Have you ever wondered why you or your friends keep making bad decisions? The answer might not be as simple as you think, because it might come down to a biological cue that affects the choices you make.
Researchers at the University of Bologna, led by Giuseppe di Pellegrino, delved into the process of associative learning and maladaptive decision making. Their findings show that the cues an individual relies on to make a decision may be the deciding factor between a good outcome and a bad one.
The study explored the decision-making process of ‘sign-trackers;, who focus on the signal that predicts a reward and move toward that. That compares to goal-trackers who ignore the signal and go straight to wherever the reward lies.
Some people rely on surrounding cues (such as images or sounds) to make decisions more often than others, known as Pavlovian learning. The study suggests those individuals have a harder time updating their beliefs and unlearning associations when the cues signify risky outcomes.
As a result, this can lead to more disadvantageous decision making over time.
The study used eye-tracking, pupillometry and computational modelling to show that those who are drawn towards reward-predicting cues update their believes slower, leading to biased decisions. They believe this provides new insight as to why certain individuals struggle with inflexible behavior, commonly seen in conditions such as addiction or compulsive disorders.
The study was conducted using a modified Pavlovian-Instrumental Transfer paradigm which incorporated three phases. Throughout all the tasks, participants were faced with an image of a slot machine which had two displays on it.
The first phase of the task was the Pavlovian learning phase, in which participants learned which visual cues predicted certain outcomes. The Instrumental learning phase enabled them to learn which actions led to specific outcomes, and finally the Transfer phase tested the Pavlovian bias to show whether the learned cues affected the chosen actions.
Results highlight that sign-trackers tend to let reward-related cues influence their behavior as only they exhibited greater pupil dilation compared to neutral cues. In this vein, cues can act as motivational magnets which capture their attention, thus inhibiting optimal decision-making.
Over time, goal-trackers were found to rapidly adapt their values, which sign-trackers were much slower to adjust and persisted with maladaptive bias. Consequently, sign-trackers relied heavily on outdated cue values, leading them to make poor choices.
For people with compulsive disorders or addiction, the association between cues and choice outcomes can promote poor decisions. The patterns noted in the study mirror the behaviors that are frequently seen among addicts whereby cues continue to drive behavior despite adverse consequences.
The report suggests that future research could analyze clinical populations suffering from psychiatric disorders and furthermore explore whether pharmacological or cognitive interventions could modulate learning rates.
Regrets, We All Have Them
A nationally representative study by Morrison and Roese in 2011 asked 270 Americans about their most significant regrets in life.
They found that the most prevalent responses centered around romance, 19 percent, family, 16 percent, and education, 14 percent.
A further 13 percent of participants held regrets over their career, while 9 percent of people regretted their decisions about parenting.
Do you have a tip on a science story that Newsweek should be covering? Do you have a question about psychology? Let us know via [email protected].
References
Degni, L. A. E., Mattioni, L., Danti, C., Bernardi, V., Finotti, G., Badioli, M., Starita, F., Soltani, A., di Pellegrino, G., & Garofalo, S. (2025). Reduced Pavlovian value updating alters decision-making in sign-trackers. Journal of Neuroscience. https://doi.org/10.1523/JNEUROSCI.1465-25.2025
Morrison, M., & Roese, N. J. (2011). Regrets of the Typical American: Findings From a Nationally Representative Sample. Social Psychological and Personality Science, 2(6), 576–583. https://doi.org/10.1177/1948550611401756
Redish, A. D. (2004). Addiction as a Computational Process Gone Awry. Science, 306(5703), 1944–1947. https://doi.org/10.1126/science.1102384