The cough-anal reflex has been described as the contraction of the external anal sphincter occurring when coughing (Meagher et al., 1993). Literature is however unclear in discriminating whether the action on the sphincter is a reflex response due to the abdominal and pelvic floor dynamics, or an integral component of the cough itself. According to the past studies, any maneuver causing an increase in abdominal pressure, like cough, Valsalva etc., determine a reflex contraction of perineal muscles (Parks et al., 1962; Bors, 1966; Deindl et al., 1993; Bø and Stien, 1994). Some authors even provided evidences that during voluntary cough the external anal sphincter is recruited after intercostals and abdominal muscles, suggesting that this response is mediated by a polysynaptic reflex (Chan et al., 2004). In both these hypotheses, a precise cause-effect relationship is identified, in which the reflex contraction of the sphincters prevent unwanted leakages or sagging of the muscular wall of the pelvis. On the contrary, recent literature reports that the sphincters contraction actually precede the rise in abdominal pressure during cough expiratory effort (Constantinou and Govan, 1982; Thind et al., 1990; Thind and Lose, 1992), also showing that the increase in anal or vaginal closing pressure precedes the cough motor act (Sapsford and Hodges, 2001) and that the external anal sphincter is recruited in advance of intercostals muscles (Deffieux et al., 2006). Based on their findings, Sapsford and Hodges (2001) and Deffieux et al. (2006) forwarded the alternative idea of a pre-programmed central control of the external anal sphincter activity for granting continence during cough.

From another perspective, if one examines the abdominal-thoracic compartment, a forceful increase of the internal pressure may be induced both by the respiratory muscles (diaphragm relaxation and internal intercostal muscles contraction) which promote the compression of the thoracic cavity, and/or by the abdominal wall muscles (contraction of rectus abdominis, transverse abdominis, external oblique muscle and internal oblique muscle) which press the abdominal organs cranially, reducing the volume of the thoracic cavity to help expel air, vomit, urine and feces from the body (Iscoe, 1998).

A different temporal and spatial recruitment of the wall muscles and of sphincters may thus lead to different behavioral results: expulsion of the air during speaking, singing, yelling or coughing (in this case perineal muscles and perineal sphincters should contract in synergy); vomit, when abdominal pressure increases and the lower esophageal sphincter is relaxed; defecation and micturition, when perineal muscles are contracted, to stiffen the pelvic floor, while perineal sphincters are relaxed, to offer an escape route to excretions. Not to mention the complex mechanism of foetal delivery, to which cooperates the uterine contraction.

Returning to the cough-anal reflex/response question, an important family of pre-programmed muscle contractions has been well documented when studying voluntary movements: these anticipatory adjustments play a fundamental role in whole body and segmental postural control (Bouisset and Do, 2008; Cavallari et al., 2016), so that they are referred to as anticipatory postural adjustments (APAs). By analogy, we propose that also the sphincter control may belong to the family of the pre-programmed actions, and in this view, the cough-anal reflex may be interpreted as an APA on the thoraco-abdominal compartment.

In humans, there are several examples of pre-programmed, anticipatory actions which accompany a willed movement. Indeed, APAs are usually described as unconscious muscular activities preceding the primary movement and aim to counterbalance the perturbation caused by it (for a review, see Massion, 1992; Bouisset and Do, 2008). APAs are so intimately linked to the motor program that the involuntary neural command to postural muscles is shared with the voluntary command to prime mover muscles (Bruttini et al., 2014). The importance of APAs is apparent when considering motion of one single limb (e.g., Belen’kii et al., 1967 for the upper limb; Alexeief and Naidel, 1972 for the lower limb), in this context, the main goal of APAs is to minimize the changes in the body center of mass, to keep its projection within the support area, and to counteract the self-initiated postural perturbation. APAs also contribute to initiate the displacement of the body center of mass when starting gait (Brenière et al., 1987) or whole body reaching movements (Stapley et al., 1998, 1999). These activities, called inter-limb APAs, set one or more fixation chains spreading over several muscles of different limbs. APAs may also precede movements of tiny masses, like a brisk flexion of the index-finger when performed in a seated position (Caronni and Cavallari, 2009). In this case, they are named intra-limb APAs as the fixation chain is limited to muscles acting on proximal joints (Cavallari et al., 2016). Taking into account the small mass of the moving segments, and the fact that this movement cannot affect the whole-body balance, the role of intra-limb APA has been attributed to the precision of the final movement (Bruttini et al., 2016). Note also that despite the different aim, inter- and intra-limb APAs share many behavioral properties: they are distributed to muscles of the same limb in which the movement occurs, precede the onset of the voluntary movement, are polarized according to the task direction in space, are scaled to the amplitude of the perturbation and adapt to changes in the postural requirement of the task (Cavallari et al., 2016).

The control of body position in space develops with different intensity during life span (Assaiante et al., 2005; Soberaa et al., 2011). As an example, Zaino and McCoy (2008) showed that young healthy children (6–8 years old) exhibit much higher variability of posture control than older healthy children (10–12 years old). It is also reported that the age 7–9 years is an important period of their life in which children master postural control (Massion, 1998). Moreover, Schmitz et al. (1999) showed that children 3–4 years old develop APA, although they show coexistence of both adult-like and immature patterns, concluding that this anticipatory activities are being set up and that children are progressively mastering them.

It is also well known that all children have wetting and/or soiling accidents at one time or another. Achievement of urinary continence is an important developmental step that most children attain with the assistance of their parents and caregivers, however literature reports that about 15%–20% of children become partially toilet trained but continue to have wetting accidents even after the age 5 (Issenman et al., 1999). The first awareness of bladder and rectal functions usually occur between one and two years of age. The neural mechanisms involved in the storage and periodic elimination of urine undergo marked changes during prenatal and postnatal development (de Groat, 2002; Jansson et al., 2005). In the first years of life, voiding is controlled by a primitive spinal reflex pathway. Voluntary control over striated muscle sphincter usually occurs by the age of 3 years (Sillén, 2004). As the human CNS matures, higher brain centers contribute in modulating reflex voiding (see Figure 7 of Fowler et al., 2008). In adults, injury or disease of the nervous system can lead to the re-emergence of primitive reflexes Geirsson et al., 1993; Jiang et al., 2002).

Changes in brain structure are continuous throughout life (Chugani et al., 1987; Giedd et al., 1996). By the age 2, the brain has reached 75% of its adult weight (Carmichael, 1990) and the processes of synaptic pruning and cell death are most active during these early years (Huttenlocher, 1979; Huttenlocher et al., 1982). During the school-age years, strong signs of brain maturation are appreciable, especially in its connectivity (Lebel et al., 2008). MRI measures of the structure in fibers tracts correlate with behavioral indices that also change in this period (Schmithorst and Yuan, 2010; Tamnes et al., 2010a,b). Later changes involve the associative neocortex, which continues to develop well into the third decade (Yakovlev and Lecours, 1967), and the corpus callosum, which connects all major subdivisions of the cerebrum (Pujol et al., 1993).

Thus, before school-age all kinds of anticipatory actions, preceding primary movements, seems to be immature, either because the whole system is not still developed or because a part of it is not properly functioning. The achievement of a complete action (anticipatory and willed; postural or non-postural), therefore requires repetition learning (Draganski et al., 2004) and memory (Takeuchi et al., 2010).

As shown above, APAs and cough-anal response share many common features: they are both anticipatory with respect to the primary movement, so that they cannot be reflex responses but must be pre-programmed; they both prevent unwanted mechanical consequences (postural perturbation or sphincter leaking); they are both scaled to the amplitude of the perturbation; and finally, they seem to follow the same ontogenetic steps.

Indeed, a conceptual similarity may be envisaged among inter-limb APAs, intra-limb APAs and the anal response. In fact, in all cases a single predictive homeostatic principle (Moore-Ede, 1986; Freddolino and Tavazoie, 2012) seems to be followed, namely that anticipatory collateral actions are needed to correctly perform the primary movement.

In the case of inter-limb APAs, that minimize the changes of the body center of mass and that counteract the self-initiated postural perturbation, this principle provides the maintenance of stability (equilibrium) of the whole body during an action (Massion, 1992; Bouisset and Do, 2008). These APAs spread over several muscles of different limbs, creating one or more long fixation chains.

In the case of intra-limb APAs, that maintain the stability of the different segments of a single limb, it means to successfully carry through a precise and coordinated movement (Caronni and Cavallari, 2009; Caronni et al., 2013; Bruttini et al., 2016; Cavallari et al., 2016). These APAs spread over several muscles of the same limbs, creating short fixation chains.

In the case of the anal or urinary responses, the anticipatory contraction of the sphincter permits to prevent fecal or urinary leakage during an increase of abdominal pressure, due to a sudden action. A behavior that should be defined anticipatory adjustment as well.

A final comment deserves the possible neural networks that may be involved in the control of postural and sphincter muscles. Indeed, some analogies may be directly outlined, while other may be actually tested. First, as mentioned in the “Cough” section, animal experiments showed that the afferent inputs to the pontine and medullary nuclei are able to trigger a reconfiguration of the respiratory central pattern generator, changing its breathing activity into a cough pattern (Baekey et al., 2001, 2003; Canning et al., 2004; McGovern et al., 2012a). In parallel, it has been shown that the automatic gait pattern of the locomotion central pattern generator may be influenced by pontine neurons (for a review, see Takakusaki, 2017). However, the advanced pre-programmed control of posture required for skilled and goal-directed movements involve many cortical areas, as well as basal ganglia and cerebellum (see Takakusaki, 2017). In this regard, the involvement of supplementary motor area and cerebellum in APAs control has been experimentally demonstrated by Bolzoni et al. (2015) and Bruttini et al. (2015). It would then be interesting to test if magnetic or DC stimulation of these structures would produce comparable effects on the anticipatory recruitment of postural muscles and of external sphincters, during both voluntary movements and respiratory/cough tasks.

On these basis, it is proposed that a common pre-programmed (i.e., feed-forward) mechanism may govern several kinds of anticipatory actions. In fact, in all cases the brain seems to control in the same way the general predictive homeostasis of the body, accomplishing movements either voluntary- or urge-driven. Thus, Anticipatory Sphincter Adjustments are likely the visceral counterpart of APAs. Of course, this speculative conclusion is just an hypothesis, which needs further experimental testing. Indeed, it could be interesting to investigate the EMG activity of visceral muscles, such as pelvic floor muscles, during other movements like hiccupping or chest vs. diaphragmatic breathing, so as to ascertain if such activities are actually anticipatory also in these conditions. Other evidences may come from testing if stimulation of the supplementary motor area or the cerebellum would affect in parallel the activation of both postural muscles and external sphincters. Indeed such finding would strengthen the analogy between the cough-anal response and the APAs.

PC proposed the idea. PC and RE collected and critically analyzed the literature. All authors contributed in writing and discussing the manuscript. All authors approved the final version and agree to be accountable for all aspects of this work.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.