Cooperation among humans, a question of age

One of the first experimental studies in the world to analyse the evolution of cooperative attitudes in different age groups

Young people between the ages of ten and sixteen tend to be more unpredictable when it comes to cooperation, in contrast to other age groups, while those over the age of sixty-six are the most cooperative, according to an article by scientists from the Barcelona, Zaragoza and Carlos III (Madrid) Universities published in the journal Nature Communications.

The new research, one of the world’s first experimental studies analysing the evolution of cooperative attitudes in different age groups, was carried out by Josep Perelló and Mario Gutiérrez, professors from the OpenSystems research group, part of the Fundamental Physics Department of the University of Barcelona, Anxo Sánchez, of the Complex Systems Interdisciplinary Group (GISC), part of the Mathematics Department of the Carlos III University in Madrid and Carlos Gracia Lázaro and Yamir Moreno, researchers in the Complex Systems and Networks Group (COSNET Lab) at the Institute for Bio-computation and Physics of Complex Systems (BIFI), part of the University of Zaragoza.


The prisoner’s dilemma, to cooperate or not to cooperate

The experiment is presented as a game using a web interface developed by researchers at the BIFI Institute of the University of Zaragoza and was carried out using 168 people between the ages of ten and eighty-seven who were randomly selected during the DAU Barcelona Board Games Festival held in the Barcelona Art Factory, Fabra i Coats, in December, 2012.

The experiment was carried out within the framework of the Barcelona Lab platform, promoted by the Creativity and Innovation Directorate of the Culture Institute of Barcelona (ICUB). The team installed a portable laboratory containing a dozen computers and selected volunteers among the festival-goers until a statistically relevant sample was obtained. The experiment was later repeated in order to confirm the results, which were corroborated by fifty-three twelve- and thirteen-year-old students from the Caspe Jesuit College, Barcelona. The study employed a virtual version of the prisoner’s dilemma, a game theory problem used for studying human behaviour and, in this instance, cooperation among people.

The prisoner’s dilemma plays a central role in this experiment, which continues the work of previous studies carried out by some of the same researchers, in that it provides a means of “asking” people for their cooperation. The participants are divided into groups of four, according to different age groups, in addition to a control group. During twenty-five consecutive rounds, group members were obliged to choose between cooperating and not cooperating with their group partners, with differing rewards for each action. When two people interact, the greater benefit comes when both collaborate. If one collaborates and the other does not, the latter receives a greater reward than the former, while if neither cooperates, neither receives a reward. The participants were also provided with information about their opponent’s actions and the rewards obtained by each of them. Upon finishing the rounds, the total points obtained by each participant was converted to cash, which was given to them immediately (in the case of minors, the money was given to the parents).


More unpredictable decisions among the younger participants

The most notable and novel results of the experiment reveal differentiated behaviour in the youngest age group. “In general, when it comes to deciding whether or not to collaborate, people bear in mind what the others have done. This is known as conditional cooperation. However, our experiments show that adults also take into account their own previous actions. In other words, their actions are more predictable and tend towards maintaining cooperation”, explains Yamir Moreno.

Behaviour among the younger participants, in contrast, does not follow this pattern. “According to our study, younger children are less predictable in their decision making. They don’t follow a fixed strategy and are essentially conditional co-operators as they pay more attention to those around them. Children tend to be dependent on the other players and react according to their responses rather than being conditioned by their own previous actions, making it more difficult to generate a cooperative environment”, explains Mario Gutiérrez Roig.

On the other end of the scale, the results revealed a further peculiarity, as pointed out by Anxo Sánchez. “Those over sixty-five appear to be more cooperative than those in other age groups, although we don’t have a lot of statistical information in this case and further testing would be required”. This would suggest, as other studies have indicated, that “bringing forward retirement ages may not be beneficial for businesses and it may be interesting to look for ways to keep this age group employed or, alternatively, in a situation where they may continue to be cooperative”, according to Sánchez.

In the second experiment, carried out in order to corroborate the results, the conclusions reached with students of the Caspe Jesuit College were the same. “The children were more cooperative, but evolutionary and cultural behaviour changes over the course of the life cycle and a tendency towards cooperation is a quality that may be learned”, affirms Gracia Lázaro.

The results also have implications regarding strategies that can be used to foment collaboration within this age group. “Specific strategies that are different to those of adults   would have to be developed in order to encourage the transition towards more persistent pro-social behaviour and to help the children understand the need for some degree of perseverance. In relation to education, for example, this could be translated into the establishment of clearer regulations governing teamwork, regulations that facilitate the reaching of agreements that are beneficial for all concerned”, explains Josep Perelló.

In other, previous experiments, it had already been observed that children between the ages of six and ten develop cooperativism, and this particular study pinpoints the moment when this changes: adolescence. “These causes are not clear but we think that what may happen is that in earlier phases they begin to develop a “theory of others”, as it is known in psychology circles, which allows them to empathise and be altruistic. But, on reaching adulthood, they may go through a phase in which understanding the other person puts them in a position where they may take advantage of them”, according to Sánchez. “The idea is somewhat intuitive and further experiments would be required to establish the causes of this change”, added Yamir Moreno.


A unique space for research into human behaviour

The study, carried out in the unique space that is the DAU Barcelona, is the result of collaboration between the research team and ICUB, itself the result of the creation of Barcelona Lab and the Citizens’ Science Group. Thanks, then, to the collaboration of OpenSystems, COSNET Lab and GISC, it was possible to obtain a more representative sample that those commonly used in behavioural studies. As researcher Josep Perelló explains, “this type of study usually uses samples consisting of social sciences and economics students, with the corresponding implications regarding subjects with university-level education and a certain financial status. In addition”, he continues, “they may even be influenced by the economic theories they are learning in class. Our sample is more diverse in terms of age and socio-educational status, which makes our conclusions more generalised. The idea is for DAU Barcelona (directed by Oriol Mengis and organised by ICUB) to also become an experimental space where it will be possible to carry out human behavioural studies using games that are both played and designed for the benefit of society”.

The idea is to facilitate experimental practises in which citizen participation is paramount. For Perelló, “this scientific paper is the first relevant result to emerge from this collaboration. Games are a great tool for investigating human behaviour. Experiments such as this one allow the public to participate in scientific studies from the word go and help to explain both how science works and the phases that are involved in scientific method”, concludes Mario Gutiérrez Roig.

Publication date: Wednesday, 16 July 2014
  • Share
  • Leave a comment
  • All comments

Send us a comment and share your experience of this event. If you have any other doubts or suggestions, you can click this link.

* The content of this field is kept private and will not be shown.

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.


  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.