How much of what we call personality is actually stable, and how much is just context?

Personality psychology rests on a foundational assumption: that people have stable, consistent traits that predict how they will behave across different situations and over time. This assumption is both partly correct and more complicated than most personality frameworks admit.

The partly correct part first. Research over the past half century has established reasonably well that broad personality dimensions, the Big Five: openness, conscientiousness, extraversion, agreeableness, and emotional stability, do show meaningful stability over time, particularly after early adulthood. If you are markedly high in conscientiousness at thirty, you are likely to be higher than average at sixty. If you score low in agreeableness now, you probably did so ten years ago and will do so in ten years' time. Identical twins raised apart show more personality similarity than fraternal twins raised together, suggesting a substantial heritable component. The traits are real in the sense that they are measuring something consistent.

But the stability is not as absolute as popular accounts of personality suggest. Significant changes in personality, not just mood or behaviour but the underlying traits themselves, do occur over the lifespan. People on average become more conscientious and more agreeable as they age, and this happens too consistently across cultures and birth cohorts to be explained purely by individual life events. Personality is not fixed at birth or even at adolescence.

More fundamentally, the consistency of trait-based personality across situations is lower than people intuitively assume. This is the insight most associated with the psychologist Walter Mischel, who in the late 1960s reviewed research on the predictive validity of personality assessments and found that the correlation between measured personality traits and actual behaviour in specific situations was typically modest. The same person who is reliably helpful with friends is often considerably less helpful with strangers. Someone who is conscientious about work may be extremely casual about personal finances. The extravert at parties may be reflective and quiet in meetings.

Mischel's work provoked a long and productive controversy. It did not establish that personality is meaningless, it established that the relationship between stable traits and specific behaviour is more conditional than simple trait theory implies. Both the traits and the situational factors matter, and their interaction is complicated.

This has an important implication. The popular notion of "being yourself", which assumes there is a single, authentic self to be consistent with, may be asking for something that does not quite exist in the way the phrase implies. Who you are is partly a function of who you are with, what role you are in, what is at stake, and what the context demands. This is not hypocrisy or inauthenticity. It is how complex social creatures function in complex social environments.

Consider the introvert/extravert dimension, one of the most well-replicated in personality research. People who score high on introversion genuinely find large social gatherings effortful in a way that high-scorers on extraversion do not. But even strong introverts can behave in highly extraverted ways when the situation demands it, when they are presenting to an audience they care about, when they are entertaining people they love, when the context activates a different mode of engagement. The underlying trait does not vanish. But behaviour is more variable than the trait alone would predict.

Culture adds another layer. Displays of personality traits that are socially rewarded vary by cultural context. Assertiveness reads very differently in cultures with different norms around hierarchy and direct communication. Emotional expressiveness that is read as warmth in one context is read as instability in another. This does not mean personality traits do not exist cross-culturally, they appear to, but it means that how they are expressed and how they are read are both filtered through context.

What does all this mean practically? A few things.

First, it suggests some caution about personality typing as a basis for major decisions, hiring, relationship compatibility, role allocation. Personality assessments give real information, but the predictive power of that information for specific behaviours in specific contexts is lower than it looks. People are more changeable and more context-sensitive than the frameworks suggest.

Second, it suggests that environments matter more than we typically acknowledge. If personality were entirely fixed, changing the environment would be beside the point. But if behaviour is partly a function of situational factors, then designing environments that bring out better versions of people's traits, environments with appropriate incentives, norms, role models, and social structures, is not wishful thinking. It is a lever.

Third, it suggests that the experience of feeling like a different person in different contexts is not a sign of fragmentation or inauthenticity. The executive who is measured and controlled at work and emotional and playful at home is not being fake in either place. They are different expressions of a more complex underlying person, activated by different contexts. The question is not which is the real one. They both are.

The stable core is real. But it has room inside it.