Gcal: space of possible gamesg: a specific gameA(g): set of admissible actions in gamegpi: policypi_local: locally optimal policypi_alt: feasible alternative policy
x_t: aggregate state at timeto_t: future optionalityl_t: lock-inz_t: remaining state variablesF: state dynamicsPhi: permanence or game-change dynamics
r(x, a, g): local payoffU(x_t, a_t, g_t): instantaneous structural utilityJ^pi(x_0, g_0): discounted structural value of policypidelta: intertemporal discount factorlambda: optionality weightmu: lock-in weightc: generic costchi: explicit exit or reconfiguration costeta: weight of the explicit exit cost in the structural functional
exploit: local-gain-maximizing policypreserve: structurally prudent policykappa_E: optionality survival factor underexploitkappa_P: closing factor for the gap to full optionality underpreserveDelta r = r_exploit - r_preserveGamma: structural optionality premiumlambda_*: critical threshold for structural weight
p_stay(r): probability of remaining in the game given observed payoffalpha, beta, gamma: parameters of the logistic permanence functione: perceived exit costq: proxy for local competencetau_bad: time spent in the bad game