The brains of cetaceans—dolphins and whales—differ from those of other mammals in a number of ways, but one of the most striking differences is the size of the hippocampus. As a general rule, the larger the size of a mammal's brain, the smaller the fraction of it that the hippocampus occupies, so dolphins and whales would be expected to have a small hippocampus in any case. But the cetacean hippocampus isn't just small; it is so tiny that it barely exists.

The relative size of the cetacean hippocampus was recently quantified by a group of researchers led by Paul Manger (Patzke et al, 2013). They examined data on the size of the hippocampus in several hundred species of mammals, including several species of whales and dolphins. They found that a plot of hippocampus volume versus total brain volume yields points that are almost all clustered tightly around a smooth curve—but the points for cetaceans are outliers. Their hippocampal volumes are only 8 percent to 20 percent of what would be expected on the basis of total brain size. No other type of mammal comes close to matching that. Even the hippopotamus—the nearest living relative of cetaceans—has a hippocampus size close to what the main trend line predicts

Manger's group also found another striking difference between the hippocampus of cetaceans and other mammals. The hippocampus is one of only two brain areas that are known to show neurogenesis (creation of new neurons) in adult mammals (the other is the olfactory bulb). Manger and his colleagues examined the hippocampus of 71 species of mammals and found evidence for adult neurogenesis in all of them except the ones from cetaceans.

Manger has long been known as a skeptic about the intelligence of dolphins and whales, and he interprets the data as additional evidence that their brains are not as sophisticated as the brains of other mammals. It is interesting, though, to think about the findings in light of current theories of hippocampal function. At a general level, the small size of the cetacean hippocampus and the absence of adult neurogenesis both suggest the same conclusion: whatever function the hippocampus performs, dolphins and whales don't have as much need for it as other mammals.

But what function is that, exactly? Executive summary of the answer: we can guess but we don't know. There are two main functions that theorists associate with the hippocampus, and both involve memory. The first is long-term episodic memory; the second is spatial memory. I will consider each in turn.

In humans, destruction of the hippocampus causes inability to form new memories and major loss of memories from the recent past, but memories from the distant past, such as childhood, are usually intact. A patient with this type of amnesia can meet with the same doctor daily for years without ever learning the doctor's name or even knowing that he had met this person before.

It is important to clarify that hippocampal amnesia does not affect all types of memory. Procedural learning (the ability to learn new tasks) and working memory (the ability to hold information in mind briefly) remain intact. The signature of hippocampal amnesia is an absence of conscious awareness of having experienced an event, not an absence of memory in general.

So what about dolphins and whales? Is it possible that their diminutive hippocampus means that they lack conscious memory of past events? Unfortunately there is not enough data to answer the question. There have been several experimental studies of memory in dolphins, but not the right sort of memory.

There is one sort of memory that dolphins clearly possess to a high degree (Herman, 2010). A dolphin can be instructed, using a gesture-language, to perform a series of actions, and then a short time later it can be instructed to “repeat” them. Dolphins are quite good at that. But the delay between the first performance and the repetition is generally only a matter of seconds, so this has to be classified as working memory rather than episodic memory.

As far as episodic memory is concerned, the most relevant information comes from a study last year that found that dolphins are capable of recognizing the vocal calls of other dolphins they have previously known, even after an absence of decades (Bruck, 2013). If recognizing a vocal call is equivalent to learning a person's name, then this ability would be unexpected in hippocampal amnesia. But it would be dangerous to draw strong conclusions from a single piece of information such as this.

In summary, dolphin episodic memory is a fertile area for future experimental work, and we can't yet predict how it will turn out.

What about spatial memory?

One of the most popular theories of hippocampal function over the years has been the "cognitive map hypothesis" (O'Keefe and Nadel, 1978), which says that the hippocampus plays a central role in the formation of mental maps, allowing the brain to represent the layout of an environment and the location of objects and features.

It is interesting to note that cetaceans in their natural environment probably have less need for cognitive maps than any other type of mammals. They are the only mammals that spend most of their lives at the surface of the open sea. On the open sea there is no meaningful concept of a “place”. Spatial relationships are very important, of course, but they are relationships between pairs of objects, not between an object and a place in a stable environment. (I am talking about small-scale maps here. Dolphins and whales clearly would benefit from being able to learn the general features of a large area of space—the layout of coastlines, the locations of islands, reefs and river outlets, etc. But there are reasons to believe that large-scale maps of that type rely on different neural mechanisms from small-scale cognitive maps.)

The question, then, is whether dolphins and whales are able to form mental maps of an environment and remember the locations of objects. It should be reasonably straightforward to test that experimentally, because even though the natural environment of free dolphins is unstructured, captive dolphins spend their lives in tanks with limited size and fixed features. It makes sense to ask how well they understand the spatial layout of the tanks they live in. Unfortunately here too experimental data is almost entirely lacking. Scientists who study spatial cognition have worked out a number of systematic ways of testing spatial memory, but none of them have been attempted using dolphins.

Practically the only relevant information comes from a study carried out by Kelly Jaakkola and her colleagues (Jaakkola et al, 2010). They wanted to know the extent to which a dolphin could maintain a memory of the location of a hidden object. Their procedure involved a small stuffed toy and three buckets spaced a foot apart at the edge of the dolphin's pool. The toy was either (1) placed in one of the buckets in full view of the dolphin; (2) placed in one of the buckets and then moved to a different bucket; (3) placed in a container and then invisibly transferred to one of the buckets; (4) placed in a bucket, after which that bucket was transposed with another one. The general finding was that the dolphins succeeded in conditions 1 and 2 at above-chance levels, but not in conditions 3 and 4. Even where they succeeded, performance was erratic and a great deal of careful training was required. The authors concluded that dolphins have only a limited ability to mentally track the locations of hidden objects—roughly as good as dogs, but not as good as great apes.

Beyond that, the only information available is anecdotal. Louis Herman and Paul Forestell carried out an experiment in the 1980s in which they asked dolphins to report the presence or absence of named objects in their pool (Herman and Forestell, 1986). In many cases the dolphins would take time to explore the pool before responding. In some cases the dolphins would respond in a way that suggested awareness, including awareness of what part of the pool the object was located in, but that aspect was not examined formally.

The bottom line is that we really have no solid information at this point about whether dolphins can form cognitive maps. And the cumulative bottom line is that we can speculate about whether the tiny hippocampus of dolphins and whales brings about a weakness in episodic memory or spatial memory, but at this point we just simply don't know the answer.