Multi-modal Stochastic Resonance to Enhance Astronaut Perceptual Performance: Experimental Design

Astronauts on future deep space exploration missions will need to maintain excellent performance to ensure safety and mission success. To enhance neurocognitive and task performance, stochastic resonance (SR) may be used to enhance “information transfer” to the brain across multiple sensory modalities, reducing the cognitive workload associated with complex tasks. SR is achieved when white noise is added to a perceptual channel (e.g., auditory, vestibular, visual, etc.), and resonates with the underlying primary signal. Our overarching study seeks to examine SR in the context of improved perceptual thresholds; specifically, we hypothesize humans can perceive smaller magnitude stimuli with the addition of SR white noise. In this paper, we introduce a novel approach to experimental design aimed at reducing the time necessary for accurate and precise perceptual threshold measurement. In the two-interval task used for this study, the subject must indicate which of the two intervals they thought the stimulus occurred in (stimulus detection). A stimulus detection task has been chosen to avoid biases in the subject's decision criteria. Each two interval sequence is called a trial; by collecting binary (correct, not correct) data on a given subject (at a given SR noise level) over a series of trials, a threshold can be estimated by fitting a psychometric curve. Measurement of thresholds can be exhausting for subjects, especially if many thresholds are measured in quick succession. Therefore, to avoid fatigue confounding performance it is important that the number of trials, and thus time, are minimized. To this effect, we use computational methods to simulate threshold measurement. The simulations seek to show how many trials are necessary for a threshold measurement accurate and precise enough to discern the presence of SR. With such simulations, we have developed an experimental design that is likely to identify SR if it exists while minimizing the number of trials required.