tl; dr: Studying how animals assign values and allocate resources contextualizes human goal-directed behavior. Here I introduce some factual claims about animals, and a suggestion for how terms from economics can be applied to their behavior.  

Acquisitiveness 

Other things being equal, animals prefer to get more of whatever they need or want, rather than less. Given a choice between one berry and two berries, a bird will usually choose two berries. In the wild this is generally an adaptive preference, because resources tend to be scarce, so getting as much as one can at every opportunity tends to increase the overall probability and degree of survival and thriving. 

The preference for more good stuff is governed by diminishing returns, however. The more one already has of something, the less value is added by having one more of that thing. A bird might slightly prefer 11 berries to 10 berries, but negligibly so (which is perhaps why most birds cannot even distinguish a pile of 10 things from a pile of 11). The value added by getting one more thing is called its marginal value (a.k.a. marginal utility). So the marginal value of x depends on how much x one already has.

Laziness 

Other things being equal, animals prefer to get whatever they need or want with as little effort as possible. This is intelligent, because time and energy are limited resources, and are often limiting resources, in the sense that the total available time and energy are insufficient to get everything animals need and want. In laboratory research on motivation and reward, the tendency toward laziness is captured by effort discounting: animals measurably subtract from the value of goods the effort or labor they have to invest to get them. 

For example, if a rat has a choice between pressing a lever 10 times to get a food pellet vs. just eating a food pellet out of a bowl, it will usually eat the one in the bowl. But if the food pellet earned by pressing the lever is sufficiently larger, the rat will choose the lever. The amount of extra food you have to offer to get the rat to choose the lever is a measure of the cost the rat assigns to that effort. If L = number of lever presses and F = amount of food obtained in grams, we can write this as: $1.5F+10L=1.0F+0L$, thus $L=-0.05F$.

Impatience and Risk Aversion

Effort isn’t the only cost animals prefer to conserve. Animals also reduce the value they assign to things according to how long they have to wait for them (delay discounting), and how uncertain they are of getting them (risk discounting). In the wild, other things being equal, it’s usually adaptive to prefer now to later, and to prefer a sure thing to a gamble. You can measure this by how much more you’d have to offer an animal for them to choose a delayed reward over an immediate one, or to choose an uncertain reward over a guaranteed one. Note that these preferences are adaptive other things being equal; this is compatible with the claim that it is also adaptive to tolerate some delay or some risk in exchange for higher returns, and it allows for individual differences in how much to discount for delay or risk.

Like marginal values, marginal costs also vary as the underlying quantity accumulates. For example, you might be willing to do a pushup to earn $1; but after 100 pushups, you might not be willing to do another one even for $100; so the marginal cost of effort might increase with accumulated effort (aka fatigue). On the other hand, you might be willing to pay a premium to get an order delivered today vs. tomorrow, but would not pay that same premium to reduce a 90-day delay to 89 days. Likewise, for animals the marginal cost of effort might increase with total effort, whereas the marginal cost of delay might diminish with total delay.

Tokens vs. Real Goods 

If a person has $\$10M$, he probably doesn’t care much about getting another $\$10$; but he rarely feels that getting another $\$10$ would make him worse off. The equivalent in animals is dopamine: there is no amount of dopamine beyond which an animal would prefer not to get more dopamine. This is because money and dopamine are not real values. You can’t eat or drink them, they don’t keep you warm or produce offspring. They are tokens that stand in for real values. Money is valuable only inasmuch as it represents the capacity to acquire things that you can actually use. Dopamine is the brain’s universal signal that something an animal did was good for it. Dopamine is valuable only inasmuch as it correlates with obtaining actual survival (or reproductive fitness) values. Under normal circumstances animal brains are hard-wired to ensure that all the actions they need to perform to survive, thrive and reproduce also produce dopamine in the brain, and behavior is motivated by an innate drive to optimize getting dopamine.  If dopamine signaling gets hijacked by a false signal - such as addictive drugs or self-stimulating electrodes - an animal will literally do the thing that gets them dopamine to the exclusion of all else, until they die. The drive to get dopamine is insatiable.

In contrast, almost all needs for real goods (food, water, shelter, etc.) are satiable. There is some amount of x minimally necessary to survive or leave progeny, and some larger amount that would further increase thriving or fecundity. But there is some maximal amount beyond which having more is worthless or could literally make one worse off^[1]. Therefore marginal value is not just diminishing (approaching $0$), at some point it reverses (becomes negative). 

Moreover, unlike tokens, real goods are often non-exchangeable. There is no amount of water I can give an animal that would compensate for having no food. Therefore, the marginal values of actual goods must be tracked separately by the animal. Third, most real goods are depletable, meaning they get used up and must be constantly renewed. Even if an animal reaches complete satiety for food and water and sleep today, it will still want food and water and sleep again tomorrow.

Money and dopamine are tokens that provide a common currency in which to value multiple real values or goals. This is probably crucial to figuring out how to allocate a single pool of limited resources (time, energy) across multiple values or goals. But the valuation of tokens is indirect, and more subject to miscalibration. To understand goal-directed behavior it is crucial not to confuse tokens (which are insatiable and interchangeable) with real values (which are satiable and non-interchangeable).

Dynamics

The marginal value of x decreases as an animal accumulates x, up to the point where the animal is either sated (prefers not to have more x, even if it is freely available), or values something else more highly (would rather invest its time getting more y, than more x). This causes the animal to switch goals, stop pursuing x and start pursuing something else for a while^[2]. However, as soon as it stops accumulating x, its reserve of x begins to deplete, so the marginal value of x starts to grow again. At some point in the future, the marginal value of x will be high enough that the animal will resume pursuing x. 

For example, a cat whose food bowl is always kept filled will eat until he is sated, and then go play or nap or type on your keyboard or whatever; but sooner or later he’ll go back to his food bowl. Thus, marginal values tend to cycle (sometimes but not necessarily periodically), and animals alternate between goals. The dynamics can get complicated because the marginal value of x often depends on the absolute value of y. (For example, if food is dry or salty, the marginal value of a sip of water increases with the total amount of food one has already consumed).

To sum up

If we want to understand what motivates animals to do stuff to get stuff, we need to remember:

• Multiplicity: animals need many different things to survive and thrive 
• Cost: time and energy are limited resources which need to be strategically allocated across these multiple needs
• Inexchangeability: although animals can trade off one value against another, each need requires separate accounting 
• Marginal value: the value of having one more x, as a function of how much x one already has (also applies to costs)
• Diminishing returns: for many real needs, marginal value declines with total value
• Satiability: most real needs can be sated (at some x, the value of +1x <= 0) 
• Depletability: most needs get "used up" and need to be regularly replenished 
• Tokens: dopamine and money are not valuable in themselves, they are tokens
• Dynamics: animals' priorities and motivations vary (and many oscillate) over time 

1. ^{^}

   For example, water is a value terrestrial animals must take action to get. Dehydration impairs function severely, and for most animals is rapidly fatal. But overhydration is also physiologically harmful, and in extreme can be fatal.

2. ^{^}

   Note that we can include things like rest or hedonic pleasure or play on the list of an animal's values, such that even if the animal chooses to chase a sunbeam or take a nap in the sun, this can be seen as allocating its limited resource of time to one of its other values.