Somatotopic mismatch of hand representation following stroke: is recovery possible?

Examining central biases in somatosensory localization: Evidence from brain-damaged individuals

How does the brain localize touch under conditions of uncertainty caused by brain damage? By testing single cases, previous work found mislocalization of touch toward the center of the hand. We investigated whether such central bias changes as a function of uncertainty in somatosensory system. Fifty-one brain-damaged individuals were presented with a tactile detection task to establish their tactile threshold, and a tactile localization task in which they localized suprathreshold stimuli presented at different locations on the hand. We predicted that with increased somatosensory uncertainty, indexed by higher detection thresholds, participants would more likely to localize the stimuli toward the center of the hand. Consistent with this prediction, participants' localization errors were biased towards the center of the hand and, importantly, this bias increased as detection threshold increased. These findings provide evidence that instead of showing random errors, uncertainty leads to systematic localization errors toward the center of the hand. We discuss these findings under different frameworks as potential mechanisms to explain biases in tactile localization subsequent to brain damage.

Intact tactile detection yet biased tactile localization in a hand-centered frame of reference: Evidence from a dissociation

We examined the performance of an individual with subcortical damage, but an intact somatosensory thalamocortical pathway, to examine the functional architecture of tactile detection and tactile localization processes. Consistent with the intact somatosensory thalamocortical pathway, tactile detection on the contralesional hand was well within the normal range. Despite intact detection, the individual demonstrated substantial localization biases. Across all localization experiments, he consistently localized tactile stimuli to the left side in space relative to the long axis of his hand. This was observed when the contralesional hand was palm up, palm down, rotated 90° relative to the trunk, and when making verbal responses. Furthermore, control experiments demonstrated that this response pattern was unlikely a motor response error. These findings indicate that tactile localization on the body is influenced by proprioceptive information specifically in a hand-centered frame of reference. Furthermore, this also provides evidence that aspects of tactile localization are mediated by pathways outside of the primary somatosensory thalamocortical pathway.

Automated and Quantitative Assessment of Tactile Mislocalization After Stroke

Topesthesia, the recognition of tactile stimulation location on the skin, can be severely affected by neurological injuries, such as stroke. Despite topesthesia being crucial for manipulating objects and interacting with the environment during activities of daily living, deficits cannot be quantitatively captured with current clinical assessments and are, as a consequence, not well-understood. The present work describes a novel automated assessment tool for tactile mislocalization in neurological patients with somatosensory deficits. We present two cases of ischemic stroke patients, describe their tactile localization deficits with the automated assessment, and compare the results to a standard manual clinical assessment. Using the automated assessment tool, it was possible to identify, locate, precisely quantify, and depict the patients' deficits in topesthesia. In comparison, the clinical assessment was not sensitive enough and some deficits would remain undetected due to ceiling effects. In addition, an MRI structural analysis of the lesion supported the existence of somatosensory deficits. This novel and quantitative assessment may not only help to raise awareness of the implications of deficits in topesthesia, but would also allow monitoring recovery throughout the rehabilitation process, informing treatment design, and objectively evaluating treatment efficacy.

Biases in tactile localization by pointing: compression for weak stimuli and centering for distributions of stimuli

Weak electrocutaneous stimuli applied to the forearm are erroneously localized toward its middle (Steenbergen P, Buitenweg JR, Trojan J, Veltink PH. Exp Brain Res 232: 597-607, 2014). We asked whether mechanical touch stimuli exhibit a similar bias and whether the bias is toward the middle of the forearm or toward the middle of the recent stimulus distribution. In experiments 1 and 2, participants ( n = 12 and n = 10) localized by pointing von Frey filaments applied to four locations on the dorsal forearm. Individually adjusted weak and strong stimuli ( experiment 1) or two levels of strong stimuli ( experiment 2) were presented in single sessions in random order. Weaker stimuli were localized with greater variability than the strong, with compression toward the middle of the forearm. Responses to the two levels of strong stimuli did not differ. In experiment 3, participants ( n = 16) were presented two spatially offset stimulus distributions (8 cm center-to-center), each offset from the forearm middle, on 2 different days. Out of four target locations comprising each distribution, two were shared. Responses to weak stimuli were compressed compared with responses to strong stimuli. Importantly, biases for the common targets had opposite directions, each being toward the middle of the distribution within which targets were presented. Responses to strong stimuli also exhibited a distribution-dependent bias, a 2-cm overall shift across the forearm midpoint. We conclude that touch localization is subject to intensity-dependent biases determined by the recent history of stimulation and possibly also by the available or perceived response space. NEW & NOTEWORTHY Recent findings show that weak electrical stimuli applied to the forearm are mislocalized toward the forearm middle, relative to strong stimuli. We found that weak mechanical stimuli are similarly mislocalized. The bias changed if, as a group, stimuli were not centered on the forearm middle: weak stimuli gravitated toward the center of prior stimulation. Localization of strong stimuli was also biased, consistent with the tendency to center responses within the available response space.

What can errors tell us about body representations?

In this review, we examine how tactile misperceptions provide evidence regarding body representations. First, we propose that tactile detection and localization are serial processes, in contrast to parallel processing hypotheses based on patients with numbsense. Second, we discuss how information in primary somatosensory maps projects to body size and shape representations to localize touch on the skin surface, and how responses after use-dependent plasticity reflect changes in this mapping. Third, we review situations in which our body representations are inconsistent with our actual body shape, specifically discussing phantom limb phenomena and anesthetization. We discuss problems with the traditional remapping hypothesis in amputees, factors that modulate perceived body size and shape, and how changes in perceived body form influence tactile localization. Finally, we review studies in which brain-damaged individuals perceive touch on the opposite side of the body, and demonstrate how interhemispheric mechanisms can give rise to these anomalous percepts.