In psychology, an interaction occurs when the effect of one factor on behavior or mental processes depends on the level of another factor. Rather than two influences operating independently, they combine in ways that neither one alone can explain. This concept shows up across nearly every branch of psychology, from personality theory and clinical diagnosis to perception, genetics, and research statistics.
The Core Idea Behind Interactions
The simplest way to understand an interaction is this: if changing one variable has a different effect depending on the value of a second variable, those two variables interact. A medication might reduce anxiety in people who also receive therapy but have no measurable effect on people who don’t. The medication’s impact depends on whether therapy is present. That dependency is the interaction.
This stands in contrast to what psychologists call main effects, where each factor influences an outcome on its own, regardless of what else is happening. If a medication reduces anxiety by the same amount whether or not someone is in therapy, there’s no interaction between the two. The moment the medication works differently across therapy conditions, the two factors can no longer be understood in isolation. You have to consider them together.
Kurt Lewin’s Formula: Behavior as an Interaction
The idea that behavior emerges from interacting forces has deep roots in psychology. In the 1930s, Kurt Lewin proposed a deceptively simple formula: behavior is a function of the person and the environment. The “person” side includes personality traits, emotions, motivations, attitudes, and cognitive habits. The “environment” side covers everything external: temperature, social pressure, the presence of other people, cultural expectations, incentives, and physical surroundings.
Lewin’s insight was that these two categories don’t just add up. They multiply. A person who is naturally assertive might speak up freely in a relaxed group setting but stay silent in a rigid hierarchy. The trait hasn’t changed, and neither has the general category of “social situation,” but the specific combination produces entirely different behavior. This multiplicative view became the foundation for what psychologists call interactionism.
The Person-Situation Debate
For decades, personality psychologists and social psychologists argued over whether internal traits or external situations were more powerful in shaping behavior. In 1968, Walter Mischel pointed out that the typical correlation between a personality test score and actual behavior in a specific situation was only about 0.30, meaning traits accounted for roughly 10% of what people did. Personality seemed weak.
But situations weren’t much stronger. When researchers converted classic social psychology experiments into the same metric, situational factors correlated with behavior at about 0.45, explaining only around 20% of behavioral variance. Neither side had a convincing claim to dominance. The resolution came through interactionism: the recognition that personal and environmental factors shape each other simultaneously. As K.S. Bowers put it, situations are as much a function of the person as the person’s behavior is a function of the situation. An anxious person doesn’t just react to a crowded room differently than a confident person does. They also perceive, interpret, and even select different situations in the first place.
Gene-Environment Interactions
One of the most studied forms of interaction in psychology involves genes and environment. A genetic predisposition alone is often not enough to produce a psychological disorder. It typically requires certain environmental conditions to trigger expression. A landmark study of Danish children born to mothers with schizophrenia illustrates this clearly: institutional rearing was associated with elevated risk for schizophrenia, but only among children who already carried the genetic vulnerability. Children without that predisposition who were raised in the same institutional settings did not show the same elevated risk.
Psychologists also distinguish between gene-environment interaction (where genes moderate the effect of environment) and gene-environment correlation, which comes in three forms. Passive correlation happens when parents pass along both genes and matching environments: anxious parents may transmit a genetic tendency toward anxiety while also creating an overprotective home, doubling the child’s exposure. Evocative correlation occurs when a child’s genetically influenced traits pull certain responses from the world around them. A child who is temperamentally withdrawn might elicit more hovering from caregivers, reinforcing the very pattern. Active correlation describes people selecting environments that match their genetic tendencies, like a naturally curious child gravitating toward books and intellectually stimulating friends.
The Diathesis-Stress Model
Clinical psychology relies heavily on interaction thinking through the diathesis-stress model. “Diathesis” refers to a biological or psychological vulnerability, and “stress” refers to life events or chronic pressures. The model holds that the diathesis alone is not sufficient to cause a disorder. It needs a stressor to activate it.
What makes this a true interaction rather than a simple sum is that the same stressor affects people differently depending on their level of vulnerability. Someone with a strong diathesis may develop depression after a relatively minor setback, like a job rejection. Someone with little or no predisposition might endure severe, prolonged stress without developing the disorder at all. The relationship is inverse: the greater the vulnerability, the less stress is needed. If the diathesis is absent, even extreme stress may not produce the condition. This is why two people can go through the same difficult experience and come out with very different psychological outcomes.
Interactions in Perception and Cognition
Your brain constantly processes information from two directions at once. Bottom-up processing builds perception from raw sensory data: light, sound, texture, pressure. Top-down processing applies your existing knowledge, expectations, and goals to interpret that data. These two streams interact dynamically rather than operating as separate pipelines.
Research in visual neuroscience has shown that when bottom-up processes successfully organize sensory input (for example, when objects in a scene naturally group together), top-down attention has less work to do. But when the sensory scene is ambiguous or cluttered and bottom-up processes leave competition between stimuli unresolved, attention kicks in more strongly to compensate. The amount of attentional effort scales inversely with how much the automatic, stimulus-driven system has already sorted out. This means perception isn’t purely driven by what’s in front of you or purely driven by what you expect to see. It’s an ongoing negotiation between the two, with each system adjusting based on what the other has accomplished.
Interactions in Research Design
When psychologists run experiments with two or more variables, they specifically test for interaction effects using statistical methods like factorial analysis of variance. An interaction effect means the impact of one variable changes based on the levels of another variable. If you’re studying whether a new teaching method improves test scores, you might find it works brilliantly for students with high prior knowledge but makes no difference for beginners. The teaching method interacts with prior knowledge.
Researchers visualize interactions using interaction plots, where lines representing different groups are graphed against an outcome. When the lines are roughly parallel, there’s no meaningful interaction: both groups respond similarly to the variable. When the lines diverge, converge, or cross, an interaction is present. A crossover (or disordinal) interaction is especially striking. It means the direction of the effect actually reverses depending on the group. One teaching method might be better for one type of student and worse for another, with the lines literally crossing on the graph. An ordinal interaction is subtler: one group always outperforms the other, but the gap between them changes across conditions.
This matters practically because when a significant interaction exists, you can’t meaningfully interpret the individual variables in isolation. Saying “the teaching method improved scores” would be misleading if it only helped certain students. The interaction tells a more complete and accurate story.
Why Interactions Matter
Interactions push psychology beyond simple cause-and-effect thinking. They explain why the same medication works for one person and not another, why identical twins raised apart can diverge psychologically, and why a personality trait that predicts behavior in one context fails to predict it in another. Nearly every meaningful question in psychology, from “why did this person develop depression?” to “why does this student thrive with one teacher but struggle with another?”, involves some form of interaction between variables that cannot be understood alone.