Assessing attention under varied acoustic conditions using eye movement and behavioral analysis

Introduction. Sound can both enhance and disturb concentration. Most studies look exclusively at white noise or the effects of listening to music in general. In this study, we aim to fill two gaps: to investigate the effects of brown noise and to compare the effects of instrumental classical music and songs with lyrics on attention. Methods. The experiment involved 30 adult participants. Participants performed five tasks with different auditory stimuli (silence, white noise, brown noise, music with words and classical music). The loudness of all stimuli was ~65 dB. In each task, the participant read a text, mentally counted the words in the specified categories and answered a question about the plot after reading. All texts were of similar length, each containing words from two different semantic categories, with between 5 and 25 words in each category. The order of presentation of the auditory conditions was balanced using a 5 × 5 Latin square, with a 3-minute break between tasks. Participants came to the experiment twice on different days, with a mirrored sequence of auditory conditions on the second visit. The tasks were presented using a custom computer program. The data were collected in two ways: eye movement recording using “Tobii Pro” to count the number and duration of gaze fixations in the text and lateral areas; and video analysis using cameras in the lab and “Viso” and “ObserverXT” software to assess eye blinks, head/body movements and frequency of facial expressions. Results. The results showed no statistically significant differences between the different sound conditions (p > 0.05). Conclusions. This may be explained by the greater influence of individual personal traits, sensory sensitivity or method limitations. Also, a cognitively demanding task may have placed a maximum load on the cognitive systems and rendered the auditory background irrelevant in the context of the overall processes (“ceiling”; effect).