The authors introduce the institutional stance as the mental process of representing other people with respect to their roles rather than their individual minds (as unfolds in the mentalistic stance). In this framework, an institution is “a type of cognitive representation consisting of a set of inter-connected roles with a level of persistence and continuity,” which may or may not be formally codified. Here I extend the application of the institutional stance analysis to marriage, which may not be an obvious target given that many of the vivid examples in the target article are hierarchical (including parent–child structures) or occupation-bound. My argument is that some institutions – here, marriage, or monogamous long-term romantic partnerships, more broadly – support the creation of such well-articulated models and role-bound expectations of our counterparts, that the roles come to replace the person in front of us, and in so doing, strip them of their full humanity.

At least two features of long-term partnerships could contribute to this phenomenon. First, it may be that over time, the role-based models we develop of our partners are sufficient to support coordinated action in the context of this institution (e.g., division of labor in the home). In the absence of any major system failures, the motivation to devote continued attention and processing resources – as is required by theory of mind (Saxe, Reference Saxe2006) – may decline. Second, the demands of life beyond the partnership may constrain the cognitive resources we have available to engage in the costly process of continuous mentalizing. Thus, even when partners are motivated, they may not have sufficient bandwidth to follow through on adopting a mentalistic stance. The challenge is that circumstances and people change; the models we develop for our partners-as-roles may degrade in their fitness over time as models of our partners-as-minds. In this way, we cease to see clearly those ostensibly closest to us.

This idea is inspired by work on the Continuum Model of impression formation (Fiske & Neuberg, Reference Fiske, Neuberg and Zanna1990) and other related dual-process models of social cognition. According to the Continuum Model, the default representation we have when we meet a new person is largely based on the stereotypes we associate with them (e.g., on the basis of visual features, such as skin tone or size). However, with sufficient motivation and capacity, we can push past the stereotype to build more elaborate, individualized representations: filling the representations in with the person’s specific traits and attributes. Examples of conditions which activate the motive to elaborate include needing to build an accurate representation of the other in support of some aim (e.g., finding a study group partner) and personal relevance of the target for one’s own outcomes (e.g., determining whether someone might make a good roommate). Cognitive capacity simply refers to having sufficient mental resources to devote to the task of individuating. Accordingly, situational constraints like time pressure and stress undermine individuation, resulting in more automatic, stereotype-based judgments of others.

Critically, creating outcome dependency between people is one of the most robust interventions for activating individuation in lab experiments (Fiske, Lin, & Neuberg, Reference Fiske, Lin, Neuberg, Chaiken and Trope1999). Are not long-term relationships the paradigmatic instance of outcome dependency? While romantic relationships may be incredibly motivating with respect to engaging individuating processes, including theory of mind, early on, these more effortful processes may give way over time to habits, rules, and routines within the context of a long-term relationship. Add to this the challenges and stresses of daily life – work and domestic demands, raising children, taking care of sick parents and pets – and the likelihood of relying on role-based models and the institutional stance (not unlike stereotypes) only increases.

There are two important implications of the Continuum Model and other related dual-process models of social cognition, more generally (Chaiken & Trope, Reference Chaiken and Trope1999). First, because individuation requires both motivation and capacity, category-based cognitive processing tends to prevail. Second, people attend more to category-consistent than inconsistent information. Therefore, within the marriage case study, once role-based expectations are well established, the institutional stance may dominate mentalizing processes, making it particularly hard to detect when and why our partner’s minds believe, desire, or prefer something different than the role-based model suggests.

This case study raises several interesting questions for the nature of the relationship between the institutional and mentalistic stances. First, is it the case that the institutional stance dominates the mentalistic stance in long-term relationships (in the sense that it is adopted more readily as a default)? Second, are institutional and mentalizing stances mutually inhibiting? And related, if there is a bistability in the representation of our partners along these lines, what triggers top-down control over which representation is appropriate or active in a given context? And finally, are there any relationships with a minimal level of persistence and continuity that do not eventually become institutions by this definition?