There is a finding in the regret research that surprises most people when they hear it. The conversations adults in their seventies most consistently regret are not, in most cases, the ones they remember having badly. They are not the angry exchanges, the words said in haste, the responses they wish they had handled differently. The substantive regrets, on the available data, are about the conversations they never had. The small ordinary ones they kept postponing across years and decades, often without noticing they were postponing them, until the opportunity for them quietly closed.
This pattern is among the more carefully documented findings in the empirical psychology of regret. The basic shape, that regrets of inaction come to outweigh regrets of action across time, has been replicated across cultures, across age groups, and across multiple methodologies. By the time a person is in their seventies, the temporal pattern has had its full effect. The conversations they regret most are the ones that, at the time, did not seem urgent enough to have.
We are writers and parents, not researchers in the psychology of regret or aging. What follows is a reading of the empirical work on action and inaction regret and the related qualitative research on late-life regret, not life advice. The article describes a pattern documented in the research; it does not prescribe how anyone should feel about the conversations they have or have not had.
What the regret research actually shows
The foundational empirical work on this comes from the Cornell psychologist Thomas Gilovich and his collaborator Victoria Medvec. In a 1995 paper in Psychological Review titled “The experience of regret: what, when, and why,” Gilovich and Medvec argued that regrets follow a particular temporal pattern. Actions, or errors of commission, generate more regret in the short term. Inactions, or errors of omission, produce more regret in the long run.
The 1995 paper proposed several mechanisms for why this happens. The pain of regrettable actions diminishes relatively quickly, partly because the person can rationalize, contextualize, and integrate the action into their understanding of who they are. The pain of regrettable inactions, by contrast, tends to grow over time, partly because the imagined alternatives expand. The person who said something they regret in anger has, in most cases, a single specific moment to come to terms with. The person who never said something at all has, by contrast, an open-ended set of imagined conversations that might have happened.
Gilovich and colleagues have replicated and extended the finding across multiple studies. In a 2003 paper in the Journal of Cross-Cultural Psychology, they showed that the pattern holds across the United States, China, Japan, and Russia. The temporal preponderance of inaction regrets over action regrets is, on the available evidence, a fairly universal feature of how human regret works across time.
What the conversations actually are
The specific conversations the late-life research has documented as the most regretted tend to share a structural feature.
They are not, in most accounts, dramatic conversations. They are small. The person who never told their father what their relationship had meant. The sibling who never asked the older sibling what the family had really been like. The friend who never said why the friendship had mattered. The adult child who never raised, with the parent who was still alive at the time, the conversation about what had been difficult between them. The parent who never told the adult child the things they had been quietly proud of. The conversations have a particular ordinariness. They are not the conversations a person would have planned to have. They are the conversations that, at the time, could have happened in five minutes if either party had thought of it.
The Cornell gerontologist Karl Pillemer’s Legacy Project, which interviewed over 1,200 Americans aged 65 and older for his book 30 Lessons for Living, documented this directly. The conversations the interviewees described regretting were, in most cases, ordinary in their content. The regret was not, in most cases, about the missed opportunity to make a grand statement. It was about the missed opportunity to say something small and true that the other person could have heard.
Why we postpone them
The reason these conversations get postponed is, in the research, fairly simple.
They do not, at the time, present themselves as urgent. The opportunity is not visibly closing. The other person is not, in the moment, going anywhere. The conversation can, by every available logic, happen next week, next month, when the family is together again, when there is more time, when the circumstances are right. The postponing is not, in most cases, a deliberate avoidance. It is the structural consequence of the assumption that the opportunity will remain available.
What the regret research suggests is that this assumption is the substantive variable. The conversations that get had are the ones the person believed they had to have. The conversations that get postponed are the ones the person believed they could safely defer. The deferral is, in many cases, indefinite. The closure of the opportunity, when it arrives, is rarely announced. The other person becomes ill. The relationship becomes strained. The person dies. The window simply ends. The conversation does not, in any of these cases, retroactively become possible.
What this might mean for the reader currently postponing
The research is not, in any clean sense, a prescription. It does not say that anyone currently postponing a conversation should have it immediately. Many conversations are reasonably postponed. Some are postponed because the timing is, in some cases, actually wrong. Others are postponed because the work of having them is, at this moment in the person’s life, more than they have available to do.
What the research does suggest, with reasonable consistency, is that the assessment the postponer is making about which conversations matter most is likely to look different from their seventy-year-old self’s perspective. The conversations that feel optional now are often the ones that, in retrospect, feel substantive. The conversations that feel difficult now are often the ones that, in retrospect, would have been worth the difficulty. This is not, in itself, a comfort to anyone deciding what to do about it. It is information about how the assessment tends to evolve.
Most cultural conversations about regret still focus on the things people did and wish they had not done. The empirical work on late-life regret consistently points in the other direction. The substantive long-term regrets are about the things never done, and most of those are about specific small conversations that were never had. The assumption that the opportunity will remain available is, in the research, the structural feature that produces the regret. The assumption is rarely tested while it is still true.
