David Stephens and Dack Anderson discussed the theory of short-term benefits as guide to animal feeding decisions. In a series of experiments conducted in an operant laboratory, blue jay birds were the subjects of the study. The foraging animals were tested in two situations namely the self control paradigm and the patch-use problems.
The article details how natural selection among the foraging blue jays favour short term rules which result to long term consequences.
The paper discusses how to combine time and amount in a suitable currency that mirrors the fitness consequences of different time/ amount sequences and determines the decision rules foragers apply to select the currency maximizing time/ amount sequence from a determined set of possibilities and investigating the relationship between decision rules and currencies.
Foragers’ farsighted way in choosing is contradicted by laboratory experiments where the blue jays made short sighted decisions. The logic of farsightedness is best illustrated in the traditional studies of Stephens and Krebs (1986) where long term rate maximizing currency is infinite gain/ infinite time however this logic cannot be generally valid because food gained after starvation is not of any use thus the expression is too farsighted.
Two groups of scholars presented an alternative to this proposal where the best rate is attained one at a time thus being short sighted (Bateson and Kacelnik, 1996; Baum and Rachlin, 1969) but Stephens and Krebs (1986) argued that when the maximum rate is taken from each chance, time may be exhausted on a mediocre alternative.
The authors concluded that both long term and short term models presented are of extremes thus introducing a discount rate which measures the comparative loss of value in order to create intermediate term models.
In this notion the long term model is determined when the discount rate is zero since the value of the gain is the same even if it was obtained early or late in a sequence but as the discount rate becomes larger the approach becomes that of the short term model. The discount rates are expected to be non-zero.
According to the authors, the long term model is the most sensible because it provides a rational estimate of rate even in a restricted time horizon. It is farsighted enough to pass over bad options and rational discount rates are small. Short sightedness is evident in the notion that animals often make prejudiced decisions which was concluded in the self control experiment.
In a self control situation long term rate model predicts that animals will choose Gain1 over inter-trial interval plus delay before the small amount plus post feeding delay over the second alternative which is post feeding delay over inter-trial interval plus delay before the large amount as long as the former is greater than the later with consideration to temporal rates such as delay to food, post feeding delay and inter-trial delay.
However these temporal rates are not treated equally by animals according to experiments. Evolutionary models can accommodate small discounting outcomes but they challenge the predictions of self control experiments. The latter show that short term hindrances have large effects. Scholars interpret such disagreements by disregarding the model and accepting the data.
The blue jays were given a binary choice in the self control situation while the same subjects were given a choice to continue or leave where they follow a future patch alternative in the patch-use exploitation. The latter shows the effects of travel time on rate maximizing patch residence time.
In the patch-use situation, the jays’ choice has comparable consequences but the subjects choose whether to wait in the patch and gather (G2 – G1) extra units of food or to leave at once after attaining a small amount (G1). In order to determine the different patterns of choices in the self control and patch-use contexts, the authors considered such in the context of 50 and 5 discount rate levels of delay to small.
Results of the experiment concluded that when the delay to small was of ample amount (50) preferences of the blue jays were not affected by the ITI thus large outcome in the patch-use context was more favourable but when the delay to small was few (5), outcome was more complex.
In the control situation, the jays’ favour for large decreased together with the ITI while in the patch-use situation the subjects’ favour increased for the large together with the ITI. The jays preferred large when the ITI was small in the self control situation while in the patch-use situation, the foraging animals preferred large when the ITI was large.
As predicted in the long term theory, the patch-use situation showed that the jays favoured large delayed outcomes as ITI increased but in the self control situation the outcomes are not the same. The authors presented situations that prove how results from both experiments (patch-use and self control) lead to long term rate maximizing.
In the long term model situation results show that the patch-use context is in line with the long term rate model while the self control situation contradicts. According to the short term rate model, results of the self control treatment of binary choice conform to short term rate models while the outcomes of patch-use exploitation conforms to the long term rate currency.
Both the self control and patch use situations were compared in the same rule different outcome analysis. This analysis resulted in the conclusion that in the patch-use experiment, short term rates were obtained with long term rate maximization but in the self control situation the conclusion differs.
The authors concluded that short term rule always agrees with the difference in long term rates according to the patch-use exploitation situation though such rule produces outcomes in relation to short sighted behaviour in the self control situation thus the subjects use the same short term rule in deciding for both cases but producing different outcomes.
For the adaptive short term rules scenario, it has been found out that the natural process of selection among foraging animals has favoured short term rate comparisons for they measure long term rate maximization in natural decision contexts. The hypothesis deals with the usage of the same rule in patch-use and self control contexts and the representation of patch condition as natural foraging problems.
The long term rate currency of foraging theory has been contradicted by self control studies and risk sensitive preferences. According to the risk sensitivity studies of Kacelnik and Bateson (1996), foraging animals favour changeable postponements even though unlimited time horizon rate models forecast indifference to unpredictability.
The authors of the article argue that the evidence of self control of short term rules is not evidence of short term currency because in natural decision context, short term rules can have long term consequences.
Finally the authors theorized that pragmatic patterns of favour can be accounted by long term rate currency which includes reasonable levels of discounting. Foraging species favour attaining short term rewards with a minimal advantage over long term ones in abundance.
The usual technique of elicitation of the inter-temporal discount rate has been critically challenged by scholars over the years when empirical studies discovered systematic violations of assumptions such as time consistency of preferences, stationary, constant discounting and utility maximization (Kalenscher and Pennartz, 2007).
The discounted utility theory has limited descriptive validity because it fails to effectively depict the authenticity of inter-temporal choice behaviour.
“Common difference and immediacy effects and the fact that preference reversals occur after deferring all choice alternatives into the future by the same interval, violate assumptions of consistent choice: (Kalenscher and Pennartz, 2007).
Foraging animals’ preferences do not depend on the proportion of gains and time in between such achievements but rather on the waiting time prior to the rewards. The technique becomes incoherent with the original and most favourable foraging formulation where animals are seen to sacrifice rather than maximize.
The inter-temporal choice notion that it can be compressed into a single discount function is challenged by other variances such as the sign, magnitude and framing effects. When making inter-temporal decisions, foraging animals break vital assumptions in discount rates.
Other limitations of such technique include the inability of long term theory to predict consistently the increase of favour for large together with ITI because choice in proportion cannot exceed 100% thus broad conditions exist where ITI has no effect.
In the patch-use situation, ITI is hard to detect when favour for large is high, another drawback is with regards to the linear relationship in focusing on temporal elements, where in reality the relationship between the amount and value can never be linear and lastly the authors’ assumption of animals’ preferences of short term rules over long term rules are not mutually exclusive.
Critics of the usual technique of elicitation of the inter-temporal discount rate methods can be traced to their inability to provide consistent outcomes over time. Systematic violations also occur with regards to assumptions such as time consistency of preferences, stationary, constant discounting and utility maximization.
These factors gravely affect results thus the method is questionable in providing valid outcomes in experiments. Hyperbolic discounting is inconsistent because it only measures the discounted outcomes of the near future and rather than the far future. The pattern of discounting is also inconsistent because how one subject’s decisions may be based on different aspects one considers than the other.
In human behaviour discounting is influenced by the person’s expectations, moods, sign, glucose levels and the like which may differ from one person to another.
Experimental ecology has provided a method in testing the inter-temporal choice behaviours of human beings though discounting can only provide outcomes of the near future and is limited to predict results in the far future, experimental ecology enables scholars to study patterns of choice.
Patterns can be explained through discounting methods offered by the experiment in distinguishing the delay and interval of discounting. The experiment can help scholars analyse human behaviours in terms of people’s decisions and the future consequences they regard in making such choices by considering the variability of discount rates the experiment may provide.
Experimental ecology provides scholars close to accurate approximations of outcomes through a series of controlled examinations of hypothesis and theories. It enables scholars to manipulate evidences and test models where distractions in their natural habitats may disrupt findings. If a model disagrees with data, the former will be rejected and the latter will be followed accordingly.
In determining the feeding decisions of foraging animals such as blue jays, Stephen and Anderson compared the self control and patch-use experiments where in the former the jays were given a binary choice, obtain gains in a short period of time even if it is a few in quantity or wait for a period of time and obtain more gains while the in the latter case jays are given a choice to attain a gain and leave or continue and receive more gains.
In both experiments the approach is similar but outcomes vary. In the self control approach the authors found out that it tends to lean on the short term model and that the patch-use situation is associated with the long term model.
Techniques in the elicitation of inter-temporal discount rates help predict outcomes in behaviours of foraging animals in their decisions towards feeding but it presents limitations such as the violations of the theory’s assumptions and the consistency of predictions which may present as a problem in synthesizing conclusions for experiments.
Furthermore, the assumption of linear relationships in the experiment deemed itself inconsistent because in reality relationships can never be linear. Though experimental ecology has its critics, contributions of such in the field can be attributed to its ability to control external factors that may disrupt results of the study and through experimentations theories and hypothesis can be proven through actual examinations.
Reference List
Bateson, M and Kacclnik, A 1996. “Rate currencies and the foraging starling: the fallacy of the averages revisited”, Behavioural Ecology, vol. 7, no. 3, pp. 341-352.
Baum, W and Rachlin, H 1969. “Choice as time allocation”, Journal of the Experimental Analysis of Behaviour, vol.12, no. 6, pp.861-874.
Kacclnik, A and Bateson, M 1996, “Risky theories – the effects of variance on foraging decisions”, American Zoology, vol. 36, pp.402-434.
Kalenscher, T and Pennartz, C 2007, “Is a bird in the hand worth two in the future? The neuroeconomics of inter-temporal decision making”, Progressive in Neurobiology, pp.1-32. Web.
Stephen, D and Krebs, J 1986, Foraging theory: Monographs in behaviour and ecology, Princeton University Press, New Jersey.