Studies of typing behavior differentiate between two basic strategies: visually guided typing and touch-typing (Long, 1976a; Long, Nimmo-Smith, & Whitefield, 1982; West, 1969). The visually guided strategy tends to be less efficient, as it requires alternation of the gaze between the source text and the keyboard in order to see which keys to press. Touch-typing tends to be more efficient, as visual search of the keys is replaced by knowledge of key location and pro-prioceptive feedback. This mode of behavior thus enables simultaneous reading and typing. Interestingly, many typists (including the authors of the present paper) who use keyboards regularly and who have even participated in touch-typing training do not use touch-typing. Previous research suggests that one reason for this phenomenon is the difficult transition from touch-typing training to everyday use of the skill (Baddeley & Longman, 1978; Cooper, 1983). It appears that in many cases people who undergo touch-typing training do not continue to touch-type afterward but revert to their former visually guided typing style.
This observation is supported by studies that show a dramatic decrease in touch-typing ability soon after training (Baddeley & Longman, 1978; Larochelle, 1983). In the study conducted by Baddeley and Longman, trainees practiced touch-typing for 60 to 80 h. Retention of typing skills for trainees who had not used the typewriter following training was assessed after periods of 1, 3, or 9 months after training. The results showed a decrease of about 30% in words per minute (wpm), starting as early as 1 month after training. The main goal of the present research is to highlight one factor that may contribute to deterioration of touch-typing skill following training and to propose a method that addresses its influence.
The factor considered here is melioration (Herrnstein & Vaughan, 1980), the sensitivity of behavior to local rates of reinforcement. In many cases, melioration predicts the tendency to allocate behavior in the direction of alternatives that produce better immediate performance and to underweight delayed performance (see Herrnstein, Loewenstein, Prelec, & Vaughan, 1993). Although this effect has been known to psychologists for more than 100 years (e.g., see Hilgard & Marquis, 1940; Skinner, 1936) and has been extensively studied and modeled (see Davison & Nevin, 1999; Herrnstein et al., 1993; Herrnstein & Prelec, 1991), we feel that in the current context its contribution has been undervalued.
The paper is organized as follows: The first section discusses the two competing strategies of typing: touch-typing and visually guided typing. The second section presents a simple model of the effect of reinforcement on the transition from touch-typing training to everyday use. The model leads to the derivation of a simple manipulation, presented in the third section, which is expected to reduce the likelihood of transition failures. The last section contains an experiment that evaluates the suggested solution.
TOUCH-TYPING VERSUS VISUALLY GUIDED TYPING
One main distinction of the touch-typing strategy appears to be the ability to look at the screen while typing and to devote a minimal level of visual search to the keyboard (Cooper, 1983). This ability is gained through the memorization of key positions and finger trajectories, which makes touch-typing a difficult skill to acquire. Other differences between touch-typing and visually guided typing include touch typists' (a) use of all fingers of both hands, as opposed to the use of one hand of only some of the fingers; (b) fixed assignment of fingers to keys; (c) reduced arm movements; and (d) fixed locations of the palms (Crooks, 1964).
The performance of touch typists, relative to the performance of visually guided typists, is contingent on the level of touch-typing skill. For expert touch typists the ability to type without having to visually search for each key is the basis of their capacity to work on different keys simultaneously, thus increasing their typing speed (Cooper, 1983). …