A study published this year by researchers at Tsinghua University examined hazard recognition among 113 construction workers. Participants were shown photographs of real work sites, some containing hazards and some safe, and asked to indicate whether danger was present. While they responded, the researchers recorded eye movements, brain activity, and emotional state.

What matters here is how performance was produced. Positive emotion did not function as a straightforward advantage or disadvantage. It altered the conditions under which the scene was perceived — and more specifically, what within that scene registered as significant.

The threshold shifts

Among workers with less than ten years of experience, and among those with more than twenty, the pattern was strikingly similar. In a more positive emotional state, vigilance declined and hazards were missed more often. The workers still looked, and they still knew what they were looking at. What changed was the threshold at which something counted as a signal requiring response.

A loose cable, an incomplete barrier, a structural irregularity — each remains fully present in the visual field while failing to become consequential. Nothing disappears in the literal sense. Yet something becomes unavailable. The decisive boundary runs between what reaches recognition and what stays below it: perceptually available, behaviorally inactive.

The group that broke the pattern

Workers with moderate experience — roughly ten to twenty years — showed a different tendency. In this group, a more positive emotional state corresponded with improved hazard detection. And this same group also tended to operate from a lower emotional baseline, meaning positive states were less frequent and emerged under different internal conditions.

The same emotional shift produced different perceptual outcomes depending on experiential history. What first appears to be a property of “positive mood” begins to look more like a change in sensitivity whose direction depends on where the system is already situated.

One thing remains consistent: the boundary between signal and background is not fixed. It moves with the condition of the perceiver.

Why this happens

Norbert Schwarz and Gerald Clore argued that affect functions as information, shaping whether cognition stays broad and heuristic or becomes narrower and more analytical. Later work by Karen Gasper and Clore suggested that positive mood often broadens perceptual scope toward the general whole, while less positive states narrow attention toward particulars. When things feel coherent, stable, and without immediate demand, quieter cues become less likely to cross into recognition — through the absence of any trigger strong enough to make them stand out.

At the level of the nervous system, this resonates with Joseph LeDoux’s work on threat processing. Organisms do not respond to stimuli in absolute terms. They respond through thresholds shaped by context, expectation, and state. Whether a signal becomes salient depends partly on whether the system is already prepared to treat it as meaningful.

Beyond the construction site

Fatigue in someone close to us often accumulates without a clear moment of arrival. Disengagement thins gradually. A person does not vanish — they become less distinct. The signal persists, yet its intensity relative to the current threshold is insufficient to bring it into focus. It continues unregistered, because it does not cross the level at which attention reorganizes itself.

This is closer to a condition than a choice — a state in which there seems to be no reason to look more carefully, and therefore nothing compels the system to do so, even while the conditions that would justify attention are already present, only too quiet to interrupt the existing frame.

What changes when you see the mechanism

Understanding this does not lower the threshold. It does not restore what has already been missed. It does alter the position from which not-noticing can be understood. Once the lapse becomes visible as part of a mechanism rather than a failure of effort, the absence of recognition stops being neutral.

The only possible adjustment is a subtle one: to look most carefully at the moments when there seems to be no need to look at all. Not out of suspicion, and not as discipline — out of something closer to care. It is often when everything appears in order that the quietest signals are least likely to get through.

---

Sources

• Wang, J., Liang, M., Liao, P.-C., Chong, H.-Y. & Chen, G. (2026). A trimodal approach for construction safety management: Combining emotion with eye movement and electroencephalogram. IEEE Transactions on Engineering Management, 73, 2583–2596.
• Schwarz, N. & Clore, G.L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45(3), 513–523.
• Gasper, K. & Clore, G.L. (2002). Attending to the big picture: Mood and global versus local processing of visual information. Psychological Science, 13(1), 34–40.
• LeDoux, J.E. (1996). The Emotional Brain. Simon & Schuster.

This text was developed with the assistance of AI and edited by the author.