Introduction: Eye fixations can be distributed in three ways: randomly, in clusters, and regularly. However, there is always a continuum among these types, because these spatial patterns are the result of a process evolving over time. The focus of the present work was to study the changes over time observed in the Nearest Neighbor Index (NNI), an index derived from the spatial distribution of eye fixations that has been reported to be sensitive to variations in mental workload. Of particular interest are periodic changes in the ultradian timescale (an ultradian rhythm is a recurrent period or cycle repeated throughout a 24-h circadian day).
Methods: Data from a previously reported experiment were further analyzed using temporal spectral analysis, which is one of the most commonly used techniques for studying measurements collected at regularly spaced intervals of time.
Results: An ultradian rhythm with a periodicity between 2 and 15 min was found, which is compatible with results obtained by analyzing reaction times in prolonged vigilance tasks.
Discussion: The identification of a periodicity in the allocation of mental resources should be considered in the design of automation support that is dynamically matched to mental workload.