84
Hans Winkler
Morton (1986, 1998) proposed that birds have to learn these signals and
familiarize themselves with the songs of territorial neighbors in order to be able to
compare the incoming more or less degraded signal with some internal
representation of the original one. Alternatively, they also may use general
distance-dependent features for ranging and would not need to be familiar with the
songs (Morton and Derrickson 1996; Wiley and Godard 1996; Naguib 1997a).
3.2 Signal Design and Broadcasting Tactics
The ultimate goal of research on the ecology of communication is to understand to
what degree ecological factors contribute to the diversity of avian acoustical
signals and what portion of the variance in their features they explain.
Environmental influences on sound transmission, the requirements for signal
reception and processing by the birds' sensory and cognitive systems, and the
number of alternatives that must be coded by the signal set have to be respected if
one wishes to come up with design rules for a set of signals (Morton 1980; Endler
1993; Bradbury and Vehrencamp 1998). Signal range and locatability are the
ecologically most relevant aspects when discussing the specific design principles
(Bradbury and Vehrencamp 1998).
The range of signals should be maximal for mate attraction, large to moderate
for need signals, moderate for territorial defence, moderate to short for alarm calls,
and short for courtship and other intimate signals (Bradbury and Vehrencamp
1998).
The range of territorial signals should increase with territory size and its "active
space", that distance from the sender over which the signal can be detected by
potential receiver's (Marten and Marler 1977; Brenowitz 1982), should at least
include the territories of nearby neighbors (Bradbury and Vehrencamp 1998). If
the sender does not care about the receiver's espionage or even has some
interest to cooperate in that respect, it should mix degradable elements into its
song (Sorjonen 1983), while in the former case a greater uniformity in song
element features is predicted. To sum it up, it can be said that optimizing
transmission distance is not necessarily equivalent to maximizing it (e.g., Lemon
et al. 1981).
The following design principles can be deduced from present knowledge on
sound transmission, if range maximising is assumed as in the acoustic adaptation
hypothesis (Morton 1975; see also Romer, this Vol.): in terms of frequency,
signals should be low pitched if the sender is very close to the ground. If signal-tonoise ratio is low, energy should be concentrated in a narrow frequency band and
signal elements should become louder and longer (so that receivers can integrate
over longer periods). Acoustic signals intended for close interactions only should
be of low amplitude, unstructured, and involve frequencies that are quickly
attenuated.
Hans Winkler
Morton (1986, 1998) proposed that birds have to learn these signals and
familiarize themselves with the songs of territorial neighbors in order to be able to
compare the incoming more or less degraded signal with some internal
representation of the original one. Alternatively, they also may use general
distance-dependent features for ranging and would not need to be familiar with the
songs (Morton and Derrickson 1996; Wiley and Godard 1996; Naguib 1997a).
3.2 Signal Design and Broadcasting Tactics
The ultimate goal of research on the ecology of communication is to understand to
what degree ecological factors contribute to the diversity of avian acoustical
signals and what portion of the variance in their features they explain.
Environmental influences on sound transmission, the requirements for signal
reception and processing by the birds' sensory and cognitive systems, and the
number of alternatives that must be coded by the signal set have to be respected if
one wishes to come up with design rules for a set of signals (Morton 1980; Endler
1993; Bradbury and Vehrencamp 1998). Signal range and locatability are the
ecologically most relevant aspects when discussing the specific design principles
(Bradbury and Vehrencamp 1998).
The range of signals should be maximal for mate attraction, large to moderate
for need signals, moderate for territorial defence, moderate to short for alarm calls,
and short for courtship and other intimate signals (Bradbury and Vehrencamp
1998).
The range of territorial signals should increase with territory size and its "active
space", that distance from the sender over which the signal can be detected by
potential receiver's (Marten and Marler 1977; Brenowitz 1982), should at least
include the territories of nearby neighbors (Bradbury and Vehrencamp 1998). If
the sender does not care about the receiver's espionage or even has some
interest to cooperate in that respect, it should mix degradable elements into its
song (Sorjonen 1983), while in the former case a greater uniformity in song
element features is predicted. To sum it up, it can be said that optimizing
transmission distance is not necessarily equivalent to maximizing it (e.g., Lemon
et al. 1981).
The following design principles can be deduced from present knowledge on
sound transmission, if range maximising is assumed as in the acoustic adaptation
hypothesis (Morton 1975; see also Romer, this Vol.): in terms of frequency,
signals should be low pitched if the sender is very close to the ground. If signal-tonoise ratio is low, energy should be concentrated in a narrow frequency band and
signal elements should become louder and longer (so that receivers can integrate
over longer periods). Acoustic signals intended for close interactions only should
be of low amplitude, unstructured, and involve frequencies that are quickly
attenuated.