Summary: Some people rely strongly on visual and auditory cues when making choices. This heightened cue sensitivity can produce persistent, maladaptive decisions: when cue–outcome relationships change, these individuals have difficulty updating their beliefs and may continue to act on outdated signals even when those signals predict more risky or disadvantageous outcomes.
New research led by Giuseppe di Pellegrino at the University of Bologna shows that elevated cue-driven learning may leave certain individuals more vulnerable to rigid decision patterns commonly associated with addiction, compulsive disorders, and anxiety. The study clarifies how subtle environmental signals can exert disproportionate influence on behavior and why some people struggle to break harmful habits despite changing circumstances.
Key facts
- Cue sensitivity: A subset of people depend more heavily on surrounding visual or auditory cues to guide choices.
- Poor belief updating: When cue–outcome contingencies worsen or become risky, these individuals are slower to revise their expectations.
- Maladaptive decision loops: This reduced updating of cue values contributes to persistent, disadvantageous choices and may help explain compulsive or addictive behaviors.
Source: SfN
When people learn that particular visuals or sounds predict specific outcomes, those cues can become decision guides. Over time, cue–outcome associations shape expectations and can bias choices toward or away from certain options. For many individuals this process is adaptive, supporting efficient decisions. But in some people, the same process becomes maladaptive: strong cue reactivity combined with slow updating of cue values produces persistent choices that no longer match the current contingencies.
Di Pellegrino and colleagues investigated these individual differences using behavioral testing, eye-tracking and computational modeling. Their work, reported in the Journal of Neuroscience, identifies distinct learner profiles and shows that people who orient more to Pavlovian cues—often called sign-trackers—display reduced performance under conditions where those cues are no longer reliable guides for action.
In the study’s experimental design, participants first learned cue–outcome pairings (Pavlovian phase), then learned which actions led to rewards (instrumental phase), and finally performed a transfer test where Pavlovian cues could influence instrumental choices. Eye gaze data were used to classify participants as sign-trackers (who focus on cues) or goal-trackers (who focus more on outcomes or actions). While both groups learned the basic contingencies, sign-trackers were more likely to select options guided by previously learned cues and were slower to adjust when those cues became associated with worse outcomes.
Computational modeling of behavior revealed that participants dynamically balanced values coming from Pavlovian and instrumental systems. Crucially, the poorer performance of sign-trackers was not explained by an overweighting of Pavlovian values relative to action values. Instead, it reflected slower updating of the Pavlovian cue values during the transfer stage—meaning sign-trackers held onto outdated cue expectations longer than others, and this maintained disadvantageous choices.
Implications
These results offer a clearer computational account of inflexible decision making driven by environmental signals. Heightened cue reactivity combined with reduced Pavlovian updating can create decision loops that are difficult to break. That pattern resembles the behavioral dynamics observed in addiction and some compulsive or anxiety-related disorders, where individuals repeatedly choose actions prompted by cues despite negative consequences.
Understanding these mechanisms has practical value: it points toward targeted assessment of cue sensitivity and belief-updating processes in clinical populations, and it suggests potential intervention strategies aimed at improving the flexibility of cue–value updating. Future work by the team aims to test these processes directly in patients with compulsive disorders, addiction, or anxiety to determine whether heightened cue sensitivity predicts greater susceptibility to maladaptive decision patterns in those groups.
Key questions answered
Q: Why do some people rely heavily on visual or auditory cues when making decisions?
A: They exhibit stronger associative (Pavlovian) learning responses that make surrounding cues powerful guides for choices, often more so than internal deliberation or action-based knowledge.
Q: What happens when those cues start signaling risky or disadvantageous outcomes?
A: People with heightened cue sensitivity tend to update their beliefs about those cues more slowly and may continue following the cues despite worse outcomes.
Q: How does this relate to addiction, compulsive disorders, or anxiety?
A: Elevated cue reactivity and impaired cue-value updating can promote rigid, habitual decision patterns similar to those seen in compulsive and addictive behaviors, potentially making these individuals more susceptible to harm.
Editorial notes
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full by our staff.
- Additional context was added by the editorial team to clarify methods and implications.
About this neuroscience research news
Author: SfN Media
Source: SfN
Contact: SfN Media – SfN
Image: The image is credited to Neuroscience News
Original research: Closed access. “Reduced Pavlovian Value Updating Alters Decision-Making in Sign-Trackers” by Giuseppe di Pellegrino et al., Journal of Neuroscience. DOI referenced in the original publication.
Abstract
Reduced Pavlovian Value Updating Alters Decision-Making in Sign-Trackers
Adaptive reward-guided behavior depends both on learning which actions produce rewards (instrumental learning) and on learning which cues predict those rewards (Pavlovian conditioning). How the brain updates and arbitrates between these systems—especially when Pavlovian cues are irrelevant or misleading for the decision at hand—remains poorly understood.
The researchers addressed this question with a Pavlovian-to-Instrumental Transfer task in 60 human participants (30 females), combining eye-tracking, pupillometry, and computational modeling. The task included a Pavlovian phase (learning conditioned stimulus–outcome associations), an instrumental phase (learning response–outcome associations), and a transfer phase (testing Pavlovian bias on instrumental responses).
Eye gaze patterns were used to classify participants as sign-trackers or goal-trackers. Both groups learned the basic contingencies, but sign-trackers showed lower performance in the presence of Pavlovian cues because they favored options aligned with their cue–outcome associations. Computational modeling indicated that participants dynamically combined values from Pavlovian and instrumental systems; the reduced performance in sign-trackers resulted from slower updating of Pavlovian cue values during the transfer phase rather than an excessive weighting of Pavlovian values over instrumental action values.
Overall, the study provides a computational framework for understanding inflexible, cue-driven decision making and points to possible intervention targets for disorders characterized by maladaptive cue reactivity.