Acute damage to the corpus callosum (CC) may generate a cluster of clear-cut inter-hemispheric disconnection (IHD) “split-brain” symptoms reflecting the inability to convey sensory information to contralateral motor or linguistic output areas (Tomasch, 1954; Sperry, 1968a). In rare cases it may also produce diagonistic dyspraxia (DD; Akelaitis, 1945), a fascinating neuropsychological syndrome involving involuntary inter-manual conflicts where one hand acts at cross purposes with the other. DD has remained the most representative “alien hand” syndrome, a term encompassing a variety of seemingly goal-directed movements occurring after a brain insult, performed without the intention of the actor (Scepkowski and Cronin-Golomb, 2003; Biran and Chatterjee, 2004; Berlucchi, 2012). Although DD has been observed in cases of acquired CC damage, to our knowledge it has never been described in developmentally-based CC dysgenesis.

Agenesis of the CC (AgCC), occurring in 1/4000 live births (Paul et al., 2007), may be partial or complete; may be found in isolation (primary AgCC) or as part of a wider developmental disorder (Paul et al., 2007); and is commonly associated with epilepsy (Taylor and David, 1998). In contrast to acquired lesions of the CC, primary AgCC has an overall limited and subtle impact on cognition and subjects with AgCC do not display clinically relevant IHD symptoms (Sperry, 1968b). To account for this puzzling fact—known as “Sperry’s paradox” (Sperry, 1968a)—compensatory structural pathways have been suggested. Neuroimaging has recently provided support for macrostructural changes in AgCC, evidencing intra- and interhemispheric white matter tracts providing bilateral links via the posterior and anterior commissures (Tovar-Moll et al., 2014). These are likely to compose, at least partly, the set of compensatory pathways preserving inter-hemispheric transfer. These observations are corroborated by a resting-state functional magnetic resonance imaging (rs-fMRI) study showing preserved functional connectivity (FC) between homotopic cortices in AgCC subjects (Tyszka et al., 2011). Additionally, intrinsic connectivity networks (ICNs) in the AgCC group were similar to those identified in controls, suggesting that global functional architecture remains largely preserved in AgCC. Paralleling the clinically-defined Sperry’s paradox, these findings stand in sharp contrast with the drastic disruption of inter-hemispheric FC observed following surgical interruption of the CC (Johnston et al., 2008).

Here, we report on an epileptic AgCC patient who developed pervasive DD behaviors following status epilepticus (SE), with gradual improvement over 2-years. Patient AM offers a rare opportunity to longitudinally study the functional underpinning of DD, without the major impact produced by acute callosal section/lesions on connectomics. By examining rs-fMRI connectivity via edgewise comparison of FC and graph theoretical modeling, we demonstrate large scale disorganization and reorganization paralleling the emergence and extinction of IHD and DD symptoms.

Patient AM represents a unique case of DD in a developmentally acallosal patient, highlighting a novel complication of comorbid AgCC and epilepsy, as well as serving as a test case for the use of resting-state analyses in a little-understood disorder.

According to recent theoretical accounts, “alien” movements occur because affordances supplied by the environment reflexively generate motor primitives that are not inhibited by an intended action (Frith et al., 2000; Biran et al., 2006). The patient is aware that an action is produced (via sensory feedback), but the actual action is performed despite no intention and no postulation (feed-forward intention) of a predicted bodily change. Failure of inhibitory signals arising from the supplementary motor area (SMA) acting on a sensorimotor (SM) fronto-parietal system involved in the production and coordination of motor primitives has also been posited, either because of defective inhibition (e.g., lesions of the SMA) or an impairment of its inter-hemispheric transfer (e.g., lesions of the CC; Frith et al., 2000; Biran et al., 2006). Other accounts stress a form of “split-brain” phenomena: movements are purposeful and reasonable given each hemisphere’s competences, but the “intention” is inaccessible and thus “alien” due to failed interhemispheric integration (Verleger et al., 2011).

Activation-fMRI studies have attempted to delineate the neural basis of alien hand syndromes (Assal et al., 2007; Schaefer et al., 2010). Collectively, they suggest that both intentional and non-intentional movements involve a common set of regions, principally primary motor, supplementary motor and pre-motor areas, but differ in the activation of the inferior frontal gyrus (IFG). However, conclusions regarding the role of the IFG could not be drawn, since the activation studies are in apparent disagreement: associating the IFG with either intentional (Assal et al., 2007) or non-intentional movements (Schaefer et al., 2010).

Given its minimal demands on the individual being scanned, a resting-state approach offers an adjunct to traditional imaging approaches that need not rely on specific symptoms or tasks that may be difficult in DD (Scepkowski and Cronin-Golomb, 2003; Berlucchi, 2012). Our results (Figure 1A) implicate inter-hemispheric disruption between a set of frontal regions identified in the above activation studies (Assal et al., 2007; Schaefer et al., 2010) in both intentional and non-intentional movements (superior mesial, pre-central and SMA regions), as well as those that distinguish them (i.e., the IFG). Additionally, graph analysis implicates reduced efficiency in the salience network (Figure 2). The salience network is a recently described large-scale network that comprises frontoinsular regions and the anterior cingulate cortex (ACC), along with limbic and subcortical structures (Seeley et al., 2007). Comparisons between intrinsic connectivity and task-based co-activation derived from many thousands of functional datasets indicate the salience network encompasses a wide range of processes including integration of bodily-related information and conflict/error monitoring (Smith et al., 2009), positioning it remarkably well for the transitional role between extero-/interoception and cognition originally posited by Seeley et al. (2007): as a sensory integrator and filter capable of “tagging” information as relevant or irrelevant for higher-level “executive” networks. Concordantly, salience network disruption observed in frontotemporal dementia is thought to be specifically implicated in disinhibited behaviors (Farb et al., 2013; Zhou and Seeley, 2014). Thus, our results may reflect a disruption of a system involved in the ability to select the relevant sensory stimuli, and/or inhibit non-intended actions.

Furthermore, patients experience non-volitional movements as not only unplanned but also as distinctly “other” or “alien” (Scepkowski and Cronin-Golomb, 2003; Biran and Chatterjee, 2004; Biran et al., 2006). Both edgewise and ICN results indicate disruption of an inter-hemispheric network that task-based neuroimaging implicates in distinguishing self/other-generated actions, including dorsomedial frontal cortices and one of the salience network’s key nodes (the insular region; Farrer and Frith, 2002; Farrer et al., 2003; Sperduti et al., 2011). Interestingly, corticobasal degeneration can also result in volitional experience and movement aberrations falling under the rubric of “alien hand” (c.f. Schaefer et al., 2016). In a group of patients suffering from this condition, Wolpe et al. (2014) revealed modified FC of the pre-SMA with a set of fronto-parietal regions that included the anterior component of the salience network.

Taken together, the normalization of connectivity and network organization with remission of DD may reflect a regained ability to detect and disregard irrelevant SM programs primed by environmental affordances, and a restored sense of authorship in self-generated actions. Enduring modifications within a right hemisphere could reflect remaining difficulties in tool-use pantomimes (Supplementary Figure 1), consistent with the role of FP regions in complex actions relying on stored information such as tool use (Wheaton and Hallett, 2007).

We suggest that disconnection symptoms and DD were likely prompted by SE, unbalancing the inter-hemispheric transfer that may have previously transited via an intact anterior commissure which subserved the patient’s previously “normal” behavioral profile (Franz, 2012; Winter and Franz, 2014). Comorbities in patient AM, while potential sources of vulnerability seem unlikely to account for the acute onset and trajectory described here. Ongoing seizure activity cannot explain AM’s DD, since seizures have fully resolved. Despite distributed network effects of epilepsy (Bettus et al., 2009, 2010; Guye et al., 2010; Ridley et al., 2015), changes involving the regions identified here are for the most part identified in children and young adults and represent developmental and non-acute processes (Ibrahim et al., 2014; Luo et al., 2014; Li et al., 2015; Wei et al., 2015). Likewise, despite drastic and life-long structural modifications (Kasprian et al., 2013; Jakab et al., 2015) AgCC it is not equivalent to callosotomy (Owen et al., 2013), and recent evidence suggests acallosal brains can support a largely “normal” cognitive and intrinsic FC repertoire in the form of preserved homotopic connectivity and bilaterally symmetric ICN architecture (Tyszka et al., 2011; Tovar-Moll et al., 2014). This conclusion is bolstered by the fact that partitioning at gross anatomical scales (Supplementary Figure 4) does not yield equivalent graph metric changes, underlining the relevance of the functional network partitions employed here.

However, despite the forgoing, larger homogenous samples—including separate longitudinal control, AgCC and epilepsy patient groups—are certainly necessary to address the limitations of this study and confirm Patient AM as a general model of alien hand syndromes. Furthermore, the test-retest reliability of connectomics metrics places a lower limit on the interpretability of data, though the stability of our data over time as suggested by Figure 1B militate against this being the main determinant of our results. Future work will also need to speak to known sources of variance in connectomics, including age, gender, handedness, hemispheric dominance, medication, and vigilance state (Bettus et al., 2009; Liu et al., 2009; Tian et al., 2011; Vlooswijk et al., 2011; Tagliazucchi and Laufs, 2014).

Our data suggest that the salience network and interhemispheric connectivity play a role in supporting the experience and control of volitional movements and the ability to select the most relevant among internal and extrapersonal stimuli in order to produce appropriate behavior. Bearing methodological and sampling caveats in mind, the current work represents a promising first indication that rs-fMRI is relevant to the understanding of alien hand syndromes over time and, more broadly, of neuropsychiatric manifestations that may benefit in the same way from an adjunct to structural and task-based functional imaging. In this spirit, we hope that AM’s case will serve to marshal resources in pursuit of a better understanding of this difficult and rare disorder.

BR, OF, MG, J-PR, MB and JW contributed to the design of the work, acquisition analysis and interpretation of data. ALeT, ET, SaA and SoA contributed to the acquisition, analysis and interpretation of data. All authors revised, contributed to and approved the final manuscript. All co-authors agree on all aspects of the work and ensure that questions related to the accuracy and integrity of any part of the submitted work are appropriately investigated and resolved.

The funding sources (“PHRC-I 2013” EPI-SODIUM (grant number 2014-27)) had no involvement in study design; collection, analysis and interpretation of data; in the writing of the report; in the decision to submit the article for publication.

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