It was the best of times, it was the worst of times: Evidence for switch cost beyond stimulus-response tasks.

To achieve any goal, a configuration of cognitive processes (i.e., "task set") is needed. Navigating between goals requires switching between task sets. Such switching entails a cognitive toll ("switch cost") that is manifested as a performance cost: Switching between tasks takes longer than repeating the same task. The study of the toll of navigating between task goals has been confined to stimulus-response tasks, characterized by bare-bones features. We argue that this has prevented researchers from exploring how far task-switching can reach. Task goals (e.g., choosing a vacation destination) are not confined to stimulus-response tasks (e.g., deciding on the parity of a number) and extend to more complex decisions. We combine two lines of research, that of task-switching and that of choosing between positive options versus choosing between negative options. To the extent that choosing between positive options differs mechanistically (i.e., activates different task sets) from choosing between negative ones (as previous research suggests), switching between positive and negative choices (vs. repeating them) should induce a switch cost. In a reanalysis of the data of six studies (N = 694), we found a consistent pattern of a switch cost between positive and negative choices, supported by both a frequentist and a Bayesian meta-analysis. We further replicated the effect in two preregistered, well-powered studies (N = 241 and N = 205). We discuss the implications of our findings to different accounts of switch cost. We propose that switch cost can be used to implicitly differentiate between different concepts, as an example of where task-switching research can venture. (PsycInfo Database Record (c) 2026 APA, all rights reserved).