causal mechanisms to a centrally stored representation in the brain. The “bio-
informational” theory postulates that MP effects reflect an interaction of three different
factors: the environment in which the movement in question is performed (“stimulus”
information), what is felt as the movement occurs (“response” information) and the
perceived importance of this skill to the performer (“meaning” information). Let us now
outline and evaluate each of these theories briefly (but see Murphy and Martin, 2002, for
a more detailed review) before proposing a possible compromise between these rival
models of mental practice.
Neuromuscular theories of mental practiceThe earliest theories of mental rehearsal (e.g., Carpenter’s, 1894, ideo-motor principle;
Washburn, 1916) contained two key propositions. First, they suggested that imagination
of any physical action tends to elicit a pattern of faint and localised muscle movements.
Second, they claimed that such muscular activity can provide kinaesthetic feedback to the
performer which enables him or her to make adjustments to this skill in future trials. This
version of neuromuscular theory was supported by Jacobson (1932) who suggested that
visualisation causes tiny “innervations” to occur in the muscles that are actually used in
the physical performance of the skill being rehearsed covertly. Such minute subliminal
muscular activity was held to be similar to, but of a lower magnitude than, that produced
by actual physical execution of the movements involved. A more recent term for this
theory is the “inflow explanation” approach (Kohl and Roenker, 1983) whereby the
covert efferent activity patterns elicited by imagery are held to “facilitate appropriate
conceptualizing for future imagery trials” (p. 180).
In order to corroborate neuromuscular theories of MP, evidence would have to be
found which shows that there is a strong positive relationship between the muscular
activity elicited by imagery of a given skill and that detected during the actual
performance of this skill. Unfortunately, there is very little empirical support for
neuromuscular theories of mental practice. For example, there is no convincing evidence
that the faint muscular activity which occurs during imagery of a given skill is similar to
that recorded during its overt performance. Thus Shaw (1938) found that increased
electromyographic (EMG) activity during motor imagery was distributed across a variety
of muscle groups in the body—including some which were not directly related to the
imagined action. In other words, the muscular innervations elicited by imagery may
merely reflect generalised arousal processes. Furthermore, doubts have surfaced about
the type of muscular activity elicited by imagery. Thus despite using nuclear magnetic
resonance (NMR) spectroscopy to monitor what happens in people’s muscles during
imaginary performance of a specific skill, Decety, Jeannerod, Durozard, and Baverel
(1993) could not detect any change in relevant muscular metabolic indices. Finally, in a
recent test of some predictions from neuromuscular theory, Slade et al. (2002) reported
that the EMG pattern of activation in biceps and triceps for two types of imagined
movements (namely, dumbbell and “manipulandum” curls) did not match the EMG
pattern detected during actual movement. The authors of this study concluded that it
added to “the mounting research evidence against the psychoneuromuscular theory” (p.
164). On the basis of the preceding evidence, Murphy and Martin (2002) concluded that
there is little or no empirical support for a relationship between the muscular activity
Using imagination in sport: mental imagery and mental practice in athletes 137