Summary: Changes of mind can arise from increased uncertainty about an initial decision rather than from new sensory information arriving after the choice.
Source: University of Ulster
People sometimes reverse a decision shortly after making it. For example, a shopper who initially reaches for one brand may switch to another before leaving the aisle, or a driver navigating in poor weather may misread a sign and alter course as a result. Researchers have long attributed these rapid reversals—often called changes-of-mind—to additional sensory evidence that was still being processed after the initial choice. In many traditional accounts, the brain continues to accumulate noisy sensory input and this delayed information can tip the balance and produce a reversal.
Recent research published in PLoS Computational Biology challenges that interpretation by demonstrating that changes-of-mind can occur even when no new post-decision sensory evidence is presented. The study used a well-established perceptual decision task in which participants judge the dominant direction of motion in a cloud of randomly moving dots. Unexpectedly, the investigators found that participants sometimes changed their decision despite there being no additional external evidence available after the first response.
Dr. Arkady Zgonnikov, who led the experimental component while supported by the Irish Research Council at the National University of Ireland (NUI) Galway, noted that changes-of-mind in the experiment were often preceded by prolonged deliberation during the initial decision. “Participants occasionally changed their minds after committing to an initial choice even though no extra sensory information was presented,” he said. “We were surprised to see that these reversals were often associated with long initial response times, which points to a heightened internal sense of conflict or uncertainty during the first decision.” The experimental results were independently replicated by Dr. Stefan Scherbaum and Martin Schoemann at Technische Universität Dresden, strengthening the reliability of the findings.
To explain these behavioral observations, the team developed a neurocomputational model that reproduced the pattern of results. Dr. Nadim Atiya, who led the modeling work while a Ph.D. student at Ulster University, explained that the model links changes-of-mind to an internal monitoring of uncertainty. “Our simulations indicate that elevated decision uncertainty can delay the initial response and lead to prolonged deliberation before committing to a first choice,” he said. In this account, heightened uncertainty at the moment of the initial decision increases the likelihood that the person will later change their mind, even when no new external cues are available.
Senior and corresponding authors Dr. Denis O’Hora from NUI Galway’s School of Psychology and Dr. KongFatt Wong-Lin from Ulster University’s Intelligent Systems Research Centre emphasized the broader implications. They noted that the study integrates novel behavioral experiments with computational modeling to shed light on decision confidence and metacognition. “By linking changes-of-mind to an internal uncertainty-monitoring process, this work speaks to fundamental aspects of human metacognition and our conscious awareness of mental states,” they said. Understanding how we monitor and respond to uncertainty has implications for research into consciousness and cognitive control.
Neuroscience News thanks KongFatt Wong-Lin from the University of Ulster for submitting this research news for inclusion.
Source:
University of Ulster
Media Contacts:
KongFatt Wong-Lin – University of Ulster
Image Source:
The image is in the public domain.
Original Research: Open access
“Changes-of-mind in the absence of new post-decision evidence”. Atiya NAA, Zgonnikov A, O’Hora D, Schoemann M, Scherbaum S, Wong-Lin K. University of Ulster Computational Biology. DOI: 10.1371/journal.pcbi.1007149.
Abstract
Changes-of-mind in the absence of new post-decision evidence
Human decisions are sometimes followed by rapid reversals. While such changes-of-mind have been viewed as evidence of cognitive flexibility, the underlying mechanisms are not fully understood. Prior work focused on reversals driven by the continued accumulation of noisy sensory input after an initial choice. In contrast, this study demonstrates that changes-of-mind can arise without additional external evidence. In a motion discrimination task, many reversals occurred in trials where the initial response time was long, suggesting a different causal mechanism. A neural circuit model that includes an uncertainty-monitoring population shows how top-down signals related to decision uncertainty can produce these reversals. This framework accounts for the prolonged initial decision times and the observed motor response trajectories, providing a computational explanation for changes-of-mind that do not rely on new post-decision sensory information.