The study begins from a puzzle that has long divided the literature on anxiety and effort. Cognitively demanding tasks are widely understood to be perceived as costly, and this sense of cost can lead to effort avoidance, particularly in populations whose motivation is impaired. When researchers have examined where anxiety fits into this picture, however, the results have refused to line up neatly. Some studies suggest that anxiety raises the subjective price of mental effort, making people more likely to disengage. Others indicate that anxiety can actually motivate engagement, sharpening focus or driving people to invest more when the stakes feel high. Both patterns have empirical support, which means something important must be missing from the way the field has framed the question.

Sayalı and her colleagues proposed that the missing ingredient is the interaction between two different kinds of anxiety. State anxiety refers to the transient, situational feeling of tension and apprehension that rises and falls with circumstances. Trait anxiety, by contrast, describes a stable disposition to experience such states frequently and intensely. The team hypothesized that state anxiety would enhance task engagement as task difficulty increased, and that this effect would diminish in people with higher trait anxiety. In other words, a jolt of situational anxiety might push people to try harder on harder tasks, but only up to a point, and the strength of that push would depend on the anxious disposition a person brings to the laboratory with them.

To test this, the researchers ran two online experiments with independent samples. In both, they assessed participants’ self-reported anxiety and manipulated their affective state experimentally, allowing them to examine how induced changes in anxious mood altered engagement with tasks of varying difficulty. The design is notable for its scale and its ecological reach: rather than bringing volunteers into a physical laboratory, the team collected data online, which permitted a much larger second sample and a more diverse participant pool than many effort-based decision studies typically achieve.

The results of the first experiment, conducted with 107 participants, came as a surprise. Contrary to the hypothesis, participants with low trait anxiety reported reduced task engagement when they were under high state anxiety, but this effect appeared only for easy tasks. When the tasks were easy, a spike in situational anxiety seemed to push low-trait-anxious individuals away from the task, prompting disengagement precisely where the effort required was smallest. Meanwhile, this pattern was attenuated in participants with higher trait anxiety, whose engagement was less sensitive to the state anxiety manipulation. The predicted interaction between state and trait anxiety was there, but its direction and its boundary conditions were not what the team expected.

Experiment 2, with a substantially larger sample of 572 participants, put the finding to a replication test. The key interaction partially replicated: the negative effect of state anxiety on engagement was again attenuated in individuals with higher trait anxiety. But there was an important difference in scope. In the second experiment, the negative effect of state anxiety on engagement was observed across all task difficulty levels rather than being restricted to easy tasks. The core signature held, namely that higher trait anxiety blunted the impact of state anxiety on disengagement, yet the first experiment’s restriction of the effect to easy tasks did not generalize to the larger sample.

What does it mean that trait anxiety appears to buffer, or at least dampen, the disengaging effect of state anxiety? The authors draw a provocative conclusion from their data: trait anxiety may reflect a reduced impact of state anxiety on disengagement. On this reading, people high in trait anxiety are not simply people who experience more anxious states; they are people whose engagement with cognitive work is less perturbed by those states when they arise. Someone who lives with a chronically anxious disposition may have, in effect, a decoupling between the transient feeling of anxiety and the motivational machinery that decides whether a task is worth continuing. For low-trait individuals, by contrast, a surge of state anxiety arrives as an unfamiliar signal that can knock them off even a lightly demanding task.

This framing helps explain why the earlier literature seemed contradictory. Studies that measured only state anxiety, or only trait anxiety, or that examined tasks of a single difficulty level, would each capture a different slice of what is fundamentally an interactive system. A study of anxious states in a low-trait sample might find that anxiety increases effort cost and promotes avoidance. A study in a high-trait sample might find that anxiety motivates engagement, because the state manipulation simply has less traction on disengagement in those individuals. Task difficulty adds a further layer, with the first experiment suggesting the disengagement effect may be most visible when the task demands little and the perceived cost-benefit balance is otherwise favorable.

The work also speaks to a broader theoretical conversation about cognitive effort as something the brain evaluates like a cost, weighing the demands of a task against its value. Effort avoidance is a well-documented phenomenon, and it is often amplified in populations with motivational impairments. Anxiety has been a candidate modulator of this cost calculus, but the present findings suggest it cannot be treated as a single dial. The same anxious state can carry different motivational meaning depending on the dispositional context in which it occurs, and the same trait disposition can alter how strongly moment-to-moment mood steers behavior. Any model of effort-based decision making that treats anxiety as a unitary input is likely to be incomplete.

There are, of course, limits to what these two experiments can establish. The manipulations were delivered online, the engagement measures were self-reported, and the partial replication indicates that some details of the phenomenon, such as the role of task difficulty, remain unsettled. The authors are transparent that their hypothesis was not confirmed as stated, and the divergence between the two experiments on the difficulty question is itself informative, showing that the easy-task specificity seen in the smaller sample was not robust. Still, the consistency of the trait anxiety attenuation across both samples gives the central finding considerable weight, and the large second sample lends confidence that the interaction is not a statistical fluke.

The implications reach beyond the laboratory. If momentary anxiety differentially disrupts engagement depending on a person’s trait anxiety, then interventions aimed at keeping anxious people on task may need to be tailored to disposition rather than applied uniformly. Educational settings, workplaces, and clinical contexts all involve cognitively demanding work performed under emotionally loaded conditions, and the assumption that anxiety always undermines effort may lead well-intentioned support strategies astray for some individuals. For the researchers themselves, the study is a reminder of why hypothesis-driven science matters even when the hypothesis fails: the team predicted an interaction, found one with a different shape, and in doing so uncovered a candidate redefinition of trait anxiety, not as the tendency to feel anxious more often, but as the tendency for those feelings to matter less when it comes time to decide whether to keep working.

Subject of Research: The interaction between state and trait anxiety in cognitive effort evaluation and task engagement

Article Title: Anxiety state-related task disengagement varies with trait anxiety

Article References: Sayalı, C., Heling, E., & Cools, R. (2026). Anxiety state-related task disengagement varies with trait anxiety. Communications Psychology. https://doi.org/10.1038/s44271-026-00545-1

Image Credits: AI Generated

DOI: 10.1038/s44271-026-00545-1

Keywords: anxiety, trait anxiety, state anxiety, cognitive effort, task engagement, effort avoidance, motivation, cognitive control, psychology, decision making, affective state, task difficulty