Chapter 6: Communication
6.1 Communication Occurs When A Specialized Signal From One
Individual Influences The Behaviour of Another
1. Signal: an individual that produces a signal
2. Signal Receiver: an individual that detects a signal
3. Communication: the purpose in which specialised signal produced by one
individuals affects the behaviour of another.
4. Signal: a packet of energy or matter generated from a display or action of a
signaller that travels to a receiver.
HONEYBEES AND THE WAGGLE DANCE
5. When an individual honeybee scout finds a rich food source, it flies back to the
hive and recruits others to help exploit the food. Such communication allows the
colony to rapidly exploit the food resource before competitors do.
6. Von Frish manipulated the location of a distant food source and observed the
behaviour of both scouts and new recruits.
1. Waggle Dance: behaviour performed by a honeybee scout that recruits
workers to a food source.
2. During the waggle dance, the scout moves in figureeight pattern on a vertical
wall of the honeycomb. During the linear movement of the dance, the scout
vigorously wags it body, and the duration of the wagging, von Frish argued,
indicated the distance to the food.
3. Subsequent work indicated that every 75 msec of waggling translates into a
distance of approximately 100m from the hive. The waggle dance also
describes the direction of the food source, relative to an imaginary line that
runs from the hive to the sun.
1. For instance, if the sun is on the horizon (as at dawn) and the scout’s
linear movement is 30 left of vertical, then the food is 30 left of the sun.
4. Von Frisch recognized that odour is used to locate nearby food resources, but
he believed that the waggle dance provides the crucial signal, especially for
food located larger distances away distances more than, say, 500m.
1. However most recruits take much longer to fly to the food source than
would be expected based on a straightline route.
7. Wenner and Johnson proposed that the bees doing the waggle dance present the
odour of the food source, not a precise direction and distance; this behaviour then
stimulates recruits to search for food on their own.
1. They suggested that new recruits fly downwind a few hundred meters before
starting their search and use olfaction as a method of locating food by flying
upwind, in a zigzag pattern, back toward the hive.
1. Many insects that rely on only odour cues to find food do fly back and
forth into prevailing wind.
ODOUR OR THE DANCE IN BEES
8. One way to find out if the waggle dance or odour is important is to track the
actual movements of new recruits. Riley and colleagues examined the flight paths of recruits that viewed the waggle dance from a hive placed in a large, flat,
1. The field contained few natural sources of nectar or pollen, so the only food
available was located at a feeding station. The station was placed 200m due
east of the hive and contained 0.21M sucrose solutions with no artificial
2. The researchers recorded wind speed and direction at 10second intervals
from four locations surrounding the field and at the hive to determine their
effect on the flight of the bees.
3. Bees were required to move through a clear plastic tube to enter and leave the
hive, which facilitated bee capture.
1. This setup allowed the researcher to mark most of the bees in the colony
with small numbered tags on their backs. They could then record the
identity and movement of bees into and out of the colony.
4. At the start of the experiment, individual scouts were allowed to visit the
feeder and their identity was recorded. Their waggle dance indicated that
food was located due east of the hive. The researchers captured recruits who
had never previously visited the feeding station and placed small
transponders on their backs that transmitted continuous information about
1. These individuals were released either at the hive or at one of three
locations 200 to 250m southwest of the hive, and their movements were
5. Most of the recruits released at the hive almost immediately began flying east
toward the food station. However, only two of the 9 recruits actually found the
6. After 200m, most individuals began to follow a more circuitous flight path,
which the researchers interpreted as searching behaviour for the food. In
addition, the 17 bees transported 200250m southwest of the hive and then
released tended to fly due east for 200m before also adopting a circuitous
7. The wind data indicated that no odour from the feeding station was available
to the bees at their release sites. There was no evidence that individuals flew
downwind and then returned in a zigzag pattern to find the food. Instead, the
bees had to fly across the prevailing wind to maintain a due east heading.
8. The flight paths of new recruits confirmed von Frisch’s hypothesis that the
waggle dance appears to signal the distance and direction of the food.
1. In addition, the data demonstrates that recruits can travel to the area of a
novel food source without odour cues. However, they also show that the
dance signal is not sufficient to allow a recruit to locate a food source
2. Additional work has demonstrated that new recruits use odours to
pinpoint the exact location of the food. In sum, the waggle dance in a behavioural signal that allows individuals to travel to the area of a distant
food source, but bees then use local odours cues to find the food itself.
AUDITORY SIGNALS: ALARM CALLS
9. Alarm calls: unique vocalizations produced by social animals when a predator is
10. Cheney and Seyfarth found that the alarm calls of vervet monkey differ in how
they affect the receivers. Vervet monkeys are attacked by a variety of predators,
including leopards, eagles, and snakes; when a vervet spots one of these animals,
it gives an alarm call that alerts others in its group.
1. However, individuals produce different alarm calls for different types of
predator. Each predator represents a threat that requires a different
1. For example, leopards attack from the ground. When a leopard is spotted,
a “bark” call is given, and vervets escape attack by moving up into a tree.
“Cough” calls are given when an eagle is sighted. Eagles attack from
above. Avoiding their attack requires moving down from treetops and into
dense bushes. Snakes, in contrast, often hide in dense grass. When a
python or cobra is observed, vervets give a “chutter” call that causes
individuals to stand erect and look into grass clumps.
TITMOUSE ALARM CALLS
11. Courter and Ritchison examined alarm calls in the tufted titmouse, a small
songbird. Titmouse produces an alarm call in response to a perched avian
1. Vocalization can be visualized in a sonogram, which characterizes
frequencies of sound as a function of time. The titmouse call is composed of
three basic notes, called, Z, A, and D, which are typically produced in
sequence. However, individuals often vary the number of D notes.
1. This call recruits other birds (both con and heterospecifics) to approach
and mob the predator. Mobbing behaviour, in which birds produce loud
vocalizations, often harasses a predator and can drive it away from an
2. Titmice cooccur with several potential avian predators. Small raptors like
Eastern screech owls, sharpshinned hawks, and Cooper’s hawks are known
as predators of titmice, while large raptors like great horned rarely attack
1. The size of a predator thus correlates negatively with risk of a titmouse. 12. Courter and Ritchison conducted a study to determine whether the alarm calls
differ in responses that correlate with the size and degree of threat posed by a
perched avian predator.
1. The experiment consisted of six treatments, each presenting a different
predator, and a control. At several sites, the researchers established a feeding
station that contained sunflower sees, a food readily consumed by titmice.
2. Lifelike models of six different birds were placed on platforms 1 m from the
feeding station; one control presented nothing on the platform. Three models
were highrisk (Eastern screech owl, sharpskinned hawk, and Cooper’s
hawk), two were lowrisk predators (great horned owl and redtailed hawk),
and one was a nonpredators control bird, a ruffed grouse.
3. They examined whether titmice would produce different alarm calls in
response to these different models.
1. As soon as the titmice were within 25m of the feeding station, the models
were uncovered. The researchers recorded titmice behaviour for six
minutes and recorded all alarm calls.
4. Alarm calling varied was higher in response to small, highrisk predators
compared to the number produced in response to large lowrisk predators and
5. The smaller the predator, the longer the mobbing response: mobbing
behaviour lasted for approximately 250 seconds for the smallbodied, high
risk predators, compared to only 150 seconds for the larger, low0risk
predators and less that 50 seconds for the controls.
1. These results indicate that titmice produce different alarm calls that lead
to differences in the behaviour of receivers, which correlate with varying
levels of threat. This finding illustrates the rapid speed and flexibility of
auditory signals in response to changes in the environment.
INFORMATION OR INFLUENCE?
13. We do not if the vervet alarm calls may not mean “leopard present” or “snake
present,” but rather “climb a tree” or “look down”
14. Rendall, Owen and Ryan contend that the assumption that signals encode specific
information is problematic for two reasons.
1. First, it can imply a languagelike meaning of communication that, as has
been challenging to document.
2. Second, it can encourage attempts to characterize the information encoded in
a signal, rather than focusing on factors that shape signal properties. These
researchers suggest that it is more productive to study signals as traits that
have evolved to influence the behaviour of others.
1. Not everyone agrees. Others argue that removing the concept of
information form the study of communication is unwarranted and this
discussion continues to stimulate further work.
1. What we can say is that signals influence behaviour as if they
contained information about the location of a food source, the threat
of a predator, or the phenotype of a signallers. 6.2 Signals Are Perceived By Sensory Systems and Influenced By The
15. Sensory receptors: nerve endings that respond to an internal or external
1. These receptors cells then transmit information along axons to the central
nervous system, where it is processed and an appropriate response in
16. Sensory systems vary greatly across taxa and have coevolved with many kinds of
signals, including chemical, visual, auditory, vibrational, and even electrical
17. Chemoreceptors: a sensory receptor that detects chemical stimuli.
18. Chemical signals are compounds secreted into the environment and detected by
odourbinding proteins in sensory structures such as the antennae of
invertebrates and the olfactory system of vertebrates.
19. Olfaction: the detection of airborne chemical stimuli.
20. Gustation: the detection of dissolved chemicals, often within the mouth.
21. Olfactory signals can be transmitted in water or air, are long lasting, and can
travel long distance. They can also be deposited on a substrate, as illustrated by
the territorial scent marking of mammals and the foraging trails of ants.
1. Because these signals can be produced in variable amounts, animals can vary
their strength. In addition, chemical signals can travel around environmental
barriers such as dense vegetation. Once transmitted, however, chemical
signals cannot be modified.
22. Odorant: gaseous compound that is perceived as odorous.
23. Volatile: a chemical that can evaporate that is, become gaseous.
24. Pheromones: volatile organic compounds that are speciesspecific and affect the
behaviour of another individual of the same species.
25. These pheromones are often composed of speciesspecific hydrocarbons found in
the cuticle. In ants, sentries at the nest entrance often use tactile cues such as
antennal contact to determine whether individuals should be admitted to the nest
1. However, this type of signal means that an intruder could get to the nest
entrance and perhaps even in the nest before being detected.
26. Brandstaetter and colleagues wondered whether cuticular hydrocarbons might be
volatile; perhaps ants could detect them at some distance, instead of using tactile
cues. The researcher collected the hydrocarbon signal from the postpharyngeal gland of workers from colonies of the Florida carpenter ant and dissolved it in a
1. They then placed the signal close to an ant, but at a distance from which it
could not touch the signal. They observed that when the signal came from a
nonnest mate, the ants responded with more aggressive behaviours, just as
they do when they receive tactile cues. The researchers concluded that there
does seem to be a volatile component to this signal.
27. Photoreceptors: a specialized neuron that is sensitive to light.
28. Rod: a type of photoreceptor, sensitive to dim light levels.
29. Cones: a type of photoreceptor for colour vision under bright light conditions.
1. The number and ratio of rods and cones vary across species and depend on
the light levels of the environment. This variation leads to a wide diversity of
wavelengths of light that are detected across species.
30. Visual signals can be detected quickly, allowing a rapid response. In clear skies
or clear water, they can also be visible at fairly large distances. However, their
perception requires sufficient light levels. Thus, in deep ocean or murky water
and at night, conditions favour the evolution of signals detected by other sensory
1. In addition, obstacles can block visual signals; so again, they do not transmit
well in dense media like murky water and think vegetation. Therefore, diurnal
organisms that interact over relatively short distances in open habitats most
often use visual signals.
HABITAT STRUCTURE AND VISUAL SIGNALS IN BIRDS
31. Marchetti examined the evolution of bright plumage in Old World, warblers that
breed in the forests of Kashmir. These birds live in a variety of forest habitats that
vary in light intensity.
1. She examined several species in the genus Phylloscopus, as well as the closely
related gold crest. These small birds are mostly dull green but differ in the
number of bright colour patches (from zero to four) that occur on their wings,
crown, rump, and tail.
2. Males compete for territories using display behaviour such as wing flashes,
head tripping, and rump flashing. She found that these behaviours correspond
well to the presence of different colour patches.
1. For example, only birds with a crown stripe display headdripping
behaviour. This movement allows the birds to make themselves
temporarily more conspicuous.
3. She hypothesized that colour patterns were morphological signals used in
competition for territories and that brightness has influenced the
evolution of these signals. 1. She reasoned that because the amount of light influences visual signal
perception, birds living in darker environments, like coniferous forests,
should have more colour, patches than birds living in open, treeless areas.
Selection would favour the evolution of bright visual signals in darker
4. She recorded the species of birds present an their colour patterns at five
locations along an elevational transect. These locations varied in vegetation,
from an open, bright habitat above the tree line to relatively dark coniferous
forest. At each location, she recorded the brightness of the habitat based on
the habitat based on a still camera’s automatic shutter speed, measured while
holding the camera’s aperture and film constant.
5. Shutter speed is affected by light level and can be converted into footcandles
as a measure of illumination. She found a strong negative correlation
between habitat brightness and the number of colour patches on male birds,
6. Marchetti manipulated the colour patterns of one species, P. inornatus, which
has two wing bars. She captured males while they were establishing
territories before females arrived to breed and assigned them to one of three
groups. “Control” birds had their wing bars coloured with clear paint,
“reduced” birds had their wing bars made larger with yellow paint.
1. Marchetti then released these males and measured the effects of the
manipulation on their territory size.
7. Males who had their wings bars experimentally enlarged obtained the largest
territories, while males with reduced colour patches ended up with the
smallest territories. To discount the possibility that wing bar size rather than
brightness of the colours patch affects territory size.
1. She then released these males and measured the effects of the
manipulation on their territory size.
8. Males who had their win bars experimentally enlarged obtained the largest
territories, while males with reduced colour patches ended up with the
smallest territories. To discount the possibility that wing bar size rather than
brightness of the colour patch affects territory size, Marchetti conducted on
final experiment: she added a novel colour patch to the crown of males but
did not enlarge their wing bars. These males too obtained the largest
9. Marchetti concluded that males rely on colour patches as a visual signal in
defending territories. Selection appears to have favoured the evolution of
multiple colourful patches in darker environments as a way to enhance signal
32. Auditory receptors: specialized nerve cells that detect auditory signals. Two
properties facilitate the evolution of auditory signals. 1. First these signals can bypass obstacles: sound waves can travel around
objects. That means that auditory signals can sometimes be perceived by
receivers when visual signals cannot.
2. Second, auditory signals can be modified rapidly. They can be “turned on and
off” depending on conditions (say, when a predator is nearby), and they can
be produced at different amplitudes (decibel level) to overcome background
noise. This feature makes them more versatile than chemical signals.
1. For example, Cynx and colleagues recorded the song of individual zebra
finches while varying the amplitude of white noise played from a speaker.
3. They found that birds produced louder songs when the white noise was louder.
Auditory signals can be detected at great distance in both air and water, as
well as in habitats with different vegetation.
1. Because they can be turned off quickly, they provide excellent and
relatively safe longdistance communication. They can also be rapidly
modified in response to current conditions.
HABITAT STRUCTURE AND BOWERBIRD VOCAL SIGNALS
33. To be effective, signals must be able to be detected over background noise: they
must stand out from their environment, as was seen in the previously mentioned
research on zebra finch song production.
34. Auditory signals can also travel over long distances, but they degrade (lose
quality) and attenuate (lose intensity) as they travel; furthermore, the rates of
degradation and attenuation are affected by habitat structure.
1. For example, higher frequencies attenuate more rapidly in dense vegetation,
which should favour the use of lower frequencies, because selection should
favour individuals whose vocalizations are best transmitted through their
35. Nicholls and Goldizen tested this hypothesis by comparing advertisement calls of
satin bowerbirds. Satin bowerbirds live in a variety of habitats in eastern
Australia, ranging from dense rainforests to open woodlands. Males construct a
stick bower on the ground, from which they display to attract mates.
1. They also produce a loud call to attract females, and previous work noted
geographical differences in these calls. They wondered whether such
variation was de to differences in the vegetation structure across populations.
2. To test this prediction that call frequency should be lower in more densely
vegetated habitats, the researchers recorded male advertisement calls from 18
locations that varied in habitat and vegetation structure.
3. They used sonograms to characterize the average call structure at each site,
including the minimum frequency, maximum frequency, dominant (or loudest)
frequency, and call duration.
4. To characterize the vegetation, they counted the number of tree stem wider
than 5 cm at 1m above the ground that occurred along four 40m transects.
Bowerbird vocalizes at a height of a few meters off the ground, and stems of
this diameter will scatter sounds to attenuate a call. 5. Call structure varied across sites and was related to habitat type. Minimum
frequency and dominant frequency were negatively correlated with tree
densitythat is, they were lower n sites with more trees.
6. This finding appears to support the prediction that selection favours vocal
signals that minimize attenuation. One possibility was that morphological
differences across population, such as body size, could account for this
correlation, but after examining the data, the researchers ruled this
7. Another possibility is that the differences in calls are associated with genetic
differences among populations. However, data indicate that populations in
different habitats that produce different calls were not necessarily genetically
divergent populations often produced similar calls.
8. One final possibility concerns learning: juveniles may simply learn those
habitatspecific calls that they hear best.
36. These examples show how the environment can affect the evolution of signals.
Recall that the function of communication is to influence the behaviour of another
individual. As such, the evolution of signals is strongly influenced by the fitness
interests of both signaller and receiver.
5.3 Signals Can Accurately Indicate Signaler Phenotype And
37. For signals to evolve, they must increase the fitness of the signaler. Because
signals affect the behaviour of another individual, receivers will also affect
selection on signals.
38. In particular, divergent interests can lead to two outcomes.
1. In the first, the signal will evolve to become an accurate indicator of signaler
phenotype or the environment.
2. In the second, the signal will evolve to become an inaccurate indicator of
SIGNALS AS ACCURATE INDICATORS: THEORY
39. Three conditions favour the evolution of signals as accurate, or “honest”
1. First, the fitness interests of the signaler and the receiver may be similar. Both
parties will then benefit from an accurate relationship between the signal and
the signaler phenotype or environmental conditions.
2. Second, we can expect signals to be accurate indicators when they cannot be
1. For example, suppose that a signal is a function of body size. In contests
over resources or territories, males often produce vocal signals. The
frequency of a vocalization, or its pitch, is often negatively related to body
3. Finally, and perhaps most commonly, signals may be accurate indicators
because they are costly to produce or maintain.
1. For example, the long horns of beetles used in aggressive interactions,
the wide eyespan of stalkeyed flies, and the wattle size and colour oh
pheasants used to attract mates are physiologically costly to produce: only males with access to abundant resources possess the largest, most
extreme form of these traits.
APOSEMATIC COLOURATION IN FROGS
40. Aposematic colouration: brightly coloured morphology in a species that stands
out from the environment and is associated with noxious chemicals or poisons
that make them unpalatable or dangerous prey.
41. Saporito and colleagues conducted an experiment to investigate whether the
bright body colouration in the dendrobatid frog functions as an aposematic
1. Saporito’s team examined predation attempts on clay models molded to
resemble the toxic O. pumilio and nontoxic, brown leaflitter frog similar to
frogs of the genus Craugastor. Each model measured approximately 20mm in
lengththe average size of O. pumilio in the study region in Costa Rice.
2. To quantify predation, the researchers placed 800 frog models on eithe