Showing posts with label Crowd Sourcing. Show all posts
Showing posts with label Crowd Sourcing. Show all posts

Flock Sourcing


Attracting birds to our yards by feeding them is a common morning activity for many people, especially during the winter. The small, pretty ones are more welcome than the big dull ones. We have many arrangements that make it easier for our favored guests to get the food and keeping unwelcome ones like crows or squirrels out.

We admire the agility and ingenuity of our pretty friends in getting the treats we put out for them. In the UK, the Tit family, especially Blue Tits and Great Tits are particularly loved, even when they have pecked through the tops of the milk bottles on the doorstep (in places where doorstep deliveries still take place.) Morand-Ferron and Quinn of Oxford U are fond of watching these pretty birds and report a study of their problem solving abilities using some automated feeding puzzles (1).

There were two main findings. The first was that the problem of getting into the food was solved quicker if the flock size was larger – crowd sourcing wins again. Secondly, and this is good news for the old fogies amongst us, the solution was reached faster if at least one experienced “problem solver” was included in the flock. This fits nicely with the studies on people that suggests that diverse groups, but which include at least one experienced person in the area, are best at problem solving.

The usual model of group living is that large groups tend to limit access to the food supply, as it has to be spread thinner. Now the good news is that the “pool of competence” that occurs in the larger flocks, improves the problem solving ability and hence, the exploitation of the food sources so that everybody eats better.

Don’t you just love it when common sense and working together for the general good work out better than “the devil take the hindmost” philosophy of the "every man for himself" approach?

  1. http://www.pnas.org/content/early/2011/09/08/1111560108.abstract

Swarming to a Conclusion


The concept of swarm intelligence in humans has grown, and is one of the latest business focused self-improvement topics. It follows naturally from the concept of crowd sourcing and the Wisdom of Crowds (1). Both the democratic system and the jury system can be considered as having a basis in the swarm intelligence concept. Now, we all take pride in our individuality, and we will sort the problems of the world out over a coffee, a glass of wine or a pint of beer, depending on our habits. Few of us actually want the opportunity to put our ideas into practice though, and rule the country let alone the world.

A key requirement of the human swarm to be successful is diversity (1). A single individual with a high level of expertise may get stuck with preconceived ideas and old, outdated practices. A swarm of people with varied experience will come up with a wide range of possible solutions. Some of them may be novel and powerful, sweeping away the old ideas.

A new study of the phenomenon is published in J. Animal Behavior by Krause et al (2) and they confirm the diversity concept. They conclude that adding diversity is better than adding expertise to a team (swarm). This is counter-intuitive and will be hotly debated over the lattes, but there are caveats.

They found that a group of individuals who were classified as underperformers can out-compete a group made up of high performing individuals.   This is not a surprising result to anyone who has had to work with a team of prima donnas, and who would rather opt for a job of herding cats.

The caveats mentioned above are in the type of problem that is requiring a solution. Fixing my car would not require a swarm, although sometimes I feel that I am paying for one. Electing a president would.



  1. The Wisdom Of Crowds, James Surowiecki Random House, New York, 2004.
  2. Krause et al., 10.1016/j.anbehav.2010.12.018 |