Avian Visual Cognition


Evolution of the Avian Visual System

Scott Husband & Toru Shimizu
Department of Psychology, University of South Florida

Next Section:
Evolution of the Amniote Brian



This chapter provides an overview of the properties of the avian visual system and what selective pressures are thought to have operated on its evolution.  The evolutionary perspective is invaluable in approaching the questions of structure and function in avian vision.  Such a perspective provides insights on the capabilities and limitations of avian vision, and how different species utilize information to navigate, forage, mate, et cetera.  A brief review of amniote (reptiles, birds, and mammals) evolution is outlined, with descriptions of the many similarities in brain organization and basic visual pathways common to all amniotes. Such similarities include: the importance of the tectum as a visual structure for maintaining topographic representation of sensory space, and the possession of dorsal thalamic zones in both birds and mammals which receive ascending visual input from the retina either directly (lemnothalamic pathways) or indirectly (via the tectum, as in the collothalamic pathways).  Consideration of these similarities, however, must coincide with appreciation for obvious differences in brain structure between birds and mammals (i.e. dorsal ventricular ridge versus the six-layered isocortex of mammals).  The chapter includes general ideas regarding brain evolution, such as forebrain expansion and changes in the ratio of midbrain and forebrain structures.

Dinosaur and avian evolution are considered, along with a discussion of the two major competing hypotheses regarding avian evolution.  The two major competing hypotheses are the "basal archosaur hypothesis" (which proposes that the first birds descended directly from ancestral reptiles about 230 million years ago) and the "theropod dinosaur hypothesis" (which advocates a much later entry of birds, with derivation from dinosaurs some 100 million years after the time proposed by the basal archosaur hypothesis).  Debates regarding the origin of flight (arboreal vs cursorial theories) and the metabolism of the dinosaurs from which birds arose (ecto- vs endo-thermy) are also mentioned.

The avian retina contains interesting differences from that of primates.  Such variations include retinal morphology (e.g., avian double cones), more numerous photopigments, more horizontal and amacrine cells, richer intraretinal connections, and more complex ganglion cell response properties.  Comparisons of avian and primate retinae are discussed in terms of their differing evolutionary paths from common amniotic ancestry (e.g., periods of nocturnality during the reign of dinosaurs; Bottleneck theory).  Visual processing beyond the retina is discussed with comparisons between the collo- and lemno-thalamic visual pathways in amniote brains.  Issues related to lateral versus frontal eyes and binocular vision are addressed.  The development of flight undoubtedly had an important effect on the transition from ancient reptile to avian brains.  Useful clues regarding this are derived from comparisons of early flying reptiles like pterosaurs and the brains of modern birds in terms of optic lobe, cerebellar, and forebrain developments

Chapter Outline & Navigation

I.   Introduction 
          Some terms and concepts

II.  Evolution of the Amniote Brain 
Amniote evolution
          Amniote brains
          DVR and cortex
          Brain evolution

III. Dinosaurs Among Us?
The mystery of avian ancestry   
          The theropod dinosaur hypothesis
          The basal archosaur hypothesis 

IV. Evolution of Retinal Structures
Retinal morphology and processing
Avian retina
Comparing avian and primate retinae

V.  Taking Flight: Post Retinal Processing
Collo & lemno thalamic visual pathways
          The flying brain 
          The brains of pterosaurs and birds

VI. References

I. Introduction

This chapter provides a general review of the specializations and evolution of the avian visual system, focusing on the selective pressures which drove the development of its current form.  Since the vast majority of data is available from studies of the pigeon, our discussion will focus on this body of literature.  Other birds (e.g., owls, falcons, etc.) will be included as applicable to the particular topic.  Scientists interested in the visual perception and cognition of birds should find it useful to consider what drove the evolution of the avian visual system.  Consideration and application of such ideas should, at some level, guide investigations of visual search, concept formation, et cetera.  It is perhaps conceptually more tractable to think that our avian subjects must see the world (at least somewhat) as we do.  But, we must understand as best we can, armed with psychophysical, anatomical, cognitive, and other data, how the organism perceives and operates in its visual Umvelt (subjective world).  Familiarity with the process of visual system evolution in birds can assist in appreciating the biases and capabilities of avian subjects.  This is especially true in the design of experiments in the artificial situations of the laboratory and interpretation of data derived from such investigations. 

Birds are highly visually dependent organisms, possessing visual capabilities comparable (and in some cases, superior) to those of another visually-dependent vertebrate, the primates.  There are many basic similarities in the visual pathways of birds and mammals.  These commonalities in visual systems are rooted both in distant ancestral phylogenetic relationships and the operation of convergent evolution.  

Unique specializations of avian visual systems (as compared to primates) include the existence of double cones, different photopigment absorption spectra, the presence of oil droplets, centrifugal efferents to the retina, and an emphasis on the collothalamic (tectofugal) visual pathway.  Evidence from several sources indicates the avian visual system was driven by alterations of the basic reptilian plan, primarily in support of flight behaviors.  The development of the collothalamic pathways encouraged the development of the avian telencephalon as the site for more complex social and feeding behaviors. 

Some Terms and Concepts 

Darwin, in the Origin of Species (1859), called eyes "organs of extreme perfection."  Indeed, many of the opponents to Darwin's evolutionary theories argued that such a complex entity could not have arisen from accident, that they must have been divinely "designed", not driven by a tinkering process occurring over earth's history.  Darwin spent considerable time working out the dynamics of evolution, elucidating how natural selection could arrive at the "perfection" which naturalists of the day saw in many biological systems. 

The variety of organisms on earth was a considerable source of both inspiration and data for Darwin's ideas on how organisms adapted to their environment.  The vertebrates which rely on vision reflect this diversity in the many differences in their visual systems.  The photoreceptors, retinal structure, and other components of the visual system appear tailored to their unique lifestyles.  However, despite the array of ways in which organisms employ vision in their survival, there are remarkable similarities in the underlying plan of how these visual systems operate. 

The underlying similarities found in the diversity of organisms which employ vision are fundamentally driven by stimulus-organism interactions.  From the pineal eyespot under the skin of amphibians to capabilities of the primate eye, the evolution of visual functions are driven by the content of the photic stimulus (i.e., wave and particle properties of light) and the physical and biochemical constraints of living organisms.  This is true of all sensory systems, and is the fundamental principle on which perceptual systems evolve and the constraints under which they operate. 

Similarities in organisms' visual systems are also attributable to the common (albeit in some cases long removed) evolutionary ancestry shared by all vertebrates.  In particular, amniotes (reptiles, birds, and mammals) share some striking similarities in the organization of their visual pathways.  Ascertaining how much of these similarities in their visual systems derive from common ancestry is a difficult task.  The stem reptiles from which all amniotes derived (known as the order Captorhinida) became extinct about 300 million years ago (Carroll, 1988).  The study of modern birds, reptiles, and mammals can indicate which brain areas and visual behaviors are “holdovers” from ancestral reptilian characteristics.  Conversely, important specializations of the brain for vision can be discovered by comparing birds and primates, which generally have more sophisticated visual capabilities than most reptiles.  Therefore, traits shared by avian and primate visual systems, but not reptiles, may indicate specializations in the brain for their excellent visual capabilities.   Looking at both the similarities (driven by the stimulus and constrained by biology) and varieties (resulting from genetic mutations and natural selection pressures of the environment) of visual systems enables an appreciation of how visual systems originated and have changed across evolutionary time.  

Clues to the evolution of the avian brain and visual system must rely on relatively rare fossil finds and cranial endocasts.  One cannot perform experiments to look at the variables of natural selection operating across evolutionary time (although interesting simulations have been performed in the realm of computers and artificial life).  The reconstruction of phylogenetic relationships is difficult, since fossilized remains are either non-existent or yet to be discovered.  This is even more prominent in the evolution of birds, whose delicate bones do not usually withstand the passing of geologic time.  In addition, the soft tissues do not fossilize, so any reconstruction of the eye or brain must be based on cranial structure.  However, cranial endocasts can be formed naturally by the deposition and petrification of sediments in the brain cavity.  These remains provide some basis for insights into brain structures and their relative development in extinct organisms.  Other than fossil analysis, the approach of comparative anatomy offers clues to the origins and adaptations of the brain and visual system which led to modern forms.  

The conditions which existed in the brains of amniote ancestors are difficult to ascertain.  Keeping in mind the possible convergences due to evolution under similar environmental pressures, anaysis of the basic features common to all modern amniotes can be enlightening.  For example, some general features of reptilian brains (the closest relatives of modern birds) may indicate the condition found in ancestral avian forms.  Crocodilians, which appear to have undergone comparatively few major changes over the course of evolution, are a valuable species with which to compare birds.  Both crocodiles and birds evolved from the archosaurs, or ruling reptiles. Their phylogenetic relationship is determined from both fossil evidence and the reconstruction of their lineages through methods like cladistics.  Sets of shared, similar traits are analyzed and compared; those organisms which numerically possess the greatest number of shared characters are then considered more closely related (Butler & Hodos, 1996).  Organisms are grouped based on the number and kinds of shared traits they possess.  It is assumed in cladistics that two groups of animals which share a new or evolutionarily “derived” trait are more closely related to each other than either is to its common ancestor.  The common ancestor would be in possession of an older structure from which the derived trait evolved.  The identification of shared traits are represented in a tree-like diagram or cladogram.  Each branch reflects the emergence from an ancestral form of a group having new, derived characteristics not present in the earlier group.  Caution must be taken, however, not to fall into the scala naturae fallacy.  This view of evolution is one of a continuous, escalating progression from simpler "primitive"  forms to more complex and advanced ones (Butler & Hodos, 1996).  Phylogenetic scales, a kind of Aristotelian staircase from fish and amphibians, through reptiles and mammals, to the “pinnacle” of evolution, modern man, is simplistic, anthropocentric, and clearly false based on current scientific data.  No living reptile can constitute a “primitive” form from which birds derived, just as no modern monkey or ape can represent a direct evolutionary stage leading to humans.   Modern organisms are not primitive forms of others; all extant organisms have evolved to survive under their unique selective environmental pressures.  

In discussing traits of the ancestral amniote brain, and how it has subsequently changed through evolution, a distinction is made between homology and homoplasy.  Characteristics which appear similar in structure and/or function in two organisms may derive from the condition of the presumed ancestor of the two forms.  This is referred to as homology, which has been defined by Campbell and Hodos (1970): "(structures and characters) are homologous when they could, in principle, be traced back through a genealogical series to a stipulated common ancestral precursor irrespective of morphological similarity."

Aside from homology, one must keep in mind that convergence due to similar environmental demands over time can lead to remarkable similarities between two organisms.  Many basic similarities in vertebrate brains may reasonably be attributed to homology.  However, evolution may produce similar morphologies and other traits in organisms which do not share a close common ancestor from which the particular trait derived.  In these cases, the trait has risen independently in the two organisms, probably by virtue of their similar ecological niches and lifestyles.  Similar traits in brain structure or behaviors driven by the adaptation to selection pressures, but not derived from common ancestry, are referred to as homoplastic.  For example, homoplasy exists when comparing the relationship of a human hand to that of a raccoon.  The raccoon hand resembles a human hand, but their common mammalian ancestor most likely possessed a forepaw, like that seen in rodents (Butler & Hodos, 1996).  Thus, the similarities seen in raccoon and human hands probably came about due to convergence, with both species independently undergoing changes in forepaw structure in order to better use them as means to manipulate their enviornment.  Traits which two organisms share can also be analogous.  Such characters may have quite different morphology and phylogeny, but serve very similar functions (Butler & Hodos,1996).  For example the wings of both a dragonfly and a bat are analogous since they are used for flight, but is very unlikely they share any common morphological ancestry.  Hence, the determination of phylogenetic relationships is difficult, and evidence from a wide variety of studies must be brought to bear to arrive at the best approximation of such relations between organisms.  

In the next sections of this chapter, we will look at the evidence indicating the condition of the ancestral amniote brain, its evolution in birds and mammals, and consider the similarities and differences found in the brains and visual systems of modern birds and mammals. 

Next Section:  Evolution of the Amniote Brain