Access to this full-text is provided by Wiley.
Content available from The Anatomical Record
This content is subject to copyright. Terms and conditions apply.
SPECIAL ISSUE ARTICLE
Behavioral and anatomical evidence for electroreception in
the bottlenose dolphin (Tursiops truncatus)
Tim Hüttner
1,2
| Lorenzo von Fersen
2
| Lars Miersch
1
| Nicole U. Czech
1
|
Guido Dehnhardt
1
1
Sensory & Cognitive Ecology, University
of Rostock, Institute for Biosciences,
Rostock, Germany
2
Nuremberg Zoo, Nuremberg, Germany
Correspondence
Tim Hüttner and Guido Dehnhardt,
Institute for Biosciences, University of
Rostock, Albert-Einstein-Strasse 3, 18059
Rostock, Germany.
Email: huettner.tim@gmail.com (T. H.) and
guido.dehnhardt@uni-rostock.de (G. D.)
Funding information
This work was supported by a grant of the
VolkswagenStiftung to Guido Dehnhardt;
Nuremberg Zoo Germany, and Verein der
Tiergartenfreunde Nürnberg e.V.
Abstract
In the order of cetacean, the ability to detect bioelectric fields has, up to now,
only been demonstrated in the Guiana dolphin (Sotalia guianensis) and is
suggested to facilitate benthic feeding. As this foraging strategy has also
been reported for bottlenose dolphins (Tursiops truncatus), we studied
electroreception in this species by combining an anatomical analysis of
“vibrissal crypts”as potential electroreceptors from neonate and adult animals
with a behavioral experiment. In the latter, four bottlenose dolphins were
trained on a go/no-go paradigm with acoustic stimuli and afterward tested for
stimulus generalization within and across modalities using acoustic, optical,
mechanical, and electric stimuli. While neonates still possess almost complete
vibrissal follicles including a hair shaft, hair papilla, and cavernous sinus, adult
bottlenose dolphins lack these features. Thus, their “vibrissal crypts”show a
similar postnatal morphological transformation from a mechanoreceptor to an
electroreceptor as in Sotalia. However, innervation density was high and
almost equal in both, neonate as well as adult animals. In the stimulus gener-
alization tests the dolphins transferred the go/no-go response within and
across modalities. Although all dolphins responded spontaneously to the first
presentation of a weak electric field, only three of them showed perfect trans-
fer in this modality by responding continuously to electric field amplitudes of
1.5 mV cm
1
, successively reduced to 0.5 mV cm
1
. Electroreception can
explain short-range prey detection in crater-feeding bottlenose dolphins. The
fact that this is the second odontocete species with experimental evidence for
electroreception suggests that it might be widespread in this marine mammal
group.
KEYWORDS
dolphin cognition, electroreception, stimulus generalization, vibrissal crypts
Received: 1 April 2021 Revised: 12 June 2021 Accepted: 18 July 2021
DOI: 10.1002/ar.24773
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2021 The Authors. The Anatomical Record published by Wiley Periodicals LLC on behalf of American Association for Anatomy.
592 Anat Rec. 2022;305:592–608.wileyonlinelibrary.com/journal/ar
1|INTRODUCTION
Studying the sensory capabilities of marine mammals
provided a deep understanding of how whales and dol-
phins, pinnipeds, and sirenians perceive their environ-
ment. Especially in order to forage effectively or to avoid
predators, assessing and integrating all available informa-
tion on a multimodal scale is essential (Nachtigall, 1986;
Torres, 2017). In addition to species-specific adaptations
of sensory systems like vision or audition to either a fully
aquatic (cetaceans and sirenians) or a semi aquatic (pin-
nipeds) lifestyle, marine mammals developed new sen-
sory abilities beyond the classical mammalian modalities
to gather all available sensory cues.
In seals and manatees for instance, well-developed
vibrissae represent an effective haptic system allowing the
animals to identify for example, benthic food items by
active touch (Bauer et al., 2012; Dehnhardt &
Dücker, 1996; Dehnhardt & Kaminski, 1995; Grant,
Wieskotten, Wengst, Prescott, & Dehnhardt, 2013). How-
ever, in both of these orders the respective vibrissal system
also serves hydrodynamic reception for the detection of
object-generated water movements or currents, allowing
orientation and prey detection under conditions when
vision is restricted (Dehnhardt, Mauck, Hanke, &
Bleckmann, 2001; Gaspard III et al., 2013, 2017; Hanke
et al., 2010; Krüger, Hanke, Miersch, & Dehnhardt, 2018;
Wieskotten, Dehnhardt, Mauck, Miersch, & Hanke, 2010).
Adapted to similar foraging niches such as seals and
sea lions, odontocetes are well known for their use of
active echolocation to acoustically detect objects under
water (Au, 1993; Nachtigall & Patterson, 1980; Norris,
Prescott, Asa-Dorian, & Perkins, 1961). However,
although this highly sophisticated sensory system can
explain many behavioral achievements in toothed
whales, it may not be sufficient for all foraging strategies
in these marine predators. For example, it was demon-
strated in a combined behavioral and anatomical study
that the marine Guiana dolphin (Sotalia guianensis) pos-
sesses passive electroreception, the ability to detect weak
electric fields from animate sources like those their prey
fish inevitably generate due to ion flow at mucous mem-
branes such as the gills (Wilkens & Hofmann, 2005). The
Guiana dolphin's detection threshold was below
5μVcm
1
, an absolute sensitivity even lower than that
known from the platypus, an electrosensory specialist
(Gregory, Iggo, McIntyre, & Proske, 1987). The hairless
vibrissal follicles on the upper rostrum were identified as
the electrosensory units. While newborn toothed whales
still possess vibrissae on their upper rostrum, they
quickly lose them postnatally and only the hairless crypts
remain visible (Japha, 1912; Ling, 1977). The morphologi-
cal study by Czech-Damal et al. (2012) revealed that these
“crypts”in adult S. guianensis lack many of the charac-
teristic structures of vibrissal follicle-sinus complexes
(F-SCs, Rice, Mance, & Munger, 1986) described in terres-
trial mammals like cats and rats. Instead, they consist of a
richly innervated, ampulla-shaped invagination of the epi-
dermal integument, thus resembling the structure of other
electroreceptors, such as the ampullae of Lorenzini in
sharks and rays, or the mucous gland electroreceptors of
the platypus (Manger, Collins, & Pettigrew, 1998;
Manger & Pettigrew, 1996; Murray, 1974).
Passive electroreception allows benthic feeders like
sharks or the monotreme platypus primarily to search for
and locate prey hidden in the sediment (Dehnhardt,
Miersch, Marshall, von Fersen, & Hüttner, 2020;
Kalmijn, 1974; Manger & Pettigrew, 1996). As benthic
feeding is also a foraging strategy in the Guiana dolphin
(Rossi-Santos & Wedekin, 2006), it is suggested that the
electroreceptors on the dolphin's rostrum facilitate prey
detection while digging for prey in the sediment and thus
at least supplement echolocation during benthic feeding.
Behavioral observations of bottom-feeding bottlenose dol-
phins (Kaplan, Goodrich, Melillo-Sweeting, & Reiss, 2019;
Mann & Sargeant, 2003; Rossbach & Herzing, 1997) raise
the question of whether this and also other delphinid spe-
cies possess passive electroreception enabling the animals
to detect the weak electric fields generated by their
prey fish.
In the present study we addressed this question with
the approach of Czech-Damal et al. (2012) by combining
an anatomical study of vibrissal crypts of bottlenose dol-
phins with a behavioral experiment testing for elec-
troreception. Recently, a light microscopic study by
Gerussi et al. (2020) analyzed the functional morphology
of vibrissal crypts in neonate and adult bottlenose dol-
phins (Tursiops truncatus). First of all, they concluded
that, in agreement with the results from Sotalia,the
vibrissal structures are strongly innervated and thus repre-
sent functional sensory units. However, deviating from the
findings in Sotalia they describe the structure to be almost
identical to that of F-SCs in terrestrial mammals, with a
vibrissal hair shaft inside the follicle identified in newborn
as well as adult Tursiops. Gerussi et al. (2020) could not
preclude that vibrissal crypts are electroreceptive in adult
bottlenose dolphins but suggested that their results were
in favor of a mechanosensory or proprioceptive function.
Using light microscopic techniques, we also examined the
vibrissal follicles of neonate and adult bottlenose dolphins.
However, as the emphasis of the present study is on the
verification of electroreception in this species, we will
focus on aspects relevant for electroreception in adult
bottlenose dolphins, in particular the presence of a
vibrissal hair shaft inside the follicle and morphological
similarities to the vibrissal crypts of Sotalia.
HÜTTNER ET AL.593
In addition to the detailed qualitative description of
the innervation of the vibrissae structures of T. truncatus
provided by Gerussi et al. (2020), we quantified the inner-
vation in order to directly compare our data with the data
from S. guianensis. In the behavioral study bottlenose
dolphins were tested for their ability to detect weak elec-
tric DC fields in a stimulus generalization test. We
decided for the stimulus generalization paradigm because
testing for a novel sensory modality never addressed in
an animal could entail different problems. A negative
behavioral result, that is, if a dolphin fails to respond to a
novel stimulus spontaneously, may not be due to the fact
that the animal did not sense the stimulus. Instead, the
animal may have failed to learn the correct response
behavior since it did not have any previous experience
with the novel stimulus and may have focused on other
sensory input instead. Also, the experimental procedure
itself, or limited experience with the specific task could
prevent an animal to respond correctly (Hanke &
Dehnhardt, 2013; Scholtyssek, Kelber, &
Dehnhardt, 2015). That is why we first established a
go/no-go task using only acoustic stimuli. However, prior
to testing the bottlenose dolphins' ability to detect electric
fields, it was tested, whether the dolphins could general-
ize the go/no-go task within and across modalities to new
stimulus dimensions by using novel acoustic, as well as
optical and mechanical stimuli.
2|ANATOMICAL STUDY
2.1 |Material and methods
Vibrissal crypts were obtained from three neonate
(Figure 1a) and three adult (Figure 1b) bottlenose dolphins
(T. truncatus). Samples of a 1-week-old female neonate
(Figure 1a) and a 35-year-old female adult animal were sup-
plied by Nuremberg Zoo, Germany. Two neonates and two
adults of unknown age and gender were provided by Prof.
Dr. Oelschlaeger, Dr. Senckenbergische Anatomie, Frank-
furt a. M., Germany.
The material was fixed in 7–10% formaldehyde and
stored in 4% formaldehyde. For histological purposes
vibrissal crypts were dissected macroscopically, dehydrated,
and embedded in Paraplast Plus (Sherwood Medical,
St. Louis, MO). They were cut at 7 μm on a Leica RM 2135
rotary microtome into sagittal and transversal sections and
attached to SuperFrostPlus glass slides (Menzel-Gläser,
Braunschweig, Germany). In total, seven vibrissal crypts of
neonate and eight of adult bottlenose dolphins were
processed.
For general histology, the dewaxed and hydrated sec-
tions were stained with standard Masson–Goldner
trichrome staining technique. For quantitative determi-
nation of innervation sections were stained with nerve
fiber-specific polyclonal rabbit anti-PGP9.5 (UltraClone
Ltd., Cambridge, UK). To block endogenous peroxidase,
they were pretreated for 40 min in methanol containing
0.5% hydrogen peroxide. After several rinses in 0.01 M
phosphate-buffered saline (PBS), pH 7.4, sections were
incubated for 48 hr at 4C with the primary antibody
diluted in PBS containing 0.5% Triton X-100, 0.01%
sodium azide, and 1% bovine serum albumin (dilutions
1:3000). After rinsing in PBS, Histofine Simple Stain
MAX PO (R) (Nichirei Corporation, Tokyo, Japan) was
used as secondary antibody with room temperature incu-
bation for 30 min and rinsed again. For visualization
AEC Simple Stain Solution (Nichirei Corporation, Tokyo,
Japan) was applied for 15–20 min. Sections were washed
in tap water and in de-ionized water, and coverslipped
FIGURE 1 Appearance and arrangement of follicle-sinus
complexes (F-SCs; neonate) and vibrissal crypts (adult) on the
rostrum of bottlenose dolphins (Tursiops truncatus). (a) Neonate
animal (fixated material) with caudally curled vibrissal hair shafts
1–5. F-SCs 3 and 5 lost the vibrissae during fixation,
Scalebar =1 cm. (b) Vibrissal crypts (1–7) without visible vibrissal
hair shafts of one of the adult subjects of this study while stationing
in the apparatus
594 HÜTTNER ET AL.
using Aquatex (Merck, Darmstadt, Germany) as the
mounting medium. The results were analyzed and photo-
graphed using an Axiophot light microscope (Carl Zeiss,
Oberkochen, Germany) equipped with a SemiCam digital
camera (PCO, Kelheim, Germany). Image editing was
performed digitally with Adobe Photoshop CS and sche-
matic drawings were generated in Adobe Illustrator.
Axon density was quantified in cross sections underneath
the base of the vibrissal crypts.
3|BEHAVIORAL STUDY
3.1 |Material and methods
3.1.1 | Subjects
Subjects were four bottlenose dolphins (T. truncatus):
Kai, a subadult 5-year-old male, as well as Dolly and
Donna, two 10-year-old females, had no previous experi-
mental experience. Anke, a 33-year-old female, took part
in experiments on lateralization (Kilian, von Fersen, &
Güntürkün, 2000) and the ability to form equivalence
classes (von Fersen & Delius, 2000). The dolphins lived
together in a group of 7–10 bottlenose dolphins at
Nuremberg Zoo, Germany. The entire enclosure con-
sisted of a complex of six outdoor and three indoor pools
of various sizes and depths with a total water volume of
approximately 7 million liters of saltwater. Experimental
sessions were carried out in a round indoor pool (diame-
ter approximately 12 m) with one animal at a time, once
per day, usually 5 days per week. Approximately 20% of
the daily diet was fed to the dolphins during one experi-
mental session (1–1.5 kg of herring, sprat, capelin, squid,
and mackerel).
3.1.2 | Experimental setup
A cubic-formed experimental apparatus for multimodal
testing was constructed from PVC tubes. Prior to the
beginning of each experimental session, it was lifted into
the water and was mounted to the edge of the pool
(Figure 2a). Besides the experimenter, an assisting trainer
handled the dolphin during an experimental session. The
trainer was positioned on the opposite side of the pool
and was out of the dolphin's sight during trials. For a
trial, upon a hand signal given by the trainer, the dolphin
entered the apparatus headfirst through a square opening
of 50-cm side length at the front side of the apparatus
(Figure 2a,b). Then, the dolphin stationed itself on a tar-
get (a plastic ball) in the center of the apparatus, approxi-
mately 1 m below the water surface (Figure 2a,b). The
target was fixed at a distance of 50 cm to the square open-
ing. Additionally, a U-shaped resting platform was
installed in front of the target, for the animal to place its
lower jaw on. Both, the target, and the resting platform
ensured a consistent position of the dolphin's head and
rostrum in every trial (Figure 2b).
The experimenter sat on land behind the submerged
apparatus, out of sight of the stationing dolphin. Addi-
tionally, in order to prevent any unintentional cueing by
the experimenter, a visual cover (white tarp screen,
110 cm 80 cm, Figure 2a) was attached to the PVC-
structure above the animal's station screening the experi-
menter from the dolphins during the trials. The experi-
menter observed the dolphins' behavior during each trial
via a monitor screen that was connected to an underwa-
ter camera (WoSports
®
Fish Finder, Figure 2a) fixed to
FIGURE 2 (a) Experimental setup for the presentation of
different stimulus types (1: light on/off, 2: water jet, 3: electric
stimulus, 4: pure tones, 5: air bubbles; melodica sounds not shown)
addressing four different sensory modalities. (b) Close-up view of a
dolphin inside the apparatus. The dolphin enters the apparatus and
touches the target while resting the lower rostrum on the U-shaped
resting platform
HÜTTNER ET AL.595
one of the tubes on the right side of the apparatus,
pointed at the stationing dolphin.
3.1.3 | Experimental phases and stimuli
A total of 55 different stimuli were presented during the
three phases of this study. The stimuli were chosen to
address a dolphins hearing, vision, the tactile, as well as
the hypothesized electric sense (Table 1).
Phase 1: Acquisition the go/no-go task
Three different training stimuli were used to establish
the go/no-go behavior within the auditory modality.
Each session consisted of 20 trials. While a standard
dog clicker was presented underwater, a standard
police whistle and a metal rattle were presented above-
water. During a session, training stimuli were pres-
ented in randomized order. In sessions of 20 trials the
learning criterion in this phase was defined as ≥80%
correct responses (combined hits across all go-trials
presenting one of the three training stimuli and correct
rejections, binomial test, p< .0001) and a false alarm
rate of ≤20% over three consecutive sessions (binomial
test, p<.01).
Phase 2: Auditory go/no-go generalization
In the second experimental phase, additional acoustic
stimuli were used to test the dolphins' ability to general-
ize the previously learned go/no-go task within the audi-
tory modality (see Table 1). At first, 25 different melodica
sounds (Thomann Melodica, 37 keys, range: f-f”“) were
introduced that were manually played by the experi-
menter for 3 s. The 25 stimuli included 15 single tones
between 0.3 and 1.4 kHz with a frequency difference of
≥0.04 kHz. Additionally, 10 double stimuli were used
with two keys played at once (0.3—1.2 kHz). Frequency
differences between the two pressed keys of each double
stimulus ranged from 0.05 to 0.18 kHz.
Twenty-one digitally produced pure tones were pres-
ented for 3 s likewise. The pure tones were produced run-
ning version 2.2.1 of Audacity
®
recording and editing
software (Audacity Team, 2017). The pure tones were
generated between 8 and 15 kHz at intervals of 0.5 kHz
and could thus be clearly discriminated by bottlenose dol-
phins (Thompson & Herman, 1975). They were presented
via an underwater speaker (DNH Aqua 30, Norway) that
was placed in the center of the experimental apparatus
above the visual cover, approximately 30 cm below the
water surface (Figure 2a). The underwater speaker was
connected to a second speaker (Bose Roommate II
Powered Speaker, Bose Corporation, USA) powered by a
12 V battery pack (GP Recyko 210AAHCB 2050 mAh)
placed on land next to experimenter. The pure tones were
played simultaneously on both speakers to control stimu-
lus presentation via a small audio player (iPod Nano, 3rd
generation, 2007, Apple Inc., USA) that was connected to
the second speaker. Last, a bike-bell was attached to one
of the tubes of the apparatus underwater and served as
another acoustic stimulus producer (Table 1). A string
was tied to the bell, enabling the experimenter to ring the
bell from his position behind the apparatus.
The criterion for a successful generalization of the
go/no-go behavior to the novel acoustic stimuli was
defined as an averaged first trial performance of ≥80%
hits (binomial test, p< .05) over all different sounds of
each novel stimulus type (melodica sounds: 25 different
sounds; pure tones: 21 different sounds). The dolphins
also needed to demonstrate an overall performance of
≥80% correct responses (hits and correct rejections com-
bined), and a false alarm rate of ≤20% during these ses-
sions in which the novel stimuli were introduced to fulfill
the criterion. The novel stimuli were interspersed 2–4
times per session in a randomized order. The melodica
sounds were introduced first and, session-by-session, a
TABLE 1 Stimulus types used during the experiment
addressing different sensory modalities
Stimulus
category
Stimulus
name Stimulus description
Acquisition of the go/no-go task
Acoustic
stimuli
Training
stimuli
Dog clicker, police whistle,
metal rattle
Auditory go/no-go generalization
Acoustic
stimuli
Melodica
sounds
25 different melodica sounds
played by the experimenter
Pure tones 21 digitally produced sine
wave pure tones
Bike bell Bike bell ringed underwater
Go/no-go across-modality transfer
Optical
stimuli
Light ON LED lights switched on
Light OFF LED lights switched off
(switched on before the
dolphin enters the
apparatus)
Mechanical
stimuli
Water jet Water jet directed at the
upper rostrum
Air bubbles Air bubble flow directed at
the lower rostrum
Electrical
stimuli
DC electric
fields
Weak DC electric fields
(≤1.5 mV cm
1
) were
generated above the
dolphins' rostrums
596 HÜTTNER ET AL.
new melodica sound was added to the stimulus set and
presented instead of one of the training stimuli until all
training stimuli trials were replaced by a melodica sound.
In the next step, as far as the dolphins reached the
learning criterion, the sine wave pure tones were intro-
duced. The novel pure tones were also presented two to
four times per session. Now, melodica sounds, and pure
tones were presented in a randomly mixed sequence.
After a dolphin reached the learning criterion with the
pure tones, the bike bell was introduced. However, only
Anke and Dolly were tested with this third acoustic
stimulus type.
Phase 3: Go/no-go across modality transfer
To test for transfer of the go/no-go task across modalities,
stimuli addressing the dolphins' visual, tactile, and
electrosensory modality were presented. Novel stimulus
types were randomly interspersed two to four times
among the go trials of each session. They were mixed
with the familiar acoustic stimuli of phase 2 and pres-
ented in a random, counterbalanced sequence. Thus, the
dolphins were confronted with an increasing number of
different stimulus types presented across different modal-
ities within each session. In all transfer tests, the criterion
for a successful transfer was defined as a correct response
at first trial, and an averaged hit rate of ≥80% hits over
the first 15 trials of each novel stimulus (binomial test,
p< .05). The dolphins also needed to reach an overall
performance of ≥80% correct responses and a false alarm
rate of ≤20% during these sessions. First, the visual and
tactile modality was addressed to test if the dolphins are
able to transfer the go/no-go task across modalities
already know in this species. At last, weak electric fields
were introduced as stimuli, this way testing the dolphins
for this novel sensory modality. Regarding the visual
modality of the dolphin, three serially connected under-
water LED lamps were used to create two different opti-
cal stimuli, “light-on”and “light-off”(Table 1). One of
the three LEDs were installed on each side of the appara-
tus next to the square opening and thus located on both
sides of a dolphin's head (Figure 2). The third lamp was
fixed on top of the apparatus and was directed upwards.
It served as a control light for the experimenter and was
not visible for the dolphins. For the “light ON”stimulus
the LEDs were switched on for 3 s by the experimenter
after the subject stationed properly and then switched off
again. For the “Light OFF”stimulus the LED lights were
already switched on before the dolphin entered the appa-
ratus and then were switched off to generate a reversed
stimulus.
Next, two mechanical stimuli were used to address
the dolphins' tactile modality. The first consisted of a
weak water jet that was directed toward the upper
rostrum of the dolphin. Therefore, a thin water hose
(0.5 cm in diameter) that was connected to the facilities'
general water supply using a ball-shaped valve was
installed above the target. The opening of the water hose
was located approximately 10 cm above the upper ros-
trum of a dolphin in station. The experimenter manually
opened the valve to produce the water jet. The strength
of the water jet was consistent during all trials. Stimulus
duration was defined as 3 s and, as well as the strength of
the water jet, was controlled by the experimenter. The
second mechanical stimulus consisted of a flow of air
bubbles directed toward the lower jaw of the dolphin.
Therefore, a flexible tube (0.3 cm in diameter) was fixed
to a PVC tube below the resting platform. The tube was
connected to a common aquarium air pump. Switching
on the pump created an upward flow of air bubbles rising
toward the water surface that way touching the dolphin's
lower rostrum. Again, stimulus duration was 3 s, con-
trolled by the experimenter by switching the pump on
and off.
To test whether bottlenose dolphins possess elec-
troreception, weak DC electric fields were generated by a
custom-made battery-powered electric field generator
(EFG, version 2.0, 2014, University of Rostock) acting as
a constant current source. Two copper wire electrodes
(1 cm long, 2 mm in diameter and 1 cm apart) embedded
in a PVC tube were connected to the electric field genera-
tor to form an electric circuit. The electrodes were
installed in the center of the apparatus above the resting
platform (Figure 2) with a distance of approximately
10 cm from the upper rostrum of a stationing dolphin.
Positioned directly in front of the melon, the electrodes
were not visible by the dolphins as soon as they were in
station. Even though the visual fields of both eyes overlap
up to 20
–30(Mass & Supin, 2018), the distinctive anat-
omy of the bottlenose dolphin's head, and the lateral
position of the eyes create a small blind area directly in
front of the melon above the rostrum (Cozzi,
Huggenberger, & Oelschläger, 2017; Dral, 1975; Xitco,
Gory, & Kuczaj II, 2004).
Electric stimulus duration was controlled by a timer,
stimulus duration was set to 3 s for the experiment. The
EFG was powered by three 12 V batteries, two of which
were serially connected and served as the power-source
for stimulus generation. Electric fields were generated
using either one 12 V battery or both (24 V). A multiturn
wirewound potentiometer (VISHAY SPECTROL, Model
534, 10 turns, 10 kΩ, VISHAY Intertechnology, Malvern,
PA) was used to modulate stimulus amplitude. The
experimenter monitored and recorded applied currents
manually using a digital TrueRMS multimeter (Voltcraft
VC870, Conrad Electronics SE, Germany) connected to
the EFG.
HÜTTNER ET AL.597
Electric field amplitude was measured and adjusted
to a predefined stimulus strength before each session.
Stimulus strength was also measured after each session
to control stimulus stability. Two nonpolarizable
Ag/AgCl electrodes (1 cm long, 0.1 mm thick, 1 cm
apart) were used as measuring electrodes. They were
fixed to the apparatus and placed below the stimulus
electrodes at the same location as the vibrissal crypts of
a stationing dolphin. The measure electrodes were con-
nected to a custom-made electric field detector (EFD,
version 2.0, 2014, University of Rostock) based on an
instrumental amplifier (AD620 amplifier, Analogue
Devices, Norwood, MA). A TrueRMS multimeter (UNI-T
UT61D, Uni-Trend Technology, China) was connected
to the EFD to monitor the electric signal. The
multimeter was connected to a battery-powered laptop
via USB running a measuring and recording software
(UT61CD Interface Program 3.03, Uni-Trend Technol-
ogy, China). Recording interval was approximately 1 Hz.
Electric field amplitude was calculated in real-time by
the experimenter using Microsoft Excel (2013, 2016).
Electric fields of 1.5 mV cm
1
were initially used during
the experiment.
3.2 |General experimental procedure
All sessions were run with a trainer handling the dolphin
and the experimenter handling stimulus presentation.
Each session consisted of 20 pseudo-randomly mixed go
and no-go trials following the rules of Gellermann (1933)
and Holt and Schusterman (2002). At the beginning of a
trial, the trainer sent the dolphin to swim into the appa-
ratus. Stimulus presentation started approximately 3 s
after the dolphin entered the apparatus and touched the
target. During go trials (stimulus present trials) the dol-
phin had to respond to the stimulus by leaving the appa-
ratus within 5 s after the stimulus onset (hit). Irrespective
of stimulus type, stimulus duration in all experimental
sessions was 3 s. The experimenter used secondary rein-
forcement for correct responses (hits) by giving a short
single whistle-sound. As soon as it returned to the
trainer, it was reinforced with fish. During no-go trials
(stimulus-absent trials) the dolphin had to stay in the
apparatus and continue to station on the target until a
predefined time limit of 12 s expired. This correct rejection
was also immediately secondarily reinforced by the
experimenter with a short whistle sound. Then the dol-
phin was allowed to leave the apparatus and was also
rewarded with fish by the trainer. Correct responses were
always reinforced. False responses, that is, no responses
during go trials (miss) or go responses during no-go trials
(false alarm) were signaled to the dolphin with three
short consecutive whistle-sounds by the experimenter
and were not reinforced.
In order to exclude the possibility that the dolphins
could have completed the transfer tests by relying on
potential acoustic secondary cues associated with the
onset of the optical, mechanical, or electric stimuli, an
acoustic analysis was carried out. A hydrophone
(Neptune D140) connected to a battery-powered laptop
running a custom-made LabVIEW sound recording soft-
ware (“Biologger using NI USB-63x6”, programmed by
Alain Moriat, National Instruments, USA; courtesy of
Magnus Wahlberg, University of Southern Denmark)
was placed in the center of the experimental apparatus at
the dolphin's head position. Stimuli were presented in
the exact same manner as during an experimental ses-
sion. However, irrespective of the stimulus type pres-
ented, no audible acoustic artifacts were revealed. We
can therefore rule out that secondary acoustic cues indi-
cated the presence of nonacoustic stimuli that the dol-
phins could have used to master the transfer tests.
4|RESULTS
4.1 |Anatomical study
Aligned along the rostral-caudal axis, the vibrissal crypts
are visible as oval pores on the skin surface of the upper
rostrum. They consist of an asymmetric, downwardly
tapering, saber-shaped invagination of the epidermal
integument (Figures 3, 4, 6). In newborns, the overall size
varies from 4.1 to 6.2 mm in length and 0.3 to 1.1 mm in
width. Adults possess larger crypts that vary from 5.8 to
8.2 mm in length, and 0.2 to 2.6 mm in overall width.
Figure 1a shows vibrissal crypts of a 1-week-old new-
born animal each containing a vibrissal hair shaft visible
at the skin surface. The external parts of the hairs were
curled caudally and were 18 mm long. The vibrissal
crypts of adult bottlenose dolphins lack vibrissal hairs
but exhibit an expanded lumen that varies between 0.6
and 2.4 mm in total lumen width and 5.4–7.5 mm in total
length. The lumen possesses a crater-shaped opening and
is filled with a dense meshwork of keratinous fibers and
swirled corneocytes (Figure 3a,b). This loose and spongy
structure in adult Tursiops truncatus differs considerably
from the solid, self-contained structure of the hair shaft
present in neonate animals what becomes particularly
clear when comparing the transversal sections in
Figure 4a,b
While neonates possess an intact hair papilla and a
keratinizing zone, in adults the hair papilla contains an
agglomeration of fat cells (Figure 5a,b), but no
keratinizing zone can be observed. Neonates as well as
598 HÜTTNER ET AL.
adult vibrissal crypts are wrapped with a thick layer of
loose connective tissue enclosed by a capsule of dense
connective tissue. In neonates a dermally derived cavern-
ous sinus is found that is embedded in the layer of loose
connective tissue surrounding the crypt (Figure 5a,b).
Filled with red blood cells, the sinus spaces are segre-
gated by connective tissue trabeculae. In adults, by con-
trast, the cavernous sinus is reduced to individual blood-
filled lacunae. Nevertheless, overall blood supply is not
reduced in adult dolphins and small arterioles run
upward through the loose connective tissue along the
vibrissae crypts. In both, neonates and adults, differences
are found between the pattern and density of the overall
blood supply of the vibrissal crypts and the surrounding
tissues. The cavernous sinus and hair papilla are supplied
with blood by muscle-bound blood vessels that reach and
traverse the vibrissal crypts basally and laterally.
4.1.1 | Innervation
PGP 9.5 staining showed a rich innervation of vibrissal
crypts in both, neonate, and adult bottlenose dolphins.
The count of axons in a plane below the hair papilla
resulted in 290 axons per vibrissal crypt (n=2 crypts) in
neonates and 245 axons per crypt (n=2 crypts) in adults.
Differences between the crypts could not be determined
in neonates as well as in adults.
In both newborns and adults, each crypt is innervated
by large bundles of the deep vibrissal nerve that approach
the base of a crypt from caudally, where they branch out
into many smaller bundles (Figure 3). Most nerves then
FIGURE 3 (a) Longitudinal section through a vibrissal crypt in
adult bottlenose dolphins (Masson–Goldner trichrome). Black
arrows indicate dermal layer inter-spersed with adiposal cells. Note
the expanded Lumen (L) filled with a dense meshwork of
keratinous fibers and corneocytes. Large arteries (a) and nerve
bundles (Nb) approach and ascend. Scalebar =1.5 mm. (b)
Schematic drawing of the structure and innervation of vibrissal
crypts in adult bottlenose dolphins. In adults, the cavernous sinus
(CS) is reduced, and the hair papilla (HP) transforms to an afc. The
L widens with age and instead of a vibrissal hair shaft (VHS) a
meshwork of keratinous fibers and corneocytes is situated in the
L. Arrows indicate dermal layer interspersed with adiposal cells. A,
arteries, afc, agglomeration of fat cells; C, capsule; CS, cavernous
sinus; EP, epidermis; HP, hair papilla; L, Lumen; lct, loose
connective tissue; Nb, nerve bundles; SVN, superficial vibrissal
nerve; VHS, vibrissal hair shaft
FIGURE 4 Cross section through a vibrissal crypt above the hair bulb in an adult (a) and a neonate (b) bottlenose dolphin (Masson–
Goldner trichrome). (a) Note the dense meshwork of corneocytes and keratinous fibers in the lumen (L). (b) Note the wide sinus spaces of
the CS as well as the dense structure of the VHS. Scalebar =0.5 mm. (c), capsule; CS cavernous sinus; L, Lumen; Nb, nerve bundles; VC,
vibrissal crypt; VHS, vibrissal hair shaft
HÜTTNER ET AL.599
penetrate the capsule of a vibrissal crypt basally and lat-
erally at the level of the hair papilla and spread upward
through the layer of loose connective tissue along the
internal margin of the capsule. In addition, two branches
of the superficial vibrissal nerve (SVN) ascend apically
and then branch to penetrate the capsule as well as to
innervate the epidermis. In neonates, some branches
additionally traverse the capsule midway on the caudal
and rostral side.
4.2 |Behavioral study
4.2.1 | Phase 1: Acquisition of the go/no-
go task
All dolphins performed highly significant during the last
three sessions to reach the learning criterion (see
Table 2). Anke completed three consecutive sessions with
20 trials the fastest and reached the learning criterion
after 109 trials, whereas Dolly and Donna needed
820 and 1,058 trials, respectively. Kai never performed
more than 10 trials per sessions due to persistent poor
motivation. He reached the criterion of ≥80% correct
responses over three consecutive sessions with 10 trials
after 210 trials.
4.2.2 | Phase 2: Auditory go/no-go
generalization
For each new stimulus type, the dolphins first trial per-
formance on each individual sound, their overall perfor-
mance (hits and correct rejections combined) and false
alarm rate over all sessions during which the new sounds
were introduced was analyzed. Also, the dolphins' all-
time hit rate on all trials of each new stimulus type that
were continuously presented until the end of the study
was calculated.
The dolphins' pooled first trial performances over all
melodica sounds were highly significant as the dolphins
responded correctly to at least 23 of the 25 different
sounds on first trials (see Table 3). Only Anke failed to
respond correctly on the very first melodica sound trial,
the other three dolphins immediately transferred the
go/no-go task to the new stimulus type. The dolphin's
overall performances and false alarm rates during these
sessions were also significant (7–22 sessions; see
Table 3). Although Kai's false alarm rate over the 22 ses-
sions during which the melodica sounds were presented
for the first time was with 21.6% above the predefined
criterion, his performance on no-go trials was still signif-
icantly different from chance level (binomial test,
p< .0001; see Table 3). Thus, it was decided to continue
with the experiment, also to avoid increased frustration.
All dolphins continued to respond well across all mel-
odica sounds presented until the end of the study (see
Figure 6).
Presented with the 21 different pure tones, Anke,
Dolly, and Donna immediately responded correctly on
FIGURE 5 Hair papilla in a neonate and an adult bottlenose
dolphin (Masson–Goldner trichrome). (a) Neonate with an intact
HP and kz. (b) Adult. Former HP transformed into an afc.
Scalebar =200 μm. afc, agglomeration of fat cells; HP, hair papilla;
kz, keratinizing zone; L, lumen
TABLE 2 The dolphins'
performances (overall performance, hit
rate, and false alarm rate) during the
last three sessions to train the go/no-go
paradigm using the three training
stimuli
Go/no-go acquisition
Subject Overall performance Hit rate False alarm rate
Anke 85.0** 86.7** 16.7**
Dolly 95.0*** 91.7*** 0.0***
Donna 91.7*** 96.7*** 13.3***
Kai 96.7*** 100.0*** 6.7**
Level of significance:
**
p< .001;
***
p< .0001.
600 HÜTTNER ET AL.
first trials, respectively (Table 3). The three dolphins'
overall session performance was highly significant with
at least 95.8% correct responses and no more than 8.9%
false alarms across all sessions during which the 21 pure
tones were introduced (Anke: 11 sessions, Dolly: 14 ses-
sions; Donna: 6 sessions; Table 3). All dolphins further
maintained a significant hit rate across all pure tones
presented until the end of the study (see Figure 6).
Anke and Dolly immediately responded correctly also
to the bike bell at first trials and upheld a significant per-
formance over the first 15 trials with no errors (Table 3).
Both dolphins demonstrated a high overall performance
as well as low false alarm rate during these sessions
(Anke: 89.0% hits, 16.0% false alarms, Dolly: 95.5% hits,
8.1% false alarms). Their hit rate over all bike bell trials
(101 trials each) until the end of the study remained
highly significant. 99.0% (Anke, 100 hits, binomial test,
p< .0001) and 100.0% (Dolly, binomial test, p< .0001),
respectively (Figure 6).
Due to management constraints, Kai and Donna were
not able to participate in all transfer tests. Kai's ability to
generalize the task within the auditory modality was only
tested using the melodica sounds. Donna did not perform
tests on auditory generalization using the bike bell and
on transfer tests across modalities using the light off-
stimulus and the air bubbles.
TABLE 3 The dolphins' go/no-go performances during acoustic generalization tests. For the 25 melodica sounds and the 21 pure tones
≥80% correct go-responses (hits) on cumulated first trials was defined as a successful generalization. For the bike bell, the dolphins' hit rates
over the first 15 go-trials were calculated. Additionally, the four subjects were required to show an overall performance of ≥80% correct
choices (hits and correct rejections combined), as well as false alarm rates ≤20% over sessions during which the novel sounds were
introduced
Acoustic go/no-go generalization
Subject
Number of
sessions
Number of
first trials
First trial
performance
Overall session
performance (%)
Overall session false
alarm rate (%)
Melodica sounds
Anke 7 25 92.0*** 87.9*** 18.6***
Dolly 16 25 100.00*** 94.5*** 9.3***
Donna 13 25 92.0*** 96.2*** 3.7***
Kai 22 25 96.0*** 83.3*** 21.6***
Pure tones
Anke 11 21 100.0*** 96.8*** 8.9***
Dolly 14 21 100.0*** 97.9*** 4.5***
Donna 6 21 100.0*** 95.8*** 6.7***
Bike bell
Anke 5 15 100.0*** 89.0*** 16.0***
Dolly 8 15 100.0*** 95.5*** 8.1***
Level of significance:
**
p< .001;
***
p< .0001.
FIGURE 6 The dolphins' all-time hit rate over all trials
presented throughout the course of this study for each stimulus
type. Except Kai on electric field trials of 1.5 mV cm
1
, all dolphins
demonstrated significant performances of at least 80% hits (dotted
line) on all stimulus types
HÜTTNER ET AL.601
4.2.3 | Phase 3: Go/no-go across-modality
transfer
Visual modality
Dolly, Donna, and Kai performed well on the first optical
stimulus (light-on) transfer tests. They all responded cor-
rectly at first trials and continued to respond well over
the first 15 trials (see Table 4). Overall performances
and false alarm rates during these sessions were also sig-
nificant during these sessions (five–eight sessions,
Table 4). While Anke also showed a significant overall
performance and a low false alarm rate across all ses-
sions during which the light-on stimulus was presented
for the first 15 times, her averaged hit rate over the first
TABLE 4 The dolphins' performance over the first 15 trials with each novel stimulus type during across modality transfer tests. A
successful generalization was defined as an averaged hit rate ≥80% at these first 15 trials while maintaining an overall performance (hits &
correct rejections combined) ≥80% and a false alarm rate ≤20%
Go/no-go across-modality transfer
Subject
Number of
sessions
Hit rate first 15
presentations (%)
Overall session
performance (%)
Overall session false
alarm rate (%)
Visual modality
Light on
Anke 5 26.7 86.0*** 4.0***
Dolly 5 100.0*** 98.1*** 4.3***
Donna 6 100.0*** 95.0*** 10.0***
Kai 8 93.3** 91.3*** 15.0***
Light off
Anke 8 100.0*** 95.0*** 9.4***
Dolly 5 40.0 84.0*** 12.5***
Donna —— — —
Kai 7 100.0*** 78.8*** 27.3*
Tactile modality
Water jet
Anke 5 100.0*** 97.0*** 0.0***
Dolly 4 100.0*** 96.3*** 0.0***
Donna 5 100.0*** 93.3*** 13.3***
Kai 8 100.0*** 86.3*** 15.0***
Air bubbles
Anke 6 86.7* 92.2*** 8.5***
Dolly 5 100.0*** 94.0*** 15.0***
Donna 7 100.0*** 95.7*** 7.1***
Kai 6 100.0*** 80.3*** 27.7*
Electrosensory modality
1.5 mV cm
1
Anke 4 93.3** 87.5*** 22.5**
Dolly 4 100.0*** 95.0*** 3.1***
Donna 4 100.0*** 96.3*** 7.5***
Kai 5 60.0 68.6 36.0
Level of significance:
*
p< .01;
**
p< .001;
***
p< 0.0001.
602 HÜTTNER ET AL.
15 optical stimulus trials was below chance level at only
26.7% (Table 4). However, in the following Anke's per-
formance increased and Anke learned to respond well to
the light-on stimulus. At the end of the study, after
490 trials, her hit rate had increased to 94.5%. The other
three dolphins' all-time hit rate across all light-on trials
also remained at a highly significant level (Figure 6).
Kai, Anke, and Dolly performed additional transfer
tests within the visual modality using the reversed optical
stimulus (light-off). Kai and Anke both showed a highly
significant performance with correct responses at first tri-
als (Table 3), no errors over the first 15 presentations
(Table 4) and demonstrated significant performances in
regard to overall performance and false alarm rate
(Table 4). Dolly responded correctly on the first trial but
over the first 15 trials her performance dropped below
chance level (binomial test, p=.8491, Table 4). As the
reversed optical stimulus was continued to be inter-
spersed randomly, her performance increased to 85.0%
until the end of the study (185 trials, Figure 6). Anke also
continued to respond well to the light-off stimulus (94.9%
hits, Table 4). With Kai, only 15 trials could be
carried out.
Tactile modality
All dolphins immediately transferred the go/no-go
behavior to the tactile modality. First trial performance
was correct for all four dolphins (Table 4) for both the
water jet and the air bubbles, and all subjects sustained
a highly significant hit rate over the first 15 trial
(Table 4) accompanied by a significant overall perfor-
mance and false alarm rate during these sessions
(Table 4). They also maintained a significant hit rate of
at least 96.5% correct responses over all trials presented
until the end of the study. Dolly and Donna even
showed a 100.0% hit rate for of both stimuli (Dolly:
248 water jet trials, 231 air bubble trials; Donna: 47 water
jet trials, 54 air bubble trials).
Electrosensory modality
In the last stage of this study, all four dolphins responded
correctly at first trials to weak DC electric fields of
1.5 mV cm
1
. Anke, Dolly, and Donna continued to
show a highly significant hit rate over the first 15 trials
(Table 4). Kai's performance, however, dropped to chance
level after the first 15 trials (binomial test, p=.6072,
Table 4). This poor performance by Kai after he had pre-
viously reacted spontaneously to the first electric field
stimuli can be explained by his generally poor motivation
in the experiment already before testing on elec-
troreception started. Anke, Dolly, and Donna still
maintained a highly significant performance after 32 trials
with an electric field of 1.5 mV cm
1
(Figure 6). In the
following the electric field amplitude was reduced to 1.0
and 0.5 mV cm
1
, respectively, with 32–34 trials each.
The performance of the three dolphins remained almost
unaltered by these changes in stimulus strength (see
Figure 6).
5|DISCUSSION
5.1 |Morphology
Our anatomical results demonstrate that while neonate
bottlenose dolphins possess a vibrissal hair shaft, a func-
tional hair papilla and a cavernous sinus, adult animals
lack these features. These modifications suggest func-
tional differences of vibrissal crypts in neonates and
adults.
The vibrissal hairs in the 1-week-old neonate reached
a maximum overall length of 18 mm which is consistent
with a former study by Eberle and Regan (1998). But in
contrast to straight vibrissae in most terrestrial mammals
which are circular in cross section, they are curled back-
wards displaying an oval form in cross section (Figures 1a
and 4b) as for example, in pinniped species of Otariidae
and Odobenidae (Dehnhardt & Hanke, 2018). The neonate
vibrissal crypts were innervated by a mean number of
290 axons per crypt which exceeds the number of axons
found in F-SCs of most terrestrial mammals (Halata, 1975;
Halata & Munger, 1980; Rice et al., 1986; Rice, Fundin,
Arvidsson, Aldskogius, & Johansson, 1997). The combina-
tion of the vibrissal hair as a mechanical leverage with a
cavernous sinus and a high innervation density supported
a mechanosensory function of the vibrissal crypts in neo-
nates as suggested by Gerussi et al. (2020). Vibrissae in
neonate toothed whales may be important for successful
nursing right after birth (Gerussi et al., 2020) as it has been
reported for suckling rats that are unable to attach to the
maternal nipple without their vibrissae (Kenyon, Keeble, &
Cronin, 1982). However, due to the mechanical properties
of the vibrissal hair being oval in cross section as in pinni-
peds, it may also respond to water movements impinging
on the hair from all directions. As neonate bottlenose dol-
phinstendtoswiminechelonformationwiththeir
mothers during the first weeks of life (Chirighin, 1987;
Cockcroft & Ross, 1990; Eastcott & Dickinson, 1987;
Mann & Smuts, 1999; Reid, Mann, Weiner, &
Hecker, 1995; Reiss, 1988; Tavolga & Essapian, 1957) their
vibrissal hairs could be used to perceive drifting out of the
mother's laminar boundary layer and thus to maintain close
contact with her. Considering that analyses of the melon tis-
sue indicate that dolphins are not born with the biochemi-
cal properties for functional echolocation (Gardner &
Varanasi, 2003) and that this capability as well as the
HÜTTNER ET AL.603
improvement of motor coordination develops over time
(Mann & Smuts, 1999), it is conceivable that the neonate
vibrissal crypts assume an important mechanosensory and
hydrodynamic role until sensory-motor ability has
improved.
Contrary to the study of Gerussi et al. (2020) our
results indicate a profound structural modification of
vibrissal crypts in adult bottlenose dolphins and show
many basic morphological similarities to those found in
the electrosensitive adult Guiana dolphin (Czech-Damal
et al., 2012). As in Sotalia, vibrissal crypts display a kind
of ampullary shape with an expanded lumen and a
crater-shaped opening. Instead of possessing a vibrissal
hair the lumen is filled with a dense meshwork of
corneocytes and keratinous fibers. If this dense mesh-
work is not pulled apart in the course of tissue prepara-
tion or during the cutting process (see Figure 3a), the
impression of a solid structure and thus a hair shaft can
easily arise. However, Figure 4 clearly shows the spongy
composition of this structure in adult dolphins as com-
pared to the solid hair in neonates.
However, whether the rest of internal part of the
neonate vibrissal hair shaft remains inside the lumen
and decays over time or if it is additionally shed needs
to be investigated in more detail. Since the vibrissal
crypts of adult bottlenose dolphins are comprised of an
invagination of the skin, it would be certainly possible
that the dense mixture of corneocytes and keratinous
fibers inside the lumen consists of the remains of the
vibrissal hair shaft as well as desquamated corneocytes
of the stratum corneum. But the general high desqua-
mation rate in toothed whales (Brown, Geraci, Hicks,
St. Aubin, & Schroeder, 1983; Hicks, St Aubin, Geraci, &
Brown, 1985; St. Aubin, Smith, & Geraci, 1990) suggests
that the meshwork trapped in the crypt consists mere
of shedded corneocytes which from vibrissal crypts in
live S. guianensis as well as T. truncatus are excreted
from time to time (Dehnhardt, personal observation).
Thus, it is reasonable to suppose that they are also dis-
posed of the hypothesized conductive biogel which is
assumed to enhance their sensitivity to voltage
gradients.
5.2 |Behavioral experiment
5.2.1 | Stimulus generalization
The results of the behavioral experiments show that
bottlenose dolphins can learn and generalize a specific
response paradigm. After the dolphins were trained with
an auditory go/no-go task they transferred it both within
and across modalities to novel acoustic, as well as novel
optical, mechanical, and also to weak electric fields of
1.5 mV cm-1.
Clearest evidence for generalization is a successful
first trial transfer with novel stimuli and the maintenance
of transfer performance equal to training performance
(Daniel et al., 2016; Herman, Hovancik, Gory, &
Bradshaw, 1989; Herrnstein, 1979; Katz, Wright, &
Bodily, 2007; Wright & Katz, 2007). When new sounds
were presented (melodica and pure tones), the dolphins
immediately left the apparatus according to the previ-
ously established go/no-go task and showed a high per-
formance at first trials with a new stimulus as well as
during the further 15 trials presented with a new stimu-
lus. These findings demonstrate that the dolphins had
learned to generalize the go/no-go task within the audi-
tory modality.
Bottlenose dolphins have been shown to be capable of
identifying objects only perceived through echolocation
with the identical object only perceived visually and vice
versa (Herman, Pack, & Hoffmann-Kuhnt, 1998; Pack &
Herman, 1995; Pack, Herman, & Hoffmann-Kuhnt, 2004;
Pack, Herman, Hoffmann-Kuhnt, & Branstetter, 2002).
Although we did not test for intermodal transfer of stimulus
perception, the dolphins in the present study showed the
transfer of an abstract rule like “Go, whenever you perceive
something”across modalities. The robust transfer with a
high first trial performance indicates that the dolphins not
only generalized the go/no-go task but had learned the
go/no-go rule on a conceptual level. During the first two
phases, transfer performance could have been mediated by
physical similarities between the stimuli as all stimuli were
presented within the same modality (Daniel et al., 2016;
Katz et al., 2007; Wright & Katz, 2006). In the third phase,
however, the dolphins were confronted with novel stimulus
types presented within new modalities. Instead of
responding to common features, the dolphins had to form a
more abstract relation between the novel stimuli and the
corresponding “go response”to be able to respond to the
new optical, mechanical, or electric stimuli. Whether an
animal learned a concept is measured by its ability to trans-
fer a learned relation to novel stimuli at the first trial
(Castro & Wasserman, 2017; Lazareva & Wasserman, 2018;
Roitblat & von Fersen, 1992; Thompson, 1995; Zentall, Gal-
izio, & Critchfied, 2002; Zentall & Hogan, 1975). Only at
first experience with a new stimulus situation, learning can-
notaffectananimal'sperformance.Consequently,asuc-
cessful first trial transfer can only be achieved through the
formation of a concept (Pack, Herman, & Roitblat, 1991;
Roitblat & von Fersen, 1992; Vonk & Jett, 2018; Vonk,
Jett, & Mosteller, 2012; Wright & Katz, 2006). However, due
to the small number of novel stimuli within each modality,
our first trial data can only hint at the formation of an
abstract concept in the dolphins of this study.
604 HÜTTNER ET AL.
Furthermore, maintaining a performance equal to
training performance after new stimuli are introduced is
considered a hallmark for the formation of an abstract
concept (Herman, Pack, & Wood, 1994; Thomas &
Boyd, 1973; Thomas & Noble, 1988; Thompson, 1995;
Zentall & Hogan, 1975). The fact that all dolphins
sustained a significant performance equal to training per-
formance over all transfer stimuli presented until the end
of the study, provides more evidence that the dolphins
learned the go/no-go task at a conceptual level.
5.2.2 | Electroreception
After the dolphins in this study had shown that they can
reliably apply the go/no-go rule to acoustic, visual, and
tactile stimuli, a perfect starting point was given for the
tests on electroreception. Accordingly, all four animals
responded just as spontaneously with “go”when they
were presented with a weak electric field of 1.5 mV cm
1
for the first time as in the preceding tasks. Although, we
could not determine a detection threshold for the
bottlenose dolphins in this study, the spontaneous and
reliable response of three dolphins to decreasing stimulus
amplitudes as low as 0.5 mV cm
1
(Figure 6) indicates
that, like Sotalia, the animals can perceive significantly
weaker field strengths. Also, our data on vibrissal crypt
structure and innervation density suggests an electro-
receptive sensitivity in Tursiops similar to that found in
Sotalia. This was confirmed in a follow-up study in which
detection thresholds were determined for two bottlenose
dolphins (Hüttner et al. in prep.).
It would be basically conceivable that the dolphins in
this study perceived the electric stimuli via different skin
areas than the vibrissal crypts. Although we were not
able to carry out a control experiment in this regard, the
exclusion test in the study by Czech-Damal et al. (2012)
showed beyond any doubt that the vibrissal crypts are the
electroreceptive units. The similarity in structure and
innervation of the vibrissal crypts of both species suggests
that this is also the case with the bottlenose dolphin.
The ability to sense weak bioelectric fields can aid ben-
thic foraging (Czech-Damal et al., 2012; Dehnhardt
et al., 2020), a foraging strategy also described for
bottlenose dolphins as crater-feeding (Rossbach &
Herzing, 1997). Especially in bottlenose dolphins that dis-
play a variety of different foraging behaviors associated
with specific habitats or ecological niches, electroreception
might increase foraging success when foraging on bottom-
dwelling prey (Hoese, 1971; Lewis & Schroeder, 2003).
While from a distance vision and echolocation may be
used for a first localization of potential prey, elec-
troreception provides important sensory information as
soon as a dolphin gets close to prey fish buried in the sedi-
ment. Prey induced electric fields are strongest around the
head and gills of a fish and have been recorded to be as
high as 0.5 mV cm
1
(Kalmijn, 1974). Thus, based on the
detection threshold of the Guiana dolphin determined by
Czech-Damal et al. (2012) and our hypothesis that
bottlenose dolphins perceive electric DC fields well below
0.5 mV cm
1
, the electroreceptive abilities of these two
species appear well suited to detect prey-generated electric
fields. However, the ability to find prey is not only a func-
tion of the absolute sensitivity of the system, but is also
based on other characteristics, in particular its spatial reso-
lution. As spatial resolution can be considered as a func-
tion of the number of separated electroreceptors it should
be lower in dolphins as compared to the platypus with its
thousands of electroreceptive units on the bill, indicating
that electroreception is part of a synergy of different
modalities when a dolphin searches for benthic prey
(Dehnhardt et al., 2020).
So far, there is only little information about an elec-
tric sense in cetaceans. While beside the present study
experimental proof is only available for S. guianensis
(Czech-Damal et al., 2012), a marine dolphin from the
coasts of Central and South America, Dehnhardt
et al. (2020) discussed indirect evidence for potential elec-
troreception in Kogia breviceps and Physeter macro-
cephalus, both of which are known for benthic feeding
strategies (Oelschläger, Ridgway, & Knauth, 2010). In
Kogia, vibrissal crypt mouths are closed and thus are
hard to be seen as long as the head is above the water
surface (C. Marshall, personal communication). Only as
soon as the animal submerges, vibrissal crypts appear to
open widely, which is an interesting analogy to the ele-
ctroreceptors in platypus (Dehnhardt & Mauck, 2008).
Sperm whales have often become entangled in submarine
telecommunication cables, which rarely happened after
the cables were redesigned and as a result no longer emit-
ted electric fields (Wood & Carter, 2008). Although the
present study establishes passive electroreception as a
sensory modality in T. truncatus, further experimental
and field studies on various toothed whale species are
necessary in order to better understand the system and
the benefits the animals can derive from this sensory
ability.
ACKNOWLEDGEMENTS
We thank Nuremberg Zoo, Germany for making their
facility available for this research. The authors would also
like to thank all colleagues at Nuremberg Zoo, including
the animal care, technical, veterinary, and management
staff for their support. Above all, many thanks to Armin
Fritz and all the trainers at the dolphinarium for their
ideas and support during the realization of this project.
HÜTTNER ET AL.605
Thanks to Gertrud Klauer for her valuable input during
the histological analysis and the entire team of the
Marine Science Center, Rostock for support. Open access
funding enabled and organized by Projekt DEAL.
AUTHOR CONTRIBUTIONS
Tim Hüttner: Conceptualization (lead); Data curation
(lead); Formal analysis (lead); Investigation (lead); Meth-
odology (lead); Validation (lead); Visualization (lead);
Writing-original draft (lead); Writing-review & editing
(lead). Lorenzo von Fersen: Funding acquisition
(equal); resources (equal); supervision (equal); writing –
review and editing (supporting). Lars Miersch: Method-
ology (equal); software (equal). Nicole Czech: Formal
analysis (equal); investigation (equal); methodology
(equal); writing –original draft (equal). Guido
Dehnhardt: Funding acquisition (equal); project admin-
istration (lead); supervision (equal); writing –review and
editing (equal).
ORCID
Tim Hüttner https://orcid.org/0000-0001-7179-2728
REFERENCES
Au, W. W. L. (1993). The sonar of dolphins. New York, NY: Springer
New York.
Bauer, G. B., Gaspard, J. C., III, Colbert, D. E., Leach, J. B.,
Stamper, S. A., Mann, D., & Reep, R. (2012). Tactile discrimina-
tion of textures by Florida manatees (Trichechus manatus
latirostris). Marine Mammal Science,28, E456–E471.
Brown, W. R., Geraci, J. R., Hicks, B. D., St. Aubin, D. J., &
Schroeder, J. P. (1983). Epidermal cell proliferation in the
bottlenose dolphin (Tursiops truncatus). Canadian Journal of
Zoology,61, 1587–1590.
Castro, L., & Wasserman, E. A. (2017). Perceptual and abstract cate-
gory learning in pigeons. In H. Cohen & C. Lefebvre (Eds.),
Handbook of categorization in cognitive science (2nd ed.,
pp. 709–732). Amsterdam, The Netherlands: Elsevier.
Chirighin, L. (1987). Mother-calf relationships and calf develop-
ment in the captive bottlenose dolphin (Tursiops truncatus).
Aquatic Mammals,13,5–15.
Cockcroft, V. G., & Ross, G. (1990). Observations on the early devel-
opment of a captive bottlenose dolphin calf. In S. Leatherwood
(Ed.), The bottlenose dolphin (pp. 461–478). San Diego, CA:
Academic Press.
Cozzi, B., Huggenberger, S., & Oelschläger, H. (2017). Chapter 5 -
Head and senses. In B. Cozzi, S. Huggenberger, & H. Oelschläger
(Eds.), Anatomy of dolphins: Insights into body structure and func-
tion (pp. 133–196). Amsterdam, The Netherlands: Elsevier.
Czech-Damal, N. U., Liebschner, A., Miersch, L., Klauer, G.,
Hanke, F. D., Marshall, C. D., …Hanke, W. (2012). Ele-
ctroreception in the Guiana dolphin (Sotalia guianensis). Pro-
ceedings of the Royal Society of London B: Biological Sciences,
279, 663–668.
Daniel, T. A., Goodman, A. M., Thompkins, A. M., Forloines, M. R.,
Lazarowski, L., & Katz, J. S. (2016). Generalization cannot
explain abstract-concept learning. In M. C. Olmstead (Ed.), Ani-
mal cognition: Principles, evolution and development (pp. 131–
145). New York, NY: Nova Science Publishers Inc.
Dehnhardt, G., & Dücker, G. (1996). Tactual discrimination of size
and shape by a California Sea lion (Zalophus californianus).
Animal Learning & Behavior,24, 366–374.
Dehnhardt, G., & Hanke, F. D. (2018). Whiskers. In B. G. Würsig,
J. G. M. Thewissen, & K. M. Kovacs (Eds.), Encyclopedia of marine
mammals (pp. 1074–1077). London, England: Academic Press.
Dehnhardt, G., & Kaminski, A. (1995). Sensitivity of the mystacial
vibrissae of harbour seals (Phoca vitulina) for size differences of
actively touched objects. Journal of Experimental Biology,198,
2317–2323.
Dehnhardt, G., & Mauck, B. (2008). Mechanoreception in secondar-
ily aquatic vertebrates. In J. G. M. Thewissen & S. Nummela
(Eds.), Sensory evolution on the threshold: Adaptations in sec-
ondarily aquatic vertebrates (pp. 294–314). Berkeley, CA: Uni-
versity of California Press.
Dehnhardt, G., Mauck, B., Hanke, W., & Bleckmann, H. (2001).
Hydrodynamic trail-following in harbor seals (Phoca vitulina).
Science,293, 102–104.
Dehnhardt, G., Miersch, L., Marshall, C. D., von Fersen, L., &
Hüttner, T. (2020). Passive electroreception in mammals. In B.
Fritzsch (Ed.), The senses: A comprehensive reference (2nd ed.,
pp. 385–392). Oxford, England: Elsevier.
Dral, A. D. G. (1975). Vision in Cetacea. The Journal of Zoo Animal
Medicine,6,17–21.
Eastcott, A., & Dickinson, T. (1987). Underwater observations of
the suckling and social behaviour of a new-born bottlenosed
dolphin (Tursiops truncatus). Aquatic Mammals,13,51–56.
Eberle, A., & Regan, G. T. (1998). The hair of the bottlenose dolphin
revisited. The Journal of the Alabama Academy of Science,
69, 47.
Gardner, S. C., & Varanasi, U. (2003). Isovaleric acid accumulation
in odontocete melon during development. Naturwissenschaften,
90, 528–531.
Gaspard, J. C., III, Bauer, G. B., Mann, D. A., Boerner, K.,
Denum, L., Frances, C., & Reep, R. L. (2017). Detection of
hydrodynamic stimuli by the postcranial body of Florida mana-
tees (Trichechus manatus latirostris). Journal of Comparative
Physiology. A,203, 111–120.
Gaspard, J. C., III, Bauer, G. B., Reep, R. L., Dziuk, K., Read, L., &
Mann, D. A. (2013). Detection of hydrodynamic stimuli by the
Florida manatee (Trichechus manatus latirostris). Journal of
Comparative Physiology. A,199, 441–450.
Gellermann, L. W. (1933). Chance orders of alternating stimuli in
visual discrimination experiments. The Pedagogical Seminary
and Journal of Genetic Psychology,42, 206–208.
Gerussi, T., Graïc, J.-M., de Vreese, S., Grandis, A., Tagliavia, C., de
Silva, M., …Cozzi, B. (2020). The follicle-sinus complex of the
bottlenose dolphin (Tursiops truncatus). Functional anatomy
and possible evolutional significance of its somato-sensory
innervation. Journal of Anatomy,238, 942–955.
Grant, R., Wieskotten, S., Wengst, N., Prescott, T., & Dehnhardt, G.
(2013). Vibrissal touch sensing in the harbor seal (Phoca
vitulina): How do seals judge size? Journal of Comparative Phys-
iology. A,199, 521–533.
Gregory, J. E., Iggo, A., McIntyre, A. K., & Proske, U. (1987). Ele-
ctroreceptors in the platypus. Nature,326, 386–387.
606 HÜTTNER ET AL.
Halata, Z. (1975). The mechanoreceptors of the mammalian skin
ultrastructure and morphological classification. Advances in
Anatomy, Embryology, and Cell Biology,50,3–77.
Halata, Z., & Munger, B. L. (1980). Sensory nerve endings in rhesus
monkey sinus hairs. The Journal of Comparative Neurology,
192, 645–663.
Hanke, W., & Dehnhardt, G. (2013). Sensory biology of aquatic
mammals. Journal of Comparative Physiology. A,199, 417–420.
Hanke, W., Witte, M., Miersch, L., Brede, M., Oeffner, J.,
Michael, M., …Dehnhardt, G. (2010). Harbor seal vibrissa mor-
phology suppresses vortex-induced vibrations. Journal of Exper-
imental Biology,213, 2665–2672.
Herman, L. M., Hovancik, J. R., Gory, J. D., & Bradshaw, G. L.
(1989). Generalization of visual matching by a bottlenosed dol-
phin (Tursiops truncatus): Evidence for invariance of cognitive
performance with visual and auditory materials. Journal of
Experimental Psychology. Animal Behavior Processes,15,
124–136.
Herman, L. M., Pack, A. A., & Hoffmann-Kuhnt, M. (1998). Seeing
through sound: Dolphins (Tursiops truncatus) perceive the spa-
tial structure of objects through echolocation. Journal of Com-
parative Psychology,112, 292–305.
Herman, L. M., Pack, A. A., & Wood, A. M. (1994). Bottlenose dol-
phins can generalize rules and develop abstract concepts.
Marine Mammal Science,10,70–80.
Herrnstein, R. J. (1979). Acquisition, generalization, and discrimi-
nation reversal of a natural concept. Journal of Experimental
Psychology. Animal Behavior Processes,5, 116–129.
Hicks, B. D., St Aubin, D. J., Geraci, J. R., & Brown, W. R. (1985).
Epidermal growth in the bottlenose dolphin, Tursiops
truncatus.Journal of Investigative Dermatology,85,60–63.
Hoese, H. D. (1971). Dolphin feeding out of water in a salt marsh.
Journal of Mammalogy,52, 222–223.
Holt, M. M., & Schusterman, R. J. (2002). Seals, sequences, and sig-
nal detection. Marine Mammal Science,18, 994–998.
Japha, A. (1912). Die Haare der Waltiere. Zool. Jb. Abteilung für
Anatomie und Ontogenie,32,1–42.
Kalmijn, A. J. (1974). The detection of electric fields from inanimate
and animate sources other than Electric organs. In T. H. Bull-
ock, A. Fessard, P. H. Hartline, A. J. Kalmijn, P. Laurent, R. W.
Murray, et al. (Eds.), Electroreceptors and other specialized
receptors in lower Vertrebrates (Vol. 3, pp. 147–200). Berlin,
Germany: Springer Berlin Heidelberg.
Kaplan, J. D., Goodrich, S. Y., Melillo-Sweeting, K., & Reiss, D.
(2019). Behavioural laterality in foraging bottlenose dolphins
(Tursiops truncatus). Royal Society Open Science,6, 190929.
Katz, J. S., Wright, A. A., & Bodily, K. D. (2007). Issues in the com-
parative cognition of abstract-concept learning. Comparative
Cognition and Behavior Reviews,2,79–92.
Kenyon, C. A., Keeble, S., & Cronin, P. (1982). The role of perioral
sensation in nipple attachment by weanling rat pups. Develop-
mental Psychobiology,15, 409–421.
Kilian, A., von Fersen, L., & Güntürkün, O. (2000). Lateralization
of visuospatial processing in the bottlenose dolphin (Tursiops
truncatus). Behavioural Brain Research,116, 211–215.
Krüger, Y., Hanke, W., Miersch, L., & Dehnhardt, G. (2018). Detec-
tion and direction discrimination of single vortex rings by har-
bour seals (Phoca vitulina). The Journal of Experimental
Biology,221.
Lazareva, O. F., & Wasserman, E. A. (2018). Categories and con-
cepts in animals. In J. Byrne (Ed.), Learning and memory: A
comprehensive reference (2nd ed., pp. 111–139). Amsterdam,
The Netherlands: Elsevier.
Lewis, J. S., & Schroeder, W. W. (2003). Mud plume feeding, a
unique foraging behavior of the bottlenose dolphin in the Flor-
ida keys. Gulf of Mexico Science,21,92–97.
Ling, J. K. (1977). Vibrissae of marine mammals. In R. J. Harrison
(Ed.), Functional anatomy of marine mammals (pp. 387–415).
London, England: Academic Press.
Manger, P. R., Collins, R., & Pettigrew, J. D. (1998). The develop-
ment of the electroreceptors of the platypus (Ornithorhynchus
anatinus). Philosophical Transactions of the Royal Society, B:
Biological Sciences,353, 1171–1186.
Manger, P. R., & Pettigrew, J. D. (1996). Ultrastructure, number,
distribution and innervation of electroreceptors and mechano-
receptors in the bill skin of the platypus, Ornithorhynchus ana-
tinus.Brain, Behavior and Evolution,48,27–54.
Mann, J., & Sargeant, B. (2003). Like mother, like calf: The ontog-
eny of foraging traditions in wild Indian Ocean bottlenose dol-
phins (Tursiops sp.). In D. M. Fragaszy & S. Perry (Eds.), The
biology of traditions: Models and evidence (pp. 236–266). Cam-
bridge, England: Cambridge University Press.
Mann, J., & Smuts, B. (1999). Behavioral development in wild bottlenose
dolphin newborns (Tursiops sp.). Behaviour,136,529–566.
Mass, A. M., & Supin, A. Y. (2018). Vision. In B. G. Würsig, J. G. M.
Thewissen, & K. M. Kovacs (Eds.), Encyclopedia of marine
mammals (pp. 1035–1044). London, England: Academic Press.
Murray, R. W. (1974). The Ampullae of Lorenzini. In T. H. Bullock,
A. Fessard, P. H. Hartline, A. J. Kalmijn, P. Laurent, R. W.
Murray, et al. (Eds.), Electroreceptors and other specialized
receptors in lower vertrebrates (Vol. 3, pp. 125–146). Berlin,
Germany: Springer Berlin Heidelberg.
Nachtigall, P. E. (1986). Vision, audition, and chemoreception in
dolphins and other marine mammals. In R. J. Schusterman,
J. A. Thomas, & F. G. Wood (Eds.), Dolphin cognition and
behavior: A comparative approach (pp. 79–113). New York, NY:
Psychology Press.
Nachtigall, P. E., & Patterson, S. A. (1980). Echolocation sameness-
difference discrimination by the Atlantic bottlenose dolphin
(Tursiops truncatus). The Journal of the Acoustical Society of
America,68, S98–S98.
Norris, K. S., Prescott, J. H., Asa-Dorian, P. V., & Perkins, P. (1961).
An experimental demonstration of Echo-location behavior in
the porpoise, Tursiops truncatus (Montagu). The Biological Bul-
letin,120, 163–176.
Oelschläger, H. H. A., Ridgway, S. H., & Knauth, M. (2010). Cetacean
brain evolution: Dwarf sperm whale (Kogia sima) and common
dolphin (Delphinus delphis) - An investigation with high-
resolution 3D MRI. Brain, Behavior and Evolution,75,33–62.
Pack, A. A., & Herman, L. M. (1995). Sensory integration in the
bottlenosed dolphin: Immediate recognition of complex shapes
across the senses of echolocation and vision. The Journal of the
Acoustical Society of America,98, 722–733.
Pack, A. A., Herman, L. M., & Hoffmann-Kuhnt, M. (2004). Dol-
phin echolocation shape perception: From sound to object. In
J.A.Thomas,C.F.Moss,&M.Vater(Eds.),Echolocation in
bats and dolphins (pp. 288–308). Chicago, IL: University of
Chicago Press.
HÜTTNER ET AL.607
Pack, A. A., Herman, L. M., Hoffmann-Kuhnt, M., &
Branstetter, B. K. (2002). The object behind the echo: Dolphins
(Tursiops truncatus) perceive object shape globally through
echolocation. Behavioural Processes,58,1–26.
Pack, A. A., Herman, L. M., & Roitblat, H. L. (1991). Generalization of
visual matching and delayed matching by a California Sea lion
(Zalophus californianus). Animal Learning & Behavior,19,37–48.
Reid, K., Mann, J., Weiner, J. R., & Hecker, N. (1995). Infant develop-
ment in two aquarium bottlenose dolphins. Zoo Biology,14,
135–147.
Reiss, D. (1988). Observations on the development of echolocation
in young bottlenose dolphins. In P. E. Nachtigall & P. W. B.
Moore (Eds.), Animal sonar: Processes and performance
(pp. 121–127). Boston, MA: Springer.
Rice, F. L., Fundin, B. T., Arvidsson, J., Aldskogius, H., &
Johansson, O. (1997). Comprehensive immunofluorescence and
lectin binding analysis of vibrissal follicle sinus complex inner-
vation in the mystacial pad of the rat. The Journal of Compara-
tive Neurology,385, 149–184.
Rice, F. L., Mance, A., & Munger, B. L. (1986). A comparative light
microscopic analysis of the sensory innervation of the mystacial
pad. I. Innervation of vibrissal follicle-sinus complexes. The
Journal of Comparative Neurology,252, 154–174.
Roitblat, H. L., & von Fersen, L. (1992). Comparative cognition:
Representations and processes in learning and memory.
Annual Review of Psychology,43, 671–710.
Rossbach, K. A., & Herzing, D. L. (1997). Underwater observations
of benthic-feeding bottlenose dolphins (Tursiops truncatus)
near Grand Bahama Island, Bahamas. Marine Mammal Sci-
ence,13, 498–504.
Rossi-Santos, M. R., & Wedekin, L. L. (2006). Evidence of bottom
contact behavior by estuarine dolphins (Sotalia guianensis)on
the eastern coast of Brazil. Aquatic Mammals,32, 140–144.
Scholtyssek, C., Kelber, A., & Dehnhardt, G. (2015). Why do seals
have cones? Behavioural evidence for colour-blindness in har-
bour seals. Animal Cognition,18, 551–560.
St. Aubin, D. J., Smith, T. G., & Geraci, J. R. (1990). Seasonal epi-
dermal molt in beluga whales, Delphinapterus leucas.Canadian
Journal of Zoology,68, 359–367.
Tavolga, M. C., & Essapian, F. S. (1957). The behavior of the bottle-
nosed dolphin (Tursiops truncatus): Mating, pregnancy, parturi-
tion and mother-infant behavior. Zoologica; Scientific Contribu-
tions of the New York Zoological Society,42,11–31.
Thomas, R. K., & Boyd, M. G. (1973). A comparison of Cebus
albifrons and Saimiri sciureus on oddity performance. Animal
Learning & Behavior,1, 151–153.
Thomas, R. K., & Noble, L. M. (1988). Visual and olfactory oddity
learning in rats: What evidence is necessary to show conceptual
behavior? Animal Learning & Behavior,16, 157–163.
Thompson, R. K., & Herman, L. M. (1975). Underwater frequency
discrimination in the bottlenosed dolphin (1-140 kHz) and the
human (1-8 kHz). The Journal of the Acoustical Society of Amer-
ica,57, 943–948.
Thompson, R. K. R. (1995). Natural and relational concepts in ani-
mals. In H. L. Roitblat & J.-A. Meyer (Eds.), Comparative
approaches to cognitive science (pp. 175–224). Cambridge, MA:
The MIT Press.
Torres, L. G. (2017). A sense of scale: Foraging cetaceans' use of
scale-dependent multimodal sensory systems. Marine Mammal
Science,33, 1170–1193.
von Fersen, L., & Delius, J. D. (2000). Acquired equivalences
between auditory stimuli in dolphins (Tursiops truncatus). Ani-
mal Cognition,3,79–83.
Vonk, J., & Jett, S. E. (2018). “Bear-ly”learning: Limits of abstraction
in black bear cognition. Animal Behavior and Cognition,5,68–78.
Vonk, J., Jett, S. E., & Mosteller, K. W. (2012). Concept formation
in American black bears, Ursus americanus.Animal Behaviour,
84, 953–964.
Wieskotten, S., Dehnhardt, G., Mauck, B., Miersch, L., &
Hanke, W. (2010). Hydrodynamic determination of the moving
direction of an artificial fin by a harbour seal (Phoca vitulina).
Journal of Experimental Biology,213, 2194–2200.
Wilkens, L. A., & Hofmann, M. H. (2005). Behavior of animals with
passive, low-frequency electrosensory systems. In T. H. Bullock,
C. D. Hopkins, A. N. Popper, & R. R. Fay (Eds.), Ele-
ctroreception (Vol. 21, pp. 229–263). New York, NY: Springer.
Wood, M. P., & Carter, L. (2008). Whale entanglements with sub-
marine telecommunication cables. IEEE Journal of Oceanic
Engineering,33, 445–450.
Wright, A. A., & Katz, J. S. (2006). Mechanisms of same/different
concept learning in primates and avians. Behavioural Processes,
72, 234–254.
Wright, A. A., & Katz, J. S. (2007). Generalization hypothesis of
abstract-concept learning: Learning strategies and related issues
in Macaca mulatta,Cebus apella, and Columba livia.Journal of
Comparative Psychology,121, 387–397.
Xitco, M. J., Gory, J. D., & Kuczaj, S. A., II. (2004). Dolphin pointing
is linked to the attentional behavior of a receiver. Animal Cog-
nition,7, 231–238.
Zentall, T. R., Galizio, M., & Critchfied, T. S. (2002). Categorization,
concept learning, and behavior analysis: An introduction. Jour-
nal of the Experimental Analysis of Behavior,78, 237–248.
Zentall, T. R., & Hogan, D. E. (1975). Concept learning in the
pigeon: Transfer to new matching and nonmatching stimuli.
The American Journal of Psychology,88, 233–244.
How to cite this article: Hüttner, T., von Fersen,
L., Miersch, L., Czech, N. U., & Dehnhardt, G.
(2022). Behavioral and anatomical evidence for
electroreception in the bottlenose dolphin
(Tursiops truncatus). The Anatomical Record,305
(3), 592–608. https://doi.org/10.1002/ar.24773
608 HÜTTNER ET AL.
Content uploaded by Tim Hüttner
Author content
All content in this area was uploaded by Tim Hüttner on Sep 25, 2021
Content may be subject to copyright.