Abstract: Attempts to include psychological constraints in models of foraging behaviour differ in their assumptions concerning the accuracy of estimation of environmental parameters. Psychologists model estimation error as increasing linearly with the magnitude of a stimulus (Weber's Law), whereas behavioural ecologists either ignore error or assume it to be independent of stimulus magnitude. Studies on the estimation of time intervals have confirmed Weber's Law, but there are few data on the accuracy of estimation of amounts of food. Since the currency of most foraging models is the amount of food acquired per unit of time spent foraging, information on estimation of amount is required. Here, a titration method was used in which starlings chose between two cues. One colour signalled a standard food reward, and the other a reward that adjusted in magnitude according to the birds' choices: it increased when the standard was preferred and decreased when the adjusting option was preferred. There were two standards of 3 and 9 units of food, each of which was delivered at two rates to control for possible effects of rate of reinforcement on discrimination. The observed value of the adjusting option oscillated around a mean value slightly larger than that of the standard. The amplitude and period of these oscillations were larger when the standard was larger, independent of the rate of reinforcement. Also, molecular analysis showed that the probability of choosing the currently larger alternative increased as the relative difference between the adjusting option and standard increased. These results are consistent with Weber's Law applying to starlings' memories for amounts of food.

Abstract: We used European starlings, Sturnus vulgaris, to investigate the relationship between the cost paid to obtain food rewards and preference between stimuli associated with the resulting rewards. In no-choice trials either 16 1-m flights (high effort) or four 1-m flights (low effort) gave access to differently coloured keys. Pecking at these keys resulted in identical food rewards. When subjects were given choices between the coloured keys in choice trials without having paid any effort, the majority preferred the coloured key that was paired with the higher level of work in no-choice trials. We relate our findings to results in animal behaviour, psychology and economics, and give a theoretical account that has implications for phenomena variously recognized as the `sunk cost fallacy' (the tendency to invest more in something after much has already been invested), `work ethics' (valuing an option more as a result of physical effort), `cognitive dissonance' (making mental effort to overlook or re-evaluate information that does not accord with a dominant internal representation) and the `Concorde Fallacy' (the readiness to forego more fitness for something that has been responsible for greater fitness compromise in the past).

Abstract: Many interpretations of the adaptive value of group living involve tranfer of knowledge. However, according to learning theory, being in a pair with a knowledgeable partner can have paradoxical consequences. Obtaining food by following a skilled companion may reduce the ability of naive individuals to learn about clues that signal the occurrence of food. This study examined the relation between learning and following in paris of zebra finches. Knowledgeable partners were trained to obtain food from a computer-controlled dispenser by using the information provided by a signal. For non-knowledgeable partners, the signal was irrelevant and could not be used to predict foraging opportunities. The rate of learning about the signal by naive birds that shared the experience of either knowledgeable or nonknowledgeable tutors was then examined. Naive birds learned more slowly as a result of being in a pair with a knowledgeable than a non-knowledgeable partner. Well-informed mates acted as a reliable cue to predict foraging opportunities, and thus overshadowed the independent signal. The knowledge of a partner influences learning rates in naive individuals, but in the opposite direction to that predicted by earlier accounts of learning in social contexts.

Abstract: An experiment with starlings was conducted to investigate the effect of variability in inter-reinforcement intervals on foraging decisions. The experimental design simulated an environment in which food was distributed in patches. Patches contained zero to four food items which could be collected by pecking at a key. All patches ended with sudden depletion. The time elapsed since the last reinforcement was the only way to detect the depletion of the patch. Once a patch was depleted, a new patch could be reached by completion of a travel requirement of 20 flights between two perches. Key pecks within a patch and the time of the last response in a patch (giving-in time) were recorded. The level of variability in the inter-reinforcement intervals was varied between different conditions. An increase in inter-reinforcement interval variability resulted in a flattening of response rate functions and giving-in time distributions, and in more asymmetry of the response functions, but not of the giving-in time distributions. Two theoretical models of decision making are presented, which differ in the assumptions about memory constraints. In one case, all inter-reinforcement intervals are remembered but in the other, only the intervals with extreme values are remembered. Both models accommodate response rates as a function of trial time, but only the second is compatible with the observed departure decision. Our results are compatible with net rate maximization.