Questions 1–5.

Questions 6–10. Select the TWO correct answer choices.

Questions 11–15 are based on the passage below.

The "marshmallow test," a famous study of delayed gratification conducted in the late 1960s, offered preschool children a choice: eat one marshmallow immediately, or wait alone in a room for several minutes to receive two. Follow-up studies, tracking the same children years later, reported that those who waited longer tended to have better academic outcomes and health markers in adolescence, a correlation widely interpreted as evidence that self-control in early childhood shapes lifelong success. This interpretation shaped decades of parenting advice and educational policy. A 2018 replication attempt, however, using a much larger and more socioeconomically diverse sample, found that the relationship between waiting time and later outcomes largely disappeared once the child's family background was statistically accounted for. Children from more affluent households, the replication suggested, had less reason to distrust that a second marshmallow would actually arrive, while children accustomed to scarcity had rationally learned to take a certain reward now rather than risk a promised one later. The delay itself, on this account, may reflect a child's environment and circumstances at least as much as an innate capacity for self-control.

Questions 16–20 are based on the passage below.

Confirmation bias — the tendency to seek, interpret, and remember information in ways that support one's existing beliefs — is often described as a flaw in human reasoning, a bug to be corrected through training in critical thinking. Some cognitive scientists have recently challenged this framing, arguing that confirmation bias may be better understood as a byproduct of an otherwise adaptive feature of cognition: efficient, belief-consistent processing that allows individuals to make rapid decisions without constantly re-evaluating settled questions from scratch. On this view, treating every belief as perpetually open to revision would be cognitively exhausting and would have made rapid decision-making, historically necessary for survival, effectively impossible. This does not mean confirmation bias is harmless; in contexts such as scientific research or jury deliberation, where accuracy matters more than speed, the same tendency that once conferred an advantage becomes a liability, entrenching errors rather than correcting them. The practical implication, proponents of this view suggest, is not to expect people to abandon the tendency altogether, but to design institutions and procedures — peer review, structured deliberation, blind analysis — that compensate for a bias unlikely to disappear through willpower alone.