03 Brain-Computer Interfaces

Question 1

What are the premotor cortices? (position, input, involvement)

Answer 1

• lateral part (including Broca’s area in the dominant hemisphere) and medial part (supplementary motor area and pre-SMA)
• receive input from prefrontal areas on motivation and intention (action goals) and from parietal lobe (information on circumstance)
• lateral premotor areas more involved in responses to the environment, medial premotor areas more involved in self initiated movements

Question 2

Relation premotor cortices - movement

Answer 2

• in a conditioned task, premotor neurons begin firing with the cue (intention), primary motor neurons fire at movement execution (action)
• premotor cortices assemble the movements necessary to achieve those goals and instruct the motor cortex, where simple movement patterns are stored
• premotor influence motor behavior both indirectly (via projections to primary motor cortex) and directly (>30 % of corticospinal neurons arise in premotor areas)

Question 3

Primary motor cortex (M1) anatomy

Answer 3

• anterior rim of the central sulcus and minor part of precentral gyrus
• topographical map of motor cortex comparable to sensory homunculus (yet less clear cut)
• lateral to medial: face (corticobulbar), upper extremity, trunk, lower extremity (all corticospinal)
• layer V giant pyramidal cells (Betz cells) contact contralateral (“second”) motor neurons in spinal chord
• stimulation of a single motor neuron elicits activity in 2-3 muscles (“muscle field” of neuron)
• M1 codes for simple movements, not single muscle

Question 4

What are upper and lower motor neurons? Which other structures are important for motor control?

Answer 4

• upper motor neurons: neurons from descending systems (motor cortex, brainstem centers)
• lower motor neurons: send axons out of the brainstem or spinal chord (ventral horn), innervate single muscle (grouped together into rod shaped clusters)
• other important structures for motor control: basal ganglia, cerebellum

Question 5

How are neurons in the ventral horn of spinal chord organised?

Answer 5

• lower motor neurons enter/exit spinal chord here
• neurons innervating proximal extremities are more medial, more distal extremities are more lateral

Question 6

Corticospinal and corticobulbar tract together

Answer 6

• Betz cells in layer V of M1 only account for 5% of the axons projected from motor cortex to the spinal cord
• Non Betz pyramidal neurons found in the V layer of M1 descend in the corticobulbar and corticospinal tracts
• they pass through the internal capsule, enter the cerebral peduncle at the base of the midbrain, pass through the base of the pons and coalesce to form the medullary pyramids
• corticobulbar tract terminates in the brainstem

Question 7

corticospinal tract

Answer 7

• near the caudal end of the medulla most fibers in the medullary pyramids are corticospinal axons
• About 90% of these axons decussate on the height of the caudal medulla and form the lateral corticospinal tract
• other 10% of such axons do not change sides and form the ventral corticospinal tract
• lateral corticospinal tract forms direct pathways from the cortex to the spinal cord and terminates primarily in lateral part of the ventral horn
• Some of these axons synapse directly onto α motor neurons that directly govern distal extremities (mostly hand and forearm muscles)
• Most of these axons however will terminate in pools of local circuit neurons that coordinate activities in lateral cell columns of the ventral horn

Question 8

definition Brain-Computer interface

Answer 8

• system measuring CNS activity
• converting it into artificial output
• output replaces, restores, enhances, supplements or improves natural CNS output
• -> changing/modifying ongoing interaction between CNS and its external or internal environment
• based on sensorimotor hypothesis
• BCI gives CNS additional artificial output (not neuromuscular or hormonal)

Question 9

sensorimotor hypothesis

Answer 9

entire function of CNS is to translate sensory inputs into motor outputs

Question 10

elements of BCI

Answer 10

• signal acquisition
• feature extraction
• feature translation
• device output
• (feedback)
• interactive functioning between CNS and BCI determined by operating protocol

Question 11

BCI - signal acquisition

Answer 11

• measurement of neurophysiological state of the brain
• recording interface gathers neural information reflecting intent embedded within brain activity
• acquired through particular sensory modality, e.g. scalp or intracranial electrodes, fMRI for metabolic activity

Question 12

BCI - feature extraction

Answer 12

• process of analysing digital signals and distinguishing pertinent signal characteristics (features) from irrelevant signals
• features should have strong correlations with users intent
• typical features: amplitudes or latencies of ERPs (e.g. P300), frequency power spectra (e.g., sensorimotor rhythms), firing rates of individual cortical neurons

Question 13

BCI - feature translation

Answer 13

• conversion of pertinent feature into device command using translation algorithm
• core = model (set of mathematical equations) describing relationship between intent and feature
• description can be employed in order to convert future observations to appropriate output (generalization)
• not all BCIs require separate feature extraction and translation (e.g. artificial neural networks)

Question 14

BCI - device output

Answer 14

• translated features allow for an output to operate a task-specific device
• various examples of different output devices: letter selection, cursor control, robotic arm operation
• output then again produces feedback

Question 15

Restoration of Hand Use in temporarily paralyzed monkeys

Answer 15

• Multi electrode recordings from the primary motor cortex M1 can be deployed to predict kinematic features of movement
• Neurons in M1 carry information related to the dynamics of movement as well as kinematics
• Such information can be employed to predict muscle activity (EMG) underlying complex reaching tasks
• In this study Pohlmeyer and colleagues use real time EMG predictions to control the focal electrical stimulation (FES) of multiple forearm muscles in temporarily paralyzed monkeys

Question 16

Pohlmeyer et al. (2009) - methods and results

Answer 16

• 2 Rhesus Macaque monkeys had an 100-electrode array chronically implanted into the hand area of the motor cortex
• 4 forearm muscles where artificially paralyzed
• by extracting and translating features from the EMG signal measured in M1 intended muscle movement was predicted in real time
• prediction was used to tune the strength of activation in the FES electrodes of the forearm muscles
• force exerted with FES system activated significantly different from force exerted in a blocked nerve trial

Question 17

Summary BCI

Answer 17

• Motor neurons can be categorized as upper and lower motor neurons
• primary motor cortex (M1) can map particular movements rather than muscle contractions
• 2 important tracts regarding motor control: corticobulbar and corticospinal tract
• BCI Systems usually comprises 5 elements: signal acquisition, feature extraction, feature translation, device output and Feedback
• Pohlmeyer et al. (2009) could show proof of concept that FES can be implemented with implanted electrode arrays as a BCI