Neural autopilot and context-sensitivity of habits

This paper is about the background of two new ideas from neuroeconomics for understanding habits. The main idea a two-process ‘neural autopilot’ model. model hypothesizes that contextually cued habits occur when reward habitual behavior numerically reliable (as in related models with an ‘arbitrator’). computational lightly parameterized, has essential ingredients established animal learning and cognitive neuroscience, simple enough to make nonobvious predictions. An interesting set predictions how consumers react different kinds changes prices qualities goods (‘elasticities’). Elasticity analysis expands habit marker insensitivity devaluation, other types sensitivities. second use machine discover which contextual variables seem cue habits, field data.