Sampling method is about how participants are chosen. In this study, the researchers used opportunity sampling, selecting chimpanzees that were readily available in the setting rather than drawing from the entire population. They organized the subjects into mother–child pairs, giving five pairs in total. This approach is practical and ethical in animal research, enabling study of maternal relationships with a manageable sample size. A random sample from the whole population would be ideal in theory but isn’t usually feasible in captive or field settings; stratified sampling would require predefined subgroups and a specific sampling plan within them, which isn’t described here.

White-collar refers to office-based, professional or managerial roles that involve cognitive work rather than physical labor. In Milgram’s discussions, this category is used to describe individuals in professional occupations, distinguishing them from those in manual labor (blue-collar) or other non-manual roles. Because of that, the statement is true: white-collar workers are those who work in professional occupations. The alternative that they are manual laborers would be incorrect, since manual labor falls under blue-collar, not white-collar. It isn’t left undefined in Milgram’s framework, as these occupational distinctions were part of how social roles and status were discussed.

The main idea here is that memory strength across items that fall in a mild-to-moderate range produces balanced, non-biased outcomes. When scenes are only moderately distinctive, they aren’t strong enough to guarantee recall for most items, nor weak enough to guarantee forgetting for many. That mid-range cue strength leads to a similar mix of remembered and not-remembered responses across those scenes, so the distribution of memory outcomes ends up being comparable rather than skewed in one direction. If you saw a skew toward remembering, you’d expect most items to be recalled; if toward forgetting, most would be forgotten. The option about being only familiar doesn’t fit because the pattern described is a shared, mid-range distribution, not a restriction to familiarity alone.

Dreaming was studied in a sleep lab with tight control over the environment and procedures. Researchers used EEG to identify precisely when someone was in REM sleep and then woke participants at specific times to collect dream reports. This clear manipulation of the sleep condition (awakening during REM versus non-REM) and the systematic measurement of the outcome (whether a dream could be recalled and its content) show a laboratory experiment design. The setting and methods reduce extraneous influences and allow a cause-and-effect relationship between sleep stage and dream recall to be tested, with repeated measures across nights strengthening the comparison within the same individuals. It isn’t a cross-sectional survey, since it isn’t gathering one-off data from a population; it isn’t grounded theory, since it isn’t generating theory from qualitative data; and it isn’t a longitudinal cohort, since it doesn’t follow people over long periods in their daily lives.