T. Andrew Hurly

Irrational choices in hummingbird foraging behaviour

It is conventionally assumed that, when animals evaluate alternative options, the value assigned to an option is absolute and independent of the other options available. It follows that animal choices should exhibit the rational property of regularity whereby the proportion of choices for an option cannot be increased by the addition of further options to the choice set. However, violations of regularity occur in human decision making, suggesting that humans may use comparative evaluation mechanisms whereby the value of an option is computed relative to the other options available. For example, in the asymmetrically dominated decoy effect the preference for a target option over a competitor is altered by the addition of a decoy option that is inferior to the target and competitor on one attribute, but lies between them on a second. We tested whether foraging wild rufous hummingbirds, Selasphorus rufus, would demonstrate violations of regularity in response to an asymmetrically dominated decoy. Sixteen birds chose between three artificial flower types (Target: 15 μl, 40% sucrose; Competitor: 45 μl, 30%; Decoy: 10 μl, 35%) in Binary (Target versus Competitor) and Trinary (Target versus Competitor versus Decoy) treatments. We predicted higher preference for the Target in the Trinary treatment. The birds ranked the three options in the same order in the Binary and Trinary treatments (Competitor>Target>Decoy). Seven birds showed violations of regularity, six increasing their absolute preference for the Competitor in the Trinary treatment. Overall, relative preference for the Competitor over the Target was higher in the Trinary than in the Binary treatment. These changes in preference are incompatible with an absolute evaluation mechanism.

Context-dependent, risk-sensitive foraging preferences in wild rufous hummingbirds

We tested the risk-sensitive foraging preferences of wild rufous hummingbirds, Selasphorus rufus, with three types of artificial flowers. All three flower types provided the same mean volume of 30 µl of sucrose, but differed in terms of variability of the reward: constant, low variance and high variance. In trinary comparisons, subjects preferred the low-variance reward over the constant reward, and the constant reward over the high-variance reward; a result not predicted by risk-sensitive foraging theory. However, when tested with traditional binary comparisons, hummingbirds showed conventional risk-averse behaviour and selected the constant reward over the low- or high-variance rewards. This reversal of preference represents a context-dependent foraging preference. The utility of selecting intermediate levels of risk and the source of the preference reversal are discussed relative to risk-sensitive foraging theory and the effects of local context on foraging choices. Copyright 1999 The Association for the Study of Animal Behaviour.

Absolute and Relative Pitch Production in the Song of the Black-Capped Chickadee

Black-capped Chickadee (Parus atricapillus) song consists of two notes, termed fee and bee. Frequency measures at three key points (at the start and end of fee, and at the start of bee) were obtained from the songs of a large sample of chickadees (n = 151) in the wild. In this sample, 19 birds produced songs shifted downward in frequency as well as their normal songs. Analysis of normal song revealed that fee declines in frequency in a glissando of nearly pure tone, then continues at greatly reduced amplitude at the start of bee; whereas bee, also a nearly pure tone, is always lower in frequency than either the start or end offee. The absolute pitches (frequencies) of these measures vary substantially among birds, but much less within individuals. In contrast, pitch intervals (ratios of higher to lower frequencies) for frequency changes among the three measures are highly invariant among birds. Moreover, chickadees with normal and frequency-shifted songs maintain virtually the same pitch intervals in both. This analysis suggests that the absolute and relative pitch constancies in chickadee song production may provide information for individual and species recognition, respectively.

Timing in Free-Living Rufous Hummingbirds, Selasphorus rufus

Animals organize their lives around circannual and circadian rhythms, but little is known of their use of much shorter intervals. In the laboratory, some animals can learn the specific duration (seconds or minutes) between periods of food access. It has been supposed that wild nectarivores, such as hummingbirds, might also learn short time intervals so as to avoid revisiting emptied flowers until the nectar has been replenished. We provided free-living, territorial rufous hummingbirds each with eight artificial flowers containing sucrose solution. Four flowers were refilled 10 min after the bird emptied them, and the other four were refilled 20 min after being emptied. Throughout the day, birds revisited the 10 min flowers significantly sooner than they revisited the 20 min flowers, and return visits to the flowers matched their refill schedules. Hummingbirds remembered the locations and timing of eight rewards, updating this information throughout the day. Not only is this the first time that this degree of timing ability has been shown in wild animals, but these hummingbirds also exhibit two of the fundamental aspects of episodic-like memory (where and when), the kind of memory for specific events often thought to be exclusive to humans.

SCATTERHOARDING AND LARDERHOARDING BY RED SQUIRRELS: SIZE, DISPERSION, AND ALLOCATION OF HOARDS

We quantified the degree of scatterhoarding and larderhoarding exhibited during late summer and early autumn by 10 individuals within a population of red squirrels ( Tamiasciurus hudsonicus ) in the Cypress Hills, Alberta, Canada. All squirrels made extensive and obvious larderhoards in central middens. In addition, sampling quadrats and behavioral observations revealed that ca. 43% of total stores were scatterhoarded throughout territories. Across individuals, the number of cones scatterhoarded was related inversely to the number stored in the midden. Nearest-neighbor analyses indicated that cones were distributed in a spatially clumped manner throughout each territory. However, the hoards themselves, independent of size, tended to be dispersed randomly. Cones hoarded in middens did not differ from scatterhoarded cones in terms of content of seeds, nor were they treated differently by the squirrels in terms of carrying distance. At the end of our study, seasonal hoarding was not yet complete, but squirrels already had hoarded sufficient cones to meet their energetic requirements for ≥3 months. We conclude that the characteristic pattern of storage of food by red squirrels is a deliberate mixture of distinct larderhoarding and scatterhoarding tactics.

Rufous hummingbirds' memory for flower location

Abstract We used an open-field analogue of the eight-arm radial maze to investigate the role of memory during foraging by rufous hummingbirds, Selasphorus rufus . In experiment 1 we attempted to determine whether birds were able to differentiate between flowers of the same type that they had emptied, flowers they had seen but not visited and new flowers. They were tested with three trial types, all of which involved birds visiting four rewarded flowers in the first phase of a trial. In ‘free’ trials, the bird was allowed to choose four from eight flowers. In ‘forced’ trials there were only four flowers available in phase 1 and in ‘mixed’ trials the bird could choose four from six available flowers. In all trial types eight flowers (including all those in the same locations as in phase 1) were presented to the bird on its return in phase 2. The four rewarded flowers were those not visited in phase 1. In free and mixed trials, birds were better than chance at avoiding the flowers they had emptied in phase 1. In mixed trials, birds were more likely to visit the new flowers that were unique to phase 2. In experiment 2 we tested whether flower height was a floral feature remembered by birds. Birds were given forced and free trials in which the flowers in the radial maze were presented at two heights. As performance in both trial types was better than chance we suggest that hummingbirds use flower height to remember the locations of flowers.

Spatial Learning and Memory in Birds

Behavioral ecologists, well versed in addressing functional aspects of behavior, are acknowledging more and more the attention they need also to pay to mechanistic processes. One of these is the role of cognition. Song learning and imprinting are familiar examples of behaviors for which cognition plays an important role, but attention is now turning to other behaviors and a wider diversity of species. We focus here on work that investigates the nature of spatial learning and memory in the context of behaviors such as foraging and food storing. We also briefly explore the difficulties of studying cognition in the field. The common thread to all of this work is the value of using psychological techniques as tools for assessing learning and memory abilities in order to address questions of interest to behavioral ecologists.

What, where and when: deconstructing memory

The ability of animals to remember the what, where and when of a unique past event is used as an animal equivalent to human episodic memory. We currently view episodic memory as reconstructive, with an event being remembered in the context in which it took place. Importantly, this means that the components of a what, where, when memory task should be dissociable (e.g. what would be remembered to a different degree than when). We tested this hypothesis by training hummingbirds to a memory task, where the location of a reward was specified according to colour (what), location (where), and order and time of day (when). Although hummingbirds remembered these three pieces of information together more often than expected, there was a hierarchy as to how they were remembered. When seemed to be the hardest to remember, while errors relating to what were more easily corrected. Furthermore, when appears to have been encoded as a combination of time of day and sequence information. As hummingbirds solved this task using reconstruction of different memory components (what, where and when), we suggest that similar deconstructive approaches may offer a useful way to compare episodic and episodic-like memories.

Cognitive Ecology: Foraging in Hummingbirds as a Model System

Publisher Summary The chapter presents foraging in free-living rufous hummingbirds, Selasphorus rufus, as a model system in which to assess cognitive abilities, particularly learning and memory, in animals in the field. The value of this system is twofold: first, and possibly most important, are the logistic advantages. The logistics of testing foraging in territorial rufous hummingbirds are such that this behavior appears to offer a compromise that will irritate both field zoologists and laboratory psychologists however may also continue the integration between the two groups that has been seen in song learning, imprinting, and food storing. The second advantage to using foraging in hummingbirds for investigating cognition in the field is that the birds feed on flowers that would seem to require learning and memory in at least two ways: (1) because the flowers of different species vary in color, shape, size, location and content, the birds may learn their species-specific features; and (2) some of the flowers refill once emptied, however do so over the course of several hours. By testing the hummingbirds in the field have managed, thus far, to assess their learning and memory abilities by using arrays of flowers on a spatial scale that is comparable to that they are faced with in everyday foraging. The rufous hummingbird may be a useful species in which to make such a comparison as they are fairly readily kept and tested in captivity.

Spatial relational learning in rufous hummingbirds (Selasphorus rufus)

There is increasing evidence that animals can learn abstract spatial relationships, and successfully transfer this knowledge to novel situations. In this study, rufous hummingbirds (Selasphorus rufus) were trained to feed from either the lower or the higher of two flowers. When presented with a test pair of flowers, one of which was at a novel height, they chose the flower in the appropriate spatial position rather than the flower at the correct height. This response may also have been influenced by a preference for taller flowers as acquisition of the task during experimental training occurred more readily when the reward flower was the taller of the pair. Thus, it appears that although learning abstract relationships may be a general phenomenon across contexts, and perhaps across species, the ease with which they are learned and the context in which they are subsequently used may not be the same.