{"paper_id":"2507.08394","title":"Temperature Measurement in Agent Systems","authors":["Christoph J. Börner","Ingo Hoffmann"],"paper_date":"2025-07-11","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.08394v1","page_url":"https://thequant.space/flowcharts/temperature-measurement-in-agent-systems/","scores":{"empirical_rigor":2,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a measurement equation that makes the state variable temperature directly accessible by assessing the balance of conforming and non‑conforming decision tendencies in an agent system.","quote":"It examines an agent system with two decision options in a news environment, establishes the measurement equation, and outlines the basic concept of temperature measurement.","summary":"Models for spin systems, known from statistical physics, are applied analogously in econometrics in the form of agent-based models. The models discussed in the econophysics literature all use the state variable T, which, in physics, represents the temperature of a system. However, there is little…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The agent system is ideal (J = 0), i.e., there is no pairwise equilibrium benefit among agents.","type":"model","stated":"explicit","quote":"This describes the prerequisite for an ideal agent system and J = 0 must be set.","support":"full"},{"assumption":"The per‑agent benefit for conformity is set to µ = 1 USD.","type":"model","stated":"explicit","quote":"In the following, it is assumed that the outcome of the individual decision – correct (conform) or incorrect (non-conform) – of each participant (agent) has the same value for the experimenters (observer) and set this to µ = 1 USD.","support":"full"},{"assumption":"The information cost for an agent under maximum uncertainty is set to k = 1 USD.","type":"model","stated":"explicit","quote":"Similarly, it is assumed that for the experimenters the information about an 9 agent’s exact state when the agent system is in a state of maximum uncertainty has a ﬁxed value and this is set to k = 1 USD.","support":"full"},{"assumption":"The strength of the news environment is fixed at B_C1 = 1.0 for Choice 1 and B_C2 = 0.9 for Choice 2.","type":"model","stated":"explicit","quote":"For further analysis, we set the strength of the news environment BC1 = 1.0 for Choice 1 and BC2 = 0.9 for Choice 2.","support":"full"},{"assumption":"Thermal equilibrium holds between competing subsystems, i.e., T = T1 = T2.","type":"model","stated":"explicit","quote":"Assuming an equilibrium T = T1 = T2 (phys.: thermal equilibrium), then: µB k 1 M1 = µB k 1 M2 + 1 2 Jz k M2 = M1 1 1 −1 2 J µ z B M1 (10) If the Taylor expansion 1/(1 −x) = 1 + x + x2 + . . . for x ≈0 is used on the right-hand side of the last line of equation (10), the ﬁrst-order approximation is: M2 = M1 + 1 2 J µ z B M 2 1 (11) Equation (11) shows that in equilibrium of the agent systems,","support":"full"},{"assumption":"The surplus of decisions M is small (0 < M1, M2 ≪ 1) so that linear approximations (Equations 8 and 9) are applicable.","type":"model","stated":"explicit","quote":"It is further assumed, that the average surplus of the decisions in the two agent systems is small, 0 < M1, M2 ≪1, so that the Equations (8) and (9) can be applied.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Sun et al. (2022) human decision‑making experiment (Figure 1)","provider":null,"kind":"other","access":"unknown","period":null,"universe":null,"quote":"The latter data are depicted in Figure 1 of Sun 2022’s paper and serve as a basis for further processing in our analysis. Choice Scale Pixel ø Surplus Temperature No. T0 N+ N− M T 1 1.0 5869 2805 0.353 2.709 (blue full) (red doted) 2 0.9 6042 3137 0.316 2.746 (red full) (blue doted) Table 1: Analysis of the experiment conducted by Sun et al. (2022) with regard to the measurement of temperature T. The termperature is","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"none"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"Applying the measurement equation to the Sun et al. (2022) experiment yields a temperature of approximately T = 2.73 ± 0.03 for the ideal agent system.","kind":"theoretical","quote":"Therefore, the current tem-perature of T = 2.73 with a scattering width of ±0.03 would be used for this ideal agent system in further model applications.","support":"partial"},{"finding":"In equilibrium of competing subsystems, the average decision surplus is larger in the non‑ideal system (M2 > M1), and can be reduced by increasing the conformity benefit µ.","kind":"theoretical","quote":"the average sur-plus of decisions is smaller in the ideal agent system than in the non-ideal agent system: M2 > M1.","support":"full"},{"finding":"A consistent temperature measurement in a constant news environment can be obtained from the decision surplus M via the derived equation T = µ B k⁻¹ M.","kind":"theoretical","quote":"Equation (7) or (8) then shows that a consistent temperature measurement in a constant news environ-ment with strength B can be obtained by determining the surplus of decisions M. If there is hardly any beneﬁt from the pairwise equilibrium behavior of agents, then in the limiting case J →0 the agent system is called an ideal agent system (Bouchaud, 2013; Börner et al., 2023c) and it follows for the tem-perature","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":8,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":2,"extracted_at":"2026-11-14T19:49:04Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}