Evoked Responses

Evoked responses are elicited by sensory stimulation and are usually recovered by signal averaging time-locked to stimulus delivery. Response waveforms consist of a stereotypical series of peaks and valleys observed between 2 and 250 msec poststimulus. Such features reflect the sequential and parallel activation of multiple specialized processing waystations within the targeted sensory pathway, although individual response components may represent simultaneous activity in several distinct areas. The responses are a product of the network architecture and the dynamics of participating neurons and are reasonably robust. Sensory evoked activity can often be recorded even under anesthesia.

Evoked responses have been used to probe the architecture and the information processing activities of sensory systems. Such work has provided evidence for specialized processing modules, corresponding to the discrete areas identified through invasive physiology, and has disclosed systematic projection of the sensory parameter space onto cortex. Other studies have provided evidence for early modulation of sensory responses by selective attention and demonstrated dynamic plastic changes in the functional organization of cortex, based on patterns of sensory coactivation or neglect.

1. Somatosensory Evoked Responses

The somatosensory system can be activated through tactile stimulation of almost any body surface. Particularly robust responses can be elicited by electrical stimulation, for example, of the median nerve of the forearm. The earliest responses to somatosensory stimulation are observed from 9 to 15 msec poststimulus and represent responses generated in the spinal cord and brain stem. Initial responses in primary somatosensory cortex are observed around 20 msec poststimulus. An evolving complex of responses typically lasts over 150 msec. Source modeling studies identify at least 5-6 discrete regions of activation. Studies of the organization of the primary somatosensory area (S1) disclose a systematic projection of the body surface onto cortex adjacent to the central sulcus,

Figure 7 Analysis of MEG evoked responses by Bayesian inference. (a) Extended parametric source models used for Bayesian inference. Upper panel: A source defined by the intersection of cortex with a sphere centered on cortex. Note that adjacent sides of a sulcus or gyrus are often labeled together for extended sources. Lower panels: A source defined by a patch grown on the cortical surface. A location on cortex is seeded, and adjacent bands of voxels are labeled in a series of dilation operations. (b) A series of sample solutions from the posterior probability distribution. After 1000 iterations the MCMC algorithm found the same set of three sources in almost every sample, although additional extraneous sources also appear in some solutions. These data were simulated and thus known to contain three sources in the locations suggested by Bayesian inference. (c) Interactive visualization of spatial-temporal source probability maps coregistered with anatomical MRI for the same subject. The anatomical data set was used in the Bayesian inference procedure to constrain sources to lie in cortex. (d) Source probability maps estimated for visual evoked response data. Four views of a region found to contain activity at a 95% probability level. This example is for left visual field stimulation. For right field stimulation, the most probable source is lateralized to the calcarine fissure in the left hemisphere.

Figure 7 Analysis of MEG evoked responses by Bayesian inference. (a) Extended parametric source models used for Bayesian inference. Upper panel: A source defined by the intersection of cortex with a sphere centered on cortex. Note that adjacent sides of a sulcus or gyrus are often labeled together for extended sources. Lower panels: A source defined by a patch grown on the cortical surface. A location on cortex is seeded, and adjacent bands of voxels are labeled in a series of dilation operations. (b) A series of sample solutions from the posterior probability distribution. After 1000 iterations the MCMC algorithm found the same set of three sources in almost every sample, although additional extraneous sources also appear in some solutions. These data were simulated and thus known to contain three sources in the locations suggested by Bayesian inference. (c) Interactive visualization of spatial-temporal source probability maps coregistered with anatomical MRI for the same subject. The anatomical data set was used in the Bayesian inference procedure to constrain sources to lie in cortex. (d) Source probability maps estimated for visual evoked response data. Four views of a region found to contain activity at a 95% probability level. This example is for left visual field stimulation. For right field stimulation, the most probable source is lateralized to the calcarine fissure in the left hemisphere.

Was this article helpful?

0 0
Autism

Autism

Is there a cause or cure for autism? The Complete Guide To Finally Understanding Autism. Do you have an autistic child or know someone who has autism? Do you understand the special needs of an autistic person?

Get My Free Ebook


Post a comment