The Price of Mental Effort: Being Able Is Not the Same as Choosing

A person can perform a demanding task successfully and still decline an invitation to do it again. That refusal becomes especially interesting when repeating the task would earn more money. Choosing the easier option for a smaller payment does not erase the earlier successful performance. It asks a different question: how attractive is the reward once the work attached to it is considered? In a 2013 study, Andrew Westbrook, Daria Kester and Todd Braver turned that question into a series of choices. Their first experiment, the focus here, separated what people could do from what they were willing to repeat. [1]

Keeping the material, changing the rule

The researchers used the N-back task, which requires participants to monitor a sequence and compare the current item with an item a specified number of steps earlier. In this experiment the items were letters. A one-back comparison asks about the immediately preceding letter; a four-back comparison reaches further back through the sequence. As new letters arrive, the relevant comparison must keep moving. Increasing N therefore changes the demand for maintaining and updating information, even though the basic display material remains the same. The two alternatives are not simply unrelated hobbies that somebody might happen to like differently.

Participants had substantial practice before making their choices. They were not asked to price a vaguely described difficult activity they had never encountered. Younger adults practised levels one through six, while older adults practised levels one through four. Task blocks had a fixed duration across the different loads. This matters because an offer to do something harder can otherwise also be an offer to stay at the task longer. Here, choosing the easier block did not just mean choosing a shorter block. Equal duration still left room for unequal subjective burden, which was precisely what the experiment sought to investigate. [1]

Following the point where a choice changes

During the choice stage, a one-back task was offered against a higher-load task. The harder option carried a fixed payment of two dollars. Payment for the easy option began at one dollar and was adjusted according to the participant's decisions. A choice of the harder task raised the next easy-task offer; a choice of the easy task lowered it. Successive adjustments became smaller. Six choices for each harder level provided an estimate of the amount at which the two options were approximately equal in attractiveness. One isolated preference was not treated as a precise price for thinking.

Consider an explanatory example, not an actual participant record. Someone might find two dollars for the demanding block about as attractive as one dollar for the easy block. In that comparison, the reward attached to the harder task has lost some of its subjective value. The missing dollar is not a measurement of calories consumed by the brain. Nor does it tell us what that person's employer ought to pay. It represents a trade-off within a particular set of tasks and offers. Changing the alternatives would change the question being answered, even if the person performing them stayed the same.

The choices also had consequences. Participants were told that one decision would be selected randomly and the corresponding task would be performed again. The selected task was actually repeated four times. Crucially, the payment depended on continuing to exert effort rather than achieving a specified level of accuracy. The larger reward was not simply a bet that errors would be avoided in the more difficult condition. This does not mean that participants ceased caring about mistakes. It means that the experiment's financial rule should not be confused with a performance prize that automatically penalized choosing a harder task. [1]

When the score cannot stand in for the experience

The first experiment's analysed groups included twenty-five younger adults, aged eighteen to thirty, and twenty-five older adults, aged sixty-three to eighty-eight. As N increased, the rewards for repeating the demanding task were discounted more strongly. Smaller easy-task payments could become approximately equivalent to the larger payment for a harder block. The researchers also examined models that included accuracy and mean response time. Load continued to predict subjective value after those performance measures were taken into account. The choice measure was therefore not merely another way of reporting the observed errors and speed.

Across levels two, three and four, which both groups performed, older adults had lower average subjective values. That is a comparison of group means, not a statement that every older participant avoided effort more than every younger participant. Individual differences were wide within both groups. It is also a cross-sectional comparison rather than a record of the same people's changing preferences over many years. The observation supports a difference associated with these sampled age groups. It cannot by itself establish the path that every individual will follow as they grow older. [1]

One revealing comparison used different task levels at which the groups' performance was similar. Older adults at three-back and younger adults at four-back did not show a statistically detected difference on the reported accuracy measure, yet their subjective values differed. Failing to detect an accuracy difference does not prove that every underlying process was identical. The comparison nevertheless challenges a convenient shortcut: assuming that matched performance guarantees matched subjective cost. Two similar scores need not imply two equally attractive prospects of returning to the task. Successful completion and willingness to volunteer again supply different kinds of information.

A preference the offer system could miss

Participants also completed the Need for Cognition Scale, which asks about engagement with and enjoyment of demanding thought in daily life. People who discounted the harder-task rewards less tended to score higher on this measure. The reported relationship explained around ten per cent of the variance, leaving much individual variation unaccounted for. A link between a laboratory choice and a broader self-reported tendency is therefore informative without being a complete portrait of a person. It does not convert the offers into an intelligence test, or show that a preference is fixed for life.

The offer procedure had a further limitation that makes an important conceptual difference. It did not allow participants to choose the harder activity for less money than the easier one. But an easy task can be boring. Somebody might prefer a more engaging task even without an extra reward, or with a smaller reward. Those preferences would not be fully captured by a system built to estimate how much reward is forgone to avoid demand. The authors recognized this problem and discussed beginning with equal offers, then adjusting the appropriate offer according to the initial choice. [1]

That limitation prevents a sweeping conclusion that mental demand has only negative value for everyone. It also illustrates why a numerical measure needs an account of the choices permitted by the measurement procedure. An estimate can be meaningful and still leave a direction of preference outside its range. What looks like the end of a discounting scale might conceal enthusiasm for a harder task that the available offers never gave a chance to express. Interpreting the measurement requires understanding that boundary, rather than treating the number as a transparent report of every possible motivation.

The evidence supports a modest but useful distinction. In this experiment, increasing the letter-monitoring load reduced the subjective attractiveness of its reward, and measured performance did not explain the entire association. The study did not directly establish a neural mechanism or calculate a universal wage for intellectual work. Its contribution is to make voluntary re-engagement a question in its own right. Before calling an easy-task choice laziness, it is worth asking what the person is trading off. Knowing that somebody can produce the right answers does not tell us everything about what returning to those answers feels worth.

Sources

[1] Westbrook, A., Kester, D., and Braver, T. S. (2013). What Is the Subjective Cost of Cognitive Effort? Load, Trait, and Aging Effects Revealed by Economic Preference. PLOS ONE, 8(7), e68210. CC BY open-access primary research; especially Experiment 1 methods and results, and the discussion. Access and factual review: 14 October 2026. Independently written prose; no quotations or images used.