Economic experiments in virtual reality
The experiment, as a tool for scientific research, was proposed for use in economics 70 years ago. Since then, the attitude to economic experiments has changed and has ceased to be the same unequivocal.
At the moment there are two opinions, for which you need to use behavioral experiments:
Relatively recently, the focus on the use of laboratory methods in economics has shifted from content to context. On the one hand, it is impossible to carry out a “clean” experiment without the intervention of external circumstances, since the laboratory is not a neutral structure, in which there are no unspoken rules and factors. On the other hand, behavioral and cognitive approaches in economics cannot give reasonable conclusions about the validity of a particular theory without taking into account how the context of the experiment affects behavior.
In fact, the basic principle of cognitive psychology states that all forms of activity and decision making cannot be carried out independently of the environment. So, in order to check the accuracy of the behavioral model, it is necessary to give contextual and social signals to the test subject, which will cause associations and initiate the personal heuristic abilities of the person.
')
With the development of virtual reality technology (BP), an obvious question arose whether BP could be used for behavioral experiments in economics. This proposal was made primarily because BP provides precise control over the conditions of the reproducible situation, helps to minimize external factors and guarantees anonymity.
BP is classified depending on the depth of immersion of a person:
To answer this question, consider two criteria that worry economists:
Thomas Chesney (2009) was one of the first researchers who, in his work, tried to check whether the laboratory results of a behavioral experiment coincide with those obtained in BP. The study was based on Second Life - a game, a three-dimensional virtual world with elements of a social network where you can create your own world and visit the worlds of other users. Chesney concluded that economic decisions in the virtual world are based on similar mechanisms as those observed in standard conditions.
Similar results were obtained by Martin Spenn, when he tested the game dictator based on the game World of Warcraft. The game of dictator is a classic experiment, developed by Daniel Kahneman, to test the theory of “homo economicus” - an economic, rational man. The essence of the experiment is that the “dictator” has the right to decide how much of the money offered will go to him, and how much to another participant. This experiment was conducted many times with some changes in conditions. For example, when both participants see or do not see each other, when only a “dictator” can see another participant, etc. Spenn's experiment showed a higher proportion of altruism in BP than in the laboratory.
It is worth noting the work of John Duffy "Trust in second life", where he focuses the attention of researchers on the fact that people in a simulation game are often prone to falsifying personal data such as gender and age, which casts doubt on the accuracy of the results of experiments in BP.
However, more recent studies (Sasha Fulbrunn in 2011, Ben Greiner in 2014) confirm that BP allows us to obtain results similar to laboratory tests. From this it follows that virtual platforms can be used to study interpersonal relationships and interactions in the economic spectrum, provided that a virtual platform for the experiment is carefully prepared. Moreover, to ensure the validity of the results when using low-immersion BP (LIVE), it is necessary to determine in advance which environmental parameters correspond to the objectives of the experiment.
We now turn to the next question.
In other words, how accurately BP allows to get not only the results themselves, but also to reveal cause-effect relationships.
Andreas Niklisch and Tobias Salz in World of Warcraft analyzed the pay-to-service exchange between participants with different status (game level, social status, etc.), reproducing a similar experiment of the German economist Armin Falk. Scientists have easily proven that the “social” labels of exchange participants have a great influence on their behavior and decisions and cannot be so easily undone.
Edward Kastronova and his colleagues conducted research in the game EverQuest2 to check whether the virtual economic behavior coincides with the real world.
After analyzing a large database of economic transactions in the game, the researchers came to the conclusion that the decisions taken by the participants of the virtual market regarding the production, consumption and sale of various values ​​(food, clothing, objects and materials) are consistent with those observed in the world. Scientists also note that impermanence and intermittent economic decisions in the game repeat the dynamics of the economies of developing countries and the black market.
Yannick Ferreira de Souza and Alistair Munro in 2012 decided to check how much the effect of ownership in the virtual world is using the example of the game Runescape. The effect of possession, noticed by Richard Thaler in the 1970s, characterizes the psychological phenomenon when a person values ​​things that he owns more than what he could get. De Souza and Murno compared the laboratory indicators of the behavior of experienced players and novices with those obtained in the virtual world. Contrary to the expectations of scientists, the results revealed a strong decrease in the effect of ownership for experienced players, while for beginners the results remained almost unchanged.
Peter Twig and Kevin McCabe (2014) tested the experiment of Marco Janssen and Elinor Ostrom in BP: the formation of private property rights in the context of renewable public resources. The experiment participants interacted in a virtual environment, where, as in the original, it is possible to produce heterogeneous natural resources, in the case of Twig and McCabe these were berries of different colors. Conditions also imply that the price and costs of each type of berries are different from each other, and the purpose of the participants is to maximize profits. As expected, the more the participants interacted with each other exercising control and specialization, the more likely the creation of effective private property rights in the group.
These results show how virtual worlds are able to provide a convenient platform for testing the relationships between social structures and the economic system. Nevertheless, the specifics and social features of the environment in which the experiment is conducted should be taken into account.
The data provided provide strong support for the use of BP in economic experiments. Scientists especially pin their hopes on the use of BP with a high level of immersion, where the first, perhaps awkward, attempts to conduct economic experiments have already been made (Nicole de Horace in 2015, Gurel Ozgur in 2016).
Despite the still low use of BP (HIVE) in purely economic experiments, the results of research in the field of cognitive psychology and other social sciences confirm that a higher level of immersion and involvement in the process reinforces the external validity of the study through improved control of the environment and the task itself . In my opinion, the day is not far off when BP (HIVE) will be a common method of economic experiments and simulations of economic models and theories. With proper control and care, BP can become an indispensable tool not only in behavioral economics, but also in other areas of science.
Used sources:
At the moment there are two opinions, for which you need to use behavioral experiments:
- To find out the validity of abstract models,
- Find out if the results of laboratory experiments can be applied to the world as a whole.
Relatively recently, the focus on the use of laboratory methods in economics has shifted from content to context. On the one hand, it is impossible to carry out a “clean” experiment without the intervention of external circumstances, since the laboratory is not a neutral structure, in which there are no unspoken rules and factors. On the other hand, behavioral and cognitive approaches in economics cannot give reasonable conclusions about the validity of a particular theory without taking into account how the context of the experiment affects behavior.
In fact, the basic principle of cognitive psychology states that all forms of activity and decision making cannot be carried out independently of the environment. So, in order to check the accuracy of the behavioral model, it is necessary to give contextual and social signals to the test subject, which will cause associations and initiate the personal heuristic abilities of the person.
')
With the development of virtual reality technology (BP), an obvious question arose whether BP could be used for behavioral experiments in economics. This proposal was made primarily because BP provides precise control over the conditions of the reproducible situation, helps to minimize external factors and guarantees anonymity.
The virtual reality
- This is a computer-generated environment that creates an artificial location and allows a person to make decisions and interact in real time, affecting two or more organs of perception.
BP is classified depending on the depth of immersion of a person:
- Low-Immersive VR (Low-immersive virtual environments, LIVE) based on a screen interface where users interact through graphical models. These include simulation games, such as The Sims, Second Life or World of Warcraft;
- High-immersive VR (high-immersive virtual environments, HIVE) using special devices (Oculus Rift, Samsung Gear VR, Google Cardboard, and others) that cover more channels of perception.
Is it possible to use BP to conduct economic experiments?
To answer this question, consider two criteria that worry economists:
- Is the behavior of the subjects in the BP and the laboratory very different?
- Can BP provide the necessary environment?
Thomas Chesney (2009) was one of the first researchers who, in his work, tried to check whether the laboratory results of a behavioral experiment coincide with those obtained in BP. The study was based on Second Life - a game, a three-dimensional virtual world with elements of a social network where you can create your own world and visit the worlds of other users. Chesney concluded that economic decisions in the virtual world are based on similar mechanisms as those observed in standard conditions.
Similar results were obtained by Martin Spenn, when he tested the game dictator based on the game World of Warcraft. The game of dictator is a classic experiment, developed by Daniel Kahneman, to test the theory of “homo economicus” - an economic, rational man. The essence of the experiment is that the “dictator” has the right to decide how much of the money offered will go to him, and how much to another participant. This experiment was conducted many times with some changes in conditions. For example, when both participants see or do not see each other, when only a “dictator” can see another participant, etc. Spenn's experiment showed a higher proportion of altruism in BP than in the laboratory.
It is worth noting the work of John Duffy "Trust in second life", where he focuses the attention of researchers on the fact that people in a simulation game are often prone to falsifying personal data such as gender and age, which casts doubt on the accuracy of the results of experiments in BP.
However, more recent studies (Sasha Fulbrunn in 2011, Ben Greiner in 2014) confirm that BP allows us to obtain results similar to laboratory tests. From this it follows that virtual platforms can be used to study interpersonal relationships and interactions in the economic spectrum, provided that a virtual platform for the experiment is carefully prepared. Moreover, to ensure the validity of the results when using low-immersion BP (LIVE), it is necessary to determine in advance which environmental parameters correspond to the objectives of the experiment.
We now turn to the next question.
Can BP provide sufficient conditions and environment
In other words, how accurately BP allows to get not only the results themselves, but also to reveal cause-effect relationships.
Andreas Niklisch and Tobias Salz in World of Warcraft analyzed the pay-to-service exchange between participants with different status (game level, social status, etc.), reproducing a similar experiment of the German economist Armin Falk. Scientists have easily proven that the “social” labels of exchange participants have a great influence on their behavior and decisions and cannot be so easily undone.
Edward Kastronova and his colleagues conducted research in the game EverQuest2 to check whether the virtual economic behavior coincides with the real world.
After analyzing a large database of economic transactions in the game, the researchers came to the conclusion that the decisions taken by the participants of the virtual market regarding the production, consumption and sale of various values ​​(food, clothing, objects and materials) are consistent with those observed in the world. Scientists also note that impermanence and intermittent economic decisions in the game repeat the dynamics of the economies of developing countries and the black market.
Yannick Ferreira de Souza and Alistair Munro in 2012 decided to check how much the effect of ownership in the virtual world is using the example of the game Runescape. The effect of possession, noticed by Richard Thaler in the 1970s, characterizes the psychological phenomenon when a person values ​​things that he owns more than what he could get. De Souza and Murno compared the laboratory indicators of the behavior of experienced players and novices with those obtained in the virtual world. Contrary to the expectations of scientists, the results revealed a strong decrease in the effect of ownership for experienced players, while for beginners the results remained almost unchanged.
Peter Twig and Kevin McCabe (2014) tested the experiment of Marco Janssen and Elinor Ostrom in BP: the formation of private property rights in the context of renewable public resources. The experiment participants interacted in a virtual environment, where, as in the original, it is possible to produce heterogeneous natural resources, in the case of Twig and McCabe these were berries of different colors. Conditions also imply that the price and costs of each type of berries are different from each other, and the purpose of the participants is to maximize profits. As expected, the more the participants interacted with each other exercising control and specialization, the more likely the creation of effective private property rights in the group.
These results show how virtual worlds are able to provide a convenient platform for testing the relationships between social structures and the economic system. Nevertheless, the specifics and social features of the environment in which the experiment is conducted should be taken into account.
In this way
The data provided provide strong support for the use of BP in economic experiments. Scientists especially pin their hopes on the use of BP with a high level of immersion, where the first, perhaps awkward, attempts to conduct economic experiments have already been made (Nicole de Horace in 2015, Gurel Ozgur in 2016).
Despite the still low use of BP (HIVE) in purely economic experiments, the results of research in the field of cognitive psychology and other social sciences confirm that a higher level of immersion and involvement in the process reinforces the external validity of the study through improved control of the environment and the task itself . In my opinion, the day is not far off when BP (HIVE) will be a common method of economic experiments and simulations of economic models and theories. With proper control and care, BP can become an indispensable tool not only in behavioral economics, but also in other areas of science.
Used sources:
- Chesney, Thomas, Swee – Hoon Chuah, and Robert Hoffmann, 2009. “Virtual world experimentation: An exploratory study.” Journal of Economic Behavior & Organization 72 (1), 618–635.
- Duffy, John, 2011. “Trust in second life.” Southern Economic Journal 78 (1), 53–62.
- Spann, Martin, Oliver Hinz, Il – Horn Hann, and Bernd Skiera, 2010. “Decision Making in Altruism, Fairness and Presence.” Proceedings of the European Conference on Information Systems (ECIS).
- Nicklisch, Andreas and Tobias Salz, 2008. "The Maximus Planck for Research on Collective Goods, N. 37.
- De Sousa, Yannick Ferreira and Alistair Munro, 2012. “Truck, barter and exchange field effect.” Journal of Economic Psychology 33 (3), 483–493.
- Twieg, Peter and Kevin McCabe, 2014. "The Determinants of Territorial Property Rights in a Spatial Commons Experiment." George Mason University, Department of Economics, WPN. 14–40.
- Alessandro Innocenti, 2017. “Virtual reality experiments”, Journal of Behavioral and Experimental Economics
Source: https://habr.com/ru/post/341128/
All Articles