Achieving Adaptive Plasticity in the Spinal Cord
Summary and Keywords
The traditional view of central nervous system function presumed that learning is the province of the brain. From this perspective, the spinal cord functions primarily as a conduit for incoming/outgoing neural impulses, capable of organizing simple reflexes but incapable of learning. Research has challenged this view, demonstrating that neurons within the spinal cord, isolated from the brain by means of a spinal cut (transection), can encode environmental relations and that this experience can have a lasting effect on function. The exploration of this issue has been informed by work in the learning literature that establishes the behavioral criteria and work within the pain literature that has shed light on the underlying neurobiological mechanisms. Studies have shown that spinal systems can exhibit single stimulus learning (habituation and sensitization) and are sensitive to both stimulus–stimulus (Pavlovian) and response–outcome (instrumental) relations. Regular environmental relations can both bring about an alteration in the performance of a spinally mediated response and impact the capacity to learn in future situations. The latter represents a form of behavioral metaplasticity. At the neurobiological level, neurons within the central gray matter of the spinal cord induce lasting alterations by engaging the NMDA receptor and signal pathways implicated in brain-dependent learning and memory. Of particular clinical importance, uncontrollable/unpredictable pain (nociceptive) input can induce a form of neural over-excitation within the dorsal horn (central sensitization) that impairs adaptive learning. Pain input after a contusion injury can increase tissue loss and undermines long-term recovery.
Keywords: central sensitization, Pavlovian conditioning, instrumental conditioning, NMDA receptor, brain-derived neurotrophic factor (BDNF), tumor necrosis factor (TNF), spinal cord injury, pain, metaplasticity, hemorrhage
Our view of spinal cord function has shifted dramatically over the last 50 years. In the past, discussions of pain, learning, and locomotion focused on brain function, characterizing the spinal cord as a conduit for neural impulses to and from the periphery (Patterson, 2001a). From this perspective, neurons within the spinal cord are capable of organizing some simple reflexes, but otherwise serve as a neural relay that does not, in any substantive way, alter the flow of information to or from the brain. And when experiences bring about a change in spinal function, it was assumed that these emerged in a mechanical manner (reflexively elicited in response to environmental stimulation) or at the brain’s request (through descending fibers that regulate afferent signals and reflex vigor) (Basbaum & Fields, 1984; Millan, 2002). This characterization of spinal function focused upon the ascending/descending fiber tracts that make up the white matter, the outer region of spinal cord tissue composed of neural axons (a fiber-like structure that conducts neural impulses) and the oligodendrocytes that form the insulating myelin sheath (Martin, 1996). Nestled within this white matter is a butterfly-shaped region of tissue known as the central gray matter, which is composed of neurons and supportive tissue. It has long been recognized that neurons within the gray matter can organize some simple behaviors (spinal reflexes) and modulate the incoming (afferent) and outgoing (efferent) neural signals. However, the processing potential of the central gray matter was dramatically underestimated, with new data demonstrating that this region of the central nervous system (CNS) can organize some relatively complex behaviors (e.g., stepping) and change (learn) in response to environmental relations.
The present article reviews the findings that have transformed our view of spinal cord function, beginning with a discussion of the criteria for learning (Grau, 2014). Subsequent sections review the evidence for spinal learning, the neurobiological mechanisms involved, and the implications of this work for recovery after spinal cord injury (SCI). Research suggests that spinal cord neurons can encode environmental relations and that this learning impacts plastic potential in the future, demonstrating a form of behavioral metaplasticity (Grau & Huang, 2018; Grau et al., 2014; Schmidt, Abraham, Maroun, Stork, & Richter-Levin, 2013). New findings have shown that SCI can transform how neural tissue within the spinal cord is regulated, recapitulating an earlier developmental state that promotes plasticity by lessening neural inhibition (Grau & Huang, 2018; Kaila, Price, Payne, Puskarjov, & Voipio, 2014). Of particular clinical significance, research has revealed that the activation of pain (nociceptive) fibers after SCI can sensitize nociceptive processing in the upper (dorsal) region of the spinal cord, a change that is thought to contribute to the development of chronic pain (Latremoliere & Woolf, 2009; Malcangio, 2009). Nociceptive input can also increase tissue loss at the site of injury and undermine long-term recovery after SCI (Grau et al., 2017; Turtle et al., 2019).
From Immutable to Plastic: Changing View of Spinal Cord Function
Organization of the Spinal Cord
The spinal cord extends the central nervous system from the lower brain (brainstem) along the vertebrate back, relaying afferent and efferent neural signals to the periphery, which enter and exit through nerve bundles (ganglia) between the segments (vertebrae) of the bony spine (Martin, 1996). Afferent sensory fibers enter from the dorsal side while efferent motor commands exit through the ventral (abdominal side) spine. Anatomists have organized the segments of the spinal cord into four regions, starting with the cervical region that lies directly below (caudal to) the brain. Neurons within this region coordinate movements of the upper limbs and play an essential role in some biological functions (e.g., respiration). The thoracic region lies below the cervical spinal cord, extending down the middle back to the lumbar and sacral (tail) regions of the spinal cord. Neurons within the lumbosacral portion help to organize the movement of the lower limbs. Segments within each of these regions are numbered sequentially, with the upper-most (rostral) segment designated “1.” Sensory neurons project to the inner region of the spinal cord (the central gray matter), which is composed of interneurons and non-neuronal cells (e.g., astrocytes and microglia). Here, the signals can be processed and relayed to motor neurons in the ventral gray matter or, through axons that project along the outer surface of the spinal cord (the white matter), to other regions of the spinal cord or the brain. The white matter also contains descending fibers that project from the brain to the central gray matter.
To establish the functional limits of circuits within the spinal cord, minus communication with the brain, researchers typically cut (transect) the spinal cord above (rostral to) the area of interest (e.g., Durkovic, 2001; Grau, Barstow, & Joynes, 1998; Patterson, 1976). Because cervical injuries disrupt essential bodily functions (e.g., breathing), researchers often explore this issue by transecting the thoracic spinal cord. This surgery produces a paraplegia, in which the animal retains the capacity to move about using its forelimbs, but cannot feel stimuli applied below the waist and cannot “voluntarily” move its hind limbs. The functional capacity of the lower (lumbosacral) spinal cord can then be studied by applying stimuli caudal to injury, examining how this affects neural signaling or the capacity to move a lower limb. With this preparation, researchers have shown that (noxious) stimuli that can potentially induce tissue damage (e.g., a pin prick or burning heat) elicit a withdrawal response, a spinal reflex organized by neurons within the central gray matter (Sherrington, 1906). Historically, it was assumed that these simple stimulus-response (S-R) reflexes represented the functional limit of the gray matter, perhaps waning (habituating) or growing (sensitizing) with stimulus exposure, but otherwise seemingly immutable. Research conducted over the last 50 years has challenged this view, demonstrating that neurons within the spinal cord can support some basic forms of learning (Grau, 2014; Grau et al., 1998; Groves, De Marco, & Thompson, 1969; Patterson, 2001a). In the next section, key findings and methodological issues are introduced. After this foundation is established, evidence for spinal learning is presented, detailing the capacities and limits of spinal neurons and noting the neurobiological mechanisms involved.
Noxious Stimulation Alters Pain (Nociceptive) Processing Within the Spinal Cord
The transformation in thinking about spinal cord function is due, in part, to discoveries made within the pain literature, where it was shown that peripheral stimulation can modify how nociceptive signals are processed within the dorsal horn (Cordero-Erausquin, Inquimbert, Schlichter, & Hugel, 2016; Millan, 1999; Price & Inyang, 2015; Treede, 2016). For example, Melzack and Wall suggested that tactile stimulation can engage a neural gate within the spinal cord that inhibits the transmission of pain signals to the brain (Melzack & Wall, 1965). Subsequent studies showed that electrophysiological stimulation of sensory fibers at an intensity that engages unmyelinated pain (C) fibers induces a lasting increase in neural excitability within the spinal cord (Mendell, 1966; Willis, 2001).
In the learning and memory literature, researchers have shown that strong electrophysiological stimulation of neurons within the hippocampus can induce a lasting increase in neural excitability, a phenomenon known as long-term potentiation (LTP) (Bear, 2003; Malenka, 1994) (Long-Term Poteniation and Long-Term Depression). Neurobiologists have related this phenomenon to the release of the neurotransmitter glutamate and the activation of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors on the postsynaptic cell (Collingridge & Bliss, 1987; Morris, 2013). Engaging AMPA receptors (AMPAR) allows sodium (Na+) to enter the cell, which has an excitatory (depolarizing) effect. Another key player is the N-methyl-D-aspartate (NMDA) receptor (NMDAR), which acts as a kind of gated channel. Activating the NMDAR requires both the release of glutamate from the presynaptic cell and a strong depolarization of the postsynaptic cell. If both conditions are met, calcium (Ca++) is allowed into the postsynaptic cell, which engages signal pathways that increase the number of AMPARs in the active zone of the synapse, increasing the electrophysiological response elicited by glutamate. Research has shown that, in many instances, the development of LTP depends upon the activation of the NMDAR and the movement (exocytosis and trafficking) of AMPARs to the synapse. The subsequent discovery that activation of C-fibers can induce a form of LTP within the spinal cord (Liu, Morton, Azkue, Zimmermann, & Sandkuhler, 1998; Sandkuhler, Chen, Cheng, & Randic, 1997; Sandkuhler & Liu, 1998), and that this phenomenon depends upon a form of NMDAR-mediated plasticity (Dickenson & Sullivan, 1987; Woolf & Thompson, 1991), built a bridge to the learning and memory literature and implied that spinal cord neurons had the machinery needed to learn (Ji, Kohno, Moore, & Woolf, 2003; Sandkuhler, 2000)
Subsequent research extended these observations by showing that the peripheral application of an irritant could induce a lasting sensitization of nociceptive circuits within the dorsal horn (Coderre, Katz, Vaccarino, & Melzack, 1993; Willis, 2001; Woolf, 1983) (Neuroinflammation and Neuroplasticity in Pain). This is often studied using capsaicin, the active ingredient of chili peppers. Capsaicin selectively engages C-fibers that express the transient receptor potential cation channel subfamily V member 1 (TRPV1) receptor. Peripheral application of capsaicin enhances reactivity to stimuli that engage pain fibers (hyperalgesia) and can transform how innocuous mechanical stimuli are processed, an alteration that allows myelinated fibers that normally signal touch to engage nociceptive circuits within the spinal cord. This phenotypic shift in function is thought to contribute to the development of allodynia (pain to light touch), a key component of chronic pain (Lamotte, Shain, Simone, & Tsai, 1991; Simone, Baumann, & Lamotte, 1989).
Stepping Without a Brain
Concurrent work in the locomotor literature also played a pivotal role in transforming our view of spinal cord function (Edgerton, Roy, DeLeon, Tillakaratne, & Hodgson, 1997; Edgerton, Tillakaratne, Bigbee, de Leon, & Roy, 2004; Forssberg, Grillner, Halbertsma, & Rossignol, 1980; Grillner, 1973; Rossignol, Schwab, Schwartz, & Fehlings, 2007; Shurrager & Dykman, 1951) (Plasticity of Stepping Rhythms in the Intact and Injured Mammalian Spinal Cord). Paraplegic cats were placed in a harness that supported their body weight with their hind limbs positioned over the belt of a treadmill. Height was adjusted so that, when the belt was started, it stimulated the paws. As would be expected, the animal’s legs were initially dragged back by the belt, with the simulation occasionally eliciting some reflexive behavior but no signs of stepping. But with weeks of daily training, stepping recovered, elicited by a combination of tactile feedback to the paws and stimulation of the tail. This implies that neurons within the spinal cord can organize coordinated stepping, a functional capacity that has now been linked to a neural oscillator (central pattern generator) hypothesized to lie within the rostral lumbosacral spinal cord (Rossignol & Frigon, 2011). Moreover, once coordinated stepping emerges, neurons within the spinal cord can adjust the rate of stepping to treadmill speed (Edgerton, de Leon, et al., 1997; Edgerton, Roy, et al., 1997). Interestingly, the recovery of locomotor performance was not accompanied by a change in the ability to stand with the hind limbs (Hodgson, Roy, de Leon, Dobkin, & Edgerton, 1994). Conversely, animals trained to stand did not exhibit improved locomotor capacity. Together, the findings imply that spinal neurons can coordinate a complex behavior (stepping), that training can bring about a lasting change in the capacity to exhibit this behavior (without input from the brain), and that the effect of training may be specific to the response system trained.
Learning Without a Brain
Research within the areas of pain and the physiology of locomotion suggests that exposure to environmental stimulation can have a lasting effect on spinal cord function, implying a capacity to learn. However, fully addressing this question requires a closer look at what is meant by “learning”—how this term has been traditionally defined and the criteria used to identify it (Grau, 2014; Grau & Joynes, 2001). Doing so highlights essential methods and distinctions that inform our study of the underlying processes, helping to clarify the capacities and limits of learning without a brain.
Brain-Dependent Learning Can Modify Spinal Cord Function
As detailed by Rescorla, to demonstrate learning evidence is needed that an experience at time-1 has a lasting effect when animals are tested at time-2 (Rescorla, 1988). Researchers have described this as involving two interrelated processes: (1) the initial encoding of the events (learning) that brings about a neural modification, and (2) the maintenance of the consequence of learning (the memory) over time (Domjan, 2015).
The distinction between learning and memory helps to clarify the difference between alternative methods for probing spinal function. One approach asks whether brain-dependent learning can bring about an alteration in spinal cord function (the memory). Early evidence that the brain can induce a lasting modification in spinal function was provided by DiGiorgio (1929), who discovered that cerebellar lesion can produce a postural asymmetry of the hind limbs, with one leg flexed and the other extended. If communication with the brain was cut by means of a thoracic transection within 30 minutes, the asymmetry faded. If, however, the cut was made an hour later, the asymmetry persisted, implying that the maintenance of the behavioral modification no longer depended upon brain input. Further work, reviewed in Patterson (2001b), has shown that this effect depends upon an intraspinal modification.
Wolpaw and colleagues have taken this line of work one step further, demonstrating that brain-dependent learning can induce a lasting modification in spinal function (Wolpaw, 2010; Wolpaw & Carp, 1990). They have explored this issue using the Hoffmann reflex (H-reflex), an electrical analog of the spinal stretch reflex (a muscle contraction elicited by stretching the muscle). Animals can be trained to exhibit either an increase or a decrease in H-reflex magnitude using an appetitive reward (e.g., a sip of fruit juice). For example, an animal might be given fruit juice (the reinforcer) if it exhibits a stronger muscle contraction (up conditioning) when the H-reflex is elicited. The capacity to do this improves with training, bringing about a lasting change in performance. Notice that, in this paradigm, brain systems must integrate the relation between the appetitive reward (the outcome) and the change in performance (the trained response). Here, the process of learning depends upon neural mechanisms within the brain. But what maintains the behavioral modification over time? What underlies the memory? One possibility is that this too is mediated by the brain, which is able to modulate the performance of the H-reflex through descending fibers. Alternatively, repeatedly inducing a particular change in H-reflex performance could bring about a modification within the spinal cord that helps to maintain the required behavioral effect over time. To explore these alternatives, Wolpaw trained his animals and then cut communication with the brain by transecting the spinal cord. He found that, after extended training, the modification in H-reflex magnitude survived a spinal cut, implying that brain-dependent learning can induce a lasting modification (memory) in spinal function (Wolpaw & Carp, 1990).
Criteria for Learning
The present article will focus on paradigms where both the learning and the memory are mediated by processes outside the brain, cases wherein stimuli applied caudal to a spinal transection at time-1 induce a change in performance when animals are tested at time-2. Why is testing at time-2 required? At issue is another key distinction, the difference between learning and performance. Exposure to environmental events often induces a momentary alteration in the capacity to detect stimulation (e.g., sensory receptor adaptation) or perform the target response (e.g., motor fatigue) that may depend upon changes within non-neuronal cells (e.g., the sensory receptor or muscle fibers) (Domjan, 2015). In these cases, the consequences of stimulation generally fade soon after the training has ended—the experience does not have a lasting effect. A learning theorist would ascribe changes of this sort to alterations in performance, not to learning. To invoke the concept of learning requires evidence that the consequence of training outlasts the training experience and yields a behavioral modification that is evident when animals are subsequently tested under common conditions (Grau, 2014; Rescorla, 1988). There are two interrelated issues here. The first concerns our definition of lasting, for which no standard answer has been given. In practice, researchers generally require evidence that the behavioral effect lasts 24 hours or longer.
The second issue concerns the requirement that animals be tested under common conditions. To demonstrate learning, researchers typically compare two groups that differ in whether they experienced a particular event or relation. For example, a hind limb flexion can be elicited in spinally transected animals using electrical stimulation applied to the skin or a cutaneous nerve (Thompson & Spencer, 1966). If the stimulus is repeatedly applied, the magnitude of this response will decline (habituate) over trials. Using this paradigm, researchers examined whether the rate of habituation varies as a function of stimulus intensity. Early evidence appeared to suggest that the magnitude of the decline (habituation) observed across (n) trials was greater when animals were exposed to a relatively weak stimulus (Groves, Lee, & Thompson, 1969). But notice that, at the end of training, the animals differ in two ways: their prior experience on trials n-1 and the intensity of the test stimulus (on trial n). To evaluate which factor is critical, subjects must be tested under common conditions using stimuli of the same intensity. When this was done, the results supported the opposite conclusion, showing that exposure to more intense stimulation brings about a greater decrement in response vigor (Davis & Wagner, 1968). There are many forms of stimulation and training that can bring a change in behavior. To claim that any one of these represents an example of learning requires evidence that the experience has a lasting effect that is evident when animals are tested under common conditions. And to claim that this represents an example of learning and memory within the spinal cord, it must be shown that the initial encoding of the events, and the preservation of the behavioral modification over time, can occur without input from the brain.
Environmental events and relations can impact biological systems in a variety of ways, affecting both neuronal and non-neuronal systems. Indeed, SCI can activate astrocytes and microglia that interact with neural systems to regulate plasticity (Haydon, 2001; Perea, Navarrete, & Araque, 2009; Vichaya, Baumbauer, Carcoba, Grau, & Meagher, 2009). While this is acknowledged, the focus within the learning and memory literature is generally upon neurally mediated changes. It is also recognized that common neurobiological processes contribute to adaptive neural plasticity during development, after injury, and in response to environmental events/relations. Further, SCI may allow learning by recapitulating an earlier developmental state (Grau & Huang, 2018; Huang, Lee, Murphy, Garraway, & Grau, 2016).
In summary, three criteria must be met to demonstrate learning, providing evidence that: (1) the behavioral modification depends upon a form of neural plasticity; (2) the modification depends on the organism’s experiential history; and (3) the modification outlasts (extends beyond) the environmental contingencies used to induce it (i.e., it has a lasting effect on performance) (Grau, 2014).
Why Does Learning Matter?
Evidence for spinally mediated learning is reviewed in the sections that follow. But first it is worth considering a key question: why does learning matter? Why should we care whether or not a neural mechanism can support learning?
One reason is that it shapes our view of how learning is represented within the CNS. Is it a process that is mediated by a particular neural structure or a capacity that is widely distributed across the CNS? More generally, one might ask whether the capacity for learning is an inherent feature of neurons, a function that these cells evolved to perform.
Additionally, learning is critical to treatment after SCI. The recognition that training can bring about a lasting modification in spinal cord function fueled interest in the application of treadmill training and physical therapy (PT) (Dietz & Harkema, 2004; Wernig, Muller, Nanassy, & Cagol, 1995). Likewise, research demonstrating that noxious input can induce a lasting sensitization of nociceptive circuits within the spinal cord has shed light on the development of chronic pain after injury (Latremoliere & Woolf, 2009; Mendell & Wall, 1965; Willis, 2001). The implication is that an understanding of learning will illuminate the types of training that should promote adaptive plasticity, while minimizing the induction of processes that undermine function and well-being.
Learning is also relevant to treatments designed to foster neural regeneration and the re-wiring of spared fibers. Procedures are needed to encourage adaptive connectivity, to promote function and avoid the types of mis-wiring that might adversely affect recovery (e.g., by enhancing pain or spasticity). It is suggested that learning can be used to tune the pattern of neural connections to foster long-term recovery. And here too, views regarding the distribution of learning across the CNS will impact our theory of how PT impacts neural function. It is, of course, recognized that PT will engage brain-dependent processes that can foster adaptive responses. The work reviewed in the present article suggests that training can also impact spinal function by building upon surviving circuits and plastic potential caudal to injury.
Evidence for Spinally Mediated Learning
Learning theorists have traditionally classified alternative forms of learning based upon the environmental events and relations used to induce the phenomenon (Figure 1) (Domjan, 2015; Grau & Joynes, 2005). For example, exposure to a single stimulus may strengthen (sensitize) or weaken (habituate) the consequent response. If the learning depends upon the relationship between two stimulus events, the learning is classified as an example of Pavlovian (classical) conditioning. If the key relation is between a behavior (the response) and an environmental stimulus (the outcome), the learning is categorized as an example of instrumental conditioning. Subsequent sections review evidence that neurons within the spinal cord can support these basic forms of learning.
Habituation and Sensitization
Sherrington was one of the first to systematically explore the functional capacities of the spinal cord—what it can do minus input from the brain (Sherrington, 1906). In a typical experiment, communication with the brain was cut by means of a thoracic transection, and behavioral reactivity was probed using stimuli applied caudal to injury. A withdrawal response can be elicited by applying a stimulus that engages pain (nociceptive) fibers to an animal’s paw or tail. With weak to moderate stimulation, the magnitude of this response will typically wane (habituate) with repeated exposure. Interestingly, Sherrington also noticed that the application of an irritant can elicit a “scratching” response, implying that neurons within the spinal cord can organize rhythmic behavior.
The classic work by Groves and Thompson extended these observations (Groves, De Marco, et al., 1969; Groves & Thompson, 1970). In their studies, a hind leg flexion was elicited in spinally transected animals using electrical stimulation applied to the skin or a cutaneous nerve. As expected, when a weak to moderate stimulus was repeatedly applied, the magnitude of the response declined across trials. Using electrophysiological techniques, they linked this habituation effect to an intraspinal modification (a reduction in interneuronal efficacy) that selectively reduced response vigor in the underlying S-R pathway. Because this modification occurs within a particular S-R pathway, habituating the animal to stimulation applied to one hind leg has little effect on responsiveness to stimulation applied to the opposite leg.
When an intense stimulus was applied to a hind leg, it increased (sensitized) behavioral reactivity, and this was true independent of whether the animals were tested on the same (ipsilateral) or opposite (contralateral) leg (Groves & Thompson, 1970). This suggests that intense stimulation can engage a process within the spinal cord that augments neural excitability independent of the site of stimulation. Groves and Thompson integrated these observations within a dual process theory of habitation and sensitization, ascribing the former to a decrement within an S-R circuit and the latter to the activation of a state system that generally augments behavioral reactivity.
The dual process model is designed to account for a non-associative phenomenon—how stimulus exposure, in the absence of any relational signals (to behavior, internal cues, or environmental events) impacts behavioral reactivity. Today, it is widely accepted that spinal neurons can support this type of learning. In recent years, the focus has shifted to the sensitization of behavioral reactivity in response to noxious stimulation that can induce a lasting increase in neural excitability within the spinal cord (Latremoliere & Woolf, 2009; Willis, 2001). Controversies in this domain have centered on the identification of the underlying neurobiological processes. While early research focused on changes within the CNS, recent studies suggest that alterations in the afferent input may also play a role (Nickel, Seifert, Lanz, & Maihofner, 2012; Yang et al., 2014).
Learning about Stimulus–Stimulus (S–S) Relations: Pavlovian Conditioning
What has proven more controversial is whether spinal cord neurons are sensitive to environmental relations, an issue that is often explored using the methods of Pavlov (Patterson, 2001a; Pavlov, 1927). In this learning paradigm, a cue (the conditioned stimulus—CS) is paired with a biologically meaningful stimulus (the unconditioned stimulus—US) that innately elicits a behavioral response (the unconditioned response—UR). Pavlovian (aka classical) conditioning is evident when a CS that has been paired with a US acquires the capacity to elicit a response (the conditioned response—CR).
To explore whether spinal cord systems are sensitive to stimulus–stimulus (S–S) relations, early researchers used moderate electrical stimulation (shock) of the tail as the CS (Shurrager & Culler, 1940). The US was an intense shock to the leg that elicited an unconditioned flexion response. In later studies, stimulation of a nerve (e.g., superficial peroneal) that elicits a leg flexion was used as the US, while stimulation of a distinct nerve (e.g., saphenous) served as the CS (Durkovic, 2001). In both cases, researchers showed that pairing the CS with the US strengthened the CS-elicited flexion response. Further, presenting the trained CS alone caused the CR to wane (extinguish).
Interestingly, more robust learning was typically observed when animals were trained soon after spinal cord transection, implying that plastic potential fades with time (Durkovic, 2001). There are two implications of this observation. One is that behavioral training may have a greater effect if started soon after injury (Brown, Woller, Moreno, Grau, & Hook, 2011). The corollary of this is that inducing changes within the spinal cord weeks to months after injury may benefit from the co-administration of treatments designed to reinstate plastic potential (Edgerton et al., 2008).
Pavlovian conditioning has also been demonstrated using cutaneous stimuli that engage nociceptive fibers. In intact animals, it is well known that a CS that predicts a painful event (e.g., an unexpected shock [the US]) acquires the capacity to elicit an antinociception (the CR) that inhibits the incoming nociceptive signal (Chance, White, Krynock, & Rosecrans, 1977; Watkins, Cobelli, & Mayer, 1982), a modulatory effect that is often linked to the release of an endogenous opioid (Fanselow, 1986; Grau, 1987; McNally, Johansen, & Blair, 2011). Evidence suggests that a CS–US relation can impact nociceptive processing within the spinal cord without input from the brain (Grau, Salinas, Illich, & Meagher, 1990). This was shown using weak shock to one hind leg as the CS and an intense tail-shock as the US. Here, the US was provided at an intensity that generated an unconditioned inhibition of a spinal reflex (tail withdrawal from a noxious thermal stimulus; the tail-flick test). To explore whether this system was sensitive to environmental relations, stimulation of one hind leg (the CS+) was paired with tail-shock (the US) while stimulation to the other leg was presented in an explicitly unpaired manner (the CS-). This differential conditioning provides a powerful, within subjects, test of whether the CS–US relation matters. After training, animals were less responsive to a noxious thermal stimulation applied to the tail during the CS+, relative to CS-, demonstrating a form of conditioned antinociception (Joynes & Grau, 1996).
In the field of learning, researchers have shown that pre-exposure to a cue alone can impair the development of a conditioned response (latent inhibition) (Domjan, 2015). Learning about a CS is also impaired when the CS is presented in combination with a cue that is more noticeable (overshadowing) or that has been previously trained (blocking). Research has shown that classical conditioning within the spinal cord exhibits each of these phenomena (Illich, Salinas, & Grau, 1994).
While it seems clear that spinal cord neurons are sensitive to the temporal relationship between stimuli, and in this way meet the criteria of Pavlovian conditioning (Joynes & Grau, 1996), there are phenomena that likely lie outside of its behavioral repertoire. This includes learning about a CS–US relationship when the events are separated by a gap in time (trace conditioning), or when a configuration of stimuli (e.g., context) signals whether the US will occur (Solomon, Vander Schaaf, Thompson, & Weisz, 1986; Sutherland, 1989). There is also no evidence that spinal conditioning is sensitive to US devaluation (Colwill & Rescorla, 1986).
A critic could take issue with the fact that the CS used in studies of spinal conditioning has some capacity to generate a CR-like response prior to training (Grau & Joynes, 2005; Patterson, 2001a). The implication is that learning does not endow the CS with the ability to generate a new response but instead alters the vigor of a biologically pre-wired S-R reflex. The key question is whether this should contradict the claim of learning. As discussed elsewhere (Grau, 2014), following this course would severely limit the significance of Pavlovian conditioning, because most models of the phenomenon build upon a biologically prepared response, and the clinical phenomena (e.g., phobias) that motivate their study typically involve relations that our nervous system has been tuned to detect. An alternative view (neurofunctionalism [Grau & Joynes, 2005]) acknowledges that the nervous system can solve environmental puzzles in multiple ways. From this perspective, given evidence that the system is sensitive to environmental relations, the critical question becomes: how is this encoded at a functional and neurobiological level? Such a view assumes that learning about S–S relations in a simpler system, such as the spinal cord or an invertebrate, builds upon the modification of pre-existing S-R habits. More sophisticated neural mechanisms (e.g., in the brain) presumably embellish this capacity, to link a wider range of events, span a gap in time, or alter performance given new information (e.g., US devaluation).
Learning About Response–Outcome (R–O) Relations: Instrumental Conditioning
While it was doubted by some that spinal cord neurons could encode Pavlovian (S–S) relations, nearly all assumed that this component of the CNS lacked the machinery needed to encode instrumental (R–O) relations (cf. Buerger & Chopin, 1976; Chopin & Buerger, 1976; Church, 1989; Church & Lerner, 1976). To explore this issue, researchers applied a noxious stimulus (the O) to one hind leg of a transected rat (Figure 2) (Grau et al., 1998). The O (aka reinforcer) typically involved electrical stimulation of the tibialis anterior muscle, which elicits a flexion response. Leg position is monitored by taping a short (e.g., 4 cm) rod to a rat’s paw and placing a salt solution under the leg (Figure 2A). Whenever the leg falls, the rod touches the underlying salt solution, completing a circuit that is monitored by the control equipment. With this simple apparatus, an R–O relation can be instituted by applying shock whenever the leg is extended. Because the application of the electrical stimulation (shock) is determined by the animal’s leg position, the shock is controllable. To determine whether this R–O relation matters, other subjects are experimentally coupled (yoked) to the animals given controllable stimulation (master). Each animal in the yoked group is paired with a subject in the master condition and receives shock at the same time and for the same duration. Thus, for the yoked animal shock occurs in an uncontrollable manner, independent of its leg position. Using this paradigm, researchers showed that spinally transected rats perform differently depending upon whether the stimulation is given in a controllable or uncontrollable manner (Buerger & Chopin, 1976; Chopin & Buerger, 1976). For example, relative to the master rats, yoked rats spent more time with their hind limb in an extended position and, on this basis, it was suggested that spinal neurons can support a simple form of instrumental conditioning.
Early evidence for spinally mediated instrumental learning was soon challenged on a number of grounds (Church, 1989; Church & Lerner, 1976). One problem is that the behavioral contingency used in training could potentially impact performance in the absence of learning (Church, 1964). For example, master animals may contact the underlying solution less simply because the onset of shock drives the leg up as soon as it is extended. In contrast, on roughly half the trials, the yoked animal’s leg would fall faster and remain extended until its master partner received a shock. For this reason alone, the yoked animal would be expected to spend more time with its leg extended and this behavioral difference would emerge even if the system responded in a mechanical manner, without any capacity for learning.
If an intact (uninjured) animal is given shock to one hind leg whenever the leg is extended, the shock will do more than drive the leg up in a mechanical manner. Over time, the animal will learn to maintain the leg in a flexed position that minimizes net shock exposure. This increase in flexion duration would not be expected to emerge if shock were given independent of leg position (Grau et al., 1998). These considerations suggest that response (flexion) duration may provide a superior measure of learning, one that is less open to alternative interpretations. With this change, and other refinements, Grau et al. (1998) showed that response-contingent (controllable) shock applied to one hind leg produced a progressive increase in response duration in master, but not yoked, rats and that this learning emerged after communication with the brain was blocked by means of a T2 transection (Figure 2B). Animals were then tested under common conditions with controllable stimulation applied to the same leg. During testing, master animals learned faster than the previously unshocked controls, demonstrating a form of positive transfer that indicates the earlier learning episode had a lasting effect. Surprisingly, animals that had received uncontrollable shock (yoked) failed to learn, and this was true even though the shock continued to elicit a flexion response. The fact that uncontrollable stimulation impairs spinal learning parallels what is often observed in brain-dependent tasks, a phenomenon known as learned helplessness (Maier & Seligman, 1976).
To study the long-term effects of exposure to uncontrollable shock, a computer program was developed that applied shock using a variable schedule that emulated the pattern of shock produced by a typical master animal (Crown, Ferguson, Joynes, & Grau, 2002b). Using this procedure, it was shown that just six minutes of variable intermittent electrical stimulation, applied to the leg or the tail, induced a modification within the lower lumbosacral spinal cord that interfered with adaptive learning for 24–48 hours.
Exposure to controllable stimulation has the opposite effect—it enables learning when animals are tested with a higher response criterion (Crown, Ferguson, Joynes, & Grau, 2002a). The response criterion was increased by raising the level of the underlying solution, so that a stronger flexion response was needed to hold the contact electrode above the solution. This change made the task so difficult that untrained animals failed to learn. However, animals that had previously experienced controllable shock were able to learn, and this was true independent of whether they were tested on the same or opposite leg. Interestingly, exposure to controllable stimulation also has a protective/restorative effect that counters the adverse consequences of uncontrollable shock. For example, a rat given controllable shock to one hind leg is behaviorally immunized from developing a learning impairment when exposed to uncontrollable shock applied to the opposite leg or tail (Crown & Grau, 2001). This too parallels what is observed in brain-dependent tasks (Maier & Seligman, 1976).
These observations imply that exposure to an R–O relation has multiple consequences. One involves a modification within the trained S-R pathway, demonstrating that exposure to controllable (response-contingent) stimulation can produce an increase in response duration, an alteration that impacts only the trained limb. The second involves a change in the capacity to learn in future situations, independent of whether testing occurs on the same or opposite limb (Crown et al., 2002a; Joynes, Ferguson, Crown, Patton, & Grau, 2003). Prior exposure to uncontrollable stimulation generally undermines the capacity to learn, whereas exposure to controllable shock has the opposite effect, inducing a process that enables learning and counters the development of the learning impairment (Crown & Grau, 2001). Effects of this type, wherein a learning experience has a general influence on the capacity to learn in future situations, have been characterized as a kind of behavioral metaplasticity (Grau & Huang, 2018; Schmidt et al., 2013).
How could spinal cord neurons learn about an R–O relation? To answer this question, we need to know what aspect of the stimulation reinforces the behavioral modification. Do animals given controllable shock exhibit an increase in flexion duration because raising the leg is reinforced by shock offset (escape) or is the key event shock onset, which the system avoids by maintaining a flexion response? The relative importance of these factors can be evaluated by disrupting the temporal relationship between the response and the hypothesized reinforcement, by independently delaying the onset or offset of shock. When this was done, it was found that delaying shock onset disrupted learning while a delaying shock offset had no effect (Grau et al., 1998). This suggests that the key outcome involves the onset of noxious stimulation.
Instrumental conditioning involves learning about an R–O relation. Having identified the effective O (shock onset), what constitutes the R? More specifically, how could spinal neurons recognize that shock begins at a particular leg position? Given the limits of the system, it has been assumed that the R is related to sensory input from proprioceptive fibers that indicate leg position and movement (Grau, 2014; Grau et al., 2012). Therefore, the key relation is between an internal cue of leg position (the R) and the onset of noxious stimulation (the O). What is intriguing is that this explanation provides a Pavlovian account of the instrumental behavior, wherein a sensory cue of leg position serves as the CS and shock onset acts as the US. From this perspective, the regular pairing of the sensory cue (the CS) with shock onset (the US) endows the CS with the capacity to elicit a stronger flexion response (the CR).
The distinction between Pavlovian and instrumental conditioning turns on the encoding of distinct environmental relations (S–S versus R–O) and it is clear that spinal cord neurons are sensitive to both. Yet, at a mechanistic level, similar processes may be involved that build upon pre-existing S-R pathways. Such a conceptual framework is consistent with the way in which others have characterized simple forms of instrumental learning in intact animals (Konorski & Miller, 1937).
There is, however, an alternative tradition in the learning literature that challenges reflexive accounts of instrumental behavior. Indeed, this was a hallmark of Skinner’s approach, which questioned whether schedules of reinforcement act by strengthening particular S-R reflexes (Skinner, 1938). Consider, for example, a rat bar-pressing for food. What was traditionally claimed is that the food reinforcer strengthens this behavior by stamping in a particular S-R connection. Here, it was assumed that some feature of the bar serves as the stimulus and that this cue gains the capacity to mechanically elicit a particular response. But to Skinner, the behavior exhibited by the rat did not seem mechanical in nature. Rather, it appeared that the rat could operate on its environment in a variety of ways. Perhaps more troubling, it is not clear what cue elicits the response. One could suppose a feature of the bar, or its surroundings, served this function, but it was typically impossible to state what the particular feature was (or even whether an individual element served this purpose). Given these observations, Skinner dispensed with the S-R account of instrumental behavior, suggesting that operant behavior was emitted, not elicited. From his perspective, the stimulus environment acts by setting the occasion to respond.
For both instrumental and operant behavior, researchers agree that performance is wedded to the underlying R–O relation. What differs is the presumed relation to elicited behavior. The term instrumental conditioning is typically used in a broad manner, to include examples of R–O learning that involve modifications of preexisting (biologically prepared) S-R pathways as well as behavior that Skinner would characterize as emitted. From this view, operant behavior represents a subset of instrumental learning, involving response systems that appear more flexible and less biologically prepared (Grau, 2014).
Evidence for spinally mediated instrumental learning comes from paradigms that rely upon an elicited response, involving a modification of a preexisting S-R pathway. The spinal cord involves a behavioral system that seemingly lacks the sort of flexibility Skinner saw in operant behavior. These observations parallel the arguments made previously regarding Pavlovian conditioning, where it was suggested that the brain brings a capacity to integrate stimulus relations across a variety of modalities and temporal relations. In contrast, spinally mediated learning is biologically constrained, acting to shape preexisting response systems. These considerations suggest that the terms operant and instrumental conditioning should not be treated as synonyms, because the former involves additional behavioral criteria (Grau, 2010). Ideally, for operant behavior neither the nature of (1) the behavioral change nor (2) the reinforcer are biologically constrained (Grau, 2014). While demonstrations of such constraint-free behavior may be hard to find (Timberlake, 1990; Timberlake & Lucas, 1989), it is clear that brain-dependent systems support a level of flexibility well beyond that of spinal systems.
Learning About Temporal Relations
The effect of uncontrollable intermittent stimulation on spinal cord function has often been studied using a computer program that emulates the shock schedule produced by an animal with behavioral control (Crown et al., 2002b). This was achieved by presenting brief (100 millisecond) pulses of electrical stimulation spaced an average of two seconds apart. Because master animals exhibit some variability in the interval between each response, the interval between the stimuli is typically varied (e.g., between 0.2 and 3.8 seconds). Using this variable time (VT) schedule it was shown that 180–900 shocks produces a lasting learning impairment.
Surprisingly, changing how the stimuli are distributed across time can transform how the stimulation affects spinal function. If the shocks are given in a regular manner, with the temporal interval fixed at two seconds, 900 shocks does not induce a learning impairment (Baumbauer et al., 2008). Instead, it has the opposite effect, inducing a form of behavioral metaplasticity that can counter the adverse effect of variable stimulation, a protective effect that lasts at least 24 hours (Baumbauer, Huie, Hughes, & Grau, 2009). Further work has shown that learning that the stimuli occur in a regular manner requires extended training. If spinally transected rats receive 180 shocks with the time between shocks fixed (a fixed time—FT—schedule), the same effect as VT stimulation is observed, leading to a learning impairment. The beneficial effect of regular stimulation does not emerge until animals receive 540–720 shocks. Interestingly, spinal systems can abstract regularity when two bouts of 360 FT shocks are given separated by 24 hours, implying a form of savings (Lee, Huang, & Grau, 2016; Lee et al., 2015). Indeed, spinal cord neurons appear capable of abstracting regularity when the locus of stimulation is varied and even when some of the stimuli are randomly omitted. The latter finding, coupled with other empirical observations, suggests that regular stimulation engages a kind of internal oscillator, potentially linked to the central pattern generator hypothesized to underlie the generation of regular stepping (Grillner & Zangger, 1979; Nishimaru & Kudo, 2000).
The fact that variable- and fixed-spaced stimulation have divergent effects on spinal function implies that spinal neurons are sensitive to temporal relations, a capacity that was previously linked to brain function (Mauk & Buonomano, 2004). Instead, sensitivity to temporal relations may be an inherent property of neural systems. Indeed, recent work has shown that even individual neurons/fiber pathways are sensitive to temporal relations (Johansson, Hesslow, & Medina, 2016; Johansson, Jirenhed, Rasmussen, Zucca, & Hesslow, 2014; Perrett, Dudek, Eagleman, Montague, & Friedlander, 2001).
The fact that FT stimulation has a distinct effect implies that it must engage a cue related to time, a physiological process tied to temporal duration. From this perspective, stimulus regularity introduces a form of temporal predictability, wherein an internal cue of time (the CS) comes to predict the occurrence of the next shock (the US), a type of Pavlovian conditioning known as temporal conditioning (Grau, 2014; Pavlov, 1927).
Interestingly, learning about temporal and instrumental relations appears to impact adaptive potential in the same way—both predictable and controllable stimulation induce a form of behavioral metaplasticity that can counter the adverse effect of uncontrollable/unpredictable stimulation (Grau & Huang, 2018; Grau et al., 2014). Conversely, an extended exposure to uncontrollable intermittent stimulation only induces a lasting learning impairment if the stimulation occurs in an unpredictable manner.
An implication of these findings is that relational learning can transform how spinal systems operate, to induce a lasting change in adaptive potential. This functional description of the factors that regulate spinal cord plasticity builds upon the power of learning theory, which seeks to identify regularities in how events are processed independent of the particulars of the training situation and response system. This view also suggests a new way to think about how alternative training regimens affect spinal function. For example, training transected animals to step in a regular manner on a treadmill may benefit spinal function because it introduces both a form of behavioral control and temporal regularity (Lee et al., 2015). A learning approach suggests that tuning the training situation to amplify these factors should bolster the benefits of training.
It has been shown that introducing an environmental relation can have a lasting effect on spinal cord function. It should be noted, however, that this does not necessarily imply that the particulars of the relation are abstracted and stored. To illustrate this point, consider the claim that temporally regular stimulation has a restorative effect and that this can be observed when the requisite number of stimuli are presented across two days. What is of interest is that this occurs independent of whether the stimulus frequency is the same or different across days (Lee et al., 2015). What appears critical here is that each bout of stimuli occur in a regular manner, not whether the stimuli occur at the same frequency. The implication is that the key variable that underlies the induction and maintenance of the metaplastic effect is tied to the experience with regularity. This is clinically important because it suggests that the benefits of physical therapy are coupled to temporal predictability and behavioral control within a session, and that varying the particulars across session may have no adverse effect. The finding also reveals another way in which spinal and brain systems differ, because it is clear that higher neural systems can store and retain specific temporal durations (Mauk & Buonomano, 2004).
NMDA Receptor (NMDAR) and Brain-Derived Neurotrophic Factor (BDNF)
It was noted previously that key receptors and signal pathways implicated in brain-dependent learning and memory are expressed within the spinal cord (Ji et al., 2003; Price & Inyang, 2015). Of particular import is the NMDAR, which can strengthen a neural connection in response to conjoint activity in the presynaptic and postsynaptic neurons (a form of Hebbian plasticity [Dudai, 1989]). Evidence suggests that this receptor plays a pivotal role in learning within the spinal cord. Supporting this, application of an NMDAR antagonist (e.g., MK-801 or APV) to the spinal cord (through an intrathecal—i.t.—catheter) blocks the development of central sensitization and learning about Pavlovian (S–S), instrumental (R–O), and temporal relations (Baumbauer, Huie, et al., 2009; Dickenson & Sullivan, 1987; Durkovic & Prokowich, 1998; Joynes, Janjua, & Grau, 2004; Ma & Woolf, 1995). Administration of MK-801 (i.t.) also blocks the development of a learning impairment in animals given unpredictable/uncontrollable stimulation (Ferguson, Crown, & Grau, 2006).
Another key target involves the metaplastic consequence of exposure to controllable stimulation, which enables learning and counters the learning deficit induced by uncontrollable/unpredictable stimulation (Crown et al., 2002a; Crown & Grau, 2001). Research has linked this restorative/protective to the expression of the neurotrophin, brain-derived neurotrophic factor (BDNF) (Gomez-Pinilla et al., 2007; Huie, Garraway, et al., 2012). This protein is of interest because prior work has shown that BDNF can foster the development of LTP and enhance brain-dependent learning/memory (Bekinschtein, Cammarota, Izquierdo, & Medina, 2008; Cunha, Brambilla, & Thomas, 2010; Kang & Schuman, 1995; S. L. Patterson et al., 1996). In spinally transected animals, exposure to controllable stimulation up-regulates the expression of BDNF in the lower (lumbosacral) spinal cord. Further, i.t. BDNF enables learning. Conversely, controllable stimulation does not enable learning in animals pretreated with an antibody (TrkB-IgG, i.t.) that disrupts BDNF signaling by sequestering it. TrkB-IgG (i.t.) also blocks the restorative effect of both controllable and predictable (FT) stimulation (Baumbauer, Huie, et al., 2009; Huie, Garraway, et al., 2012). Conversely, i.t. BDNF can substitute for training with controllable/predictable stimulation to counter the development, maintenance, and expression of the learning impairment induced by uncontrollable/unpredictable stimulation. Other treatments, such as locomotor training, intermittent hypoxia, and exercise, also promote adaptive plasticity within the spinal cord and these effects too have been linked to the expression of BDNF (Baker-Herman et al., 2004; Boyce & Mendell, 2014; Boyce, Park, Gage, & Mendell, 2012; Boyce, Tumolo, Fischer, Murray, & Lemay, 2007; Cote, Azzam, Lemay, Zhukareva, & Houle, 2011; Ollivier-Lanvin, Fischer, Tom, Houle, & Lemay, 2015; Ying et al., 2008). Taken together, the results suggest that BDNF plays a pivotal role in regulating adaptive potential within the spinal cord.
Tumor Necrosis Factor
Exposure to variable uncontrollable stimulation induces a form of behavioral metaplasticity in spinally transected rats that impairs the capacity to learn (Ferguson, Bolding, et al., 2008; Grau et al., 2014). What neurobiological mechanisms mediate this effect? A clue emerged from studies examining the impact of this stimulation on mechanical reactivity, revealing an increase in responsiveness analogous to that observed after treatments that induce nociceptive sensitization (Ferguson et al., 2006). Given this, it was hypothesized that exposure to uncontrollable/unpredictable stimulation impairs learning because it induces a diffuse state of over-excitation (Ferguson et al., 2012a). From this perspective, learning is impaired because the system is overstimulated, driven to a point wherein synaptic plasticity is saturated (Moser & Moser, 1999). With every switch in an “on” position, the capacity to selectively strengthen a particular S-R connection would be lost.
If uncontrollable/unpredictable stimulation impairs learning because it induces a form of nociceptive sensitization, then treatments known to sensitize nociceptive fibers within the spinal cord should have a similar effect. To test this, capsaicin was applied to one hind paw of spinally transected rats. Prior work established that capsaicin induces a robust nociceptive sensitization (Willis, 2001). It also impaired spinally mediated learning (Hook, Huie, & Grau, 2008). Peripheral application of other irritants (e.g., formalin, carrageenan) known to sensitize pain had the same effect (Baumbauer, Young, & Joynes, 2009; Ferguson et al., 2006; Ferguson, Huie, Crown, & Grau, 2012b). Further work has shown that this capsaicin-induced learning impairment can be blocked by exposure to controllable or predictable (FT) stimulation (Baumbauer & Grau, 2011; Hook et al., 2008). In humans, peripheral treatment with capsaicin attenuates the improvement in performance observed across days in response to locomotor training (Bouffard, Bouyer, Roy, & Mercier, 2014; Mercier, Roosink, Bouffard, & Bouyer, 2017). Because these adverse effects are linked to nociceptive sensitization, which is thought to contribute to the development of chronic pain, and because they impair adaptive spinal learning, the resultant effect on neural function has been characterized as a form of maladaptive plasticity (Ferguson et al., 2012a).
Since the 1990s, researchers have learned a great deal about the mechanisms that underlie nociceptive sensitization (Latremoliere & Woolf, 2009). This corpus of findings has provided a roadmap for exploring the neurobiological mechanisms that disrupt adaptive learning within the spinal cord. One pivotal process involves the regulation of glutamatergic neural transmission (Ji et al., 2003). As discussed previously, pretreatment with an NMDAR antagonist (i.t.) blocks the induction of the learning impairment observed after uncontrollable stimulation (Ferguson et al., 2006). Additional work has implicated metabotropic glutamate receptors (mGluR), which are of interest because they appear to exert a metaplastic effect within the brain (Abraham, 2008; Cohen, Coussens, Raymond, & Abraham, 1999). In the spinal cord, both nociceptive sensitization and the learning impairment can be blocked by pretreatment with an mGluR1 receptor antagonist (CPCCOEt, i.t.) (Ferguson, Bolding, et al., 2008). Conversely, i.t. application of an mGluR agonist (DHPG) engages a lasting learning impairment.
Another key player is the cytokine tumor necrosis factor (TNF), which is released by non-neuronal cells (microglia) within the spinal cord and promotes the trafficking of AMPARs to the active zone of the synapse (Beattie et al., 2002; Stellwagen & Malenka, 2006). TNF can further enhance excitability by increasing the expression of AMPARs that lack a subunit (GluR2), a transformation that enhances permeability to Ca++ (Ferguson, Christensen, et al., 2008). Research has shown that i.t. administration of a soluble TNF receptor (sTNFR1), which inhibits TNF function by binding free TNF, blocks the induction and expression of the learning impairment in spinally transected animals (Huie, Baumbauer, et al., 2012; Huie et al., 2015). Conversely, i.t. administration of TNF induces a lasting learning impairment. So too does activating microglia with the endotoxin lipopolysaccharide (LPS), and the expression of this effect is eliminated by sTNFR1 (Vichaya et al., 2009; Young, Baumbauer, Elliot, & Joynes, 2007). Finally, a drug (Naspm) that blocks GluR2-lacking AMPARs attenuates the expression of the learning impairment induced by either uncontrollable/unpredictable stimulation or pretreatment with TNF (Huie, Baumbauer, et al., 2012).
Taken together, these findings suggest that noxious stimulation can induce a form of maladaptive plasticity that interferes with the capacity to learn and enhances nociceptive reactivity. The former has been characterized as a kind of behavioral metaplasticity, linked to the over-excitation of nociceptive circuits within the spinal cord and involving NMDAR/mGluR-dependent changes in glutamate transmission, microglia, and the pro-inflammatory cytokine TNF (Grau & Huang, 2018; Grau et al., 2014).
To study neural plasticity within the spinal cord, minus input from the brain, researchers typically transect the spinal cord, cutting all neural communication to the brain. This is, of course, a form of SCI, which would be expected to have a local effect (spinal shock) that impacts neuronal function due to the disruption in blood flow and extracellular milieu (Bach-y-Rita & Illis, 1993; Dietz, 2010). Recognizing that this could impair adaptive plasticity within the lumbosacral spinal cord, researchers often perform a remote cut of the upper thoracic spinal cord (e.g., at the second thoracic vertebra, T2) and give animals a day to recover from surgery. Under these conditions, stimulation of the lower limbs or tail readily elicits a spinally mediated reflex and supports learning (Grau et al., 1990, 1998; Meagher, Grau, & King, 1990). Indeed, reflexive behavior is sometimes more vigorous, a behavioral effect that has been characterized as a kind of release from inhibition attributed to the loss of descending fibers that regulate neural excitability (Dietz, 2010). Recent work suggests that this increase in excitability may be related to a transformation in how the neurotransmitter gamma-aminobutyric acid (GABA) affects neural function.
The presynaptic release of GABA can impact post-synaptic neural activity by engaging the GABA-A receptor, which opens a membrane channel that allows the negatively charged ion Cl− to pass (Medina et al., 2014). The intracellular concentration of Cl− is regulated by two membrane-bound cotransporters, K+-Cl− cotransporter 2 (KCC2) and Na+-K+-Cl− cotransporter 1 (NKCC1), which control the outward and inward flow of Cl−, respectively (Kaila, Price, et al., 2014). In adult animals, the membrane-bound concentration of KCC2 is high, which helps to maintain a low intracellular concentration of Cl−. Under these conditions, engaging the GABA-A receptor allows Cl− to flow into the cell, which has a hyperpolarizing (inhibitory) effect. However, early in development, the levels of KCC2 are much lower and, for this reason, the intracellular concentration of Cl− is higher (Ben-Ari, 2002; Ben-Ari, Khalilov, Kahle, & Cherubini, 2012; Huang et al., 2016). In this situation, engaging the GABA-A receptor can have the opposite effect, allowing Cl− to exit the cell, which depolarizes (excites) the neuron. Recent research suggests that SCI reduces membrane-bound KCC2 caudal to injury, producing a depolarizing shift that recapitulates the earlier development state (Ben-Ari, 2002; Ben-Ari et al., 2012; Huang et al., 2016). This effect that has been linked to a loss of descending serotonergic fibers (Huang & Grau, 2018). By reducing membrane-bound KCC2, SCI would remove a GABA-dependent brake on neural excitation and thereby enable reflexive behavior and learning. The corollary to this is that, in the absence of SCI, spinal learning/plasticity may be inhibited by GABA and for this reason the system would appear relatively immutable (Grau & Huang, 2018).
Removing the GABA-dependent brake could also have a maladaptive consequence, for it could fuel over-excitation and the sensitization of nociceptive circuits (Grau & Huang, 2018). Indeed, recent work suggests that the release of GABAergic inhibition plays an essential role in the spinally mediated learning impairment observed after exposure to uncontrollable/unpredictable stimulation (Ferguson, Washburn, Crown, & Grau, 2003). Supporting this, pretreatment with a GABA-A receptor antagonist (bicuculline) blocks both the induction and expression of the learning deficit. Bicuculline also attenuates the development of nociceptive sensitization (Huang et al., 2016). Further, a drug treatment (i.t. bumetanide) that should lower intracellular Cl− concentrations after SCI, and thereby restore GABA-dependent inhibition, blocks the development of nociceptive sensitization. Bumetanide also attenuates the development of chronic pain and the over-excitation of motor systems (spasticity) after SCI (Boulenguez et al., 2010; Cramer et al., 2008). In addition, a drug (CLP290) that acts as a KCC2 agonist fosters recovery in animals that have received partial contralateral cuts (hemisections) of the spinal cord (Chen et al., 2018). Taken together, these results imply that SCI alters plastic potential within the spinal cord, removing a GABA-dependent brake that attenuates adaptive learning and the development of maladaptive plasticity (spasticity and chronic pain). Other work has shown that alterations in GABA function can impact plastic potential in other regions of the CNS, a phenomenon known as ionic plasticity (Grau & Huang, 2018; Kaila, Ruusuvuori, Seja, Voipio, & Puskarjov, 2014).
Implications for Recovery After Spinal Cord Injury
Pain Input After Injury (Polytrauma) Impairs Recovery
The studies described here have examined plastic potential within the spinal cord, minus input from the brain. This has been studied by surgically cutting the thoracic spinal cord and examining how stimulation applied below the injury affects spinal cord function and behavioral reactivity (Grau et al., 2017). However, relatively few SCIs involve a clear cut of the spinal cord. Instead, most involve a blow that bruises the tissue (the primary injury), triggering a sequence of acute events (e.g., tissue damage, the release of pro-inflammatory cytokines, and hemorrhage) that lead to cell death (Beattie & Bresnahan, 2000; Duprez, Wirawan, Vanden Berghe, & Vandenabeele, 2009).
To explore whether training and noxious stimulation affect recovery after a more clinically relevant contusion injury, the bone covering the spinal cord tissue is removed (a laminectomy) and an SCI is produced using a surgical device that provides a controlled impact (Gruner, 1992). With a moderate injury, rats will exhibit a near complete paralysis that begins to wane after a few days (Basso, Beattie, & Bresnahan, 1996). Over the next two weeks, there is typically some recovery of locomotor function, which can be quantified using a behavioral scale developed by Basso, Beattie, and Bresnahan (1995). This scale provides a quantitative index of locomotor function that varies from 0 (complete paralysis) to 21 (normal locomotion). Between these extremes, animals exhibit a range of behaviors (e.g., plantar placement , weight-supported stepping , coordination of limbs ) indicative of improved function. Other key indices include the recovery of bladder function and responsiveness to mechanical stimulation. The latter is of particular interest because enhanced mechanical reactivity is associated with the development of allodynia (pain to light touch) and chronic pain (Lamotte et al., 1991; Simone et al., 1989). At the end of recovery, the injured spinal cord is collected and the extent of tissue loss can be quantified.
Using these procedures, researchers have examined how pain input applied caudal to (below) a contusion injury affects long-term recovery (Grau et al., 2017). This is of interest because many contusion injuries in humans are accompanied by additional tissue damage (polytrauma) that would engage pain (nociceptive) fibers (Chu, Lee, Lin, Chou, & Yang, 2009; Saboe, Reid, Davis, Warren, & Grace, 1991) and because work using spinally transected animals has shown that noxious stimulation can induce a form of maladaptive plasticity that impairs spinal learning and sensitizes pain (Ferguson et al., 2012). To explore how nociceptive input affects recovery, rats were anesthetized and given a contusion injury. The next day, they were exposed to six minutes of uncontrollable electrical stimulation. This noxious input produced an acute disruption in locomotor function that was still evident six weeks later (Grau et al., 2004). It also delayed the recovery of bladder function, enhanced reactivity to mechanical stimulation, and increased tissue loss at the site of injury (Garraway et al., 2014; Grau et al., 2004). Importantly, intermittent shock only had these effects if the stimulation was given in an uncontrollable manner; exposure to an equal number of shocks, applied in a controllable manner, did not adversely affect recovery (Grau et al., 2004). Further work revealed that noxious input has a greater effect if it occurs within the first week of injury. In addition, chemically engaging pain fibers caudal to injury with capsaicin had the same adverse effect (Turtle et al., 2018).
Nociceptive input after injury may promote tissue loss by engaging the expression of pro-inflammatory cytokines (interleukin-1 [IL-1], IL-18, TNF) and cell signals linked to the initiation of cell death (e.g., caspase 1, 3, and 8) (Lossi, Castagna, & Merighi, 2015). In addition, nociceptive stimulation appears to engage signal pathways that initiate cell death in the endothelial cells that form the blood–spinal cord barrier (Simard, Woo, Aarabi, & Gerzanich, 2013; Turtle et al., 2019). The resultant capillary fragmentation allows blood to enter the surrounding tissue. Because hemoglobin and other blood-borne proteins are neurotoxic, this hemorrhage expands the area of tissue loss (secondary injury), a phenomenon known as progressive hemorrhagic necrosis (Simard, Kahle, & Gerzanich, 2010).
Protecting the Injured Spinal Cord
The findings reviewed here show that pain input after a spinal contusion injury can expand the area of tissue loss and undermine long-term recovery. How can this adverse effect be halted? The most obvious possibility is through the administration of a known analgesic, such as morphine.
Opioids influence neural activity by engaging receptors that are distributed throughout the nervous system, with especially high densities being observed in areas associated with pain processing and reward (Olson, Olson, Kastin, & Coy, 1982; Satoh & Minami, 1995). Because the systemic application of a drug such as morphine can completely block behavioral reactivity to painful stimuli, it was thought that this drug treatment would attenuate the adverse effect noxious stimulation has on tissue sparing and recovery after a contusion injury. However, systemic treatment with morphine prior to noxious stimulation does not attenuate hemorrhage at the site of injury or block the adverse effect nociceptive input has on recovery (Hook et al., 2007). Further work showed that spinally applied (i.t.) morphine actually increases tissue loss and expands the area of injury (Hook et al., 2009). These findings are particularly alarming, given that morphine is recommended as a standard of care after SCI (Hook et al., 2017). The adverse effect of opioid treatment appears tied to the activation of the kappa opioid receptor (Aceves, Mathai, & Hook, 2016), a finding that is consistent with earlier work (Faden, 1990). Opioids can also activate non-neuronal cells by engaging the toll-like receptor 4 (TLR4) that recognizes LPS, an action linked to the development of chronic pain (Hutchinson et al., 2007; Hutchinson et al., 2010; Watkins et al., 2007).
Other work suggests that reducing neural activity by cooling the spinal cord (hypothermia) can attenuate tissue loss (Batchelor et al., 2013; Hansebout & Hansebout, 2014). Building on this observation, researchers have explored whether chemically quieting neural activity within the spinal cord can mitigate the adverse effect of noxious stimulation. This was accomplished by applying the Na+ channel blocker lidocaine to the spinal cord (Turtle et al., 2017), the same procedure used to reduce pain during childbirth. Pretreatment with lidocaine attenuated nociception-induced hemorrhage, the activation of pro-inflammatory cytokines (IL-1ß, IL-18), and the expression of signals (e.g., caspase 3) related to cell death (Lossi et al., 2015). Most importantly, pretreatment with lidocaine also blocked the adverse effect noxious stimulation had on long-term recovery (Turtle et al., 2017).
Promoting Adaptive Plasticity During the Chronic Phase of Injury
During the acute stage of injury, peripheral inflammation and noxious stimulation applied in an uncontrollable/unpredictable manner can sensitize nociceptive neurons within the spinal cord and impair adaptive learning. After a contusion injury, this noxious input amplifies hemorrhage, expands the area of secondary injury, and undermines long-term recovery (Grau et al., 2004; Turtle et al., 2017). In spinally transected animals, the adverse effect of noxious stimulation can be prevented and reversed by exposure to controllable or predictable stimulation (Baumbauer & Grau, 2011; Baumbauer, Huie, et al., 2009; Crown & Grau, 2001; Hook et al., 2008). In contused animals, allowing behavioral control mitigates the adverse effect nociceptive stimulation has on long-term recovery (Grau et al., 2004). It has been suggested that this protective effect occurs because spinal cord neurons are wired in a manner that physiologically assumes neural inputs are related to indices of limb position (Grau et al., 2012), a biologically prepared relation that gates how the nociceptive input impacts central processes. In the absence of this relational signal, nociceptive input appears to trigger hemorrhage, initiating a tidal wave of destructive processes that cannot be undone.
Research has shown that the adverse effect of noxious input on tissue sparing wanes over time (Grau et al., 2004), possibly because surviving descending fibers reestablish some inhibitory control that helps quell over-excitation. This process may be facilitated by locomotor training that helps to reestablish GABA-dependent inhibition (Cote, Gandhi, Zambrotta, & Houle, 2014). This would be expected to benefit recovery by lessening the adverse effect of noxious stimulation. Indeed, when capsaicin is applied to humans months after injury, it interferes with retention on a locomotor task but has no long-term effect (Bouffard et al., 2014). Likewise, animal studies suggest that limb stretching, which is routinely used to help maintain flexibility after SCI, can engage nociceptive fibers and disrupt locomotor function (Caudle et al., 2015). Here too, stimulation appears to have a greater effect when applied soon after injury. Further research is needed to determine whether nociceptive input during the chronic stage of injury promotes maladaptive plasticity (e.g., pain, spasticity, and uncontrolled bouts of sympathetic activation [autonomic dysreflexia]) (Brown & Weaver, 2012; Mercier et al., 2017; Rabchevsky & Kitzman, 2011).
Evidence suggests that adaptive potential may fade with time after injury (Durkovic, 2001). For example, locomotor training can counter the development of chronic pain, and this effect is greater if training is initiated soon after injury (Detloff et al., 2016). In an animal model, exercise has been shown to promote recovery when started soon after injury (Brown et al., 2011). There is also evidence that the effect of training depends upon the nature of the task. For example, spinal learning is preserved in rats that have undergone a neonatal transection followed by locomotor training, but is lost in animals that were stand-trained (Bigbee et al., 2007).
These observations suggest that the loss of plastic potential over time has a dual-edged effect. On the one hand, quelling over-excitation will lessen the adverse effect of nociceptive input (Grau et al., 2017). On the other hand, the loss of plastic potential could impair the recovery of adaptive behavior and sensations. Under these conditions, recovery may benefit from the prudent application of procedures designed to reawaken the spinal cord, to foster adaptive plasticity (Edgerton et al., 2008). This is consistent with recent studies demonstrating that epidural stimulation benefits the recovery of voluntary behavior by driving proprioceptive fiber input (Formento et al., 2018; Harkema et al., 2011), the same pathway implicated in relational (instrumental) learning (Grau et al., 2012; Huie, Morioka, Haefeli, & Ferguson, 2017). Other recent studies suggest that adaptive plasticity can be fostered using a chemical cocktail or by presenting cortical and spinal stimulation in a paired manner (Edgerton et al., 2008; Mishra, Pal, Gupta, & Carmel, 2017; van den Brand et al., 2012).
The behavioral and neurobiological processes engaged by training after SCI will likely depend upon injury severity. A severe injury could cut communication with the brain, leading to a down-regulation in the cotransporter KCC2, which lessens GABA-dependent inhibition and enables spinally mediated learning (Ben-Ari, 2002; Ben-Ari et al., 2012; Huang et al., 2016). If descending serotonergic fibers within the dorsolateral funiculus survive (or recover), GABAergic inhibition will be preserved, quelling neural excitation and plasticity within the spinal cord (Gjersstad, Tjolsen, & Hole, 2001; Grau & Huang, 2018; Huang & Grau, 2018). Under these conditions, brain-dependent processes may guide adaptive modifications through fibers within the corticospinal tract (Chen & Wolpaw, 1997, 2002; Wolpaw, 2010). Interestingly, evidence suggests that brain systems can modify the performance of a spinal reflex by strengthening the inhibitory effect of GABAergic interneurons (Wang, Chen, Chen, Wolpaw, & Chen, 2012; Wang, Pillai, Wolpaw, & Chen, 2006, 2009). These observations imply that brain systems normally regulate spinal plasticity, a homeostatic balance that is disrupted by SCI, unveiling a plastic potential within the spinal cord that can promote adaptive behavior but also fuel chronic pain and spasticity. Relational learning, whether spinally mediated or brain dependent, may help to counter the development of these maladaptive effects.
In summary, neural mechanisms within the spinal cord are sensitive to environmental relations and this learning can counter the adverse effects of nociceptive stimulation. The overall pattern of results parallels findings from studies examining how brain-dependent processes modulate learning and stress (Maier & Seligman, 2016). These observations suggest that, during the chronic stage of injury, a key to promoting adaptive plasticity and health in the face of pain input lies with the maintenance of behavioral control and temporal predictability—that independent of whether the resultant adaptations are mediated by the spinal cord or brain, behavioral control and temporal predictability will promote adaptive plasticity. And when the brain is the driving force, it can still affect spinal function through spared descending fibers and the regulation of systemic processes (Kigerl, Mostacada, & Popovich, 2018; Thompson & Wolpaw, 2014; van den Brand et al., 2012).
Dozens of individuals have contributed to the data and conceptual advances reviewed in this article and the work has grown from their collective input. The text also benefitted from input from current trainees, including R. Baine, P. Bean, J. Davis, G. Fauss, M. Henwood, K. Hudson, T. Johnston, and M. Tarbet. Work on this article was supported, in part, by grants from the National Institutes of Health (NS091723, NS104422), the Craig H. Neilsen Foundation, and the Department of Defense (SC170241).
Abraham, W. C. (2008). Metaplasticity: Tuning synapses and networks for plasticity. Nature Reviews Neuroscience, 9, 387–399.Find this resource:
Aceves, M., Mathai, B. B., & Hook, M. A. (2016). Evaluation of the effects of specific opioid receptor agonists in a rodent model of spinal cord injury. Spinal Cord, 54, 767–777.Find this resource:
Bach-y-Rita, P., & Illis, L. S. (1993). Spinal shock: Possible role of receptor plasticity and non synaptic transmission. Paraplegia, 31, 82–87.Find this resource:
Baker-Herman, T. L., Fuller, D. D., Bavis, R. W., Zabka, A. G., Golder, F. J., Doperalski, N. J., . . . Mitchell, G. S. (2004). BDNF is necessary and sufficient for spinal respiratory plasticity following intermittent hypoxia. Nature Neuroscience, 7, 48–55.Find this resource:
Basbaum, A. I., & Fields, H. L. (1984). Endogenous pain control-systems: Brain-stem spinal pathways and endorphin circuitry. Annual Review of Neuroscience, 7, 309–338.Find this resource:
Basso, D. M., Beattie, M. S., & Bresnahan, J. C. (1995). A sensitive and reliable locomotor rating-scale for open-field testing in rats. Journal of Neurotrauma, 12, 1–21.Find this resource:
Basso, D. M., Beattie, M. S., & Bresnahan, J. C. (1996). Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Experimental Neurology, 139, 244–256.Find this resource:
Batchelor, P. E., Skeers, P., Antonic, A., Wills, T. E., Howells, D. W., Macleod, M. R., & Sena, E. S. (2013). Systematic review and meta-analysis of therapeutic hypothermia in animal models of spinal cord injury. Plos One, 8, e71317.Find this resource:
Baumbauer, K. M., & Grau, J. W. (2011). Timing in the absence of supraspinal input III: Regularly spaced cutaneous stimulation prevents and reverses the spinal learning deficit produced by peripheral inflammation. Behavioral Neuroscience, 125, 37–45.Find this resource:
Baumbauer, K. M., Hoy, K. C., Huie, J. R., Hughes, A. J., Woller, S. A., Puga, D. A., . . . Grau, J. W. (2008). Timing in the absence of supraspinal input I: Variable, but not fixed, spaced stimulation of the sciatic nerve undermines spinally-mediated instrumental learning. Neuroscience, 155, 1030–1047.Find this resource:
Baumbauer, K. M., Huie, J. R., Hughes, A. J., & Grau, J. W. (2009). Timing in the absence of supraspinal input II: Regularly spaced stimulation induces a lasting alteration in spinal function that depends on the NMDA receptor, BDNF release, and protein synthesis. Journal of Neuroscience, 29, 14383–14393.Find this resource:
Baumbauer, K. M., Young, E. E., & Joynes, R. L. (2009). Pain and learning in a spinal system: Contradictory outcomes from common origins. Brain Research Reviews, 61, 124–143.Find this resource:
Bear, M. F. (2003). Bidirectional synaptic plasticity: From theory to reality. Philosophical Transactions of the Royal Society B, 358, 649–655.Find this resource:
Beattie, E. C., Stellwagen, D., Morishita, W., Bresnahan, J. C., Ha, B. K., Von Zastrow, M., . . .Malenka, R. C. (2002). Control of synaptic strength by glial TNF alpha. Science, 295, 2282–2285.Find this resource:
Beattie, M. S., & Bresnahan, J. C. (2000). Cell death, repair, and recovery of function after spinal cord contusion injuries in rats. In R. G. Kalb & S. M. Strittmatter (Eds.), Neurobiology of spinal cord injury (pp. 1–21). Totowa, NJ: Humana Press.Find this resource:
Bekinschtein, P., Cammarota, M., Izquierdo, I., & Medina, J. H. (2008). BDNF and memory formation and storage. Neuroscientist, 14, 147–156.Find this resource:
Ben-Ari, Y. (2002). Excitatory actions of GABA during development: The nature of the nurture. Nature Reviews Neuroscience, 3, 728–739.Find this resource:
Ben-Ari, Y., Khalilov, I., Kahle, K. T., & Cherubini, E. (2012). The GABA excitatory/inhibitory shift in brain maturation and neurological disorders. Neuroscientist, 18, 467–486.Find this resource:
Bigbee, A. J., Crown, E. D., Ferguson, A. R., Roy, R. R., Tillakaratne, N. J. K., Grau, J. W., & Edgerton, V. R. (2007). Two chronic motor training paradigms differentially influence acute instrumental learning in spinally transected rats. Behavioural Brain Research, 180, 95–101.Find this resource:
Bouffard, J., Bouyer, L. J., Roy, J. S., & Mercier, C. (2014). Tonic pain experienced during locomotor training impairs retention despite normal performance during acquisition. Journal of Neuroscience, 34, 9190–9195.Find this resource:
Boulenguez, P., Liabeuf, S., Bos, R., Bras, H., Jean-Xavier, C., Brocard, C., . . . Vinay, L. (2010). Down-regulation of the potassium-chloride cotransporter KCC2 contributes to spasticity after spinal cord injury. Nature Medicine, 16, 302–307.Find this resource:
Boyce, V. S., & Mendell, L. M. (2014). Neurotrophins and spinal circuit function. Frontiers in Neural Circuits, 8, 59.Find this resource:
Boyce, V. S., Park, J., Gage, F. H., & Mendell, L. M. (2012). Differential effects of brain-derived neurotrophic factor and neurotrophin-3 on hindlimb function in paraplegic rats. European Journal of Neuroscience, 35, 221–232.Find this resource:
Boyce, V. S., Tumolo, M., Fischer, I., Murray, M., & Lemay, M. A. (2007). Neurotrophic factors promote and enhance locomotor recovery in untrained spinalized cats. Journal of Neurophysiology, 98, 1988–1996.Find this resource:
Brown, A., & Weaver, L. C. (2012). The dark side of neuroplasticity. Experimental Neurology, 235, 133–141.Find this resource:
Brown, A. K., Woller, S. A., Moreno, G., Grau, J. W., & Hook, M. A. (2011). Exercise therapy and recovery after SCI: Evidence that shows early intervention improves recovery of function. Spinal Cord, 49, 623–628.Find this resource:
Buerger, A. A., & Chopin, S. F. (1976). Instrumental avoidance conditioning in spinal vertebrates. Advances in Psychobiology, 3, 437–461.Find this resource:
Caudle, K. L., Atkinson, D. A., Brown, E. H., Donaldson, K., Seibt, E., Chea, T., . . .Magnuson, D. S. (2015). Hindlimb stretching alters locomotor function after spinal cord injury in the adult rat. Neurorehabilitation and Neural Repair, 29, 268–277.Find this resource:
Chance, W. T., White, A. C., Krynock, G. M., & Rosecrans, J. A. (1977). Autoanalgesia: Behaviorally activated antinociception. European Journal of Pharmacology, 44, 283–284.Find this resource:
Chen, B., Li, Y., Yu, B., Zhang, Z., Brommer, B., Williams, P. R., . . . He, Z. (2018). Reactivation of dormant relay pathways in injured spinal cord by KCC2 manipulations. Cell, 174, 1599.Find this resource:
Chen, X. Y., & Wolpaw, J. R. (1997). Dorsal column but not lateral column transection prevents down-conditioning of H reflex in rats. Journal of Neurophysiology, 78, 1730–1734.Find this resource:
Chen, X. Y., & Wolpaw, J. R. (2002). Probable corticospinal tract control of spinal cord plasticity in the rat. Jounral of Neurophysiology, 87, 645–652.Find this resource:
Chopin, S. F., & Buerger, A. A. (1976). Instrumental avoidance-conditioning in spinal rat. Brain Research Bulletin, 1, 177–183.Find this resource:
Chu, D., Lee, Y.-H., Lin, C.-H., Chou, P., & Yang, N.-P. (2009). Prevalence of associated injuries of spinal trauma and their effect on medical utilization among hospitalized adult subjects: A nationwide data-based study. BMC Health Services Research, 9, 137.Find this resource:
Church, R. M. (1964). Systematic effect of random error in the yoked control design. Psychological Bulletin, 62, 122–131.Find this resource:
Church, R. M. (1989). The yoked control design. In T. A. L. Nilsson (Ed.), Aversion, avoidance, and anxiety: Perspectives on aversively motivated behavior. Hillsdale, NJ: Erlbaum.Find this resource:
Church, R. M., & Lerner, N. D. (1976). Does headless roach learn to avoid. Physiological Psychology, 4, 439–442.Find this resource:
Coderre, T. J., Katz, J., Vaccarino, A. L., & Melzack, R. (1993). Contribution of central neuroplasticity to pathological pain: Review of clinical and experimental evidence. Pain, 52, 259–285.Find this resource:
Cohen, A. S., Coussens, C. M., Raymond, C. R., & Abraham, W. C. (1999). Long-lasting increase in cellular excitability associated with the priming of LTP induction in rat hippocampus. Journal of Neurophysiology, 82, 3139–3148.Find this resource:
Collingridge, G. L., & Bliss, T. V. P. (1987). NMDA receptors: Their role in long-term potentiation. Trends in Neuroscience, 10, 288–293.Find this resource:
Colwill, R. M., & Rescorla, R. A. (1986). Associative structures in instrumental learning. Psychology of Learning and Motivation-Advances in Research and Theory, 20, 55–104.Find this resource:
Cordero-Erausquin, M., Inquimbert, P., Schlichter, R., & Hugel, S. (2016). Neuronal networks and nociceptive processing in the dorsal horn of the spinal cord. Neuroscience, 338, 230–247.Find this resource:
Cote, M. P., Azzam, G. A., Lemay, M. A., Zhukareva, V., & Houle, J. D. (2011). Activity-dependent increase in neurotrophic factors is associated with an enhanced modulation of spinal reflexes after spinal cord injury. Journal of Neurotrauma, 28, 299–309.Find this resource:
Cote, M. P., Gandhi, S., Zambrotta, M., & Houle, J. D. (2014). Exercise modulates chloride homeostasis after spinal cord injury. Journal of Neuroscience, 34, 8976–8987.Find this resource:
Cramer, S. W., Baggott, C., Cain, J., Tilghman, J., Allcock, B., Miranpuri, G., . . . Resnick, D. (2008). The role of cation-dependent chloride transporters in neuropathic pain following spinal cord injury. Molecular Pain, 4, 36.Find this resource:
Crown, E. D., Ferguson, A. R., Joynes, R. L., & Grau, J. W. (2002a). Instrumental learning within the spinal cord II: Evidence for central mediation. Physiology & Behavior, 77, 259–267.Find this resource:
Crown, E. D., Ferguson, A. R., Joynes, R. L., & Grau, J. W. (2002b). Instrumental learning within the spinal cord IV: Induction and retention of the behavioral deficit observed after noncontingent shock. Behavioral Neuroscience, 116, 1032–1051.Find this resource:
Crown, E. D., & Grau, J. W. (2001). Preserving and restoring behavioral potential within the spinal cord using an instrumental training paradigm. Journal of Neurophysiology, 86, 845–855.Find this resource:
Cunha, C., Brambilla, R., & Thomas, K. L. (2010). A simple role for BDNF in learning and memory? Frontiers in Molecular Neuroscience, 3, 1.Find this resource:
Davis, M., & Wagner, A. R. (1968). Startle responsiveness after habituation to different intensities of tone. Psychonomic Science, 12, 337.Find this resource:
Detloff, M. R., Quiros-Molina, D., Javia, A. S., Daggubati, L., Nehlsen, A. D., Naqvi, A., . . .Houle, J. D. (2016). Delayed exercise is ineffective at reversing aberrant nociceptive afferent plasticity or neuropathic pain after spinal cord injury in rats. Neurorehabilitation Neural Repair, 30, 685–700.Find this resource:
Dickenson, A. H., & Sullivan, A. F. (1987). Evidence for a role of the NMDA receptor in the frequency-dependent potentiation of deep rat dorsal horn nociceptive neurons following c-fiber stimulation. Neuropharmacology, 26, 1235–1238.Find this resource:
Dietz, V. (2010). Behavior of spinal neurons deprived of supraspinal input. Nature Review Neurology, 6, 167–174.Find this resource:
Dietz, V., & Harkema, S. J. (2004). Locomotor activity in spinal cord-injured persons. Journal of Applied Physiology, 96, 1954–1960.Find this resource:
DiGiorgio, A. M. (1929). Persistenza nell’animale spinale, di asimmetrie osturali e motorie di origine cerebellare. Archivo do Fisiologia, 27, 518–580.Find this resource:
Domjan, M. (2015). Principles of learning and behavior. Stamford, CT: Cengage.Find this resource:
Dudai, Y. (1989). The neurobiology of memory: Concepts, findings, trends. Oxford: Oxford University Press.Find this resource:
Duprez, L., Wirawan, E., Vanden Berghe, T., & Vandenabeele, P. (2009). Major cell death pathways at a glance. Microbes and Infection, 11, 1050–1062.Find this resource:
Durkovic, R. G. (2001). Pavlovian conditioning of flexion reflex potentiation in spinal cat: Temporal effects following spinal transection. In M. M. Patterson & J. W. Grau (Eds.), Spinal cord plasticity: Alterations in reflex function (pp. 55–75). Boston, MA: Kluwer Academic.Find this resource:
Durkovic, R. G., & Prokowich, L. J. (1998). D-2-Amino-5-phosphonovalerate, an NMDA receptor antagonist, blocks induction of associative long-term potentiation of the flexion reflex in spinal cat. Neuroscience Letters, 257, 162–164.Find this resource:
Edgerton, V. R., Courtine, G., Gerasimenko, Y. P., Lavrov, I., Ichiyama, R. M., Fong, A. J., . . . Roy, R. R. (2008). Training locomotor networks. Brain Research Reviews, 57, 241–254.Find this resource:
Edgerton, V. R., de Leon, R. D., Tillakaratne, N., Recktenwald, M. R., Hodgson, J. A., & Roy, R. R. (1997). Use-dependent plasticity in spinal stepping and standing. Advances in Neurology, 72, 233–247.Find this resource:
Edgerton, V. R., Roy, R. R., DeLeon, R., Tillakaratne, N., & Hodgson, J. A. (1997). Does motor learning occur in the spinal cord? Neuroscientist, 3, 287–294.Find this resource:
Edgerton, V. R., Tillakaratne, N. J. K., Bigbee, A. J., de Leon, R. D., & Roy, R. R. (2004). Plasticity of the spinal neural circuitry after injury. Annual Review of Neuroscience, 27, 145–167.Find this resource:
Faden, A. I. (1990). Opioid and nonopioid mechanisms may contribute to dynorphins pathophysiological actions in spinal-cord injury. Annals of Neurology, 27, 67–74.Find this resource:
Fanselow, M. S. (1986). Conditioned fear-induced opiate analgesia: A competing motivational state theory of stress analgesia. Annals New York Academy of Sciences, 467, 40–54.Find this resource:
Ferguson, A. R., Bolding, K. A., Huie, J. R., Hook, M. A., Santillano, D. R., Miranda, R. C., & Grau, J. W. (2008). Group I metabotropic glutamate receptors control metaplasticity of spinal cord learning through a protein kinase c-dependent mechanism. Journal of Neuroscience, 28, 11939–11949.Find this resource:
Ferguson, A. R., Christensen, R. N., Gensel, J. C., Miller, B. A., Sun, F., Beattie, E. C., . . . Beattie, M. S. (2008). Cell death after spinal cord injury is exacerbated by rapid TNF alpha-induced trafficking of GluR2-lacking AMPARs to the plasma membrane. Journal of Neuroscience, 28(44), 11391–11400.Find this resource:
Ferguson, A. R., Crown, E. D., & Grau, J. W. (2006). Nociceptive plasticity inhibits adaptive learning in the spinal cord. Neuroscience, 141, 421–431.Find this resource:
Ferguson, A. R., Huie, J. R., Crown, E. D., Baumbauer, K. M., Hook, M. A., Garraway, S. M., . . . Grau, J. W. (2012a). Maladaptive spinal plasticity opposes spinal learning and recovery in spinal cord injury. Frontiers in Physiology, 3, 399.Find this resource:
Ferguson, A. R., Huie, J. R., Crown, E. D., & Grau, J. W. (2012b). Central nociceptive sensitization vs. spinal cord training: Opposing forms of plasticity that dictate function after complete spinal cord injury. Frontiers in Physiology, 3, 396.Find this resource:
Ferguson, A. R., Washburn, S. N., Crown, E. D., & Grau, J. W. (2003). GABA(A) receptor activation is involved in noncontingent shock inhibition of instrumental conditioning in spinal rats. Behavioral Neuroscience, 117, 799–812.Find this resource:
Formento, E., Minassian, K., Wagner, F., Mignardot, J. B., Le Goff-Mignardot, C. G., Rowald, A., . . . Courtine, G. (2018). Electrical spinal cord stimulation must preserve proprioception to enable locomotion in humans with spinal cord injury. Nature Neuroscience, 21, 1728–1741.Find this resource:
Forssberg, H., Grillner, S., Halbertsma, J., & Rossignol, S. (1980). The locomotion of the low spinal cat. II. Interlimb coordination. Acta Physiolica Scandinavica, 108, 283–295.Find this resource:
Garraway, S. M., Turtle, J. D., Huie, J. R., Lee, K. H., Hook, M. A., Woller, S. A., & Grau, J. W. (2011). Intermittent noxious stimulation following spinal cord contusion injury impairs locomotor recovery and reduces spinal brain-derived neurotrophic factor-tropomyosin-receptor kinase signaling in adult rats. Neuroscience, 199, 86–102.Find this resource:
Garraway, S. M., Woller, S. A., Huie, J. R., Hartman, J. J., Hook, M. A., Miranda, R. C., …Grau, J. W. (2014). Peripheral noxious stimulation reduces withdrawal threshold to mechanical stimuli after spinal cord injury: Role of tumor necrosis factor alpha and apoptosis. Pain, 155, 2344–2359.Find this resource:
Gjerstad, J., Tjolsen, A., & Hole, K. (2001). Induction of long-term potentiation of single wide dynamic range neurones in the dorsal horn is inhibited by descending pathways. Pain, 91, 263–268.Find this resource:
Gomez-Pinilla, F., Huie, J. R., Ying, Z., Ferguson, A. R., Crown, E. D., Baumbauer, K. M., . . . Grau, J. W. (2007). BDNF and learning: Evidence that instrumental training promotes learning within the spinal cord by up-regulating BDNF expression. Neuroscience, 148, 893–906.Find this resource:
Grau, J. W. (1987). The central representation of an aversive event maintains the opioid and nonopioid forms of analgesia. Behavioral Neuroscience, 101, 272–288.Find this resource:
Grau, J. W. (2010). Instrumental conditioning. In I. B. W. a. C. B. Nemeroff (Ed.), Corsini Encyclopedia of Psychology (pp. 480–481). New York, NY: John Wiley & Sons.Find this resource:
Grau, J. W. (2014). Learning from the spinal cord: How the study of spinal cord plasticity informs our view of learning. Neurobiology of Learning and Memory, 108, 155–171.Find this resource:
Grau, J. W., Barstow, D. G., & Joynes, R. L. (1998). Instrumental learning within the spinal cord: I. Behavioral properties. Behavioral Neuroscience, 112, 1366–1386.Find this resource:
Grau, J. W., & Huang, Y. J. (2018). Metaplasticity within the spinal cord: Evidence brain-derived neurotrophic factor (BDNF), tumor necrosis factor (TNF), and alterations in GABA function (ionic plasticity) modulate pain and the capacity to learn. Neurobiology of Learning & Memory, 154, 121–135.Find this resource:
Grau, J. W., Huang, Y. J., Turtle, J. D., Strain, M. M., Miranda, R. C., Garraway, S. M., & Hook, M. A. (2017). When pain hurts: Nociceptive stimulation induces a state of maladaptive plasticity and impairs recovery after spinal cord injury. Journal of Neurotrauma, 34, 1873–1890.Find this resource:
Grau, J. W., Huie, J. R., Garraway, S. M., Hook, M. A., Crown, E. D., Baumbauer, K. M., . . . Ferguson, A. R. (2012). Impact of behavioral control on the processing of nociceptive stimulation. Frontiers in Physiology, 3, 262.Find this resource:
Grau, J. W., Huie, J. R., Lee, K. H., Hoy, K. C., Huang, Y. J., Turtle, J. D., . . . Garraway, S. M. (2014). Metaplasticity and behavior: How training and inflammation affect plastic potential within the spinal cord and recovery after injury. Frontiers in Neural Circuits, 8, 100.Find this resource:
Grau, J. W., & Joynes, R. L. (2001). Pavlovian and instrumental conditioning within the spinal cord: Methodological issues. In M. M. Patterson & J. W. Grau (Eds.), Spinal cord plasticity: Alterations in reflex function (pp. 13–54). Boston, MA: Kluwer Academic.Find this resource:
Grau, J. W., Salinas, J. A., Illich, P. A., & Meagher, M. W. (1990). Associative learning and memory for an antinociceptive response in the spinalized rat. Behavioral Neuroscience, 104, 489–494.Find this resource:
Grau, J. W., Washburn, S. N., Hook, M. A., Ferguson, A. R., Crown, E. D., Garcia, G., . . . Miranda, R. C. (2004). Uncontrollable stimulation undermines recovery after spinal cord injury. Journal of Neurotrauma, 21, 1795–1817.Find this resource:
Grau, J. W., & Joynes, R. L. (2005). A neural-functionalist approach to learning. International Journal of Comparative Psychology, 18, 1–22.Find this resource:
Grillner, S. (1973). Locomotion in the spinal cat. In R. B. Stein, K. G. Pearson, R. S. Smith, & J. B. Redford (Eds.), Control of posture and locomotion (pp. 515–535). New York: Plenum.Find this resource:
Grillner, S., & Zangger, P. (1979). Central generation of locomotion in the low spinal cat. Experimental Brain Research, 34, 241–261.Find this resource:
Groves, P. M., De Marco, R., & Thompson, R. F. (1969). Habituation and sensitization of spinal interneuron activity in acute spinal cat. Brain Research, 14, 521–525.Find this resource:
Groves, P. M., Lee, D., & Thompson, R. F. (1969). Effects of stimulus frequency and intensity on habituation and sensitization in acute spinal cat. Physiology & Behavior, 4, 383–388.Find this resource:
Groves, P. M., & Thompson, R. F. (1970). Habituation a dual process theory. Psychological Review, 77, 419–450.Find this resource:
Gruner, J. A. (1992). A monitored contusion model of spinal-cord injury in the rat. Journal of Neurotrauma, 9, 123–128.Find this resource:
Hansebout, R. R., & Hansebout, C. R. (2014). Local cooling for traumatic spinal cord injury: Outcomes in 20 patients and review of the literature. Journal of Neurosurgery–Spine, 20, 550–561.Find this resource:
Harkema, S., Gerasimenko, Y., Hodes, J., Burdick, J., Angeli, C., Chen, Y., . . . Edgerton, V. R. (2011). Effect of epidural stimulation of the lumbosacral spinal cord on voluntary movement, standing, and assisted stepping after motor complete paraplegia: A case study. Lancet, 377, 1938–1947.Find this resource:
Haydon, P. G. (2001). Glia: Listening and talking to the synapse. Nature Reviews Neuroscience, 2, 185–193.Find this resource:
Hodgson, J. A., Roy, R. R., de Leon, R., Dobkin, B., & Edgerton, V. R. (1994). Can the mammalian lumbar spinal cord learn a motor task? Medicine & Science in Sports & Exercise, 26, 1491–1497.Find this resource:
Hook, M. A., Huie, J. R., & Grau, J. W. (2008). Peripheral inflammation undermines the plasticity of the isolated spinal cord. Behavioral Neuroscience, 122, 233–249.Find this resource:
Hook, M. A., Liu, G. T., Washburn, S. N., Ferguson, A. R., Bopp, A. C., Huie, J. R., & Grau, J. W. (2007). The impact of morphine after a spinal cord injury. Behavioural Brain Research, 179, 281–293.Find this resource:
Hook, M. A., Moreno, G., Woller, S., Puga, D., Hoy, K., Balden, R., & Grau, J. W. (2009). Intrathecal morphine attenuates recovery of function after a spinal cord injury. Journal of Neurotrauma, 26, 741–752.Find this resource:
Hook, M. A., Woller, S. A., Bancroft, E., Aceves, M., Funk, M. K., Hartman, J., & Garraway, S. M. (2017). Neurobiological effects of morphine after spinal cord injury. Journal of Neurotrauma, 34, 632–644.Find this resource:
Huang, Y. J., & Grau, J. W. (2018). Ionic plasticity and pain: The loss of descending serotonergic fibers after spinal cord injury transforms how GABA affects pain. Experimental Neurology, 306, 105–116.Find this resource:
Huang, Y. J., Lee, K. H., Murphy, L., Garraway, S. M., & Grau, J. W. (2016). Acute spinal cord injury (SCI) transforms how GABA affects nociceptive sensitization. Experimental Neurology, 285(Pt A), 82–95.Find this resource:
Huie, J. R., Baumbauer, K. M., Lee, K. H., Bresnahan, J. C., Beattie, M. S., Ferguson, A. R., & Grau, J. W. (2012). Glial tumor necrosis factor alpha (TNF alpha) generates metaplastic inhibition of spinal learning. Plos One, 7, e39751.Find this resource:
Huie, J. R., Garraway, S. M., Baumbauer, K. M., Hoy, K. C., Beas, B. S., Montgomery, K. S., … Grau, J. W. (2012). Brain-derived neurotrophic factor promotes adaptive plasticity within the spinal cord and mediates the beneficial effects of controllable stimulation. Neuroscience, 200, 74–90.Find this resource:
Huie, J. R., Morioka, K., Haefeli, J., & Ferguson, A. R. (2017). What is being trained? How divergent forms of plasticity compete to shape locomotor recovery after spinal cord injury. Journal of Neurotrauma, 34, 1831–1840.Find this resource:
Huie, J. R., Stuck, E. D., Lee, K. H., Irvine, K. A., Beattie, M. S., Bresnahan, J. C., … Ferguson, A. R. (2015). AMPA receptor phosphorylation and synaptic co-localization on motor neurons drive maladaptive plasticity below complete spinal cord injury. eNeuro, 2(5).Find this resource:
Hutchinson, M. R., Bland, S. T., Johnson, K. W., Rice, K. C., Maier, S. F., & Watkins, L. R. (2007). Opioid-induced glial activation: Mechanisms of activation and implications for opioid analgesia, dependence, and reward. Scientific World Journal, 7, 98–111.Find this resource:
Hutchinson, M. R., Lewis, S. S., Coats, B. D., Rezvani, N., Zhang, Y., Wieseler, J. L., . . . Watkins, L. R. (2010). Possible involvement of toll-like receptor 4/myeloid differentiation factor-2 activity of opioid inactive isomers causes spinal proinflammation and related behavioral consequences. Neuroscience, 167, 880–893.Find this resource:
Illich, P. A., Salinas, J. A., & Grau, J. W. (1994). Latent inhibition, overshadowing, and blocking of a conditioned antinociceptive response in spinalized rats. Behavioral and Neural Biology, 62, 140–150.Find this resource:
Ji, R. R., Kohno, T., Moore, K. A., & Woolf, C. J. (2003). Central sensitization and LTP: Do pain and memory share similar mechanisms? Trends in Neuroscience, 26, 696–705.Find this resource:
Johansson, F., Hesslow, G., & Medina, J. F. (2016). Mechanisms for motor timing in the cerebellar cortex. Current Opinions in Behavioral Sciences, 8, 53–59.Find this resource:
Johansson, F., Jirenhed, D. A., Rasmussen, A., Zucca, R., & Hesslow, G. (2014). Memory trace and timing mechanism localized to cerebellar Purkinje cells. Proceedings of the National Academy of Sciences USA, 111, 14930–14934.Find this resource:
Joynes, R. L., Ferguson, A. R., Crown, E. D., Patton, B. C., & Grau, J. W. (2003). Instrumental learning within the spinal cord: V. Evidence the behavioral deficit observed after noncontingent nociceptive stimulation reflects an intraspinal modification. Behavioural Brain Research, 141, 159–170.Find this resource:
Joynes, R. L., & Grau, J. W. (1996). Mechanisms of pavlovian conditioning: Role of protection from habituation in spinal conditioning. Behavioral Neuroscience, 110, 1375–1387.Find this resource:
Joynes, R. L., Janjua, K., & Grau, J. W. (2004). Instrumental learning within the spinal cord: VI The NMDA receptor antagonist, AP5, disrupts the acquisition and maintenance of an acquired flexion response. Behavioural Brain Research, 154, 431–438.Find this resource:
Kaila, K., Price, T. J., Payne, J. A., Puskarjov, M., & Voipio, J. (2014). Cation-chloride cotransporters in neuronal development, plasticity and disease. Nature Reviews Neuroscience, 15, 637–654.Find this resource:
Kaila, K., Ruusuvuori, E., Seja, P., Voipio, J., & Puskarjov, M. (2014). GABA actions and ionic plasticity in epilepsy. Current Opinions in Neurobiology, 26, 34–41.Find this resource:
Kang, H. J., & Schuman, E. M. (1995). Long-lasting neurotrophin-induced enhancement of synaptic transmission in the adult hippocampus. Science, 267, 1658–1662.Find this resource:
Kigerl, K. A., Mostacada, K., & Popovich, P. G. (2018). Gut microbiota are disease-modifying factors after traumatic spinal cord injury. Neurotherapeutics, 15, 60–67.Find this resource:
Konorski, J., & Miller, S. (1937). On two types of conditioned reflex. Journal of General Psychology, 16, 264–272.Find this resource:
Lamotte, R. H., Shain, C. N., Simone, D. A., & Tsai, E. F. P. (1991). Neurogenic hyperalgesia: Psychophysical studies of underlying mechanisms. Journal of Neurophysiology, 66, 190–211.Find this resource:
Latremoliere, A., & Woolf, C. J. (2009). Central sensitization: A generator of pain hypersensitivity by central neural plasticity. Journal of Pain, 10, 895–926.Find this resource:
Lee, K. H., Huang, Y.-J., & Grau, J. W. (2016). Learning about time within the spinal cord II: Evidence that temporal regularity is encoded by a spinal oscillator. Frontiers in Behavioral Neuroscience, 10, 14.Find this resource:
Lee, K. H., Turtle, J. D., Strain, M. M., Huang, Y.-J., Baumbauer, K. M., & Grau, J. W. (2015). Learning about time within the spinal cord: Evidence that spinal neurons can abstract and store an index of regularity. Frontiers in Behavioral Neuroscience, 9, 274.Find this resource:
Liu, X. G., Morton, C. R., Azkue, J. J., Zimmermann, M., & Sandkuhler, J. (1998). Long-term depression of C-fibre-evoked spinal field potentials by stimulation of primary afferent A delta-fibres in the adult rat. European Journal of Neuroscience, 10, 3069–3075.Find this resource:
Lossi, L., Castagna, C., & Merighi, A. (2015). Neuronal cell death: An overview of its different forms in central and peripheral neurons. Methods in Molecular Biology, 1254, 1–18.Find this resource:
Ma, Q. P., & Woolf, C. J. (1995). Noxious stimuli induce an N-methyl-D-aspartate receptor-dependent hypersensitivity of the flexion withdrawal reflex to touch: Implications for the treatment of mechanical allodynia. Pain, 61, 383–390.Find this resource:
McNally, G. P., Johansen, J. P., & Blair, H. T. (2011). Placing prediction into the fear circuit. Trends in Neuroscience, 34, 283–292.Find this resource:
Maier, S. F., & Seligman, M. E. (2016). Learned helplessness at fifty: Insights from neuroscience. Psychological Review, 123, 349–367.Find this resource:
Maier, S. F., & Seligman, M. E. P. (1976). Learned helplessness: Theory and evidence. Journal of Experimental Psychology–General, 105, 3–46.Find this resource:
Malcangio, M. (2009). Synaptic plasticity in pain. New York, NY: Springer.Find this resource:
Malenka, R. C. (1994). Synaptic plasticity in the hippocampus: LTP and LTD. Cell, 78, 535–538.Find this resource:
Martin, J. H. (1996). Neuroanatomy: Text and atlas. Stamford, CT: Appleton & Lange.Find this resource:
Mauk, M. D., & Buonomano, D. V. (2004). The neural basis of temporal processing. Annual Review of Neuroscience, 27, 307–340.Find this resource:
Meagher, M. W., Grau, J. W., & King, R. A. (1990). Role of supraspinal systems in environmentally induced antinociception: Effect of spinalization and decerebration on brief shock-induced and long shock-induced antinociception. Behavioral Neuroscience, 104, 328–338.Find this resource:
Medina, I., Friedel, P., Rivera, C., Kahle, K. T., Kourdougli, N., Uvarov, P., & Pellegrino, C. (2014). Current view on the functional regulation of the neuronal K+-Cl- cotransporter KCC2. Frontiers in Cellular Neuroscience, 8, 18.Find this resource:
Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150, 971–979.Find this resource:
Mendell, L. M. (1966). Physiological properties of unyelinated fiber projection to spinal cord. Experimental Neurology, 16, 316.Find this resource:
Mendell, L. M., & Wall, P. D. (1965). Responses of single dorsal cord cells to peripheral cutaneous unmyelinated fibres. Nature, 206, 97.Find this resource:
Mercier, C., Roosink, M., Bouffard, J., & Bouyer, L. J. (2017). Promoting gait recovery and limiting neuropathic pain after spinal cord injury: Two sides of the same coin? Neurorehabilitation and Neural Repair, 31, 315–322.Find this resource:
Millan, M. J. (1999). The induction of pain: An integrative review. Progress in Neurobiology, 57, 1–164.Find this resource:
Millan, M. J. (2002). Descending control of pain. Progress in Neurobiology, 66, 355–474.Find this resource:
Mishra, A. M., Pal, A., Gupta, D., & Carmel, J. B. (2017). Paired motor cortex and cervical epidural electrical stimulation timed to converge in the spinal cord promotes lasting increases in motor responses. Journal of Physiology-London, 595, 6953–6968.Find this resource:
Morris, R. G. M. (2013). NMDA receptors and memory encoding. Neuropharmacology, 74, 32–40.Find this resource:
Moser, E. I., & Moser, M. B. (1999). Is learning blocked by saturation of synaptic weights in the hippocampus? Neuroscience & Biobehavioral Reviews, 23, 661–672.Find this resource:
Nickel, F. T., Seifert, F., Lanz, S., & Maihofner, C. (2012). Mechanisms of neuropathic pain. European Neuropsychopharmacology, 22, 81–91.Find this resource:
Nishimaru, H., & Kudo, N. (2000). Formation of the central pattern generator for locomotion in the rat and mouse. Brain Research Bulletin, 53, 661–669.Find this resource:
Ollivier-Lanvin, K., Fischer, I., Tom, V., Houle, J. D., & Lemay, M. A. (2015). Either brain-derived neurotrophic factor or neurotrophin-3 only neurotrophin-producing grafts promote locomotor recovery in untrained spinalized cats. Neurorehabilitation and Neural Repair, 29, 90–100.Find this resource:
Olson, G. A., Olson, R. D., Kastin, A. J., & Coy, D. H. (1982). Endogenous opiates: 1981. Peptides, 3, 1039–1072.Find this resource:
Patterson, M. M. (1976). Mechanisms of classical conditioning and fixation in spinal mammals. Advances in Psychobiology, 3, 381–436.Find this resource:
Patterson, M. M. (2001a). Classical conditioning of spinal reflexes: The first seventy years. In J. E. Steinmetz, M. A. Gluck, & P. R. Solomon (Eds.), Model systems and the neurobiology of associative learning (pp. 1–22). Mahwah, NJ: Erlbaum.Find this resource:
Patterson, M. M. (2001b). Spinal fixation: Long-term alterations in spinal reflex excitability with altered or sustained sensory inputs. In M. M. Patterson & J. W. Grau (Eds.), Spinal cord plasticity: Alterations in reflex function (pp. 77–99). Boston, MA: Kluwer Academic.Find this resource:
Patterson, S. L., Abel, T., Deuel, T. A. S., Martin, K. C., Rose, J. C., & Kandel, E. R. (1996). Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron, 16, 1137–1145.Find this resource:
Pavlov, I. P. (1927). Conditioned reflexes (G. V. Anrep, Trans.). London: Oxford University Press.Find this resource:
Perea, G., Navarrete, M., & Araque, A. (2009). Tripartite synapses: Astrocytes process and control synaptic information. Trends in Neuroscience, 32, 421–431.Find this resource:
Perrett, S. P., Dudek, S. M., Eagleman, D., Montague, P. R., & Friedlander, M. J. (2001). LTD induction in adult visual cortex: Role of stimulus timing and inhibition. Journal of Neuroscience, 21, 2308–2319.Find this resource:
Price, T. J., & Inyang, K. E. (2015). Commonalities between pain and memory mechanisms and their meaning for understanding chronic pain. Progress in molecular biology and translational science, 131, 409–434.Find this resource:
Rabchevsky, A. G., & Kitzman, P. H. (2011). Latest approaches for the treatment of spasticity and autonomic dysreflexia in chronic spinal cord injury. Neurotherapeutics, 8, 274–282.Find this resource:
Rescorla, R. A. (1988). Behavioral studies of Pavlovian conditioning. Annual Review of Neuroscience, 11, 329–352.Find this resource:
Rossignol, S., & Frigon, A. (2011). Recovery of locomotion after spinal cord injury: Some facts and mechanisms. Annual Review of Neuroscience, 34, 413–440.Find this resource:
Rossignol, S., Schwab, M., Schwartz, M., & Fehlings, M. G. (2007). Spinal cord injury: Time to move? Journal of Neuroscience, 27, 11782–11792.Find this resource:
Saboe, L. A., Reid, D. C., Davis, L. A., Warren, S. A., & Grace, M. G. (1991). Spine trauma and associated injuries. Journal of Trauma–Injury Infection & Critical Care, 31, 43–48.Find this resource:
Sandkuhler, J. (2000). Learning and memory in pain pathways. Pain, 88, 113–118.Find this resource:
Sandkuhler, J., Chen, J. G., Cheng, G., & Randic, M. (1997). Low-frequency stimulation of afferent Adelta-fibers induces long-term depression at primary afferent synapses with substantia gelatinosa neurons in the rat. Journal of Neuroscience, 17, 6483–6491.Find this resource:
Sandkuhler, J., & Liu, X. G. (1998). Induction of long-term potentiation at spinal synapses by noxious stimulation or nerve injury. European Journal of Neuroscience, 10, 2476–2480.Find this resource:
Satoh, M., & Minami, M. (1995). Molecular pharmacology of the opioid receptors. Pharmacology & Therapeutics, 68, 343–364.Find this resource:
Schmidt, M. V., Abraham, W. C., Maroun, M., Stork, O., & Richter-Levin, G. (2013). Stress-induced metaplasticity: From synapses to behavior. Neuroscience, 250, 112–120.Find this resource:
Sherrington, C. S. (1906). The integrative action of the nervous system. New Haven, CT: Yale University Press.Find this resource:
Shurrager, P. S., & Culler, E. (1940). Conditioning in the spinal dog. Journal of Experimental Psychology, 26, 133–159.Find this resource:
Shurrager, P. S., & Dykman, R. A. (1951). Walking spinal carnivores. Journal of Comparative Psychology, 44, 252–262.Find this resource:
Simard, J. M., Kahle, K. T., & Gerzanich, V. (2010). Molecular mechanisms of microvascular failure in central nervous system injury: Synergistic roles of NKCC1 and SUR1/TRPM4. Journal of Neurosurgery, 113, 622–629.Find this resource:
Simard, J. M., Woo, S. K., Aarabi, B., & Gerzanich, V. (2013). The Sur1-Trpm4 channel in spinal cord injury. Spine, (Supp. 4), 002.Find this resource:
Simone, D. A., Baumann, T. K., & Lamotte, R. H. (1989). Dose-dependent pain and mechanical hyperalgesia in humans after intradermal injection of capsaicin. Pain, 38, 99–107.Find this resource:
Skinner, B. F. (1938). The behavior of organisms. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:
Solomon, P. R., Vander Schaaf, E. R., Thompson, R. F., & Weisz, D. J. (1986). Hippocampus and trace conditioning of the rabbit’s classically conditioned nictitating membrane response. Behavioral Neuroscience, 100, 729–744.Find this resource:
Stellwagen, D., & Malenka, R. C. (2006). Synaptic scaling mediated by glial TNF-alpha. Nature, 440, 1054–1059.Find this resource:
Sutherland, R. J. R., (1989). Configural association theory: The role of the hippocampus formation in learning, memory, and amnesia. Psychobiology, 17, 129–144.Find this resource:
Thompson, A. K., & Wolpaw, J. (2014). Operant conditioning of spinal reflexes: From basic science to clinical therapy. Frontiers in Integrative Neuroscience, 8, 25.Find this resource:
Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for study of neuronal substrates of behavior. Psychological Review, 73, 16.Find this resource:
Timberlake, W. (1990). Biological behaviorism. In W. O’Donohuse & R. Kitchener (Eds.), Handbook of behaviorism (pp. 244–286). New York, NY: Academic Press.Find this resource:
Timberlake, W., & Lucas, G. A. (1989). Behavior systems and learning: From misbehavior to general principles. In S. B. Klein & R. R. Mowrer (Eds.), Contemporary learning theory: Instrumental conditioning and the impact of biological constraints on learning (pp. 237–275). Hillsdale, NJ: Lawrence Erlbaum.Find this resource:
Treede, R. D. (2016). Gain control mechanisms in the nociceptive system. Pain, 157, 1199–1204.Find this resource:
Turtle, J. D., Henwood, M. K., Strain, M. M., Huang, Y. J., Miranda, R. C., & Grau, J. W. (2019). Engaging pain fibers after a spinal cord injury fosters hemorrhage and expands the area of secondary injury. Experimental Neurology, 311, 115–124.Find this resource:
Turtle, J. D., Strain, M. M., Aceves, M., Huang, Y. J., Reynolds, J. A., Hook, M. A., & Grau, J. W. (2017). Pain input impairs recovery after spinal cord injury: Treatment with lidocaine. Journal of Neurotrauma, 34, 1200–1208.Find this resource:
Turtle, J. D., Strain, M. M., Reynolds, J. A., Huang, Y. J., Lee, K. H., Henwood, M. K., . . .Grau, J. W. (2018). Pain input after spinal cord injury (SCI) undermines long-term recovery and engages signal pathways that promote cell death. Frontiers in Systems Neuroscience, 12, 27.Find this resource:
van den Brand, R., Heutschi, J., Barraud, Q., DiGiovanna, J., Bartholdi, K., Huerlimann, M., . . . Courtine, G. (2012). Restoring voluntary control of locomotion after paralyzing spinal cord injury. Science, 336, 1182–1185.Find this resource:
Vichaya, E. G., Baumbauer, K. M., Carcoba, L. M., Grau, J. W., & Meagher, M. W. (2009). Spinal glia modulate both adaptive and pathological processes. Brain Behavior and Immunity, 23, 969–976.Find this resource:
Wang, Y., Chen, Y., Chen, L., Wolpaw, J. R., & Chen, X. Y. (2012). Cortical stimulation causes long-term changes in H-reflexes and spinal motoneuron GABA receptors. Journal of Neurophysiology, 108, 2668–2678.Find this resource:
Wang, Y., Pillai, S., Wolpaw, J. R., & Chen, X. Y. (2006). Motor learning changes GABAergic terminals on spinal motoneurons in normal rats. European Journal Neuroscience, 23, 141–150.Find this resource:
Wang, Y., Pillai, S., Wolpaw, J. R., & Chen, X. Y. (2009). H-reflex down-conditioning greatly increases the number of indentifiable GABAergic interneurons in rat ventral horn. Neuroscience Letters, 452, 124–129.Find this resource:
Watkins, L. R., Cobelli, D. A., & Mayer, D. J. (1982). Classical-conditioning of front paw and hind paw footshock induced analgesia (FSIA): Naloxone reversibility and descending pathways. Brain Research, 243, 119–132.Find this resource:
Watkins, L. R., Hutchinson, M. R., Ledeboer, A., Wieseler-Frank, J., Milligan, E. D., & Maier, S. F. (2007). Glia as the “bad guys”: Implications for improving clinical pain control and the clinical utility of opioids. Brain Behavior and Immunity, 21, 131–146.Find this resource:
Wernig, A., Muller, S., Nanassy, A., & Cagol, E. (1995). Laufband therapy based on “rules of spinal locomotion” is effective in spinal cord injured persons. European Journal of Neuroscience, 7, 823–829.Find this resource:
Willis, W. D. (2001). Mechanisms of central sensitization of nociceptive dorsal horn neurons. In M. M. Patterson & J. W. Grau (Eds.), Spinal cord plasticity: Alterations in reflex function (pp. 127–161). Boston, MA: Kluwer Academic.Find this resource:
Wolpaw, J. R. (2010). What can the spinal cord teach us about learning and memory? Neuroscientist, 16, 532–549.Find this resource:
Wolpaw, J. R., & Carp, J. S. (1990). Memory traces in spinal-cord. Trends in Neuroscience, 13, 137–142.Find this resource:
Woolf, C. J. (1983). Evidence for a central component of post-injury pain hypersensitivity. Nature, 306, 686–688.Find this resource:
Woolf, C. J., & Thompson, S. W. N. (1991). The induction and maintenance of central sensitization is dependent on N-methyl-D-aspartic acid receptor activation: Implications for the treatment of postinjury pain hypersensitivity states. Pain, 44, 293–299.Find this resource:
Yang, Q., Wu, Z., Hadden, J. K., Odem, M. A., Zuo, Y., Crook, R. J., . . . Walters, E. T. (2014). Persistent pain after spinal cord injury is maintained by primary afferent activity. Journal of Neuroscience, 34, 10765–10769.Find this resource:
Ying, Z., Roy, R. R., Zhong, H., Zdunowski, S., Edgerton, V. R., & Gomez-Pinilla, F. (2008). BDNF-exercise interactions in the recovery of symmetrical stepping after a cervical hemisection in rats. Neuroscience, 155, 1070–1078.Find this resource:
Young, E. E., Baumbauer, K. M., Elliot, A., & Joynes, R. L. (2007). Lipopolysaccharide induces a spinal learning deficit that is blocked by IL-1 receptor antagonism. Brain Behavior and Immunity, 21, 748–757.Find this resource: