When a practice session is too long, there will presumably be a point where no further significant gains can be made without a break. At what point will this be?
Update: This question was originally about the length of an initial training session, but it has since been rescoped to cover training sessions in general.
The expertise literature and its discussion of deliberate practice provides substantial guidance on the question of optimal practice duration. As Ericsson et al (2006) summarised:
elite performers search continuously for optimal training activities, with the most effective duration and intensity, that will appropriately strain the targeted physiological system to induce further adaptation without causing overuse and injury.
With regards to optimal practice length, Ericsson et al (2006) stated (with key points bolded by me) that:
Although the detailed nature of deliberate practice will differ across domains and as a function of attained skill, there appear to be limits on the daily duration of deliberate practice, and this limit seems to generalize across domains of expertise. Expert performers from many domains engage in practice without rest for only around an hour, and they prefer to practice early in the morning when their minds are fresh (Ericsson et al., 1993 ). Elite musicians (Ericsson, 2002) and athletes (Ericsson, 2001, 2003 c) report that the factor that limits their deliberate practice is primarily an inability to sustain the level of concentration that is necessary. Even more interestingly, elite performers in many diverse domains have been found to practice, on the average, roughly the same amount every day, including weekends, and the amount of practice never consistently exceeds five hours per day (Ericsson, 1996; Ericsson et al., 1993 ). The limit of four to five hours of daily deliberate practice or similarly demanding activities holds true for a wide range of elite performers in different domains, such as writing by famous authors (Cowley, 195 9; Plimpton, 1977), as does their increased tendency to take recuperative naps. Furthermore, unless the daily levels of practice are restricted, such that subsequent rest and nighttime sleep allow the individuals to restore their equilibrium, individuals often encounter overtraining injuries and, eventually, incapacitating “burnout.” In some domains of sports, such as gymnastics, sprinting, and weight lifting, the max- imal effort necessary for representative performance is so great that the amount of daily deliberate practice is even further limited by factors constraining the duration of produc- tion of maximal power and strength.
Thus, based on the expertise literature, four or five one-hour sessions of deliberate practice in a day with plenty of rest between and at night would be optimal if the aim is to maximise learning on the focal task in the shortest period of time.
In general, I wouldn't define optimal duration of practice in terms of amount of the task learnt.
Also, in the real-world, the optimality criterion can vary between people. The expertise literature assumes that the goal is to maximise learning on the focal skill. However, in the real-world, a person may be learning a number of things, and also need to accomplish tangible outcomes at the same time.
Experimental psychologists have really grappled with this question, because boundary conditions make a huge difference. One valiant effort is found in a recent paper by Rawson and Dunlosky at Kent State University, who put forth a "3 x 3" recommendation: Practice to your desired criterion three times, then repeat practice at 3 regularly spaced intervals.
