Undoubtedly, experiments are an essential part of the scientific method. Through the design of experiments, the use of reliable and accurate computational methods and the analysis of data, we could eventually solve the evolutionary puzzle and validate existing theories. Therefore, in the very tradition of Science, we aim at observing the world around us, collect data

and information and then, if they are not enough to understand the phenomenon under study, design new experiments on purpose. The latter is our main goal. Below you can find a description of selected works that we consider relevant for their reported results. They are ordered from the oldest to the newest.
1.-Testing Network Reciprocity
Our first experiment was aimed at testing whether the structure of the underlying network through which interactions take place plays any role in the emergence and evolution of cooperative behavior. To this end, we designed the largest experiment conducted up to now involving humans playing a Prisoner’s Dilemma. Specifically, we ran two settings: one corresponded to a square lattice in which 625 individuals were arranged in such as way that each subject played with 4 neighbors; and the second setup, involving 604 volunteers, considered a heterogeneous (so-called Scale-free networks) graph in which the nodes represent individuals that interact following the links of the graph. In this latter setting, the number of neighbors of a given player varied from 2 to 16. Both experiments were carried out concurrently, i.e., 1229 were playing in real-time (for more information about the experimental methodology, see the SI of the paper linked below).

The hypothesis to be tested in this experiment is that of network reciprocity. During the late 2000’s, among the several mechanisms that were proposed to explain cooperative behavior among humans, the effect of the underlying network of contacts received very much attention. Essentially, theoretical models have shown that the larger the heterogeneity of the contact networks is, the higher the cooperation observed is. The main mechanism for the evolution of cooperation behind these models is based on the assumption of certain strategies, including the fact that people take into account the pay-offs of their contacts to choose their actions, i.e., to decide whether they play as a cooperator or as a defector. In contrast to theoretical provisos, some experiments that were carried out with people playing social games on regular (grid-like configurations) structures, suggested that when individuals face the decision to cooperate or act selfishly, they do not
consider neighbors’ pay-offs, but a rule called moody conditional cooperation: the probability of cooperation depends on both the subject last action and on the number of cooperators in her neighborhood (the higher the number of cooperators in the neighborhood of the player is, the higher the probability to cooperate if she did so in the last round).
By implementing the above rule in different topologies, it can be shown that the observed level of cooperation is independent of the structure of the contact network [see Gracia-Lazaro et al, Sci. Rep 2, 325 (2012)]. Thus, the only thing left was to test whether this is actually the case when playing with real subjects. The experiment was carried out on December 20th, 2011, with volunteers chosen among last year’s high school students of Aragón (Spain).
The results backed-up the above theoretical predictions, closing the research cycle: the level of cooperation reached in both networks was the same, regardless of the underlying structure of contacts, and comparable with the level of cooperation of smaller networks or unstructured populations. We also found that, as had been observed previously in smaller experiments, subjects respond to the cooperation that they observe in a reciprocal manner, being more likely to cooperate if, in the previous round, many of their neighbors and themselves did so. Our results, thus, strongly point to the conclusion that population structure has little relevance as a cooperation promoter or inhibitor among humans. Among the implications of our work (leaving aside its fundamental and far-reaching consequences), it is worth mentioning that the fact that networks do not promote cooperation is crucial in the design of organizations, indicating that efforts should focus on individual incentives to cooperate, rather than on the structure of the organization.

 C. Gracia-Lázaro, A. Ferrer, G. Ruíz, A. Tarancón, J. A. Cuesta, A. Sánchez, and Y. Moreno, “Heterogeneous networks do not promote cooperation when humans play a Prisoner’s Dilemma“, Proceedings of the National Academy of Sciences USA 109, 12922-12926 (2012).

2.- Testing the evolution of cooperation in an individual lifetime
Although cooperation and altruism in human societies have been widely studied, certain fundamental aspects have received little attention. One of them is whether the observed cooperation changes through an individual lifetime. An immediate question that follows from this matter is of methodological nature: as the vast majority of experiments are performed with subjects in their early twenties’, do the results obtained with those individuals are generalizable to other age ranges? For instance, it has been observed that children develop their social abilities in different stages and that the elderly could be more inclined to prosocial behavior. In order to provide grounded evidence in one sense or another, we have performed a series of lab-in-the-field experiments carried out over a very wide range of ages. Our experiment took place in the 1st Board Games Fair in Barcelona in December 2012, and consisted of groups of 4 participants playing simultaneously a 4-player Prisoner’s Dilemma. Participants with ages between 10 and 87 among visitors of the Fair were randomly recruited. In order to compare the behavior of subjects of different ages, we either placed them in a group with other players of similar age (we divided the age range as follows: 10–17,18–25, 26–35, 36–45, 46–55, 56–65, and 66 and over) or placed in a group with
other participants irrespective of their age (control groups). A second experiment, this time involving children aged 12-13 years took place on March, 2014 at a School also in Barcelona.

The results showed that young teenagers do not have an intrinsic strategy and that elderly people cooperated more. Specifically, we found that there are two transitions in the observed cooperative level as humans get older: children in the range 10-16 years old are not intrinsically cooperators nor defectors, their behavior being influenced by their neighborhood. In adulthood, individuals are differentiated and decisions become much more persistent. Subsequently, cooperativeness increases in the elderly. Our results imply that mechanisms usually invoked to explain human cooperation are age-independent beyond adolescence. Our findings suggest that specific strategies should be developed to foster prosocial behavior in youth and have important theoretical and practical implications in fields such as behavioral economics, evolutionary theory and developmental psychology.

 M. Gutiérrez-Roig, C. Gracia-Lázaro, J. Perelló, Y. Moreno, and A. Sánchez, “Behavioral transition with age in social dilemmas: From reciprocal youth to persistent response in adulthood”, Nature Communications  5:4362, doi:10.1038/ncomms5362 (2014).

3.- Testing (again) network reciprocity in dynamic settings
While our first experiment showed that the underlying topology plays little role in the evolution of cooperation, many other mechanisms have been proposed to explain cooperative behavior. On the one hand, several authors have proposed that the network of contacts does play a role, but in order to observe it, one should allow for the formation and removal of links, i.e., the network should be dynamical in nature (as opposed to our first experiment, in which the networks were static). On the other hand, other mechanisms such as indirect reciprocity have been less explored experimentally. These mechanisms of reciprocity (no matter whether indirect or direct) are based on the concept of reputation: in pairwise encounters between members of a population, individuals may have information about previous actions of their opponents and act accordingly. Thus, reputation would enhance cooperation by favoring strategies that are based on the opponent’s cooperative stature. In order to study the effects of reputation in social dilemmas, we have carried out a series of laboratory experiments where people interact through an iterated prisoner’s dilemma on a network, allowing subjects to break and form links.
We conducted 24 experiments involving 243 participants. Initially, the neighbors of each participant were randomly assigned and through the experiment they were allowed to break their links and propose new ones. Players could see a certain number m of previous actions of their opponents. The games consisted of 25 rounds and each player participated in two different games with different memory (m).
Indeed, our results showed that it is reputation that really fosters cooperation. While this mechanism has already been observed in unstructured populations, we find that it acts equally when interactions are given by a network that players can reconfigure dynamically. Furthermore, our observations reveal that memory also drives the network formation process, and cooperators assort more, with longer link lifetimes, the longer the past actions record. Our analysis demonstrates, for the first time, that reputation
can be very well quantified as a weighted mean of the fractions of past cooperative acts and the last action performed.

 J. A. Cuesta, C. Gracia-Lázaro, A. Ferrer, Y. Moreno, and A. Sánchez, “Reputation drives cooperative behaviour and network formation in human groups”, Scientific Reports  5:7843, doi:10.1038/srep07843 (2015).

4.- Can we define behavioral phenotypes when humans face dyadic dilemmas or games?
Socially relevant situations that involve strategic interactions are widespread among animals and humans alike. To study these situations, theoretical and experimental works have adopted a game-theoretical perspective, which has allowed to obtain valuable insights about human behavior. However, most of the results reported so far have been obtained from a population perspective and considered one specific conflicting situation at a time. This makes it difficult to extract conclusions about the consistency of individuals’ behavior when facing different situations, and more importantly, to define a comprehensive classification of the strategies underlying the observed behaviors. Here, we present the results of a lab-in-the-field experiment in which subjects face four different dyadic games, with the aim of establishing general behavioral rules dictating individuals’ actions.
By analyzing our data with an unsupervised clustering algorithm, we find that all the subjects conform, with a large degree of consistency, to a limited number of behavioral phenotypes (Envious, Optimist, Pessimist, and Trustful), with only a small fraction of undefined subjects. We also discuss the possible connections to existing interpretations based on a priori theoretical approaches. Our findings provide a relevant contribution to the experimental and theoretical efforts towards the identification of basic behavioral phenotypes in a wider set of contexts without aprioristic assumptions regarding the rules or strategies behind actions. From this perspective, our work contributes to a fact-based approach to the study of human behavior in strategic situations, that could be applied to simulating societies, policy-making scenario building and even for a variety of business applications.

  J. Poncela-Casasnovas, M. Gutierrez-Roig, C. Gracia-Lazaro, J. Vicens, J. Gomez-Gardenes, J. Perello, Y. Moreno, J Duch, and A. Sanchez, «Humans display a reduced set of consistent behavioral phenotypes in dyadic games», Science Advances 2, e1600451 (2016).

5.- What is the role of onymity in the emergence and evolution of cooperation?
One of the most elusive scientific challenges for over 150 years has been to explain why cooperation survives despite being a seemingly inferior strategy from an evolutionary point of view. Over the years, various theoretical scenarios aimed at solving the evolutionary puzzle of cooperation have been proposed, eventually identifying several cooperation-promoting mechanisms: kin selection, direct reciprocity, indirect reciprocity, network reciprocity, and group selection. Here, we report the results of repeated Prisoner’s Dilemma experiments with anonymous and onymous pairwise interactions among individuals.
We find that onymity significantly increases the frequency of cooperation and the median payoff per round relative to anonymity. Furthermore, we also show that the correlation between players’ ranks and the usage of strategies (cooperation, defection, or punishment) underwent a fundamental shift, whereby more prosocial actions are rewarded with a better ranking under onymity. Our findings prove that reducing anonymity is a valid promoter of cooperation, leading to higher payoffs for cooperators and thus suppressing an incentive–anonymity–that would ultimately favor defection.

 Z. Wang, M. Jusup, R.-W. Wang, L. Shi, Y. Iwasa, Y. Moreno and J. Kurths, «Onymity promotes cooperation in social dilemma experiments», Science Advances 3:e1601444 (2017). Supplementary Material  .

6.- How do trading networks work?
Global supply networks in agriculture, manufacturing, and services are a defining feature of the modern world. The efficiency and the distribution of surpluses across different parts of these networks depend on the choices of intermediaries. This paper conducts price formation experiments with human subjects located in large complex networks to develop a better understanding of the principles governing behavior. Our first finding is that prices are larger and that trade is significantly less efficient in small-world networks as compared to random networks.
Our second finding is that location within a network is not an important determinant of pricing. An examination of the price dynamics suggests that traders on cheapest — and hence active — paths raise prices while those off these paths lower them. We construct an agent-based model (ABM) that embodies this rule of thumb. Simulations of this ABM yield macroscopic patterns consistent with the experimental findings. Finally, we extrapolate the ABM on to significantly larger random and small-world networks and find that network topology remains a key determinant of pricing and efficiency.

 F. M. Cardoso, C. Gracia-Lázaro, F. Moisan, S. Goyal, A. Sánchez, and Y. Moreno, «Trading in Complex Networks», submitted for publication (2019).

7.- Do we identify, and eventually avoid, tipping points? Can we optimally exploit a common resource?
Common-pool resources require a dose of self-restraint to ensure sustainable exploitation, but this has often proven elusive in practice. To understand why, and characterize behaviors towards ecological systems in general, we devised a social dilemma experiment in which participants gain profit from harvesting a virtual forest vulnerable to overexploitation. Out of 16 Chinese and 15 Spanish player groups, only one group from each country converged to the forest’s maximum sustainable yield.
All other groups were overzealous, with about half of them surpassing or on the way to surpass a no-recovery threshold. Computational-statistical analyses attribute such outcomes to an interplay between three prominent player behaviors, two of which are subject to decision-making «inertia» that causes near blindness to the resource state. These behaviors, being equally pervasive among players from both nations, imply that the commons fall victim to behavioral patterns robust to confounding factors such as age, education, and culture.

M. Jusup, F. M. Cardoso, C. Gracia Lazaro, L. Chen, Z. Wang, and Y. Moreno, «Robust behavioral patterns behind the demise of the commons», submitted for publication (2019). [Preprint available upon request.]