Published online 24 March 2009 | Nature | doi:10.1038/news.2009.186

News

Classic behavioural studies flawed

Nobel prizewinner took short cuts to show that the way gulls feed is instinctive.

Herring gull chick feedingHerring gull chicks instinctively peck at red spots on their parents' bills to beg for food.Imagebroker / Alamy

One of the most famous experiments in biology isn't the solid piece of work it's usually portrayed as, say Dutch researchers who have replicated the study. Instead, it's more like an anecdote that became slightly more legendary each time its author retold the story.

The work in question was done in 1947 by the Dutch researcher Niko Tinbergen on the begging behaviour of herring-gull chicks. At the time, the dominant idea in animal behaviour was that learning was all-important. Tinbergen argued that animals come into the world with instincts already adapted to their environments.

Adult gulls have a red spot on their lower bill. Tinbergen, who shared the Nobel Prize for Physiology or Medicine in 1973, presented wild chicks with model birds bearing spots and measured how much they pecked at the model.

The story that made it into the textbooks is that chicks have a powerful innate tendency to peck at red dots, which has evolved as a way of getting their parents to feed them. The original paper, however, shows that Tinbergen found that chicks actually pecked more at a black dot than a red one.

In a follow-up paper written in 1949, Tinbergen concluded that this strange finding resulted from a mistake in his methods. He had tested red, black, blue, white and yellow spots, but he presented the 'natural' red spot much more often than any other. The chicks, he decided, became habituated to the red spot and stopped pecking at it.

Pecking away at a legacy?

Tinbergen never fully tested this idea. Instead, he did another experiment comparing red and black spots presented equally, and found that the red was indeed preferred. From this result he came up with a 'fudge factor' which, when applied to the original data, reduced the chicks' apparent preference for colours other than red. He was initially explicit about this, but by the time of his books The Study of Instinct (1951) and The Herring Gull's World (1953), he had stopped mentioning this correction, and presented the finding as if it were based on unmodified data.

"The changes from one paper to the next aren't that substantial, but if you compare the first and the last, a lot of small steps make a big one," says Carel Ten Cate of Leiden University in the Netherlands, who has analysed and repeated Tinbergen's work. "There were definitely some changes to make the result more simple and clear cut than it initially was."

Tinbergen did other experiments with gull chicks — showing for example, that they will peck even at a disembodied red spot on a stick — so he may have felt he had made his point. He may also have simplified his results for rhetorical purposes.

Ten Cate and his colleagues redid Tinbergen's original experiment and got the same result — black was more attractive than red. They also did the experiment he never did, presenting each colour equally often, and found that Tinbergen's intuition had been correct after all — the birds tended to peck more often at red spots. The work is published in Animal Behaviour1,2.

The full story

There's no hint of fraud in Tinbergen's work, says ten Cate, and we shouldn't think any less of him. "Looking through today's eyes you think this was really quite sloppy," he says. "But you can't use that hindsight; at the time this was really advanced work."

"Tinbergen shouldn't be castigated for this," agrees Rebecca Kilner, who studies bird behaviour at the University of Cambridge in the UK and was not involved in the new study.

"Tinbergen is an iconic character in the history of animal behaviour research," she adds. "He pioneered the use of simple but ingenious field experiments, and these experiments are a classic example of that approach."

Other researchers think that ten Cate's study risks sullying Tinbergen's legacy. "It's not fair to Tinbergen — any paper from 50 years ago wouldn't pass modern standards," says Johan Bolhuis, a researcher in animal behaviour at the University of Utrecht and editor of a book on Tinbergen3. "If we applied the same standards to Darwin's work, we'd say what a terrible experimenter he was."

"It'd be easy to be nasty — if you wanted to be negative and critical, you could do a fair amount of damage to Tingergen's reputation," agrees ecologist Hans Kruuk, Tinbergen's biographer4 and former student. "He'd often simplify and gloss over complications: if these publications appeared now, they'd get hammered, but the ideas are lovely."

Ten Cate argues that we shouldn't airbrush Tinbergen: "It's a question of getting the balance right between respect for a great scientist, and also making clear that his story is not as neat as some people have always thought and what the textbook [says]."

The ideal of how experimental science works is often a long way from the reality, whether you're looking at bird behaviour or particle physics, says Harry Collins, a sociologist of science at Cardiff University.

"In general, when scientific discoveries are first made they're messy and untidy, and they get cleaned up in retrospect," he says. "If you're a scientist you'll say that it all comes out in the end, because nature speaks with an unambiguous voice. Speaking as a sociologist, I'd say it's a historical process, and the judgement of history can't be anticipated." 

  • References

    1. ten Cate, C. Anim. Behav. 77, 785–794 (2009).
    2. ten Cate, C. et al. Anim. Behav. 77, 795–802 (2009).
    3. Bolhuis, J. J. & Verhulst, S. (Eds) Tinbergen's Legacy: Function and Mechanism in Behavioural Biology (Cambridge Univ. Press, 2009).
    4. Kruuk, H. Niko's Nature: The Life of Niko Tinbergen and his Science of Animal Behaviour (Oxford Univ. Press, 2003).

Comments

If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60678

    The Jacque Cousteau effect: Embellishing on a basically true scientific idea to sell it the general public. i

Commenting is now closed.