Why You Keep Making The Same Choices, Even Bad Ones

The Brain Mistakes Habit For Preference, Study Suggests

When people make a decision, most assume they’re acting on preference. What they want guides what they pick. Now, however, research published in Communications Psychology turns that assumption sideways. The study suggests the brain doesn’t just consult preferences when deciding. Repeated actions can quietly bias what people come to prefer. Choose something often enough and the brain begins treating it as more desirable, more certain, and more worth choosing again, even when objective value is the same.

This is the same force behind the coffee shop someone keeps returning to despite knowing a better one opened nearby, the grocery brand grabbed on autopilot, the familiar route taken without a second thought. Based on this research, these aren’t just habits in the loose sense of the word. They may reflect a core feature of how the brain builds preference over time, using past choices as quiet instructions for future ones.

How Repetition Hijacks Decision-Making

Researchers at TU Dresden recruited 351 male and female participants, averaging around 32 years old, through an online research platform. Each person completed decision tasks split into two phases. During the learning phase, participants made repeated choices between options carrying different reward probabilities. One context might pit a 70% chance of winning a small cash reward against a 20% chance; another might offer a 70% versus 50% split. Over dozens of trials, participants learned which options paid off more.

Then came the transfer phase. Options from different learning contexts were paired against each other for the first time, including pairings where both options carried equal expected value by design. No option was objectively better. Yet participants consistently gravitated toward whichever option they had chosen more during learning, regardless of its actual reward record.

To check whether this was a fluke, the team also reanalyzed six previously published datasets covering 350 additional participants. Across all 15 datasets and more than 700 people total, the pattern held. How often someone had chosen an option during learning reliably predicted which option they’d favor when tested later.

One experiment made the mechanism especially clear. Researchers designed a task in which one decision context appeared 50 times during learning while another appeared only 30 times. Standard thinking in the field predicted participants would prefer the option from the context with larger, more distinct value differences. Instead, participants strongly favored the option from the more frequently encountered context. In that setup, repetition outweighed reward contrast.

When Repetition Makes People Choose Losses Over Gains

Perhaps the most compelling evidence came from reanalysis of experiments where participants encountered both gain and loss contexts. In those studies, participants sometimes preferred money-losing options over money-making ones. In those cases, the loss-linked option had typically been chosen more often during learning. Repetition had quietly rewritten what participants wanted.

The influence extended beyond choice behavior itself. People who had picked an option more often rated it as more valuable after the experiment ended and expressed greater certainty about that judgment. As the researchers put it, “higher within-context repetition of an option was associated with biased choices including higher subjective valuation and lower uncertainty for repeated actions.” Pick something more, and it starts to feel like the smarter pick, even when the numbers say otherwise.

A Simple Model That Beat the Field

To account formally for what they observed, the researchers built a computational model combining just two principles: standard reward-based learning and a tendency to repeat previous choices. Every time a participant selected an option, the model increased that option’s pull toward future selection. Every time they passed on it, that pull decreased. No elaborate calculations. No assumptions about how much a person knows about outcomes across different contexts.

That model was then compared against several well-established alternatives. Across datasets, the repetition-based model consistently outperformed or matched leading alternatives. The team concluded that “the combination of these two basic principles is sufficient to explain biased choices in stable environments.”

One practical advantage stands out. Many competing models require a person to have broad knowledge of all possible outcomes across learning contexts, a condition that rarely holds in the real world. The repetition model works only with what’s always available: what was chosen and what wasn’t.

Why the Brain Builds Habits Instead of Calculating

The researchers offer a straightforward explanation for why repetition carries such outsized influence. Making the same choice repeatedly reduces the mental burden of deciding. The authors suggest repetition may become especially influential when decisions grow more complex, with more options and more varied rewards. The brain leans harder on habit precisely when decisions get harder to navigate.

More than a century ago, the psychologist Edward Thorndike proposed that repeated actions become more ingrained over time, a principle he called the Law of Exercise. This research, tested across hundreds of participants and multiple experimental designs, adds modern weight to that idea and extends it further: repetition doesn’t just reinforce behavior. It quietly revises what a person believes they actually prefer.

That’s a less comfortable idea than it might first appear. If past choices shape future preferences rather than simply reflecting them, then much of what feels like reasoned decision-making may be something closer to momentum carrying people forward, without their awareness, toward whatever they’ve already done before.

---