Optimal Foraging Theory Simulator: Marginal Value Theorem

simulator intermediate ~10 min
Loading simulation...
Optimal patch time: 28s, yielding 1.8 cal/s long-run intake rate

With patch quality of 100 cal, 15s travel time, and 0.05 depletion rate, the marginal value theorem predicts an optimal patch residence time of approximately 28 seconds, yielding a long-run intake rate of 1.8 calories per second.

Formula

E(t) = Q × (1 - exp(-d × t))
Optimal t*: dE/dt = E(t*) / (T_travel + t*)
Intake rate = E(t*) / (T_travel + t*)

The Economics of Animal Foraging

Optimal foraging theory applies economic optimization to animal behavior, predicting that natural selection shapes foraging strategies to maximize net energy intake per unit time. Just as an economist models a firm maximizing profit, behavioral ecologists model an animal maximizing the currency of Darwinian fitness — typically approximated by energy intake rate. The theory generates precise, testable predictions about which food items to eat, how long to stay in a patch, and when to move on.

The Marginal Value Theorem

Eric Charnov's marginal value theorem (1976) addresses the patch departure problem: given a depleting food source and the cost of traveling to a new patch, when should a forager leave? The answer is elegant — leave when the instantaneous rate of gain (marginal value) drops to the average rate for the entire habitat. Mathematically, for a gain function E(t) = Q(1 - e^(-dt)) in a patch of quality Q with depletion rate d, the optimal residence time t* satisfies dE/dt|_{t*} = E(t*) / (T + t*), where T is travel time.

Patch Quality and Travel Costs

The MVT makes counterintuitive predictions. When travel time between patches increases, the optimal strategy is to stay longer in each patch — even though patch quality hasn't changed. This is because the opportunity cost of leaving (wasted travel time) is higher. Conversely, in rich environments where patches are close together, animals should be more selective, leaving each patch sooner when intake rate drops. This simulator lets you manipulate these variables and see how the optimal strategy shifts.

Testing the Theory in Nature

Hundreds of empirical studies have tested MVT predictions across taxa from starlings to bumblebees to parasitoid wasps. Cowie (1977) tested great tits foraging in patches with different travel costs and found remarkable agreement with MVT predictions. However, real animals face complications the basic model ignores: predation risk, incomplete information about patch quality, and the need to sample unfamiliar patches. These factors create a rich interplay between optimization and constraint that continues to drive research in behavioral ecology.

FAQ

What is optimal foraging theory?

Optimal foraging theory (OFT) is a behavioral ecology framework that uses optimization models to predict how animals should forage to maximize their net energy intake rate. It assumes natural selection has shaped foraging behavior to be efficient, and generates testable predictions about diet choice, patch use, and movement patterns.

What is the marginal value theorem?

The marginal value theorem (MVT), formulated by Eric Charnov in 1976, predicts that an optimal forager should leave a depleting food patch when its instantaneous rate of energy gain (marginal value) drops to the average rate achievable across all patches including travel time. Graphically, the optimal time is found by drawing a tangent from the travel time point to the gain curve.

How does travel time affect patch residence?

Longer travel times between patches increase the optimal residence time within each patch. When travel is costly, it pays to extract more from each patch before moving on. This prediction has been confirmed in numerous species from bumblebees to great tits.

Do animals actually forage optimally?

Many studies show that animals approximate optimal foraging predictions remarkably well, though not perfectly. Deviations can arise from incomplete information, predation risk, cognitive limitations, or competing needs. OFT serves as a useful null model against which actual behavior can be compared.

Sources

Embed

<iframe src="https://homo-deus.com/lab/ethology/optimal-foraging/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub