It is generally accepted that neural factors play an important role in muscle strength gains. This article reviews the neural adaptations in strength, with the goal of laying the foundations for practical applications in sports medicine and rehabilitation. An increase in muscular strength without noticeable hypertrophy is the first line of evidence for neural involvement in acquisition of muscular strength. The use of surface electromyographic (SEMG) techniques reveal that strength gains in the early phase of a training regimen are associated with an increase in the amplitude of SEMG activity. This has been interpreted as an increase in neural drive, which denotes the magnitude of efferent neural output from the CNS to active muscle fibres. However, SEMG activity is a global measure of muscle activity. Underlying alterations in SEMG activity are changes in motor unit firing patterns as measured by indwelling (wire or needle) electrodes. Some studies have reported a transient increase in motor unit firing rate. Training-related increases in the rate of tension development have also been linked with an increased probability of doublet firing in individual motor units. A doublet is a very short interspike interval in a motor unit train, and usually occurs at the onset of a muscular contraction. Motor unit synchronisation is another possible mechanism for increases in muscle strength, but has yet to be definitely demonstrated. There are several lines of evidence for central control of training-related adaptation to resistive exercise. Mental practice using imagined contractions has been shown to increase the excitability of the cortical areas involved in movement and motion planning. However, training using imagined contractions is unlikely to be as effective as physical training, and it may be more applicable to rehabilitation. Retention of strength gains after dissipation of physiological effects demonstrates a strong practice effect. Bilateral contractions are associated with lower SEMG and strength compared with unilateral contractions of the same muscle group. SEMG magnitude is lower for eccentric contractions than for concentric contractions. However, resistive training can reverse these trends. The last line of evidence presented involves the notion that unilateral resistive exercise of a specific limb will also result in training effects in the unexercised contralateral limb (cross-transfer or cross-education). Peripheral involvement in training-related strength increases is much more uncertain. Changes in the sensory receptors (i.e. Golgi tendon organs) may lead to disinhibition and an increased expression of muscular force. Agonist muscle activity results in limb movement in the desired direction, while antagonist activity opposes that motion. Both decreases and increases in co-activation of the antagonist have been demonstrated. A reduction in antagonist co-activation would allow increased expression of agonist muscle force, while an increase in antagonist co-activation is important for maintaining the integrity of the joint. Thus far, it is not clear what the CNS will optimise: force production or joint integrity. The following recommendations are made by the authors based on the existing literature. Motor learning theory and imagined contractions should be incorporated into strength-training practice. Static contractions at greater muscle lengths will transfer across more joint angles. Submaximal eccentric contractions should be used when there are issues of muscle pain, detraining or limb immobilisation. The reversal of antagonists (antagonist-to-agonist) proprioceptive neuromuscular facilitation contraction pattern would be useful to increase the rate of tension development in older adults, thus serving as an important prophylactic in preventing falls. When evaluating the neural changes induced by strength training using EMG recording, antagonist EMG activity should always be measured and evaluated.