This study examined cervical neuromuscular adaptations to resistance training. The ResX group performed conventional resistance training plus head-extension exercise. Another group performed only conventional resistance training, and the control group performed no resistance exercise. Muscle use during head extension was determined by quantifying shifts in T2 in serial-transaxial magnetic resonance images of the neck. ResX was the only group that showed a training effect. Training decreased (P < 0.05) the cross-sectional area (CSA) of cervical muscle used to perform submaximal head extension by 31%. This reflected a decrease (P < 0.05) in relative use of the splenius capitis, semispinalis capitis, and semispinalis cervicis and multifidus muscles by about one-third; their percentage of CSA showing contrast shift was reduced from 60 to 40% on average. This same exercise evoked no contrast shift in the levator scapulae, longissimus capitis and cervicis, and scalenus medius and anterior muscles posttraining, yet 20% or more of their CSA was engaged pretraining. The relative CSA of cervical musculature that was used to perform maximal head extension was increased (P < 0.05) 16% by training. The findings suggest functional redundancy of neck musculature that can be modified by training; submaximal tasks can be performed despite cessation of recruitment of individual muscles, yet recruitment can be increased for maximal efforts. These results also suggest that neuromuscular adaptations to training require a specific cervical exercise.