Skip to main content

From individual cognition to collective intelligence

Periodic Reporting for period 1 - GROUPCOG (From individual cognition to collective intelligence)

Reporting period: 2016-04-01 to 2018-03-31

Collective decision-making has been the subject of scientific study for over a century yet we still know little about the circumstances under which groups make better (or indeed worse) decisions than individuals. My postdoctoral research directly addressed this longstanding question, using pigeon homing as a model system. I compared the cognitive abilities of individuals and groups by testing them on the same spatio-cognitive tasks. To investigate the emergence of collective cognition – both as a one-off occurrence and as a cumulative process over time – cutting-edge GPS devices provided high-resolution individual spatiotemporal data, allowing me to create detailed individual-based models that also build on a large body of existing knowledge about pigeons’ individual cognitive abilities.
"I broadly performed three projects.

1) Cumulative cultural evolution
I showed, for the first time, that non-human animals can accumulate knowledge over time, a phenomenon called cumulative cultural evolution. These results were published in Nature Communications in 2017.
 
2) Individual differences in homing behavior
I found that bold (i.e. exploratory) pigeons tend to fly faster than shy ones. When they are released as a flock, bold individuals tend to fly in front and act as leaders. These findings were published in Philosophical Transaction B in 2018.

3) Combining individual opinions
I investigated how pigeons combine their idiosyncratic routes when they fly as a flock. A well-known theory, known as ""many-wrongs"" principle, suggests that birds should aggregate their routes with same weight (i.e. they should take the mean of individual routes) to maximize the performance. However, I found that they do not take the exact mean, and ones who are more royal to their own routes tend to influence the flock route more. I am currently preparing a manuscript for these results.

I also have been developing state-of-art devices in order to obtain new kinds of data. For example, I was able to put a mask on pigeon's head and put a camera on it. A manuscript is currently under review in Proceedings of the National Academy of Sciences.

"
I have published five papers (including one review paper) during this fellowship. One paper showed, for the first time, that non-human animals (pigeons) can accumulate knowledge over time and increase their performance as humans do. This finding was featured in many medias, including BBC and National Geographic. My findings influenced several fields, such as psychology, economics, and computer sciences. For example, my data suggest that a group of robots can do not only conduct tasks, which require high cognitive abilities, but also learn and improve their performance collectively over time.
Image