- Always exclusively respond with "OPTION 1" or "OPTION 2". < remaining prompts unchanged > Experimental sessions For each of the GPT - GPT treatments, we had 200 groups of 2 instances of GPT, interacting over 10 announced periods. The treatments involving human participants were run in the Cologne Laboratory for Economic Research in August 2023, using our tool, with oTree (D. L.
Paper not yet in RePEc: Add citation now
Chen, Y., Liu, T. X., Shan, Y., and Zhong, S. (2023). “The Emergence of Economic Rationality of GPT”. arXiv preprint arXiv:2305.12763.
- Cowen, T. and Tabarrok, A. T. (2023). “How to learn and teach economics with large language models, including GPT”. Including GPT (March 17, 2023).
Paper not yet in RePEc: Add citation now
Dreber, A., Ellingsen, T., Johannesson, M., and Rand, D. G. (2013). “Do people care about social context? Framing effects in dictator games”. Experimental Economics, 16, pp. 349–371.
- Duffy, J., Hopkins, E., and Kornienko, T. (2021). Facing the Grim Truth: Repeated Prisoner’s Dilemma Against Robot Opponents. Tech. rep. Working Paper.
Paper not yet in RePEc: Add citation now
Dufwenberg, M., Gächter, S., and Hennig-Schmidt, H. (2011). “The framing of games and the psychology of play”. Games and Economic Behavior, 73(2), pp. 459–478.
- Engel, C. and Rand, D. G. (2014). “What does “clean” really mean? The implicit framing of decontextualized experiments”. Economics Letters, 122(3), pp. 386–389.
Paper not yet in RePEc: Add citation now
Fischbacher, U., Gächter, S., and Fehr, E. (2001). “Are people conditionally cooperative? Evidence from a public goods experiment”. Economics letters, 71(3), pp. 397–404.
- Gamma, E., Helm, R., Johnson, R., and Vlissides, J. (1994). Design Patterns: Elements of Reusable Object-Oriented Software. Addison-Wesley professional computing series. Reading, Mass: Addison-Wesley.
Paper not yet in RePEc: Add citation now
- GPT - GPT June 19, 2023 https://osf.io/zepfr/?view_only=2dbfc4a483a844bd97947eddcbb2fe25 1. In a 2x2 prisoner’s dilemma with binary action space that is represented with only the payoffs, cooperation is more frequent than if the game is framed. 2. If a 2x2 prisoner’s dilemma with binary action space is framed as a conflict, there is less cooperation than if the dilemma is framed as a joint project. 3. If a 2x2 prisoner’s dilemma with binary action space is framed such that gains from cooperation only obtain in expectation, not necessarily in realization, there is less cooperation.
Paper not yet in RePEc: Add citation now
- GPT - Human July 27, 2023 https://osf.io/njuq5/?view_only=6d77255463784e3396769d4ffcf92afb 1. (within the new wave) In a 2x2 prisoner’s dilemma with binary action space that is represented with only the payoffs, cooperation is more frequent than if the game is framed as either having a joint enemy, or competing against each other. 2. (across waves) If GPT interacts with another instance of GPT, in a 2x2 prisoner’s dilemma with binary action space, a) the cooperation rate of GPT differs whether it interacts with another instance of GPT, or with a human participant b) the cooperation rate of human players differs from the cooperation rate of GPT if two instances of GPT interact 3. (interaction) The effect of frames differs between GPT interacting with each other vs. GPT interacting with a human second mover.
Paper not yet in RePEc: Add citation now
- Gu, J., Han, Z., Chen, S., Beirami, A., He, B., Zhang, G., Liao, R., Qin, Y., Tresp, V., and Torr, P. (2023). “A systematic survey of prompt engineering on vision-language foundation models”. arXivpreprintarXiv:2307.12980.
Paper not yet in RePEc: Add citation now
- Guo, F. (2023). “GPT Agents in GameTheory Experiments”. arXiv preprint arXiv:2305.05516.
Paper not yet in RePEc: Add citation now
Horton, J. J. (2023). Large language models as simulated economic agents: What can we learn from homo silicus? Tech. rep. National Bureau of Economic Research.
- Johnson, T. and Obradovich, N. (2022). “Measuring an artificial intelligence agent’s trust in humans using machine incentives”. arXiv preprint arXiv:2212.13371.
Paper not yet in RePEc: Add citation now
- Kasberger, B., Martin, S., Normann, H.-T., and Werner, T. (2023). “Algorithmic Cooperation”. Available at SSRN 4389647.
Paper not yet in RePEc: Add citation now
- Kühberger, A. (1998). “The influence of framing on risky decisions: A meta-analysis”. Organizational behavior and human decision processes, 75(1), pp. 23–55.
Paper not yet in RePEc: Add citation now
- Mei, Q., Xie, Y., Yuan, W., and Jackson, M. O. (2023). “A Turing Test: Are AI Chatbots Behaviorally Similar to Humans?” SSRN. doi: 10.2139/ ssrn.4637354.
Paper not yet in RePEc: Add citation now
Mengel, F. (2018). “Risk and Temptation: A Meta-study on Prisoner’s Dilemma Games”. The Economic Journal, 128(616), pp. 3182–3209.
- Ockenfels, A. (1999). “Fairness, Reziprozität und Eigennutz”. Ökonomische Theorie und experimentelle Evidenz, Tübingen.
Paper not yet in RePEc: Add citation now
- OpenAI (2023). “GPT-4 technical report”. arXiv preprint arXiv:2303.08774.
Paper not yet in RePEc: Add citation now
- Our oTree add-on implements the well-known façade pattern (Gamma et al., 1994: sec. 4.5).
Paper not yet in RePEc: Add citation now
Phelps, S. and Russell, Y. I. (2023). “Investigating emergent goal-like behaviour in large language models using experimental economics”.
- REFERENCES 17 Levin, I. P., Schneider, S. L., and Gaeth, G. J. (1998). “All frames are not created equal: A typology and critical analysis of framing effects”. Organizational behavior and human decision processes, 76(2), pp. 149– 188.
Paper not yet in RePEc: Add citation now
- Russell, S. (2019). Human compatible: Artificial intelligence and the problem of control. Penguin.
Paper not yet in RePEc: Add citation now
Smith, V. L. (1976). “Experimental economics: Induced value theory”. The American Economic Review, 66(2), pp. 274–279.
- To set an attribute on all Threads of a Conversation, we use convo.all.var, not merely convo.var. Thus, we only provide a composite-like interface, not a true composite (Gamma et al., 1994: sec. 4.3). We distinguish between attributes set on the Conversation and its Threads because Python does not expose a complete Proxy-like mechanism as does JavaScript, and Threads indeed each obtain their own reference to the underlying object, or a copy if the object is sufficiently simple. For Table 3, we used Polars (www.pola.rs) to create a DataFrame from the returned list of dicts, after which we grouped by politician and time and summarized to obtain GPT-4’s mean belief about the public approval rating. All data is returned from the shorthand as a “list of dicts” to be processed by other software or exported to any spreadsheet format.
Paper not yet in RePEc: Add citation now
- Tsuchihashi, T. (2023). “Do AIs Dream of Homo Economicus? Answers from ChatGPT”. Answers from ChatGPT (June 29, 2023).
Paper not yet in RePEc: Add citation now
- Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). “Attention is all you need”. Advances in neural information processing systems, 30.
Paper not yet in RePEc: Add citation now