency structure of their ownechoes to detect and discriminate their signals from other noisesand that bats can learn to recognize novel call and echo patternswhen given sufficient time

Adaptive control of sonar beam aim directionality and intensityIn noisy environments echoes can return simultaneously frommany objects How do bats perceptually segregate object echoes incluttered environments One solution is to sequentially aim thesonar beam axis (ie to control call direction see Glossary) toinspect targets of interest (Fig 3A Surlykke et al 2009 Falk et al2011 Seibert et al 2013) Bat sonar emissions and hearing are both

directional and increasingly so at higher frequencies (Fig 2BAytekin et al 2004 Jakobsen et al 2013) Echoes returning fromobjects off-axis from the batrsquos beam aim are both weaker and low-pass filtered (see Glossary right inset in Fig 3C) The batrsquos auditorysystem can separate off-axis low-pass-filtered clutter echoes fromon-axis target echoes which prevents clutter echoes from maskingtarget echoes (Bates et al 2011) This observation is based on thefollowing Echoes are detected by populations of neurons thatrespond to different frequency components of the batrsquos FM sonarsignals (Simmons et al 1990) The latency of neural firing whichregisters echo arrival time depends on echo intensity with neuronsfiring at shorter latencies for higher amplitude echoes (Simmons andKick 1984 Simmons 1989) Because of the directionality of sonarsignal production and reception echoes returning from targets alongthe batrsquos midline are more intense than echoes returning from thebatrsquos periphery and this intensity difference is registered by thebatrsquos auditory system as differences in arrival time of echoes fromobjects along the midline and off to the batrsquos side Importantlydirectional differences in echo intensity are greater for high-

0

2

4

6

86420x (m)

y (m

)

BatMoth

Capture(t=0)

p1

p6

p11

p16

p36

Pulse

Echo

Time (s)

Time (s)

0ndash14

Freq

uenc

y (k

Hz)

0

100

p1p6 p11

Search Approach Terminalbuzz

p16p36

0

100

0 0 0

p4 p12 p28

13 10 6

A B

C D

Pulse Echo

Time after pulse (ms)

Pul

se n

o

11

6

11

16

21

26

31

360 120

Search

Approach

Terminalbuzz

Target distance (m)0

30 60 90

205 10 15

1 12

32

3

Time (ms) 5000

Freq

uenc

y (k

Hz)

20

60

Fig 2 Adaptive features of the bat echolocation attack sequence (A) Echolocation sequence of the Mexican free-tailed bat Tadarida brasiliensis attackinga moth in the field Shown are an oscillogram (top) and spectrogram (middle) of the entire sequence and (bottom) spectrograms of select calls Echoes(red) were added based on known target distances at the time of each pulse (B) The spectrogram from A is re-organized relative to the beginning of each pulsePulses are aligned on the y-axis The time period is shown from each pulse to the following pulse Note that the bat progressively shortens the pulseduration and pulse interval to ensure that each echo occurs between the end of the first pulse and the beginning of the following pulse (C) Overhead view ofT brasiliensis attacking a moth in the field Circles indicate echolocation pulses Numbers indicate selected pulses for reference across panels (D) The batCormura brevirostris is one of several species that produces search calls that alternate in frequency This bat produces calls in triplets (labeled 1ndash3) that increasein frequency from 26 to 33 kHz Note that calls from other bats are also present It has been hypothesized that frequency alternation aids in correct assignment ofcalls and echoes in cluttered environments D is reproduced with permission from Moss and Surlykke (2001)

4557


Journal

ofEx

perim

entalB

iology

frequency components of sonar signals than for low-frequencycomponents In other words at the batrsquos periphery high-frequencycomponents of echo returns are weaker and therefore registered atlonger neural response latencies than low frequencies and thiscreates a temporal misalignment of the low- and high-frequencycomponents of echo returns from objects off to the batrsquos side Thismisalignment has the effect of lsquodefocusingrsquo objects in the batrsquosperiphery (Bates et al 2011) Thus a combination of the physics ofsound transmission in the environment and the effect of soundintensity on neural response latency differentially affects sonarprocessing of low- and high-frequency target echoes arriving fromoff-axis objects Bates et al (2011) hypothesize that the sonardefocusing of off-axis clutter echoes prevents these signals frommasking target echoes in the batrsquos central lsquofield of viewrsquodetermined by its beam aim In this context it is noteworthy thatbats show spatial release (see Glossary) from masking at smallangular separations of target and clutter For example one studyreports that bats achieve complete spatial release from maskingwhen sound sources are separated by only 23 deg (Suumlmer et al2009) far better performance than is achieved by animals that do notecholocateBy adjusting call frequency or mouth aperture bats can

dynamically control the directionality of their sonar emissions(Jakobsen et al 2013) One recent study found that bats alter thesize of their mouth gape to adjust the width of their sonar beam asthey move through habitats that differ in spatial structure (Kounitskyet al 2015) This appears to be another strategy that allows bats toadaptively avoid acoustic interference from off-axis objects indifferent environmentsBats also alter their beam directionality during the last moments

of an attack on an insect Specifically late in attacks on prey batstypically decrease their calling frequency which broadens the sonarbeam (Jakobsen and Surlykke 2010) This may be an adaptiveresponse to ensure that the prey stays in the ensonified volumethrough to the end of the attack when prey might otherwisemaneuver outside the sonar beam (Corcoran and Conner 2016)

Active control of sound receptionComplementing active control of sonar emissions bats also controlthe shape separation and orientation of their pinnae Pinnamovements were first studied in high duty cycle CF bats (Griffinet al 1962) andmore recently in a low duty cycle FM bat Eptesicusfuscus (Wohlgemuth et al 2016b) Wohlgemuth et al (2016b)trained E fuscus to rest on a platform and track prey items that weremoved along different trajectories using a motorized pulley systemThis allowed the investigators to monitor sonar vocalizations andear movements with high precision as bats tracked moving preyEptesicus fuscus employ two types of pinna movement the firsttype is associated with rapid head rotations or lsquowagglesrsquo (seeGlossary) that alternate the vertical orientation of the two pinnaerelative to echo returns and the second which has been observed inboth E fuscus and high duty cycle bats involves changes in theerectness and separation between the pinnae

Regarding the first type bats produced waggles more often whentargets moved along complex trajectories Wohlgemuth et al(2016b) hypothesized that these ear movements amplify interaural-level cues and spectral cues in a manner that is analogous to visualmotion parallax where head movements are used to aid depthperception

For the second type of pinna movement erect pinnae focus theears towards echoes in front of the bat lateral ear deformationsincrease the distance between the tips of the pinnae and change theirshape which amplifies sounds coming from more-peripheralregions (Gao et al 2011) In a target-tracking study E fuscusincreased inter-pinna separation as targets approached it on aplatform broadening the batrsquos acoustic field of view when it facedthe challenge of intercepting a fast-moving target (Wohlgemuthet al 2016b) Bats also made rapid changes to inter-pinnaseparation as they tracked moving prey a behavior that mightenhance cues for sonar localization accuracy

These studies show that bats exhibit fine control over theiracoustic field of view which they change through head and earmovements under different contexts (such as distance to a target)

Beamaim

Net

Right edge

Left edge

Insect

A B

C

minus5 50

5

1025 kHz

x (m)

Atte

nuat

ion

(dB

)

y ( m

)

minus5 50

5

10

minus80minus70minus60minus50minus40minus30minus20minus10

050 kHz

x (m)

Focalobject Masker

Fig 3 Acoustic scanning behavior and spatial release from masking (A) Reconstruction of the sonar beam aim of a big brown bat as it flies through ahole in a net and then captures an insect (after Surlykke et al 2009) The bat sequentially fixates on the right and then left edges of the net opening beforedirecting its beam at the insect target (B) Directionality of big brown bat sonar at frequencies that correspond to the first (25 kHz) and second (50 kHz) harmonicsof its call respectively Attenuation is a result of the directionality of the sonar beam (after Hartley and Suthers 1989) bat hearing (after Aytekin et al 2004) andfrequency-specific attenuation of sound (Bazley 1976) (C) Spatial release from masking Echoes returning from objects near the center of the sonar beam (leftinset) return a full complement of frequencies whereas off-axis objects reflect weaker echoes that are low-pass filtered (right inset) The bat has neuralmechanisms that de-focus off-axis echoes preventing them from masking echoes from focal objects (Bates et al 2011)

4558


Journal

ofEx

perim

entalB

iology

and over millisecond time scales in coordination with sonarvocalizations These mechanisms should enhance the batrsquos abilityto extract acoustic information under noisy sensory conditions

Neural basis of echolocationTo understand mechanisms that allow bats to operate in acousticallynoisy and dynamic environments it is important to consider how thebatrsquos brain processes echoes and compares them with outgoingemissions Here we consider aspects of the batrsquos neural machinerythat are relevant to echo processing under dynamic and noisyconditions We also direct the reader to reviews of other aspects ofneural signal processing in the batrsquos sonar receiver (Suga 1990Simmons 2012 Wohlgemuth et al 2016a)The batrsquos brain is specialized for extracting sonar signal features

that are important for echolocation Specific neurons have beencharacterized that respond selectively to a restricted range of pulsendashecho delays (Suga and Orsquoneill 1979) signal durations (Cassedayet al 1994) frequency modulation rates (Razak and Fuzessery2008) and sound source directions (Valentine andMoss 1997) Thefeatures encoded by these neurons (ie their receptive fields seeGlossary) tend to cover those that the bat processes as it echolocatesin the natural environment For example individual delay-tunedneurons show strongest responses to delays from 1 to 36 ms (up toroughly 6 m of target distance) which corresponds to the batrsquosoperating range for small objects such as insect prey (Dear et al1993)Neurophysiological studies have revealed specializations in the

processing of biologically natural sound sequences in passivelylistening bats For instance research has shown that midbrainneurons are more selective to broadcasts of natural sonar emissionsthan simple computer-generated FM sweeps or noise (Wohlgemuthand Moss 2016) and are selective to the temporal dynamics ofsound stimulation (Sanderson and Simmons 2005) providingevidence that bat neural pathways are selective to acoustic featuresof their own calls More research is needed to determine the neuralbasis of this selectivity and how it changes over timeStudies have also begun to change our understanding of how the

batrsquos brain processes streams of echoes (Bartenstein et al 2014Beetz et al 2016) It is increasingly clear that bat neural pathwaysprocess not only individual pulsendashecho pairs but also streams ofpulses and echoes across a sequence For example the auditorycortex of many bat species shows topographic organization withsystematic shifts in echo delay tuning of neurons located along therostrocaudal axis (eg Suga 1990 Koumlssl et al 2014) It was longassumed that this map was static but a recent study demonstratedthat the map changes rapidly and dynamically when a sequence ofpulses and echoes is presented to a passively listening anesthetizedbat (Bartenstein et al 2014) When pulses and echoes werepresented at progressively shorter delays such as occurs whenapproaching a target (see Fig 2B) the map shifted towards a higherrepresentation of short delays The degree and direction of the shiftdepended on the sequence of pulses and echoes that were presentedThis and other recent neurophysiological (Beetz et al 2016) andbehavioral studies (Kugler et al 2016 Warnecke et al 2016)shows that bats are specialized for integrating the flow of echoes asthey return from multiple sonar pulsesMechanisms have been proposed to explain how the bat nervous

system might compute the spatial location of objects in an echoscene (Simmons 1973 2012 Simmons et al 1990 Valentine andMoss 1997) These discussions remain speculative because almostall neurophysiological studies of the bat auditory system have beenconducted with artificial sonar stimuli that simulate the batrsquos sonar

emissions and echo returns rather than echo returns from the batrsquosown sonar vocalizations Moreover studies of the bat nervoussystem have been largely conducted in passively listening and oftenanesthetized bats in the laboratory We are therefore left with thequestion of how neural responses to artificial stimuli in passivelylistening bats informs us of activity patterns that are evoked byechoes of the batrsquos sonar vocalizations No doubt the representationof noisy sonar scenes arises from the activity of populations ofneurons (Simmons 2012) Recent studies of the dynamics of echo-evoked activity in the bat sonar receiver of the free-flying activelyecholocating animal indeed demonstrate remapping and shifts in 3Dspatial tuning of midbrain auditory neurons with the batrsquos sonarinspection of objects (Kothari et al 2016) These findings can serveto motivate a broad and intense investigation of neural activitypatterns in animals that freely explore noisy sensory environments

Acoustically noisy ecological scenariosHere we examine in detail three ecological scenarios where bats arefaced with noisy environmental conditions These scenarioshighlight the flexibility that is afforded to bats by using multiplemechanisms for overcoming challenging sensory conditions

Scenario 1 echolocating conspecificsBat echolocation calls are among the most intense acoustic signalsin nature sometimes exceeding 140 dB sound pressure level SPL(see Glossary) at 01 m (Holderied et al 2003 Surlykke and Kalko2008) Bats routinely encounter conspecifics when departing from ashared roost commuting or foraging A potential challenge ariseswhen a bat must filter high-intensity conspecific calls to detect anddiscriminate echo streams that are at a much lower sound level Thisproblem has received considerable attention in the literature over thepast 15 years (eg Ulanovsky et al 2004 Gillam et al 2007Cvikel et al 2015a) Much of the discussion in the literature hasfocused on the hypothesis that like electric fish (Heiligenberg1991) bats alter the frequency of their emissions to avoid spectraloverlap with conspecific calls a behavior known as the jammingavoidance response (JAR)

Early evidence for JAR in bats came from studies of bats callingalone or in pairs in the wild (Habersetzer 1981 Ulanovsky et al2004 Ratcliffe et al 2004) Pairs of bats flying together frequentlyadjusted their peak calling frequency to maintain a 3ndash4 kHzseparation Field (Gillam et al 2007) and laboratory (Bates et al2008 Takahashi et al 2014) playback experiments later confirmedthis finding bats rapidly (in one study lt200 ms) adjust their callingfrequency to avoid spectral overlap between playbacks and the mostshallowly FM components of their calls Another study examinedthe call structure of bats flying alone or in pairs in a laboratory (Chiuet al 2009) Bats adjusted their call structure when flying nearconspecifics to a degree that was dependent on the baselinesimilarity between the two batsrsquo calls when flying alone That ispairs of bats that had similar calls when flying alone made largerchanges to their calls when flying together These studiesconclusively demonstrate that at least some bats use the JAR toavoid acoustic interference from conspecifics

Recent studies have led to an alternative hypothesis for observedfrequency changes in groups of echolocating bats (Cvikel et al2015a Goumltze et al 2016) Namely the authors hypothesize andhave found strong evidence that some bats alter call frequency as areaction to the physical presence of other bats not their acousticpresence These studies show that not all bats use JAR and thatfrequency shifts alone are not sufficient for demonstrating JAR inbats This alternative hypothesis does not explain the data from

4559


Journal

ofEx

perim

entalB

iology

some previous studies that controlled for the physical presence ofbats either by using playback experiments (Gillam et al 2007Takahashi et al 2014) or by carefully measuring the positions andorientations of the bats that were present (Chiu et al 2010) Thus itappears that some but not all bats use JARRegardless of whether they employ JAR bats are likely to use

multiple mechanisms to correctly sort conspecific calls from theirown echoes (see discussions in Ulanovsky and Moss 2008 Bateset al 2008) A batrsquos own echoes are likely to form predictablestreams (Fig 2B) and have a timendashfrequency structure and directionalcues that will differ from calls of conspecifics (Yovel et al 2009)One recent study found that Pipistrellus kuhlii solved the problem ofextreme acoustic interference from conspecifics not by adjusting callfrequency but by increasing call duration intensity and pulse rate(Amichai et al 2015) These adjustments all improve the signal-to-noise ratio (SNR see Glossary) of calls over background noise afinding that indicates the problem posed by conspecific calls (at leastwhen numerous conspecifics are present) is acoustic masking notdifferentiating onersquos own calls from those of conspecificsThere are conflicting data on how bats adjust their calling rate in

response to conspecifics Some studies indicate that bats decreasetheir calling rate when calls of one conspecific are present (Jarviset al 2010 2013 Adams et al 2017) but others have found thatbats increase their calling rate particularly when faced with calls ofnumerous bats (Amichai et al 2015 Lin et al 2016) Suppressedcalling rates have been interpreted as evidence for groupcooperation (Adams et al 2017) but alternatively this couldindicate that bats are devoting more of their attention (see Glossary)to passively listening to conspecific calls (Barber et al 2003)Collectively these studies demonstrate that bats use numerousmechanisms for separating signals and noise and their reliance onthese mechanisms can shift depending on the prevailing conditions

Scenario 2 competing with conspecifics for foodGroup foraging involves a fundamental tradeoff bats can improvesearcher efficiency by eavesdropping on the feeding calls ofconspecifics (Gillam et al 2007 Dechmann et al 2009) but thiscan increase competition for food A high density of foraging batsalso increases the complexity of the acoustic and physicalenvironment taking the batrsquos attention away from foraging(Cvikel et al 2015b) Bats may be under selective pressure tofend off competitors even though they themselves benefit fromeavesdropping on others Recent research has revealed multipleacoustic strategies that bats use during competition for foodOne such strategy is the use of food-claiming calls A recent

laboratory study showed that big brown bats make specificcommunication calls called FM bouts (FMBs) when competingwith other bats for a prey item (Wright et al 2014) FMBcalls containindividual-specific signatures and when produced they caused anincrease in the spatial separation between the bats Bats that producedmore FMBs were more likely to capture food items (Fig 4A) Fieldstudies have shown that pipistrelle bats (Pipistrellus spp) producesocial calls that might have a similar function (Barlow and Jones1997) Pipistrelles produce these calls more often when food densityis low and playbacks of the social calls had a deterrent effect onconspecifics Bats at foraging sites are frequently observed chasingconspecifics while emitting social calls (eg Miller and Degn 1981)Dominant bats could be aggressively chasing away competitors andadvertising their presence with specialized individual-specific callsThis would not only reduce competition for food but also simplify theacoustic and physical environment so that the bat can focus attentionon finding prey (Cvikel et al 2015b)

Another strategy observed in pairs of big brown bats competingfor food is lsquosilent behaviorrsquo (Chiu et al 2008) Specifically whenflying within 1 m of conspecifics paired bats routinely(approximately 40 of the time) ceased echolocating for periodsof 02ndash255 s (Fig 4B) These behaviors were almost neverobserved in bats flying alone Silence was more common whenpairs of bats had echolocation calls with similar design This couldbe interpreted in one of two ways (1) bats could use silence as amechanism for avoiding jamming from conspecifics that producesimilar calls to their own or (2) the similarity in call design betweenthe two bats could make it easier for the bat engaging in silentbehavior to use the conspecificrsquos calls and echoes for its own sonarsystem This could in turn enable a batrsquos stealth attack on the preyitem At present these hypotheses remain untested

Finally Mexican free-tailed bats use sinusoidally frequency-modulated (sinFM see Glossary) calls to jam the echolocation ofcompeting bats attempting to capture prey (Corcoran and Conner2014) Bats produce sinFM calls only when a competing bat is in theapproach and terminal buzz phase of prey capture (Fig 4C Fig 5B)When conspecifics produced sinFM calls that overlapped theirfeeding buzz bats captured prey during only 6 of attackscompared with 35 when no sinFM calls were present Playbackexperiments showed that the timing and timendashfrequency structure ofsinFM calls are important for interfering with the competitorrsquosattack 3D reconstructions of bat flight trajectories showed batsengaged in extended bouts of food competition where they tookturns jamming one another while the other bat attempted to captureprey (Fig 4C)

Studies of food competition strategies give insight into how batscope with acoustic interference First these data provide furtherevidence that bats are a potential source of acoustic interferenceeither because of the calls that theymake or because of their physicalpresence as a sound-reflecting object Second silent behaviorindicates that bats are capable of orienting by eavesdropping on thecalls (and perhaps echoes) of conspecifics Third specialized sonar-jamming calls demonstrate that despite the extraordinaryadaptations observed in echolocating bats they are notimpervious to acoustic interference particularly when trying tocapture prey Jamming signals provide insight into fundamentalconstraints on echolocation a topic we discuss further below

Scenario 3 insect noiseAside from bats chorusing insects such as katydids are one of themost common sources of ultrasound in the environment (Robinsonand Hall 2002) Playback experiments provide evidence that insectnoise is a potential source of acoustic interference for batecholocation Gillam and McCracken (2007) recorded Tbrasiliensis echolocation calls in the field in the presence of silenceor playbacks of insect noise that varied in peak frequency from165 to29 kHz Bats shifted their calling frequency upward depending on thefrequency of the playback always maintaining a 2ndash4 kHz separationbetween their calling frequency and that of the insect noise Thisfinding indicates that bats exhibit a JAR not only in response toconspecifics but also to a variety of interfering signals

Several insects including several families of moths (Blest et al1963 Barber and Kawahara 2013 Corcoran and Hristov 2014) andtiger beetles (Yager and Spangler 1997) produce bursts ofultrasonic clicks in response to the attack cries of bats Clicksproduced at relatively low rates have the primary function ofwarning bats that the insect is toxic (Hristov and Conner 2005Ratcliffe and Fullard 2005) some palatable moths also mimic thesesounds to deceive bats (Barber and Conner 2007)

4560


Journal

ofEx

perim

entalB

iology

Of particular interest here are some species of tiger moths andhawkmoths that produce clicks at high rates to jam bat echolocation(Corcoran et al 2009 Kawahara and Barber 2015) Like thejamming sinFM calls of bats these clicks are produced during thebatrsquos approach and buzz phases of echolocation Psychophysical(Miller 1991) and neurophysiological (Tougaard et al 1998)experiments show that clicks disrupt the target ranging ability ofbats by multiple orders of magnitude but to do so clicks mustoccur within 1ndash2 ms of echo returns Moths cannot anticipate whenthis windowwill occur so their solution is to click at extremely highrates (as high as 4000 clicks sminus1) that ensure some clicks will co-occur with each set of echo returns Experiments pitting bats againstjamming moths found that bats often continued prey pursuit throughthe barrage of noise but missed the prey by a distance similar to theerrors observed in psychophysical and neurophysiologicalexperiments (Corcoran et al 2011)How do the jamming signals described above interfere with bat

echolocation The specialized jamming signals of bats and mothsmight provide insight into how bats process and segregate echoes

from noise Because these signals appear to have evolvedspecifically to jam bat sonar they might contain elements thateither infiltrate or disrupt the batrsquos neural pathways Currently thisdiscussion is speculative because no studies have examined how thestructure of jamming signals affects their disruptive capacity

Moth clicks and bat sinFM calls have dramatically differentacoustic structures but they also have some common features(Fig 5) Both signals occupy a high proportion of time during thebatrsquos terminal buzz overlap spectrally with the batrsquos calls and havefrequency components that change rapidly over time Tiger mothsproduce bursts of 20ndash30 clicks at a time through the sequentialbuckling and elastic recoil of their tymbal organ (see Glossary)(Blest et al 1963) Clicks are very short (024 ms) and broadbandThe peak frequency of clicks in a series decreases and then increaseswith the sequential buckling and elastic recoil of striations on thesurface of the tymbal In comparison sinFM calls consist of one tofive relatively long (mean 65 ms) syllables that are produced as longas a competing bat continues its buzz These calls oscillate up anddown over the frequency band of conspecific buzz calls (Fig 5B)

x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)

Fig 4 Acoustic competition strategies in bats Three distinct food competition strategies have been discovered in bats (A) food claiming calls (frequency-modulated bouts FMBs) (B) silent behavior and (C) jamming calls (sinusiodally frequencymodulated sinFM) Food claiming and silence have been documentedin the big brown bat Eptesicus fuscus (Chiu et al 2008 Wright et al 2014) while jamming calls have been documented in the Mexican free-tailed bat Tadaridabrasiliensis (Corcoran andConner 2014) For each strategy plots of the echolocation and socialjamming calls of each bat (left) and an overhead view of bat flighttrajectories (right) are shown FMBs and sinFM calls are highlighted in green Blue and red linesdots indicate echolocation calls Feeding buzzes are labeled aseither lsquocapturersquo or lsquomissrsquo In A the two bats follow one another closely while echolocating and producing FMBs Bat 1 produces more FMBs and captures theinsect In B bat 2 exhibits silent behavior while following bat 1 then makes a feeding buzz to capture the insect In C the two bats alternate in producing feedingbuzzes while the other bat makes sinFM calls that jam the sonar emissions of the buzzing bat Bat 2 eventually captures the insect after bat 1 has left the areaVideo animations of each sequence are available as supplemental videos in the original publications Adapted figures are reprinted with permission

4561


Journal

ofEx

perim

entalB

iology

SinFM calls oscillate at a rate of 166 Hz which is similar to the batrsquoscalling rate of 154 Hz during the feeding buzz (Corcoran andConner 2014) This suggests that the rhythmic sinFM oscillationsmight have evolved specifically to elicit responses from neurons thatfire in response to feeding buzz callsThe acoustic structure of bat and moth jamming signals hints at

the possibility that they have specific features that infiltrate the batsonar receiver It is unlikely that bats perceive jamming signals asactual echoes because bats have highly refined echo discriminationabilities (Masters and Raver 1996 Corcoran et al 2010) A morelikely possibility is that the acoustic structure of jamming signalsactively disrupts echo processing in the batrsquos neural pathwaysFurther behavioral and neurophysiological experiments are requiredto test these hypotheses

Multi-modal sensing as a mechanism for coping with noiseA common solution to sensing in noisy environments is to usemultiple sensory modalities (Munoz and Blumstein 2012) Batsprovide numerous examples of this phenomenon both as short-termbehavioral responses and as evolutionary adaptations to specificforaging niches (Schnitzler and Kalko 1998) Echolocation is poorlysuited for detecting objects resting on vegetation or the groundbecause target and background echoes return nearly simultaneouslyBats that acquire stationary food items from surfaces (includinginsects fruit and nectar) show increased reliance on passive listening(reviewed by Jones et al 2016) olfaction (Korine and Kalko 2005)and vision (Bell 1985 Ekloumlf and Jones 2003) Bats that forage closeto vegetation tend to have larger eyes and better visual acuity than batsthat forage in open spaces (table 2 in Ekloumlf 2003) These examplesshow an increased reliance on multi-modal sensing for bats thatforage in cluttered habitatsThere is increasing evidence that bats routinely integrate echondash

acoustic and visual information to perceive their surroundings

(Horowitz et al 2004 Orbach and Fenton 2010 Boonman et al2013) A recent study showed that Egyptian fruit bats (Rousettusaegyptiacus) alter their echolocation signaling rate depending onlight levels (Danilovich et al 2015) Despite having excellentvision these bats never ceased echolocating entirely This could bebecause echolocation and vision provide complementary sensoryinformation Echolocation allows detection of small targets underlow light levels and provides better ranging ability whereas visionis effective over longer distances and provides better spatialresolution along the dimensions of azimuth and elevation(Boonman et al 2013) We propose that multimodal sensing maybe widespread in naturally behaving animals and is not only ameans for coping with uncertainty in preferred sensory modalities(Munoz and Blumstein 2012)

An open question is to what extent bats rely on vision for obstacledetection and avoidance If a bat is subject to severe acousticinterference such as when flying amongst hundreds of callingconspecifics could it utilize vision to avoid flying into vegetation orother bats (Kong et al 2016) Some studies have modified eitherlight levels (Horowitz et al 2004) or the visual conspicuousness ofobstacles (Orbach and Fenton 2010) to show that bats can usevision for obstacle avoidance However further experiments areneeded that independently control for both the visual and echo-acoustic cues of obstacles

DiscussionBats exhibit numerous adaptations to successfully operate in noisysensory environments Central to the batrsquos success is the ability todynamically coordinate signal emission and reception over fine timescales (Moss and Surlykke 2010Wohlgemuth et al 2016b) Theseadjustments optimize information acquisition and minimize theeffects of interference arising from background objects such asthe signals produced by conspecifics and insects The bat sonar

Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00

Fig 5 Sonar jamming signals of moths and bats (A) Clicks produced by the tiger moth Bertholdia trigona to jam the sonar of the big brown bat Eptesicusfuscus (Corcoran et al 2009) (B) Intraspecific sonar jamming signals (sinFM) of the Mexican free-tailed bat Tadarida brasiliensis (Corcoran and Conner 2014)Oscillograms and spectrograms are shown of the jamming signals alone (top) and spectrograms are shown of jamming signals made during a bat attacksequence (bottom) Note the distinctive timendashfrequency structures of the jamming signals and that they are both produced to overlap in time and frequency withthe attacking batrsquos feeding buzz

4562


Journal

ofEx

perim

entalB

iology

system can be considered a highly refined form of animalcommunication where the signaler and receiver are one andoperate through shared neural processes that have evolved over tensof millions of years Here we relate studies of bat echolocation innoisy environments to sensory challenges encountered by a widerange of animals

Dynamic representations of echo scenesThe batrsquos auditory system is specialized to process features of sonarpulses and echoes The neural basis of acoustic imaging by sonar isstill an area of active investigation but both behavioral (Chiu et al2009 Yovel et al 2009) and neurophysiological studies(Wohlgemuth and Moss 2016 Kothari et al 2016) indicate thatbat auditory systems have evolved to detect and discriminatefeatures of their own calls from other sounds An exciting recentdiscovery is that the receptive fields of bat auditory neurons changerapidly in ways that appear to facilitate the transformation of echostreams into perceptual representations of auditory objects(Bartenstein et al 2014 Beetz et al 2016) It has also beenreported that 3D spatial response profiles of midbrain neuronsremap to represent shorter distances with higher resolution whenfreely echolocating big brown bats adjust their echolocationbehavior to inspect sonar objects (Kothari et al 2016) Thesefindings illustrated in Fig 6 indicate that the batrsquos auditory receiver

changes dynamically on a very rapid time scale What remains to beinvestigated are the ways in which acoustic clutter or noisecontribute to dynamic neural representations We hypothesize thatneurons tracking targets in the presence of acoustic clutter sharpentheir response areas and this can be tested through systematicempirical studies

Dynamic sensory processing is important to the lives of many ifnot all animals For example in the presence of masking noisebirds and other animals adjust the frequency of their courtshipsignals to improve the SNR (Shannon et al 2016) It has beenproposed that a tradeoff exists between optimizing signaltransmission and saliency of the signal to the receiver (Patricelliand Blickley 2006) A bird that shifts its calling frequency in noisecould improve the SNR at the receiver but the female receiver mightbe less responsive to this altered signal It therefore benefitsreceivers to have flexible feature detection and recognition systemsespecially under noisy conditions Future research on sensoryrepresentation in dynamic environments may reveal the extent towhich animals other than bats encode dynamic natural stimuli

Signal interferenceA downside of selective feature recognition may be that it putsanimals at increased risk to specific types of interference which canbe exploited by other animals This appears to occur in the jammingsignals of bats and moths (Fig 5) Active sensory interference alsoappears to occur in other communication systems For examplemale oyster toadfish (Opsanus tau) produce precisely timed lsquogruntsrsquothat interfere with communication between competing males andfemales (Mensinger 2014) These grunts might reduce theperceived frequency of advertisement calls made by competitorsand thereby reduce their attractiveness to females Thus interferencesignals provide distinct opportunities for probing the inner workingsof animal communication receivers

Coordination between sender and receiverSensing requires animals to first detect and discriminate signalsfrom noise and then extract meaningful information from thosesignals Animals must have in place mechanisms for achieving eachof these sensory tasks Bats have solved this problem elegantlyagain because they actively control signal emission and receptionwith respect to behavioral state and informational need Asdiscussed above bats shift rapidly from producing signals that areoptimized for detection to signals that are optimized for localizationand feature extraction This is possible because bat echolocationoperates through an actionndashperception loop to adjust signalparameters dynamically with informational needs Because senderand receiver are the same individual in bat echolocation systemsthere is rapid and tight coordination between call production andecho processing It follows that the level of coordination betweensender and receiver in other animal communication systems shouldimpact both the timing and reliability of signal transmission andreception This proposal can be tested directly through comparativeanalyses of communication behaviors throughout the animalkingdom

Comparative studies of active sensing in noisy environmentsWhile bats and other echolocating animals actively control the timingand features of biosonar signals used to probe the environment activesensing operates in species throughout the animal kingdom(Schroeder et al 2010) Active sensing refers to the movementsanimals make to modify sensory input which in turn guides futurebehaviors Eye movements for example allow an animal to scan the

Relative neural response

Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)

TimeSearch Approach Buzz

Fig 6 Cartoon representation of dynamic echo delay response profilesof three idealized neurons shown separately in red blue and green in thebat auditory system Along the lower x-axis are spectrograms of echolocationcalls produced by an FM bat through the search approach and capture phasesof insect pursuit Solid horizontal lines below calls at each insect pursuit phaserepresent signal duration and dotted lines represent the interval betweensuccessive calls Note that call duration and interval decrease progressivelyfrom search to approach to capture phases The y-axis shows relative echodelays (target distances) over which the neurons respond The upper x-axisplots the relative response of the neurons to echo delays at each of these insectcapture phases Neurons 1 2 and 3 respond to echoes at the search andapproach phases of insect pursuit but at different echo delays neuron 1responds to the longest echo delays neuron 2 to intermediate echo delaysand neuron 3 to short echo delays At the capture phase only neuron 3responds to a subset of echoes from the calls produced at a high repetition rate(short intervals) Note that neurons 1 and 2 show shifts in responses to shorterecho delay as the bat adapts its echolocation behavior and approaches theprey At the end of the approach phase the echo delay response areas of thethree neurons are close to overlapping All three neurons show a sharpening ofecho delay tuningwith increasing call repetition rate This cartoon is based on asynthesis of data reported in Suga and OrsquoNeill (1979) Sullivan (1982) Wonget al (1992) Bartenstein et al (2014) Beetz et al (2016) Kothari et al (2016)

4563


Journal

ofEx

perim

entalB

iology

environment and represent objects across a broad panorama Thevisual stimuli acquired through eye movements are also used toinform decisions for subsequent behaviors (Land 2006) Similarlyhead and ear movements introduce changes in acoustic signalsreceived at the two ears to enhance cues for auditory localization andinfluence perception of an auditory scene (Populin and Yin 1998Wohlgemuth et al 2016b) Along related lines sniffing andwhisking serve to modulate sensory signals that can be used tobuild up information over time (Ganguly and Kleinfeld 2004Catania 2006 Towal and Hartmann 2006) We propose thatquantitative analyses of the echolocating batrsquos adaptive behaviors innoisy environments will provide the motivation for new lines ofinvestigation on active sensing in a wide range of species across theanimal kingdom Ultimately such comparative studies of activesensing will serve to differentiate between species-specificspecializations and general solutions animals employ to performnatural behavioral tasks in noisy sensory environments

AcknowledgementsWe thank William Conner and two reviewers for critical feedback on earlier drafts ofthis manuscript

Competing interestsThe authors declare no competing or financial interests

FundingThe following grants supported research conducted by the authors and thepreparation of this article Human Frontiers Science Program (RGP0040) Office ofNaval Research (N00014-12-1-0339) Air Force Office of Scientific Research(FA9550-14-1-0398) National Science Foundation Collaborative Research inComputational Neuroscience (IOS1460149) and National Science Foundation (IOS1257248)

ReferencesAdams A M Davis K and Smotherman M (2017) Suppression of emissionrates improves sonar performance by flying bats Sci Rep 7 41641

Amichai E Blumrosen G and Yovel Y (2015) Calling louder and longer howbats use biosonar under severe acoustic interference from other batsProc R Soc B 282 20152064

Aytekin M Grassi E Sahota M and Moss C F (2004) The bat head-relatedtransfer function reveals binaural cues for sound localization in azimuth andelevation J Acoust Soc Am 116 3594

Barber J R and Conner W E (2007) Acoustic mimicry in a predator preyinteraction Proc Natl Acad Sci USA 104 9331-9334

Barber J R andKawahara A Y (2013) Hawkmoths produce anti-bat ultrasoundBiol Lett 9 20130161

Barber J R Razak K A and Fuzessery Z M (2003) Can two streams ofauditory information be processed simultaneously Evidence from the gleaningbat Antrozous pallidus J Comp Physiol A 189 843-855

Barlow K E and Jones G (1997) Function of pipistrelle social calls field dataand a playback experiment Anim Behav 53 991-999

Bartenstein S K Gerstenberg N Vanderelst D Peremans H and FirzlaffU (2014) Echo-acoustic flow dynamically modifies the cortical map of targetrange in bats Nat Commun 5 4668

Bates M E Stamper S A and Simmons J A (2008) Jamming avoidanceresponse of big brown bats in target detection J Exp Biol 211 106-113

Bates M E Simmons J A and Zorikov T V (2011) Bats use echo harmonicstructure to distinguish their targets from background clutter Science 333627-630

Bazley E N (1976) Sound absorption in air at frequencies up to 100 kHz NPLAcoustics Report Ac 74 pp 1-43 Teddington UK National Physics Laboratory

Beetz M J Hechavarrıa J C and Kossl M (2016) Temporal tuning in the batauditory cortex is sharper when studied with natural echolocation sequences SciRep 6 29102

Bell G P (1985) The sensory basis of prey location by the California leaf-nosed batMacrotus californicus (Chiroptera Phyllostomidae) Behav Ecol Sociobiol 16343-347

Blest A D Collett T S and Pye J D (1963) The generation of ultrasonic signalsby a new world arctiid moth Proc R Soc B 158 196-207

Boonman A Bar-On Y Cvikel N and Yovel Y (2013) Itrsquos not black or white-onthe range of vision and echolocation in echolocating bats Front Physiol 4 248

Bradbury J W and Vehrencamp S L (2011) Principles of AnimalCommunication 2nd edn Sunderland MA Sinauer Associates Inc

Brumm H and Slabbekoorn H (2005) Acoustic communication in noise AdvStudy Behav 35 151-209

Capranica R R and Moffat J M (1983) Neurobehavioral correlates of soundcommunication in anurans In Advances in Vertebrate Neuroethology (ed J-EEwert R R Capranica and D J Ingle) pp 701-730 Boston MA Springer US

Casseday J H Ehrlich D and Covey E (1994) Neural tuning for soundduration role of inhibitory mechanisms in the inferior colliculus Science 264847-850

Catania K C (2006) Olfaction underwater ldquosniffingrdquo by semi-aquatic mammalsNature 444 1024-1025

Chiu C Xian W and Moss C F (2008) Flying in silence Echolocating batscease vocalizing to avoid sonar jamming Proc Natl Acad Sci USA 10513116-13121

Chiu C Xian W and Moss C F (2009) Adaptive echolocation behavior in batsfor the analysis of auditory scenes J Exp Biol 212 1392-1404

Corcoran A J and Conner W E (2014) Bats jamming bats food competitionthrough sonar interference Science 346 745-747

Corcoran A J and Conner W E (2016) How moths escape bats predictingoutcomes of predator-prey interactions J Exp Biol 219 2704-2715

Corcoran A J and Hristov N I (2014) Convergent evolution of anti-bat soundsJ Comp Physiol A 200 811-821

Corcoran A J Barber J R and Conner W E (2009) Tiger moth jams batsonar Science 325 325-327

Corcoran A J Conner W E and Barber J R (2010) Anti-bat tiger mothsounds Form and function Curr Zool 56 358-369

Corcoran A J Barber J R Hristov N I and Conner W E (2011) How dotiger moths jam bat sonar J Exp Biol 214 2416-2425

Cvikel N Levin E Hurme E Borissov I Boonman A Amichai E andYovel Y (2015a) On-board recordings reveal no jamming avoidance in wild batsProc R Soc B 282 20142274

Cvikel N Egert Berg K Levin E Hurme E Borissov I Boonman AAmichai E and Yovel Y (2015b) Bats aggregate to improve prey search butmight be impaired when their density becomes too high Curr Biol 25 206-211

Danilovich S Krishnan A Lee W-J Borrisov I Eitan O Kosa G MossC F and Yovel Y (2015) Bats regulate biosonar based on the availability ofvisual information Curr Biol 25 R1124-R1125

Dear S P Simmons J A and Fritz J (1993) A possible neuronal basis forrepresentation of acoustic scenes in auditory cortex of the big brown bat Nature364 620-623

Dechmann D K N Heucke S L Giuggioli L Safi K Voigt C C andWikelski M (2009) Experimental evidence for group hunting via eavesdroppingin echolocating bats Proc R Soc B 276 2721-2728

Denzinger A and Schnitzler H-U (2013) Bat guilds a concept to classify thehighly diverse foraging and echolocation behaviors of microchiropteran batsFront Physiol 4 164

Eckmeier D Geurten B R H Kress D Mertes M Kern R Egelhaaf M andBischof H-J (2008) Gaze strategy in the free flying zebra finch (Taeniopygiaguttata) PLoS ONE 3 e3956

Eklof J (2003) Vision in Echolocating Bats PhD thesis Goteborg UniversityEklof J and Jones G (2003) Use of vision in prey detection by brown long-eared

bats Plecotus auritus Anim Behav 66 949-953Elemans C P H Mead A F Jakobsen L and Ratcliffe J M (2011) Superfast

muscles set maximum call rate in echolocating bats Science 333 1885-1888Falk B Williams T Aytekin M and Moss C F (2011) Adaptive behavior for

texture discrimination by the free-flying big brown bat Eptesicus fuscus J CompPhysiol A 197 491-503

Fenton M B and Simmons N B (2015) Bats AWorld of Science and MysteryChicago IL University of Chicago Press

Fenton M B Faure P A and Ratcliffe J M (2012) Evolution of high duty cycleecholocation in bats J Exp Biol 215 2935-2944

Ganguly K and Kleinfeld D (2004) Goal-directed whisking increases phase-locking between vibrissa movement and electrical activity in primary sensorycortex in rat Proc Natl Acad Sci USA 101 12348-12353

Gao L Balakrishnan S He W Yan Z and Muller R (2011) Ear deformationsgive bats a physical mechanism for fast adaptation of ultrasonic beampatternsPhys Rev Lett 107 1-4

Gillam E H and McCracken G F (2007) Variability in the echolocation ofTadarida brasiliensis effects of geography and local acoustic environment AnimBehav 74 277-286

Gillam E H Ulanovsky N and McCracken G F (2007) Rapid jammingavoidance in biosonar Proc R Soc B 274 651-660

Gillam E H Hristov N I Kunz T H andMcCracken G F (2010) Echolocationbehavior of Brazilian free-tailed bats during dense emergence flights J Mammal91 967-975

Gotze S Koblitz J C Denzinger A and Schnitzler H-U (2016) No evidencefor spectral jamming avoidance in echolocation behavior of foraging pipistrellebats Sci Rep 6 30978

Griffin D R (1958) Listening in the Dark The Acoustic Orientation of Bats andMen Mineola NY Dover Publications Inc

4564


Journal

ofEx

perim

entalB

iology

Griffin D R Webster F A and Michael C R (1960) The echolocation of flyinginsects by bats Anim Behav 8 141-154

Griffin D R Dunning D C Cahlander D A and Webster F A (1962)Correlated orientation sounds and ear movements of horseshoe batsNature 1961185-1186

Habersetzer J (1981) Adaptive echolocation sounds in the bat Rhinopomahardwickei J Comp Physiol A 144 559-566

Hartley D J and Suthers R A (1989) The sound emission pattern of theecholocating bat Eptesicus fuscus J Acoust Soc Amer 85 1348-1351

Hartmann M J Johnson N J Towal R B and Assad C (2003) Mechanicalcharacteristics of rat vibrissae resonant frequencies and damping in isolatedwhiskers and in the awake behaving animal J Neurosci 23 6510-6519

Heiligenberg W (1991) Neural Nets in Electric Fish Cambridge MA MIT PressHiryu S Hagino T Riquimaroux H and Watanabe Y (2007) Echo-intensitycompensation in echolocating bats (Pipistrellus abramus) during flight measuredby a telemetry microphone J Acoust Soc Am 121 1749-1757

Hiryu S Bates M E Simmons J A and Riquimaroux H (2010) FMecholocating bats shift frequencies to avoid broadcast-echo ambiguity in clutterProc Natl Acad Sci USA 107 7048-7053

Holderied M W and von Helversen O (2003) Echolocation range and wingbeatperiod match in aerial-hawking bats Proc Biol Sci 270 2293-2299

Horowitz S S Cheney C A and Simmons J A (2004) Interaction ofvestibular echolocation and visual modalities guiding flight by the big brown batEptesicus fuscus J Vestib Res 14 17-32

Hristov N I and Conner W E (2005) Sound strategy acoustic aposematism inthe batndashtiger moth arms race Naturwissenschaften 92 164-169

Jakobsen L and Surlykke A (2010) Vespertilionid bats control the width of theirbiosonar sound beam dynamically during prey pursuit Proc Natl Acad Sci USA107 13930-13935

Jakobsen L Brinkloslashv S and Surlykke A (2013) Intensity and directionality ofbat echolocation signals Front Physiol 4 89

Jarvis J Bohn K M Tressler J and Smotherman M (2010) Amechanism forantiphonal echolocation by free-tailed bats Anim Behav 79 787-796

Jarvis J JacksonW and SmothermanM (2013) Groups of bats improve sonarefficiency through mutual suppression of pulse emissions Front Physiol 4 140

Jones P L Page R A and Ratcliffe J M (2016) To scream or to listen Preydetection and discrimination in animal-eating bats In Bat Bioacoustics (ed M BFenton A D Grinnell A N Popper and R R Fay) pp 93-116 New York NYSpringer-Verlag

Jung K Kalko E K V and von Helversen O (2007) Echolocation calls inCentral American emballonurid bats signal design and call frequency alternationJ Zool 272 125-137

Kalko E K V (1995) Insect pursuit prey capture and echolocation in pipestirellebats (Microchiroptera) Anim Behav 50 861-880

Kalko E K V and Schnitzler H-U (1993) Plasticity in echolocation signals ofEuropean pipistrelle bats in search flight implications for habitat use and preydetection Behav Ecol Sociobiol 33 415-428

Kawahara A Y and Barber J R (2015) Tempo and mode of antibat ultrasoundproduction and sonar jamming in the diverse hawkmoth radiation Proc NatlAcad Sci USA 1126407-6412

Kong Z Fuller N Wang S Ozcimder K Gillam E Theriault D Betke MandBaillieul J (2016) Perceptual modalities guiding bat flight in a native habitatSci Rep 6 27252

Korine C and Kalko E K V (2005) Fruit detection and discrimination by smallfruit-eating bats (Phyllostomidae) Echolocation call design and olfaction BehavEcol Sociobiol 59 12-23

Kossl M Hechavarria J C Voss C Macias S Mora E C and Vater M(2014) Neural maps for target range in the auditory cortex of echolocating batsCurr Opin Neurobiol 24 68-75

Kothari N B Wohlgemuth M J Hulgard K Surlykke A and Moss C F(2014) Timing matters sonar call groups facilitate target localization in batsFront Physiol 5 168

Kothari N B Wohlgemuth M J andMoss C F (2016) Midbrain neurons of thefree-flying echolocating bat represent three-dimensional space J Acoust SocAmer 140 2973

Kounitsky P Rydell J Amichai E Boonman A Eitan O Weiss A J andYovel Y (2015) Bats adjust their mouth gape to zoom their biosonar field of viewProc Natl Acad Sci USA 112 6724-6729

Kugler K Greiter W Luksch H Firzlaff U and Wiegrebe L (2016) Echo-acoustic flow affects flight in bats J Exp Biol 219 1793-1797

Land M F (2006) Eye movements and the control of actions in everyday life ProgRet Eye Res 25 296-324

Lin Y Abaid N and Muller R (2016) Bats adjust their pulse emission rates withswarm size in the field J Acoust Soc Am 140 4318-4325

Long G R and Schnitzler H-U (1975) Behavioural audiograms from the batRhinolophus ferrumequinum J Comp Physiol 100 211-219

Masters W M and Jacobs S C (1989) Target detection and range resolution bythe big brown bat (Eptesicus fuscus) using normal and time-reversed modelechoes J Comp Physiol A 166 65-73

Masters W M and Raver K A S (1996) The degradation of distancediscrimination in big brown bats (Eptesicus fuscus) caused by differentinterference signals J Comp Physiol A 179 703-713

Masters W M and Raver K A S (2000) Range discrimination by big brown bats(Eptesicus fuscus) using altered model echoes implications for signalprocessing J Acoust Soc Am 107 625-637

Mensinger A F (2014) Disruptive communication stealth signaling in thetoadfish J Exp Biol 217 344-350

Miller L A (1991) Arctiid moth clicks can degrade the accuracy of range differencediscrimination in echolocating big brown bats Eptesicus fuscus J Comp PhysiolA 168 571-579

Miller L A and Degn H J (1981) The acoustic behavior of four species ofvespertilionid bats studied in the field J Comp Physiol A 142 67-74

Moss C F andSchnitzler H-U (1989) Accuracy of target ranging in echolocatingbats acoustic information processing J Comp Physiol A 165 383-393

Moss C F and Schnitzler H-U (1995) Behavioral studies of auditory informationprocessing In Hearing by Bats (ed A N Popper and R R Fay) pp 87-145New York NY Springer

Moss C F and Surlykke A (2001) Auditory scene analysis by echolocation inbats J Acoust Soc Am 110 2207-2226

Moss C F and Surlykke A (2010) Probing the natural scene by echolocation inbats Front Behav Neurosci 4 1-16

Moss C F Bohn K Gilkenson H and Surlykke A (2006) Active listening forspatial orientation in a complex auditory scene PLoS Biol 4 615-626

Muller R (2004) A numerical study of the role of the tragus in the big brown bat JAcoust Soc Amer 116 3701-3712

Munoz N E and Blumstein D T (2012) Multisensory perception in uncertainenvironments Behav Ecol 23 457-462

Nelson M E and MacIver M A (2006) Sensory acquisition in active sensingsystems J Comp Physiol A 192 573-586

Neuweiler G Bruns V and Schuller G (1980) Ears adapted for the detection ofmotion or how echolocating bats have exploited the capacities of the mammalianauditory system J Acoust Soc Am 68 741-753

Orbach D N and Fenton B (2010) Vision impairs the abilities of bats to avoidcolliding with stationary obstacles PLoS ONE 5 e13912

Patricelli G L and Blickley J L (2006) Avian communication in urban noisecauses and consequences of vocal adjustment Auk 123 639-649

Popper A N and Fay R R (1995) Hearing by Bats New York NY SpringerNew York

Populin L C and Yin T C (1998) Pinna movements of the cat during soundlocalization J Neurosci 18 4233-4243

Ratcliffe J M and Fullard J H (2005) The adaptive function of tiger moth clicksagainst echolocating bats an experimental and synthetic approach J Exp Biol208 4689-4698

Ratcliffe J M Hofstede H M Avila-flores R FentonM B McCracken G FBiscardi S Blasko J Gillam E Orprecio J and Spanjer G (2004)Conspecifics influence call design in the Brazilian free-tailed bat Tadaridabrasiliensis Can J Zool 82 966-971

Ratcliffe J M Jakobsen L Kalko E K V and Surlykke A (2011) Frequencyalternation and an offbeat rhythm indicate foraging behavior in the echolocatingbat Saccopteryx bilineata J Comp Physiol A 197 413-423

Razak K A and Fuzessery Z M (2008) Facilitatory mechanisms underlyingselectivity for the direction and rate of frequencymodulated sweeps in the auditorycortex J Neurosci 28 9806-9816

Ribak G Egge A R and Swallow J G (2009) Saccadic head rotations duringwalking in the stalk-eyed fly (Cyrtodiopsis dalmanni) Proc R Soc B 2761643-1649

Robinson D J and Hall M J (2002) Sound signaling in Orthoptera Adv InsectPhys 29 151-278

Sanderson M I and Simmons J A (2005) Target representation of naturalisticecholocation sequences in single unit responses from the inferior colliculus of bigbrown bats J Acoust Soc Am 118 3352-3361

Schnitzler H-U and Denzinger A (2011) Auditory fovea and Doppler shiftcompensation Adaptations for flutter detection in echolocating bats using CF-FMsignals J Comp Physiol A 197 541-559

Schnitzler H-U and Henson O W Jr (1980) Performance of airborne animalsonar systems I Microchiroptera InAnimal Sonar Systems (ed R-G Busnel andJ F Fish) pp 109-181 New York Plenum

Schnitzler H-U andKalko E K V (1998) Howecholocating bats search and findfood In Bat Biology and Conservation (ed T H Kunz and P A Racey) pp183-196 Washington DC Smithsonian Institution Press

Schroeder C E and Lakatos P (2009) Low-frequency neuronal oscillations asinstruments of sensory selection Trends Neurosci 32 9-18

Schroeder C E Wilson D A Radman T Scharfman H and Lakatos P(2010) Dynamics of active sensing and perceptual selection Curr OpinNeurobiol 20 172-176

Seibert A-M Koblitz J C Denzinger A and Schnitzler H-U (2013)Scanning behavior in echolocating common Pipistrelle bats (Pipistrelluspipistrellus) PLoS ONE 8 e60752

4565


Journal

ofEx

perim

entalB

iology

Shannon G McKenna M F Angeloni L M Crooks K R Fristrup K MBrown E Warner K A Nelson M D White C Briggs J et al (2016) Asynthesis of two decades of research documenting the effects of noise on wildlifeBiol Rev 91 982-1005

Simmons J A (1973) The resolution of target range by echolocating batsJ Acoust Soc Am 54 157

Simmons J A (1979) Perception of echo phase information in bat sonar Science204 1336ndash1338

Simmons J A (1989) A view of the world through the batrsquos ear The formation ofacoustic images in echolocation Cognition 33 155ndash199

Simmons J A (2012) Bats use a neuronally implemented computational acousticmodel to form sonar images Curr Opin Neurobiol 22 311-319

Simmons J A and Kick S A (1984) Physiological mechanisms for spatialfiltering and image enhancement in the sonar of bats Annu Rev Physiol 46599-614

Simmons J A and Stein R A (1980) Acoustic imaging in bat sonarecholocation signals and the evolution of echolocation J Comp Physiol A135 61-84

Simmons J A Moss C F and Ferragamo M (1990) Convergence of temporaland spectral information into acoustic images of complex sonar targets perceivedby the echolocating bat Eptesicus fuscus J Comp Physiol A 166 449-470

Stevens M (2013) Sensory Ecology Behaviour and Evolution Oxford UKOxford University Press

Suga N (1990) Cortical computational maps for auditory imaging Neural Netw 33-21

Suga N and OrsquoNeill W E (1979) Neural axis representing target range in theauditory cortex of the mustache bat Science 206 351-353

Sullivan W E (1982) Neural representation of target distance in auditory cortex ofthe echolocating bat Myotis lucifugus J Neurophysiol 48 1011-1032

Sumer S Denzinger A and Schnitzler H-U (2009) Spatial unmasking in theecholocating Big Brown BatEptesicus fuscus J Comp Physiol A 195 463-472

Surlykke A (1992) Target ranging and the role of time-frequency structure ofsynthetic echoes in big brown bats Eptesicus fuscus J Comp Physiol A 17083-92

Surlykke A and Kalko E K V (2008) Echolocating bats cry out loud to detecttheir prey PLoS ONE 3 e2036

Surlykke A and Moss C F (2000) Echolocation behavior of big brown batsEptesicus fuscus in the field and the laboratory J Acoust Soc Am 1082419-2429

Surlykke A Ghose K and Moss C F (2009) Acoustic scanning of naturalscenes by echolocation in the big brown bat Eptesicus fuscus J Exp Biol 2121011-1020

Surlykke A Nachtigall P E Fay R R and Popper A N (2014) BiosonarNew York NY Springer

Takahashi E Hyomoto K Riquimaroux H Watanabe Y Ohta T and HiryuS (2014) Adaptive changes in echolocation sounds by Pipistrellus abramus inresponse to artificial jamming sounds J Exp Biol 217 2885-2891

Tarsitano M S and Andrew R (1999) Scanning and route selection in thejumping spider Portia labiata Anim Behav 58 255-265

Taylor R C and Ryan M J (2013) Interactions of multisensory componentsperceptually rescue tungara frog mating signals Science 341 273-274

Tougaard J Casseday J H and Covey E (1998) Arctiid moths and batecholocation broad-band clicks interfere with neural responses to auditory stimuliin the nuclei of the lateral lemniscus of the big brown bat J Comp Physiol A 182203-215

Towal R B and Hartmann M J (2006) Right-left asymmetries in the whiskingbehavior of rats anticipate head movements J Neurosci 26 8838ndash8846

Ulanovsky N and Moss C F (2008) What the batrsquos voice tells the batrsquos brainProc Natl Acad Sci USA 105 8491-8498

Ulanovsky N Fenton M B Tsoar A and Korine C (2004) Dynamics ofjamming avoidance in echolocating bats Proc R Soc B 271 1467-1475

Valentine D E and Moss C F (1997) Spatially selective auditory responses inthe superior colliculus of the echolocating bat J Neurosci 17 1720ndash1733

Von der Emde G and Menne D (1989) Discrimination of insect wingbeat-frequencies by the bat Rhinolophus ferrumequinum J Comp Physiol A 164663-671

Von der Emde G and Schnitzler H-U (1990) Classification of insects byecholocating greater horseshoe bats J Comp Physiol A 167 423-430

Warnecke M Lee W-J Krishnan A and Moss C F (2016) Dynamic echoinformation guides flight in the big brown bat Front Behav Neurosci 10 81

Wehner R (1987) ldquoMatched filtersrdquo-neural models of the external world J CompPhysiol A 161 511-531

Wohlgemuth M J and Moss C F (2016) Midbrain auditory selectivity to naturalsounds Proc Natl Acad Sci USA 113 2508-2513

Wohlgemuth M J Luo J and Moss C F (2016a) Three-dimensional auditorylocalization in the echolocating bat Curr Opin Neurobiol 41 78-86

Wohlgemuth M J Kothari N B and Moss C F (2016b) Action enhancesacoustic cues for 3-D target localization by echolocating bats PLoS Biol 14e1002544

Wong D Maekawa M and Tanaka H (1992) The effect of pulse repetition rateon the delay sensitivity of neurons in the auditory cortex of the FM bat Myotislucifugus J Comp Physiol A 170 393-402

Wotton J M and Simmons J A (2000) Spectral cues and perception of thevertical position of targets by the big brown bat Eptesicus fuscus J Acoust SocAm 107 1034-1041

Wotton J M Haresign T and Simmons J A (1995) Spatially dependentacoustic cues generated by the external ear of the big brown bat Eptesicusfuscus J Acoust Soc Am 98 1423-1445

Wright G S Chiu C Xian W Wilkinson G S and Moss C F (2014) Socialcalls predict foraging success in big brown bats Curr Biol 24 885-889

Yager D D and Spangler H G (1997) Behavioral response to ultrasound by thetiger beetle Cicindela marutha dow combines aerodynamic changes and soundproduction J Exp Biol 200 649-659

Yovel Y Melcon M L Franz M O Denzinger A and Schnitzler H-U (2009)The voice of bats how greater mouse-eared bats recognize individuals based ontheir echolocation calls PLoS Comput Biol 5 e1000400

4566


Journal

ofEx

perim

entalB

iology

Echolocation itself is inherently subject to noise interferencetarget echoes are often several orders of magnitude weaker thanecholocation signals and sonar calls and echoes are separated bytens of milliseconds or less (Moss and Surlykke 2010) Targetechoes can easily be masked by the batrsquos own emission or bylsquoclutterrsquo echoes (see Glossary) returning from other objects in theenvironment (Fig 1) Bats also must contend with the rustling ofwind and noise from flowing water conspecific calls and chorusinginsects that produce sounds in the ultrasound range Despite thesechallenges bats perform natural echolocation behaviors withapparent ease For example bats exit cave roosts sometimesamongst thousands of echolocating conspecifics (Gillam et al2010) capture evasive insects in a fraction of a second (Kalko1995) and navigate in darkness through dense foliage while flyingat high speed (Kong et al 2016) Tight coordination and adaptive

control of both signal emission and echo reception are central to thesuccess of bats By studying these sensory specialists we hope togain broader and deeper insight into how animals in general copewith noisy sensory environments

We begin this review by briefly summarizing the mechanismsbats use to localize target echoes and separate them frombackground clutter echoes (see also Moss and Surlykke 2010)Next we consider how bats adapt to ecological scenarios involvingnoise including interactions with conspecifics and prey We alsohighlight the importance of multi-modal sensing to the success ofbats We finish by relating these findings to common sensoryproblems faced by a variety of animals

Fundamentals of echolocation common problems andsolutionsMore than 1100 bat species use echolocation for a wide variety oftasks and in diverse habitats (Denzinger and Schnitzler 2013Fenton and Simmons 2015) Bat auditory systems are composed ofthe same basic neural architecture and pathways as those of othermammals (Popper and Fay 1995) Like other animals bats comparethe amplitude and arrival time of sounds at their two ears todetermine the azimuth of sound sources (Wohlgemuth et al2016a) Bat ears typically feature a large tragus ndash a spear-likeprojection that modifies the spectral profile of echoes arriving withrespect to sound source elevation (Muumlller 2004 Wotton et al1995) Bats can use elevation-dependent spectral cues to determinethe vertical direction of a sound source (Wotton and Simmons2000) A distinguishing feature of spatial localization byecholocation is the ability to determine target distance (rangingsee Glossary) with high precision Bats accomplish this byestimating the time delay between outgoing pulses and returningechoes (Simmons 1973) Sound travels at approximately343 m sminus1 Big brown bats can discriminate echo arrival timedifferences of about 58 micros supporting range discriminationperformance of approximately 1 cm (summarized in Moss andSchnitzler 1995 Wohlgemuth et al 2016a) Some studies haveeven found evidence that bats can detect echo arrival timing changes( jitter) of less than 1 μs (Simmons 1979 Simmons et al 1990

GlossaryActive sensingUse of active processes to influence the influx of sensory informationAttentionSelectively processing stimuli that are relevant to the current behavioraltaskCall directionThe bat callrsquos aim as indicated by the axis of greatest acoustic energyCall directionalityA measure of the width of the sonar beamClutterEchoes returning from objects that are not the focus of an echolocatinganimalrsquos attentionFrequency modulatedA call that changes in frequency over timeMaskingA situation where one sensory stimulus influences the perception of asecond stimulusLow-pass filterChanging a sound so that lower frequencies are enhanced relative tohigher frequenciesMulti-modal sensingThe use of multiple sensory modalities for sensory perceptionNoiseEnvironmental stimuli that have the potential to interfere with thesensation of biologically meaningful stimuliNarrowbandA call whose acoustic energy is concentrated within a relatively shortrange of frequenciesRangingDetermining the distance to a target objectReceptive fieldThe specific range of stimuli that elicit a response from a sensory neuronsinFMSinusoidally frequency-modulated calls specialized signals used byMexican free-tailed bats to jam the echolocation of conspecificsSNRSignal to noise ratio a measure of the ratio between the strength of thesignal that is being attended to and the competing background noiseSpatial releaseThe process of using binaural hearing cues to reduce acoustic maskingof sounds that are coming from different directionsSPLSound pressure level a measure of the amplitude of an acoustic signalTymbal organA sound-producing structure found in some insects including tiger moths(Family Noctuidae subfamily Arctiinae)WaggleRapid head movements that change the orientation of the ears and areproposed to amplify auditory cues used for target localization

Fig 1 Cartoon illustration of a noisy acoustic environment of a batA focalbat (left) emits an echolocation signal (black) Echoes (gray) returnsimultaneously from a potential prey item and a tree obstacle A second insectemits an acoustic signal (red) and a conspecific echolocates nearby (green)Note that each of these acoustic stimuli could serve as signal or as noisedepending on the focal animalrsquos current behavioral goals

4555


Journal

ofEx

perim

entalB

iology








4556


Journal

ofEx

perim

entalB

iology




0

2

4

6

86420x (m)

y (m

)

BatMoth

Capture(t=0)

p1

p6

p11

p16

p36

Pulse

Echo

Time (s)

Time (s)

0ndash14

Freq

uenc

y (k

Hz)

0

100

p1p6 p11


p16p36

0

100

0 0 0

p4 p12 p28

13 10 6

A B

C D

Pulse Echo


Pul

se n

o

11

6

11

16

21

26

31

360 120

Search

Approach

Terminalbuzz


30 60 90

205 10 15

1 12

32

3

Time (ms) 5000

Freq

uenc

y (k

Hz)

20

60


4557


Journal

ofEx

perim

entalB

iology








Beamaim

Net

Right edge

Left edge

Insect

A B

C

minus5 50

5

1025 kHz

x (m)

Atte

nuat

ion

(dB

)

y ( m

)

minus5 50

5

10


050 kHz

x (m)

Focalobject Masker


4558


Journal

ofEx

perim

entalB

iology












4559


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology








4556


Journal

ofEx

perim

entalB

iology




0

2

4

6

86420x (m)

y (m

)

BatMoth

Capture(t=0)

p1

p6

p11

p16

p36

Pulse

Echo

Time (s)

Time (s)

0ndash14

Freq

uenc

y (k

Hz)

0

100

p1p6 p11


p16p36

0

100

0 0 0

p4 p12 p28

13 10 6

A B

C D

Pulse Echo


Pul

se n

o

11

6

11

16

21

26

31

360 120

Search

Approach

Terminalbuzz


30 60 90

205 10 15

1 12

32

3

Time (ms) 5000

Freq

uenc

y (k

Hz)

20

60


4557


Journal

ofEx

perim

entalB

iology








Beamaim

Net

Right edge

Left edge

Insect

A B

C

minus5 50

5

1025 kHz

x (m)

Atte

nuat

ion

(dB

)

y ( m

)

minus5 50

5

10


050 kHz

x (m)

Focalobject Masker


4558


Journal

ofEx

perim

entalB

iology












4559


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology




0

2

4

6

86420x (m)

y (m

)

BatMoth

Capture(t=0)

p1

p6

p11

p16

p36

Pulse

Echo

Time (s)

Time (s)

0ndash14

Freq

uenc

y (k

Hz)

0

100

p1p6 p11


p16p36

0

100

0 0 0

p4 p12 p28

13 10 6

A B

C D

Pulse Echo


Pul

se n

o

11

6

11

16

21

26

31

360 120

Search

Approach

Terminalbuzz


30 60 90

205 10 15

1 12

32

3

Time (ms) 5000

Freq

uenc

y (k

Hz)

20

60


4557


Journal

ofEx

perim

entalB

iology








Beamaim

Net

Right edge

Left edge

Insect

A B

C

minus5 50

5

1025 kHz

x (m)

Atte

nuat

ion

(dB

)

y ( m

)

minus5 50

5

10


050 kHz

x (m)

Focalobject Masker


4558


Journal

ofEx

perim

entalB

iology












4559


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology








Beamaim

Net

Right edge

Left edge

Insect

A B

C

minus5 50

5

1025 kHz

x (m)

Atte

nuat

ion

(dB

)

y ( m

)

minus5 50

5

10


050 kHz

x (m)

Focalobject Masker


4558


Journal

ofEx

perim

entalB

iology












4559


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology












4559


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology











4560


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology





x (m)

y (m

)y

(m)

0 350

Insect

3

Bat 1Bat 2FMBFMB

A

B

C

CaptureBat 1

Bat 2

Bat 1

Bat 2

x (m)0 350

Insect

3Bat 1

Bat 2

SilenceCapture

0200

15

sinFM

x (m)

Miss Capture

Bat 1

Bat 2

sinFM

1 s

MissMiss

MissBat 1

Bat 2

y (m

)


4561


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology








Time (ms)

Freq

uenc

y (k

Hz)

100 200 300 400 500

20

40

60

sinFM

25 50 750

20406080

100

A B

25 50 75 1000

20406080

100

Moth clicks

20

40

60

80

Buzz Buzz

0100 200 300 4000

00


4562


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology









Rel

ativ

e ec

ho d

elay

Neuron 1

Neuron 2

Neuron 3

Freq

uenc

y (k

Hz)



4563


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology





















































4564


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology






























































4565


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology








































4566


Journal

ofEx

perim

entalB

iology

sensing in a noisy world: lessons from auditory ... › content › jexbio › 220 › 24 ›...

Documents