Training augmentation using additive sensory noise in a lunar rover navigation task

Sage O Sherman; Anna Jonsen; Quinlan Lewis; Michael Schlittenhart; Daniel Szafir; Torin K Clark; Allison P Anderson

doi:10.3389/fnins.2023.1180314

Training augmentation using additive sensory noise in a lunar rover navigation task

Front Neurosci. 2023 Jun 23:17:1180314. doi: 10.3389/fnins.2023.1180314. eCollection 2023.

Authors

Sage O Sherman¹, Anna Jonsen¹, Quinlan Lewis¹, Michael Schlittenhart¹, Daniel Szafir², Torin K Clark¹, Allison P Anderson¹

Affiliations

¹ Ann & H.J. Smead Department of Aerospace Engineering Sciences, The University of Colorado Boulder, Boulder, CO, United States.
² Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, NC, United States.

Abstract

Background: The uncertain environments of future space missions means that astronauts will need to acquire new skills rapidly; thus, a non-invasive method to enhance learning of complex tasks is desirable. Stochastic resonance (SR) is a phenomenon where adding noise improves the throughput of a weak signal. SR has been shown to improve perception and cognitive performance in certain individuals. However, the learning of operational tasks and behavioral health effects of repeated noise exposure aimed to elicit SR are unknown.

Objective: We evaluated the long-term impacts and acceptability of repeated auditory white noise (AWN) and/or noisy galvanic vestibular stimulation (nGVS) on operational learning and behavioral health.

Methods: Subjects (n = 24) participated in a time longitudinal experiment to access learning and behavioral health. Subjects were assigned to one of our four treatments: sham, AWN (55 dB SPL), nGVS (0.5 mA), and their combination to create a multi-modal SR (MMSR) condition. To assess the effects of additive noise on learning, these treatments were administered continuously during a lunar rover simulation in virtual reality. To assess behavioral health, subjects completed daily, subjective questionnaires related to their mood, sleep, stress, and their perceived acceptance of noise stimulation.

Results: We found that subjects learned the lunar rover task over time, as shown by significantly lower power required for the rover to complete traverses (p < 0.005) and increased object identification accuracy in the environment (p = 0.05), but this was not influenced by additive SR noise (p = 0.58). We found no influence of noise on mood or stress following stimulation (p > 0.09). We found marginally significant longitudinal effects of noise on behavioral health (p = 0.06) as measured by strain and sleep. We found slight differences in stimulation acceptability between treatment groups, and notably nGVS was found to be more distracting than sham (p = 0.006).

Conclusion: Our results suggest that repeatedly administering sensory noise does not improve long-term operational learning performance or affect behavioral health. We also find that repetitive noise administration is acceptable in this context. While additive noise does not improve performance in this paradigm, if it were used for other contexts, it appears acceptable without negative longitudinal effects.

Keywords: auditory white noise; long duration exploration mission; longitudinal effects; macrocognition; noisy galvanic vestibular stimulation; stochastic resonance.