Donaldson et al. (2012) describe the spread of a novel foraging tactic through part of a population of Indo-Pacific bottlenose dolphins (Tursiops sp.) in Cockburn Sound, Western Australia. Some dolphins in this population have learned to accept fish offered by humans. Initially, this behaviour was rare and had been observed in only one individual, but during the course of the study, the behaviour spread to a larger number of dolphins. The authors found that the dolphins using areas with high densities of recreational vessels and those associating more with previously conditioned dolphins were more likely to adopt this foraging tactic. There are two major implications of their findings. First, a previous study in the same population revealed that accepting fish handouts from humans lead to higher risks of boat strikes and entanglement (Donaldson, Finn & Calver, 2010), making the spread of this particular foraging tactic a potential conservation issue. Second, and in my opinion even more surprising, a possible social transmission mechanism of this foraging tactic is consistent with the data and also plausible, given the well-established cognitive capacities of toothed whales in general and bottlenose dolphins in particular (e.g. Rendell & Whitehead, 2001; Herman, 2002). In a conservation context, it might just be semantics whether cognitively more demanding social learning mechanisms (such as imitation or emulation), or those that are less demanding (such as local enhancement, stimulus enhancement or social facilitation) explain the observed patterns. The main point is that potentially maladaptive behaviours can spread through a population.

Critics will argue that the present study cannot unequivocally show that social learning of any form is responsible for the observed patterns, but rather that individual learning would be the more parsimonious explanation. From the way the data were gathered, it is quite obvious that the initial study of Cockburn Sound dolphins was not designed to look at transmission of foraging tactics. Rather, the researchers documented and analysed the spread of an innovation taking place in front of their eyes with post hoc analyses of gathered data. It appears that serendipity played a significant role in an exciting scientific discovery, similar to a report of cultural transmission of a novel song among humpback whales migrating up the east Australian coast (Noad et al., 2000).

The question remains, however, whether it really matters if it is some form of social or just individual learning that drives the spread of this foraging tactic in Cockburn Sound dolphins. The authors seem to shy away from this issue by stating that ‘[…] social learning could play a facilitative or supplementary role in the acquisition of learned behaviours involving the use of anthropogenic foods, and thus potentially influence which individuals within a population become conditioned to human interaction.’ Yet, once established, social learning can be quite a powerful and efficient agent for information transfer. If based on social transmission rather than individual learning, the spread of an innovation through a population will be much faster and the innovation has a much higher chance to be persistent in the population (Whitehead et al., 2004). Thus, conservation measures may have to be drastically altered depending on the mechanism with which they are established and maintained.

A potential major source of criticism is that the authors use only association data gathered during group surveys for an arguably very rough proxy for the potential of social transmission. Yet, everyone working on great apes or cetaceans knows how notoriously difficult it is to infer social transmission of innovations. I do agree with the authors that social transmission is likely to be involved in the spread of this behaviour. Supporting this claim, I would like to emphasize the authors are dealing with a species that is highly encephalized and long lived, and relies on extensive external inputs during development, including social ones. Thus, one has to ask whether it is still appropriate to use canalized development as the null model for behavioural variation within or between populations, as we have recently argued for great apes, cetaceans and other highly encephalized taxa (Krützen, Willems & van Schaik, 2011). Moreover, there is an ever-growing chain of evidence substantiating a cultural interpretation, based on social transmission of innovations, of geographic variation in certain elements of non-human primate and, to a lesser extent, cetacean behaviour (e.g. Whiten et al., 1999; van Schaik et al., 2003; Krützen et al., 2005; Krützen et al., 2011), as well as many other taxa.

The developmental aspects of foraging tactics in cetaceans are unfortunately only mentioned briefly in the paper, although they would shed an important light on a potentially very exciting feature. Most foraging tactics in cetaceans appear to be transmitted vertically, that is, they are usually passed on from mother to offspring (e.g. Mann & Sargeant, 2003). Great apes and toothed whales appear to differ in this regard, as behavioural variation in the former is usually best documented among populations, probably due to horizontal and vertical spread of innovations. In cetaceans, however, variation in foraging tactics has so far only been reported within populations, for instance Shark Bay dolphins (Mann & Sargeant, 2003; Krützen et al., 2005; Allen, Bejder & Krützen, 2011). If substantiated, the finding that this foraging tactic appears to spread horizontally among adult and potentially unrelated Cockburn Sound dolphins would be one of the first for cetaceans.


  1. Top of page
  2. References