Texting and tapping : a dynamical approach to multitasking.

Abstract

Jobs in various work fields (e.g., flying airplanes; Helmreich, 2000) require a high ability to successfully handle more than one task at a time, or to multitask. Researchers usually explain multitasking by having priorities in which individuals either attend to one task at a time, or one task receives more time processing than the other task. The current study approaches multitasking from a dynamical systems perspective. Fourteen general psychology students participated in the study by pressing a pedal attempting to maintain a steady beat and text messaging. Researchers recorded behavior over time (2 min. for each task and multitasking). The inputs to the data analysis were the X-Y coordinates of thumb movement (in pixels) over time and the recorded beat's deviation (in sec) from the metronome's beat over time. The patterns of behavior were recorded. Nonlinear analyses (Iterated Function Systems and a MANOVA on Hurst exponents for monofractality, and Wavelet Modulus Transform Maxima for multifractality) tested for fractal patterns which characterized both tasks in both conditions (single task or multitasking). Thumb movement's patterns during texting were not significantly different for single task and multitasking conditions, both displaying short-term correlations (brown noise). Patterns in tapping deviations were significantly different between the two conditions. Structure of deviations while only tapping was characterized by strong long-term correlations (pink noise); the structure while multitasking was also positively long-term correlated, but less strong. Results showed that texting and tapping behavior, as single tasks or during multitasking, are fractal.