Dual tasking typically comes with performance costs in at least 1 of the 2 tasks relative to their performance as single tasks. In addition, compatibility relations between stimuli and responses can affect task performance. We here focus on the (response- related) backward crosstalk effect (BCE): Response times in Task 1 are shorter if the responses in Task 1 and 2 share compatible rather than incompatible spatial features. Interestingly, the BCE shows a sequential modulation: It is larger after compatible Trials N - 1 than after incompatible Trials N - 1, although the underlying reasons remain unclear. We report 2 experiments. Whereas Experiment 1 replicated the sequential modulation, Experiment 2 did not when effector combinations changed from trial to trial. These results suggest that sequential modulations of the BCE do not occur when the context of a trial changes to a sufficient degree. Additional results from applying a diffusion model attribute the sequential modulation to less Task 2 response activation after incompatible Trials N - 1. This result offers insights into the mechanisms underlying task shielding in dual tasking.
Automation is pervasive, and humans often collaborate with it to perform various tasks. With technological advancements, it has become increasingly important to understand the characteristics of complex teams composed of multiple humans and multiple agents. In this study, we examined various team- level metrics in multioperator, multiagent teams, including trust, communication, and task completion time (TCT). We conducted a human subjects experiment with 30 teams, each consisting of 2 human and 2 agent players, performing collaborative blockmoving tasks. Each team completed 10 trials under 3 agent reliability pairing conditions: perfect (both agents had 100% reliability), mixed (1 agent had 100%, the other 50%), and imperfect (both agents had 50% reliability). Humans searched for target- colored blocks and issued commands to the agents, which followed predetermined logic to move the blocks to the drop zone. Participants' trust ratings for each teammate and for the team were collected after each trial. From the experimental data, we quantified several team metrics: team- level trust in each referent (each teammate and the team), divergence of trust in each referent, team communication frequency, commands per block delivery, divergence of team communication, and TCT. We evaluated the effect of agent reliability pairing on each of these metrics and explored the associations between them. The results indicate strong interdependence: Shorter TCT is associated with higher team- level trust in team, greater agreement in trust ratings, and more effective, convergent communication patterns shared by the humans. This study offers important guidelines for evaluating human-agent teams at a higher, team- level perspective.
We possess few laws of behavior. In experimental psychology, the ones we do have largely reflect some form of systematic growth in variation as the mean of a percept or a response grows commensurately. These linear, or often log- linear relationships provide us with important insights. Yet fewer still are behavioral laws that take more curvilinear forms in Cartesian coordinate space. The present work concerns perhaps the most famous of the latter: the inverted- U relationship. Often called the Yerkes-Dodson law, after its supposed originators, it is arguably one of the most famous psychological principles. What of its origins and the solidity of the foundations on which it is grounded, and thus its reliability and subsequent applicability? The present work examines, explores, elucidates, and explains this foundation and illustrates some of the many assumptions and flaws involved in its original promulgation and in some subsequent interpretations, extrapolations, and adoptions. Where, how, and why these associated shortcomings and shortfalls were ignored, overlooked, or blandly accepted leads us to an understanding as to how naive beliefs about behavior populate the psychological world and remain embedded in the very fabric of psychological theory. Notwithstanding these observations, collectively we persist in our belief that humans do poorly at the extremes of overload or underload and peak in our performance at some intermediary level of stress. Why this belief remains the case is the pith of the present work and illustrates how intuition can readily trump factual information in the perception of reality.
Observations that people process nominally task- irrelevant information are commonly taken to demonstrate automaticity, an inability to avoid involuntary processing. However, it has been argued that this interpretation rests on an overly narrow view of people's active goals, which are not restricted to the goal reflecting the current task but also comprise other temporary and chronic goals. This suggests that nominally task- irrelevant stimuli are processed because they satisfy other concurrently active goals, such as curiosity: the chronic drive to reduce uncertainty. We tested this possibility in two experiments by comparing standard flanker and Simon tasks with otherwise identical task versions in which the irrelevant stimuli were perfectly predictable. The compatibility effects produced by the irrelevant stimuli were reduced by 50% or more. A third experiment did or did not arouse curiosity in a flanker task by blurring the flankers in some (not analyzed) trials. As predicted, the flanker compatibility effect was increased in the group with occasionally blurred flankers. Taken altogether, these findings suggest that nominally task-irrelevant stimuli are processed not because their processing cannot be prevented but because processing them satisfies curiosity and thus meets a chronic goal of perceivers/actors.
In 2-choice task-switching paradigms, repeating the previous response is beneficial when the task repeats but leads to performance costs when the task switches. Competing accounts explain such response repetition (RR) costs in different ways. Among these, some accounts postulate a signaling mechanism such that changing the task signals the cognitive system to change the response. With 2 response alternatives, a task switch signals the lone alternative response, yielding RR costs when the previous response should instead be repeated. With 3 response alternatives, a task switch promotes a bias to switch responses but does not signal which of the 2 alternative responses should be made. Therefore, it should be associated with reduced RR costs. To investigate this hypothesis, we manipulated the number of response alternatives in a cued task-switching paradigm, with 1 session affording 2 response alternatives and another affording 3. This manipulation did not influence RR costs in the reaction time measures. However, it did influence RR costs in mean error rates, which were smaller with 3 (vs. 2) response alternatives. Because only a signaling mechanism should yield this pattern, the latter finding suggests that signaling contributes to response repetition costs. We discuss this finding in light of other empirical results, suggesting that performance improves when different elements of a task require the same transition (e.g., a repetition) compared with a different transition.
