James K. McNulty

Introduction

James K. McNulty is an American social psychologist and professor in the Department of Psychology at Florida State University. He studies how intimate relationships change over time, and he is identified with a single stubborn argument that runs through almost everything he has written: there is no such thing as a behaviour that is good or bad for a relationship in the abstract. Whether a given act helps depends on what it is a response to and what it is likely to produce next.

That position — often called the contextual view — has made McNulty the most prominent internal critic of the assumption, common to both couples therapy and positive psychology, that more positivity, more forgiveness and more constructive communication are straightforwardly better. His 2012 paper with Frank Fincham, Beyond Positive Psychology?, is his most-cited work and the clearest statement of the case. He has published more than 130 articles, holds fellowships in the Association for Psychological Science, the American Psychological Association and the Society for Personality and Social Psychology, and has received over two million dollars in grant support.

Early Life and Education

McNulty took his doctorate in social psychology at the University of Florida in 2001. The department was then a centre for the study of automaticity and social cognition, and the influence is visible throughout his later work: he has consistently been more interested in what people do without deliberation than in what they say about themselves.

His dissertation research already showed the two features that define his career — longitudinal design and an interest in uncomfortable results. Rather than asking which traits make for happy couples, he began asking why the same trait predicts opposite outcomes in different marriages.

Academic and Professional Career

After Florida, McNulty held a faculty position at the University of Tennessee, where his research group ran multi-wave studies of newlywed couples and where he received a Distinguished Faculty Award in 2011. Earlier affiliations include Florida Atlantic University. He moved to Florida State University, where he has remained, and where his laboratory continues to combine experimental, observational and longitudinal methods.

The characteristic design associated with his group is the multi-wave newlywed study: recruit couples before or shortly after marriage, bring them back at intervals over the first four or five years, and collect self-report, behavioural observation of problem-solving discussions, and implicit measures. Several of his best-known findings come from studies of this shape, and they are the reason his conclusions about change over time carry more weight than cross-sectional survey work.

More recently his research has turned to automatic partner attitudes — the associations in memory between a partner and one's evaluation of that partner, measured indirectly rather than by asking. In 2025 he gave the Psi Chi Distinguished Lecture on the topic, arguing that these automatic evaluations track interpersonal experience closely and often predict outcomes better than self-reported satisfaction does.

Major Contributions to Psychology

The contextual analysis of relationship behaviour

The paper that established McNulty's reputation, written with V. Michelle Russell and published in Journal of Personality and Social Psychology in 2010, asked a simple question with an awkward answer. Researchers had long assumed that constructive problem-solving — validation, humour, affection — is good for a marriage. In a four-year study of newlyweds, McNulty and Russell found that it was, but only for couples whose problems were relatively severe. Among couples with mild problems, high levels of constructive behaviour were followed by declines in satisfaction.

The interpretation is that effortful problem-solving signals that something is wrong. Fixing a small problem at length is not reassuring; it is a reminder. The finding is a clean illustration of the general argument — the value of a behaviour is a property of the situation, not of the behaviour.

The dark side of forgiveness

McNulty applied the same logic to forgiveness, which the field had treated as unambiguously virtuous. His 2011 paper in JPSP reported that a spouse's general tendency to forgive predicted continued psychological and physical aggression in the marriage: forgiving a partner who keeps offending removes a consequence. With Eli Finkel and colleagues he reported the corresponding result at the level of the self — the doormat effect — that forgiving erodes self-respect and self-concept clarity when the offender has not signalled that the relationship is safe.

Neither finding says forgiveness is bad. Both say that forgiveness is a response to a situation, and that in the wrong situation it makes things worse.

Expectations in early marriage

With Benjamin Karney, McNulty showed in 2004 that holding very positive expectations for a marriage is not uniformly beneficial. Optimistic expectations predicted better outcomes only when couples also had the skills to sustain them; where problems were severe and skills were poor, high expectations were followed by sharper declines. The pattern is the same shape as the later work, and it set up the theoretical argument.

Beyond positive psychology

The 2012 American Psychologist paper with Frank Fincham generalised the point beyond close relationships. If a trait or behaviour is beneficial only in certain contexts, McNulty and Fincham argued, then the research programme of positive psychology — identifying and promoting the virtues — is built on the wrong premise. What is needed is not a list of positive qualities but an account of the conditions under which each quality is functional. The paper has been cited several hundred times and remains the most cited of his career.

Automatic partner attitudes

With Michael Olson, Andrea Meltzer and Melissa Shaffer, McNulty published a 2013 report in Science showing that newlyweds' automatic partner attitudes, measured with an implicit task, predicted changes in marital satisfaction over four years above and beyond their own self-reported feelings. The paper was brief, but it opened a research line on processes that operate without a person's awareness or willingness to report them, and it is the direction his laboratory has pursued since.

Major Publications

McNulty, J. K., & Karney, B. R. (2004). Positive expectations in the early years of marriage: Should couples expect the best or brace for the worst? Journal of Personality and Social Psychology, 86(5), 729–743.
McNulty, J. K., & Russell, V. M. (2010). When "negative" behaviors are positive: A contextual analysis of the long-term effects of problem-solving behaviors on changes in relationship satisfaction. Journal of Personality and Social Psychology, 98(4), 587–604.
Luchies, L. B., Finkel, E. J., McNulty, J. K., & Kumashiro, M. (2010). The doormat effect: When forgiving erodes self-respect and self-concept clarity. Journal of Personality and Social Psychology, 98(5), 734–749.
McNulty, J. K. (2011). The dark side of forgiveness: The tendency to forgive predicts continued psychological and physical aggression in marriage. Journal of Personality and Social Psychology, 100(3), 453–467.
McNulty, J. K., & Fincham, F. D. (2012). Beyond positive psychology? Toward a contextual view of psychological processes and well-being. American Psychologist, 67(2), 101–110.
McNulty, J. K., Olson, M. A., Meltzer, A. L., & Shaffer, M. J. (2013). Though they may be unaware, newlyweds implicitly know whether their marriage will be satisfying. Science, 342(6162), 1119–1120.

Impact and Legacy

McNulty's contribution is methodological as much as substantive. He and his collaborators made longitudinal dyadic design the standard against which claims about relationships are judged, and they demonstrated repeatedly that a cross-sectional correlation between a behaviour and satisfaction tells you almost nothing about what will happen if you change the behaviour.

The contextual framework has attracted a specific and serious objection: it is very hard to falsify. If the same behaviour is predicted to be beneficial in one setting and harmful in another, then almost any outcome can be accommodated by characterising the setting after the fact. McNulty's papers generally specify moderators in advance, which mitigates the problem, but the theory does not in itself say which contexts matter or how to identify them independently of the result they are meant to explain. Critics have argued that a contextual view risks becoming a licence for post hoc interpretation.

The automatic-attitudes work has drawn the criticisms that attach to all implicit measurement. Reaction-time-based measures of evaluation have modest reliability, with test–retest coefficients well below those of good self-report scales, and meta-analytic work on their incremental predictive validity has produced small and heterogeneous estimates. There is also a direction-of-causation problem that McNulty himself acknowledges: automatic partner attitudes track what is happening in the relationship, so a measure that reflects accumulating experience may predict later satisfaction simply because it is an efficient summary of the past rather than a cause of the future.

Sampling is the third line of criticism, and the one least amenable to defence. The bulk of this research is drawn from heterosexual newlywed couples in the American South, recruited in the first years of marriage and followed for four or five years. Whether the same interaction patterns hold for same-sex couples, for long-married couples, for cohabiting partners, or outside North America and Western Europe is largely unknown. Findings about forgiveness and aggression are additionally limited by the fact that severe intimate partner violence is, fortunately, rare in these samples, which makes it difficult to carry the doormat effect and the dark side of forgiveness over into clinical and forensic contexts.

These qualifications do not overturn the central empirical claim, which has now been replicated in several independent laboratories: relationship-relevant behaviours are not uniformly good or bad. But they do mean that the contextual programme is better at explaining why simple advice fails than at generating advice to replace it.

Learn More

Harold Kelley — interdependence theory, the formal framework for studying dyads
Martin Seligman — positive psychology, the movement McNulty and Fincham argued against
John Bowlby — attachment theory, the other major account of close relationships