This study investigated the relation between cognitive load (CL) and eye movement frequency characteristics. Previously, eye gaze has been known to have more gaze deviation when the person is engaged in a cognitively demanding task. However, previous studies were difficult to generalize to other studies due to their original algorithm usage. The aim of this study is to present a generalizable method to quantify eye movements that are related to CL. CL was manipulated in 3 levels, Low, Medium, and High, using an auditory N-back task. Each participant performed a total of 12 one-minute trials. Eye movement data were recorded at 50 Hz, using Tobii 1750 eye tracker, in a time domain format, and then transformed into a frequency domain format using Fast Fourier Transformation (FFT). The frequency domain eye data were analyzed by the frequency window of 0.1 Hz. Each frequency window had 42 counts of eye movement data, representing 14 participants in 3 CL levels. The 42-count data were taken into a Spearman rank-order correlation analysis. We found a sequence of strong, positive correlation coefficients in a low frequency range of eye movements (0.3 Hz to 3.2 Hz), having an Rs average +0.63, ranging from Rs +0.51 to +0.82. By contrast, a higher frequency range (13.0 Hz to 25.0 Hz) had a lower correlation coefficient average +0.34. The result implies that a person’s CL level may be estimated by observing the low frequency range of eye movements. This index can be generalized to other studies since this CL indicator used only FFT without any original complex algorithm. (Poster 4-83)