The functional organization of anti-predator behaviour in the pied flycatcher: A study of avian visual perception

• Published in: Animal behaviour Vol. 23; no. 1; pp. 1,IN1,13,IN3,111 - 12,IN1,110,IN4,115
• Main Author: Curio, E.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.02.1975
• Subjects: Aggression; Animals; Animals, Newborn; Appetitive Behavior; Behavior, Animal; Biological Evolution; Birds; Color Perception; Escape Reaction; Eye; Habituation, Psychophysiologic; Humans; Humidity; Orientation; Predatory Behavior; Pressure; Temperature; Time Factors; Visual Perception; Vocalization, Animal
• ISSN: 0003-3472; 1095-8282
• DOI: 10.1016/0003-3472(75)90056-1

1. 1. The pied flycatcher ( Ficedula hypoleuca Pallas), a palaearctic songbird, is endangered in its breeding grounds by a number of highly diverse predators and, on occasion, by several avian nest competitors. The ecological impact of the various dangers upon the adult P.F. and/or its brood is broadly assessed and the defence methods described. 2. 2. In order to elucidate part of the stimuli which elicit two distinct anti-predator responses, mobbing and snarling attacks, enemy dummies were presented to over 2200 territorial subjects in the course of 12 years, utilizing three German populations (Berlin, Hessen, Brunswick) and one in Spain (La Granja). Comparison with mobbing responses to the live enemy give confidence in the dummy experiment as a suitable and valid method of approach. 3. 3. Mobbing movements and calls together with their characteristic interrelationships are described as a function of response strength RS. Phenomena as diverse as warming-up, after-response, and short-term fluctuations of the calling rate, all of which are a function of RS, can be causally explained by a unifying concept called ‘catch’. 4. 4. A number of external and internal factors tend to increase or diminish mobbing responsiveness. Opportunity to see live redbacked shrikes (= redbacks) ( Lanius collurio) neither detectably affected the response to shrikes nor to incomplete shrike patterns. Subjects when tested repeatedly with the same dummy stimulus tend to remain significantly at the same RS level. RS data from parents and their offspring are presented, the proper evaluation of which in terms of heritability has suffered so far from lack of suitable statistics. Evidence for spontaneity of mobbing derives from vacuum alarms of both wild and captive, hand-raised birds. 5. 5. Three similar-sized enemies, e.g. pigmy owl ( Glaucidium passerinum), male redback, and great spotted woodpecker ( Dendrocopus major) are recognized by various sign stimuli, the physical complexity and/or the degree of interaction of which increase(s) in that order. Interaction follows the principle of ‘stimulus dilation’ both in terms of RS as well as of the underlying aroused motivation M. The latter conclusion is permitted through the derivation of the functional relationship RS = f( M). Two simple formalisms are proposed that can account for ‘weighted summation’ of sign stimuli with stimulus dilation being a special case. 6. 6. Novelty can be dismissed as an explanation for the high-intensity mobbing of shrikes and owls. However, novel bird-shaped models of about shrike-size do elicit some alarm when properly oriented and minimally patterned, suggesting a new version of the ‘rarity principle’ of danger recognition. Recognition of the perched predator is at the same time specific enough not to be mistaken by harmless bird species resembling either enemy, and sufficiently unspecific as to cope with at least two shrike and several owl species. 7. 7. A varying fraction out of eighty-eight hand-raised, predator-naive birds responded, though somewhat sub-normally, when tested from 5 weeks up to nearly 3 years of age to live owls and shrikes. Live, novel control birds released some mobbing typical for them from yearling birds, but not at 8 weeks of age. It is concluded that recognition of owls, shrikes, and novel birds is innate (as defined here, see C.2.2). Whether predator-unrelated, ‘subsidiary’ experiental factors would contribute to the response predictability of birds growing up normally is doubtful as suggested by RS levels of hand-raised as compared to wild-caught subjects familiarized with captivity. The innate nature of recognition accords well with a number of observations, among which the wholesale failure to condition, by five different methods, and mobbing in both wild and captive birds to neutral and novel stimuli is the most conspicuous. 8. 8. The hypothesis of at least two perceptual channels decoding either owls or shrikes and subserving a common response mechanism received strong support from five pieces of evidence: (a) Sign stimuli of both predators do not effectively combine thus explaining at the same time the low releasing value of the redback female, (b) An identical change of both predators leads to a different effect in each case, (c) Habituation of the response to the shrike neither affects the response to the owl, nor does the latter lead to dishabituation. (d) The owl response is capable of raising the stimulus value of an otherwise ineffective owl model but not that of an otherwise ineffective shrike (or novel bird) model, (e) Factors associated with a foreign territory abolish almost entirely the response to shrikes (and novel birds), but not the owl response. Differential evaluation of identical sign stimuli, a range of distinctly different (low) releasing values, and, presumably, the susceptibility to habituation leads one to invoke a third channel in its own right that is tuned to novel birds. The functional independence of the three channels envisaged is underlined by the lack of intra-individual correlation of responses to two classes of objects. 9. 9. The adaptiveness of nest defence strategies in general is derived from the compromise achieved between protection of the brood and that of the defending adults. Owl-shrike recognition in particular is not only currently maintained but also has arisen by the pressures from owls and shrikes as evidenced by the precise fit between (a) stimulus configuration and corresponding IRM, and (b) perceptual capabilities and the geographical distribution of both enemies. In order to elucidate the way in which these exert selection pressure a more direct approach to the problem of survival value is needed. 10. 10. In an attempt at a synthesis, a coherent scheme is proposed which accounts for the results of both stimulus and response analysis. Its most prominent features are distinct though non-unitary, danger-specific perceptual channels which converge before or at the adaptively appropriate common response mechanism M. It is argued that not only recognition of at least several danger stimuli is innate (para. 7), but also the multi-channel organization of recognition itself. Factors that are apt to modify RS with the stimulus remaining constant (para. 4) either affect the channel activated or M (plus that channel?). The nature of the channels and evolutionary implications considering the ecological diversity of the impact by predators upon perceptual mechanisms of the prey are discussed. Finally, the maintenance of the IRM concept is advocated if used as a shorthand description of a rigorously defined stimulus-response correspondence and the operations performed to establish it.