**3. Muscles activation and fatigability**

Despite is expected a reduced fatigability in older adults, the findings of several studies is controversial (Allman & Rice, 2002, Avin & Frey Law, 2011).

During muscular fatigue, there are changes in the amplitude and frequency of the EMG signal (Cardozo & Gonçalves, 2003, Cardozo et al., 2011), which is dependent on the number of active motor units, their firing rates and the conduction velocity (Oliveira & Gonçalves, 2009). These changes are described in figure 2. Along these lines, EMG is widely used to highlight the muscular fatigue phenomenon in several populations, including people who suffer from back pain, athletes and, recently, older adults (Croscato et al., 2011, Fraga et al., 2011, Hunter et al., 2004, Lindström et al., 2006).

**Figure 2.** Amplitude (root mean square-RMS) and frequency (median frequency) behavior due to an isometric fatigu‐ ing protocol (Cardozo et al., unpublished data).

Hunter et al. (2004) compared the time to task failure, physiological responses (mean arterial pressure, heart rate, and rating of perceived exertion) and EMG responses at a sustained submaximal isometric contraction (20% of MVC) for elbow flexion in young and old men and women. The main finding of this study was that the time to task failure was longer with older adults, regardless of gender, and longer with young women than with young men. However, older adults had a reduced rate of increase in physiological parameters (mean arterial pressure, heart rate and rating of perceived exertion) and in EMG burst relative to younger adults. The authors speculated that changes in the EMG pattern were related to torque fluctuations. The authors concluded that motor unit activity increased most slowly during fatiguing submaxi‐ mal efforts in older adults, possibly leading to increases in the time of task failure (Hunter et al., 2004).

Lindstrom et al. (2006) assessed the EMG activation of the vastus lateralis and rectus femoris during 100 repeated maximum knee extension contractions at 90º.s-1 in young and old men and women. The authors found that older male adults were most fatigable according to the peak torque and EMG parameters (with a higher area based fatigue index and lower root mean square for the vastus lateralis in older men), but this group did not see the greatest fatigue according to the Borg scale. The authors suggested that the EMG amplitude revealed that fatigue is a combination of age-related changes in muscle and central activation failure (Lindstrom et al., 2006).

Aging leads to selective atrophy of type II fibers and increases the contribution of type I fibers to the generation of torque (Avin et al., 2011). However, even in low intensity activities (e.g., rising from a sitting position and walking) when torque is generated by the recruitment of type I fibers, older adults have a higher metabolic cost and higher fatigability than young subjects (Hortobágyi et al., 2011, Wert et al., 2010). This phenomenon is related to a declining VO2max (which occurs at a rate of approximately 8% per decade) and leads to older adults performing their daily activities at higher relative intensities (as measured by percentage of VO2max) than young people (Wilson and Tanaka, 2000). Additionally, recent studies have shown that the rate of consumption of VO2 during walking is also related to the EMG activation pattern (Peterson & Martin, 2010, Hortobágyi et al., 2011).

Peterson and Martin (2010) and Hotobágyi et al. (2011) found a moderate association between higher Cw and increased antagonist coactivation of the thigh and calf muscles in older adults (Peter & Martin, 2010, Hortobágyi et al., 2011). According to Hortobágyi et al. (2011), older adults had an 18.4% higher Cw than young adults and this higher Cw was associated with increased antagonist coactivation (Vastus Lateralis x Biceps Femoris and Tibialis Anterior x Gastrocnemius Lateralis). Peterson and Martin (2010) determined that antagonist coactivation of the thigh (vastus medialis, biceps femoris and semitendinosus) had a higher contribution to the increase in Cw than the contribution from the shank (tibialis anterior, lateral soleus and medial gastrocnemius). Both studies suggested that age-related neuromuscular adaptations in the lower limbs decrease the joint instability and that a higher antagonist coactivation is required to maintain dynamic stability during a normal gait, which increases the Cw.
