Summary: A new MIT study reshapes how we understand cooperation. It shows that keeping an internal ledger of favors is not the default human approach; rather, it is a strategy people use mainly among equals. In relationships with power differences, people tend to follow established precedents instead of tracking exact reciprocity.
Moving beyond classic game-theory setups that pair anonymous strangers, the researchers embedded realistic social roles and relationships into coordination tasks. Their experiments reveal a consistent pattern: among peers, people often use turn-taking and strict reciprocity to preserve equality. In contrast, in unequal or role-based relationships (for example, manager–employee or older–younger siblings), people stop mentally balancing exchanges and instead expect a one-way pattern of generosity to continue. Once generosity flows in a particular direction within a hierarchy, observers assume it will keep flowing that way because it reduces cognitive effort and clarifies social roles.
Key Facts
- Reassessing game-theory results: Standard behavioral-economics experiments typically pair anonymous strangers, which makes turn-taking reciprocity appear universal. This study shows that those findings reflect the artificial absence of social context.
- Reciprocity is cognitively costly: Tracking whose turn it is to return a favor requires active mental bookkeeping. People invest that effort mainly when they need to preserve an equal status relationship.
- Precedent replaces scorekeeping in hierarchies: In asymmetric relationships, people expect whichever party first performs a generous act to continue doing so, rather than expecting the recipient to reciprocate.
- Precedent can run either way: The established pattern of generosity can flow from higher to lower status or from lower to higher status; the critical factor is the precedent itself, not its direction.
- Social structure is reinforced by exchanges: Gift-giving and favors function as social instruments that maintain and signal relationship roles. Following an established pattern stabilizes expectations and reduces the need for continual negotiation.
- Toward computational models: The team is building mathematical and computational models to weigh variables such as benefits, relationship type, and cultural norms to predict when reciprocity versus precedent will occur.
- Diverse funding and focus: The project received support from the Simons Foundation Autism Research Initiative and the Patrick J. McGovern Foundation, enabling study of how social expectations vary across neurodiverse populations.
Source: MIT
Everyday reciprocity, and its limits
We commonly expect a favor to be returned: if a friend buys you coffee today, you might buy the next round. Such reciprocal patterns have been documented extensively in behavioral economics. But ethnographers and anthropologists have long pointed out that in many real-world relationships—especially those with unequal status—exact tit-for-tat reciprocity is uncommon.
MIT researchers have now demonstrated experimentally that small shifts in relational context profoundly change what people expect after an act of generosity. In relationships marked by asymmetry, observers predict that generosity will continue from the same actor rather than be returned by the recipient. For instance, if a professor routinely pays for a student’s coffee, people expect that to keep happening. Conversely, if a junior person repeatedly assists a superior with a task, that aid can become a stable expectation.
The authors suggest that keeping track of whose turn it is to reciprocate is the exception rather than the rule. It is an extra cognitive effort we choose to make when preserving equality matters; otherwise, following precedent is a simpler default that permits smoother social functioning.
“In many close or role-based relationships, you don’t expend the effort to keep track of turns,” says Rebecca Saxe, professor of Brain and Cognitive Sciences and co‑author of the study. “Following an established precedent is easier: everyone knows what to expect without monitoring past exchanges.”
The study appears in the journal Open Mind, with graduate student Alicia Chen as lead author and Saxe as senior author.
How the experiments worked
Most prior generosity experiments used strangers in decontextualized games. To measure the influence of social context, the MIT team presented participants with short scenarios and vignettes that described interactions between people in either symmetric (equal) or asymmetric (unequal) relationships. Some vignettes specified roles such as aunt–niece or manager–employee. Participants then predicted how the next interaction would unfold after a generous act occurred.
Across these scenarios, participants expected reciprocity when the relationship was symmetric—among friends, peers, or equals. When relationships were asymmetric, they instead expected the same actor to continue performing the generous act, regardless of whether the actor was higher or lower in rank.
The results suggest that strict reciprocity in laboratory studies may reflect expectations among strangers or equals, but not the wider variety of relationships people inhabit daily.
Why precedent matters
Precedent both reduces cognitive load and helps define roles. When an exchange pattern becomes established, it communicates what each person’s role is within that relationship, removing the need for constant renegotiation. Anthropologists have long described gift-giving as a means to build and maintain social structures; this study shows how expectations of continued generosity function as a mechanism for stabilizing those structures.
“Following a precedent can actively preserve relationships and hierarchies when the asymmetry reflects an underlying social reality,” Saxe explains.
The research team is now developing computational frameworks that integrate factors such as individual benefit, relationship type, and cultural norms, allowing quantitative comparison of competing theories about when people will reciprocate versus follow precedent. These models can test which variables most strongly predict behavior and help explain variations across contexts.
Funding:
This work was funded by the Simons Foundation Autism Research Initiative and the Patrick J. McGovern Foundation.
Key Questions Answered:
A: Because the mind favors cognitive efficiency. Tracking turns to keep exchanges perfectly balanced requires ongoing mental effort. In unequal relationships—such as between a boss and an employee—people default to following an established pattern of behavior. That precedent becomes the practical rule of the relationship, saving the cognitive cost of monitoring every past interaction.
A: No. A key finding is that precedent can lock in either direction. If a senior person begins paying for others, people expect that to continue. If a junior person routinely provides help or services, that pattern can become the ongoing expectation. The crucial factor is the initial precedent, not which direction it takes.
A: It challenges the idea that tit-for-tat reciprocity is a universal human default. Laboratory findings of strict reciprocity reflect interactions among strangers or equals; once social roles and hierarchies enter the picture, people tend to rely on precedent instead. This broadens our understanding of cooperation beyond the simplified contexts of many classic experiments.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by staff.
About this social neuroscience research news
Author: Sarah McDonnell
Source: MIT
Contact: Sarah McDonnell – MIT
Image: Image credited to Neuroscience News
Original Research: Open access. “Expectations of Reciprocal Generosity Are Specific to Equal Relationships” by Alicia M. Chen and Rebecca Saxe. Open Mind. DOI: 10.1162/OPMI.a.357
Abstract
Expectations of Reciprocal Generosity Are Specific to Equal Relationships
When someone behaves generously, what do observers expect next? Classic theories stress reciprocity—the idea that generosity will be returned. Ethnographic evidence, however, often shows that generosity creates a precedent, leading people to expect the same actor to act generously again. Across six online behavioral experiments (total N = 599 U.S. adults) using third‑party vignettes and first‑person incentivized games, the study tests when generosity leads to expectations of reciprocation versus expectations of continued generosity by the same actor. Participants expected reciprocity only in equal or symmetric relationships; in asymmetric relationships, they expected the original actor to continue the generous behavior. These expectations held across roles, contexts, and cost structures, suggesting that evidence for strict reciprocity may reflect a subset of interactions among strangers or equals rather than the broader array of everyday social relationships.