Description and Prediction of Long-Term Learning of a Keyboarding Task

Abstract

The goal of this study was to determine an equation (“learning function”) that describes long-term learning of a new keyboard. Five subjects learned 18 characters on a chord keyboard, then improved keying speed by inputting typical numeric keypad text for about 60 total hours. Their performance, in characters typed per minute, was recorded for every trial. Of the various functions that were considered to describe performance, the best fitting equation was a Log-Log relationship of the form CPM_i = e^b0T_i ^b1, where CPM_i is the performance in characters per minute on the i-th trial (T_i) and b₀ and b₁ are fitted coefficients.

A second goal was to investigate how many trials of performance are needed before the entire learning function can be determined. The coefficients of the Log-Log function were determined using only the first 25, 50, 75, 100, 125, 150, 175, and 200 of the initial performance points (out of about 550 total actual data points). The mean squared error (MSE) was calculated for each of these fits and compared to the MSE of the fit using all points. From the results of MSE data, it appears that at least 50 performance data points are required to reduce the prediction error to an acceptable level.