Joel Suss

Skill-based differences in option generation in a complex task: a verbal protocol analysis

In recent models of decision-making, cognitive scientists have examined the relationship between option generation and successful performance. These models suggest that those who are successful at decision-making generate few courses of action and typically choose the first, often best, option. Scientists working in the area of expert performance, on the other hand, have demonstrated that the ability to generate and prioritize task-relevant options during situation assessment is associated with successful performance. In the current study, we measured law enforcement officers’ performance and thinking in a simulated task environment to examine the option generation strategies used during decision-making in a complex domain. The number of options generated during assessment (i.e., making decisions about events in the environment) and intervention (i.e., making decisions about personal courses of action) phases of decision-making interact to produce a successful outcome. The data are explained with respect to the development of a situational representation and long-term working memory skills capable of supporting both option generation processes.

Use of an Option Generation Paradigm to Investigate Situation Assessment and Response Selection in Law Enforcement

When individuals make decisions in the natural ecology, generation and selection of a course of action is informed by their assessment of the situation. Previous option generation research—largely using complex but static tasks—has examined, separately, the decision strategies employed during the situation assessment and response phases of decision making. This research found that decision makers typically generate a small number of options, and their first option is generally a good one. In dynamic tasks, however, skilled performance involves not only comprehension of the current situation, but the ability to predict impending events. The goal of the current study is to test existing claims about the option generation strategies employed during the situation assessment and response phases of decision making, in the context of a dynamic task.

The effect of time constraint on anticipation, decision making, and option generation in complex and dynamic environments

Abstract Researchers interested in performance in complex and dynamic situations have focused on how individuals predict their opponent(s) potential courses of action (i.e., during assessment) and generate potential options about how to respond (i.e., during intervention). When generating predictive options, previous research supports the use of cognitive mechanisms that are consistent with long-term working memory (LTWM) theory (Ericsson and Kintsch in Phychol Rev 102(2):211–245, 1995; Ward et al. in J Cogn Eng Decis Mak 7:231–254, 2013). However, when generating options about how to respond, the extant research supports the use of the take-the-first (TTF) heuristic (Johnson and Raab in Organ Behav Hum Decis Process 91:215–229, 2003). While these models provide possible explanations about how options are generated in situ, often under time pressure, few researchers have tested the claims of these models experimentally by explicitly manipulating time pressure. The current research investigates the effect of time constraint on option-generation behavior during the assessment and intervention phases of decision making by employing a modified version of an established option-generation task in soccer. The results provide additional support for the use of LTWM mechanisms during assessment across both time conditions. During the intervention phase, option-generation behavior appeared consistent with TTF, but only in the non-time-constrained condition. Counter to our expectations, the implementation of time constraint resulted in a shift toward the use of LTWM-type mechanisms during the intervention phase. Modifications to the cognitive-process level descriptions of decision making during intervention are proposed, and implications for training during both phases of decision making are discussed.

Investigating Perceptual Anticipation in a Naturalistic Task using a Temporal Occlusion Paradigm A Method for Determining Optimal Occlusion Points

Perceptual anticipation has often been investigated using a video-based, temporal occlusion paradigm, especially in sport. In this paradigm, the participants’ task is to predict the outcome of the situation based only on the information prior to the occlusion. The occlusion point(s) has typically been based on objective, physically-deterministic events (e.g., the point of foot-ball contact in a soccer penalty kick). However, in other dynamic domains (e.g., law enforcement), such events are often less informative with regards to the ultimate outcome of any given action. In this paper, we describe a series of studies that employed a temporal occlusion paradigm in law enforcement. Using a series of converging methods of analysis, the purpose was to identify scenarios that discriminated between experienced and less-experienced law enforcement officers’ ability to accurately anticipate the immediate future state of the situation, and then identify the occlusion point in each discriminating scenario that maximized the experience-based difference. This method can be applied to investigations of perceptual anticipation in other dynamic and complex domains.

Cognitive Processes Supporting Recognition in Complex and Dynamic Tasks

Previous research has shown that anticipation is one of the best determinants of skill in numerous complex and dynamic domains, such as law enforcement, driving, aviation, surgery, and sport (for a review see Ward, Williams, & Hancock, 2006). Likewise, recognition ability has formed the cornerstone of much of the naturalistic decision making literature for the last 3 decades (e.g., Klein, Calderwood, & Clinton-Cirocco, 1986). In this research, we examined whether skill at anticipating the outcome of a dynamic situation would predict recognition skill, over and above domain-general measures of cognitive ability. We expected that domain-general cognition would account for some of the variance in recognition skill, but that anticipation skill would explain additional, unique variance. Counter to our expectations, anticipation skill did not explain significant unique variance. Instead, only one of the domain-general cognitive measures—spatial ability—was predictive of recognition skill, suggesting that training for improvement in recognition skill should be skill-specific.

Use of Cognitive Task Analysis to Probe Option-Generation in Law Enforcement

Option-generation paradigms have been employed successfully to investigate skill-based differences in performance, particularly in complex, dynamic, and/or uncertain domains. However, although knowledge of option-generation behavior (e.g., number of options generated, frequency with which the criterion best option is selected) is informative, the underlying basis for the observed option-generation behavior is not always apparent. To address this issue, we probed option-generation behavior using cognitive task analysis. Experienced and less-experienced law enforcement officers first observed temporally-occluded video simulations, and then completed an option-generation task. The cognitive task analysis comprised elicitation of retrospective verbal reports of thinking, followed by video-stimulated recall; analysis of these data revealed information that potentially explains the observed option-generation behavior and provided information relevant to the design of decision-making training.

Don't overlook the human! Applying the principles of cognitive systems engineering to the design of intelligent video surveillance systems

Intelligent video surveillance systems (IVSS) that can track people, identify unattended objects, and recognize activity are constantly being developed, updated, and tested against standard test stimuli. The development of these systems, however, seems to focus on technological advancements and largely ignores how automation will affect the human closed-circuit television (CCTV) operator. This paper outlines the advantages of employing a cognitive system engineering approach to the design of IVSS, reviews the literature that has examined human factors issues related to CCTV automation, and introduces a testbed for investigating the effects of automated surveillance technology on human CCTV operators.