Summary: New research challenges the idea that variability in human behavior, often considered “noise,” is simply an error that should be minimized. Researchers argue that noise provides valuable insight into cognitive processes, from decision-making to moral reasoning.
By applying a computational model, we show how different types of noise can reflect clear cognitive mechanisms rather than random mistakes. This perspective can reconstruct psychological research in areas such as medicine, decision science, and ethics and improve interventions.
Important facts
Noise as an insight: variability in human behavior can reveal underlying cognitive processes. The role of decision: Some can help the brain navigate uncertainty and drive decision-making. Applications in psychology: Understanding noise can improve cognitive models and improve real decision-making.
Source: APS
The inevitability of human behavioral fluctuations is a phenomenon, and psychologists have long recognized and developed the statistical methods explained in their analysis. Human behavior differs not only within people but also within separate instances of one individual.
The study of individual variation, known by many psychologists as “noise,” is a topic that focuses on special issues in the perspective of psychological science, published in March 2025.
Joakim Sundh and his co-authors at Uppsala University set the stage for this debate by first providing a historical review of how noise was studied in psychological research.
They argue that noise is usually studied as an external factor that needs to be minimized, rather than an internal factor that can provide information about the process under study.
“Every noise has a source, and various sources of noise can cause different representations of noise,” they write.
“If these different sources of noise are associated with different processes, the way in which the noise is expressed in the data can be used to draw conclusions about those processes.”
Sundh and his team describe their own research in an accurate/inaccurate (PNP) model developed to determine how different sources of noise are related to different psychological processes.
The PNP model is applied to three experiments to show how the distribution of noise can be studied to determine an analytical and intuitive method of inference (Sundh et al., 2025).
The authors conclude by previewing three additional contributions to the special issue, and conclude that each paper “represents a great example of the inherent possibility of informing scientific research by using noise to treat it as a source of information.”
Adam Sanborn and his team of Warwick expand this argument by exploring the purpose of noise in human thought processes.
“When asked to ask participants to perform the same task on multiple occasions, even if those opportunities are approaching in time, it generates surprisingly loud behavior,” the author writes.
Is that noise an error in our cognitive process or does it actually help us perform better?
To answer this question, Sanborn and his co-authors discuss the role of noise throughout the information processing stages, from perception to calculations to responses.
They argue that noise is more accurate as a feature of cognitive processing, rather than a glitch that confuses it (Sanborn et al., 2025).
“It can be said that cognitive ‘noise’ is not just a feature, not a bug, but also a key feature, and even underpins its ability to deal with an uncertain world of complexity where accurate analysis is not computationally possible,” Sanborn and his co-authors write.
While noisy behaviour is widely recognized as inevitable, it is often desirable for individuals to remain consistent in how information is classified. For example, doctors who evaluate the presence of skin cancer are expected to make a consistent and correct diagnosis each time.
In their research, Florian Seitz (University of Basel) and his co-authors work to better understand how individuals can reduce noise as they classify information.
To do this, we present a simulation that explains the possible variation in behavior during the process of information perception and processing.
In the simulation, Seitz and his colleagues used two general category structures. (1) A rule-based structure in which categories are based on a single function, and (2) an information integration structure in which categories are determined by multiple functions that are considered simultaneously.
Researchers argue that by integrating continuous data into experiments assessing classification, researchers can better identify the causes of behavioral noise and categorize cognitive processes during play (Seitz et al., 2025).
“Therefore, continuing assessment of people’s category beliefs may help to solve perceptual and process-related sources of behavioral variation,” Seitz and his team wrote, adding that the findings can inform both future cognitive models and applied interventions.
Inspired by recent chaotic reactions to current events like the Covid-19 pandemic and the ongoing campaign of misinformation, APS’ Michelle Leigheter and his team explore his team at the University of Illinois at Urbana-Champaign.
They use the concept of exchangeability as the basis for their discussion. This philosophical concept is used in discussion as a kind of logical equation of morality. This is the understanding that if A is better than B, if B is better than C, then A is better than C.
Regenwetter and his co-authors investigate whether participants can be explained by the principle of transitiveness despite the massive variation in behavior. Their study included 28 participants from the Urbana-Champaign area, Illinois.
Each participant was presented with a survey of 126 moral vignettes, each accompanied by a binary selection question. For each vignette, they were instructed to choose the “bad” or most “morally wrong” option.
Based on the sample, the team finds participants adhere to transitive moral thinking, suggesting that a shared set of moral principles underlies the way people judge the moral value of one item over another (Regenwetter et al., 2025).
“Despite the surprising heterogeneity of the actions of the world around us, there may be order under the confusion,” they wrote.
About this psychology and behavioral neuroscience research news
Author: Hannah Brown
Source: APS
Contact: Hannah Brown – APS
Image: Image credited to Neuroscience News
