Virtual environments—approximations or analogous representations of the real world expressed as games, simulations, immersive environments, or other software interfaces—are achieving both sophistication and accessibility not previously envisioned even five years ago. With their expansion in capabilities and accessibility, they are providing increasingly efficacious tools for psychologists to observe and shape both complex and dynamic human behavior within minimally constrained environments with rich choice sets and consequences. In previous decades their “virtual” nature made suspect the ecological validity of these environments as study platforms. However, their current capabilities—graphics, physics, expansiveness, and logging potential—and the increasing ubiquity of software environments in human life make virtual environments rivals, if not eventual successors, to traditional laboratory stimulus presentation paradigms.
The potential of virtual environments as study platforms for observing and sampling realistic human behavior in situ gives us pause to rethink traditional approaches to experimental design and interpretation of study-related statistical findings. A number of important methodological questions need be addressed before usage of virtual environments comes to full fruition: First, virtualizations of dynamic, real-world environments and psychologically relevant contexts may provide opportunities to observe and model naturalistic exploration, appraisal and choice. Yet, without the structure and systematicity of more traditional research approaches, such as stimulus repetition, ordering, and categorization by condition, how can we systematically study human behavior within these noisy, realistic environments? Innovations in data collection, labeling, post-processing and analytics will be critical to make sense of behavior observed from within fewer constraints than in traditional laboratory research. Second, when conducting experiments with rich stimulus sets, what constitutes suitable control stimuli from which to collect baseline responses and make comparisons to test stimuli? Do control or baseline stimulus sets need be more dynamic and complex as well to scale with that of test stimuli? Third, without the benefit extensive averaging of responses to repetitive stimulus presentation, or well separated stimulus presentation conditions are traditional benchmarks for effect size and statistical significance guidelines suitable for interpreting findings sampled from complex virtual environments? Fourth, what are emerging best practices for providing descriptive information and inferential findings about behavior within dynamic virtual environments when the stimuli and scenarios presented to users are not independent of their behavior, but progressively shaped by prior behavior(s)? In truly interactive environments, assumptions of normally distributed observations will likely be completely untenable. Finally, what are the range of topics that can efficaciously be examined within virtual environments?
In this Research Topic we are interested in receiving Original Research, Methods, and Review papers focusing on how virtual environments, defined broadly, can be utilized and implemented to expand our understanding of dynamic human behavior, in realistic contexts. Appropriate submissions will explore or demonstrate quantitative and methodological approaches for utilizing virtual environments in this way. Submissions with accompanying topical examinations within domain areas are appropriate. Domains of interest include, but are not limited to: social and or interpersonal dynamics; cognitive task analysis; executive function assessment, intelligence or aptitude testing; therapeutic applications; game usability testing; and training or educational applications.
Virtual environments—approximations or analogous representations of the real world expressed as games, simulations, immersive environments, or other software interfaces—are achieving both sophistication and accessibility not previously envisioned even five years ago. With their expansion in capabilities and accessibility, they are providing increasingly efficacious tools for psychologists to observe and shape both complex and dynamic human behavior within minimally constrained environments with rich choice sets and consequences. In previous decades their “virtual” nature made suspect the ecological validity of these environments as study platforms. However, their current capabilities—graphics, physics, expansiveness, and logging potential—and the increasing ubiquity of software environments in human life make virtual environments rivals, if not eventual successors, to traditional laboratory stimulus presentation paradigms.
The potential of virtual environments as study platforms for observing and sampling realistic human behavior in situ gives us pause to rethink traditional approaches to experimental design and interpretation of study-related statistical findings. A number of important methodological questions need be addressed before usage of virtual environments comes to full fruition: First, virtualizations of dynamic, real-world environments and psychologically relevant contexts may provide opportunities to observe and model naturalistic exploration, appraisal and choice. Yet, without the structure and systematicity of more traditional research approaches, such as stimulus repetition, ordering, and categorization by condition, how can we systematically study human behavior within these noisy, realistic environments? Innovations in data collection, labeling, post-processing and analytics will be critical to make sense of behavior observed from within fewer constraints than in traditional laboratory research. Second, when conducting experiments with rich stimulus sets, what constitutes suitable control stimuli from which to collect baseline responses and make comparisons to test stimuli? Do control or baseline stimulus sets need be more dynamic and complex as well to scale with that of test stimuli? Third, without the benefit extensive averaging of responses to repetitive stimulus presentation, or well separated stimulus presentation conditions are traditional benchmarks for effect size and statistical significance guidelines suitable for interpreting findings sampled from complex virtual environments? Fourth, what are emerging best practices for providing descriptive information and inferential findings about behavior within dynamic virtual environments when the stimuli and scenarios presented to users are not independent of their behavior, but progressively shaped by prior behavior(s)? In truly interactive environments, assumptions of normally distributed observations will likely be completely untenable. Finally, what are the range of topics that can efficaciously be examined within virtual environments?
In this Research Topic we are interested in receiving Original Research, Methods, and Review papers focusing on how virtual environments, defined broadly, can be utilized and implemented to expand our understanding of dynamic human behavior, in realistic contexts. Appropriate submissions will explore or demonstrate quantitative and methodological approaches for utilizing virtual environments in this way. Submissions with accompanying topical examinations within domain areas are appropriate. Domains of interest include, but are not limited to: social and or interpersonal dynamics; cognitive task analysis; executive function assessment, intelligence or aptitude testing; therapeutic applications; game usability testing; and training or educational applications.