Study 2
Question 2
Does augmented feedback ameliorate the effects of reduced somatosensory or visual feedback while wearing an unfamiliar prosthetic device?
Aim 2
Determine the neurobehavioral effects of modulation of visual and somatosensory feedback during a reach-and-grasp task.
Approach Rationale
Preventing visual tracking of a prosthetic device while in the transport phase of a reach and grasp task may increase cognitive load, and lead to variation in setting aperture size of the terminal device. Furthermore, the properties of movements are expected to change, including decreased harmonicity of component movements of the task, and decreased smoothness. Providing augmented sensory feedback may lessen these changes. Subjects will perform a reach and grasp task multiple times while electroencephalographic (EEG) and kinematic data are gathered. In one group, the subject's vision of the prosthesis and terminal device will be occluded by a barrier during the reach phase of the reach and grasp. The other group will perform the same task with full vision. Each subject will be supplied with vibrotactile feedback during either the first or last half of their trials. Changes in cortical activation assessed by EEG will indicate changes in neural activity, as well as changes in cognitive load. Variability in kinematics will indicate the effectiveness of adaptive strategies to loss of sensation and vision, and serve as a possible indicator for predicting fatigue and dissatisfaction with a prosthetic device. The efficacy of providing augmented feedback to elssen these effects will be assessed as well.
Hypothesis 2.1
Neural activity in somatosensory areas will be increased by reduced visual feedback and decreased by augmented somatosensory feedback.
Hypothesis 2.2
Decreased availability of visual or somatosensory feedback will decrease grasp aperture precision.
Hypothesis 2.3
Use of a tool rather than a prosthesis will result in smaller changes in neural activity in premotor areas than will the prosthesis, while occluded vision will result in increased neural activity in parietal areas.
Methods
Subjects
Eighteen (12 female) right-handed neurologically-healthy adults participated in the study. All subjects provided written, informed consent, and all methods were approved by the Georgia Institute of Technology Institutional Review Board. Subjects completed the Edinburgh Handedness Inventory to assess hand dominance \citep{Oldfield1971}. Subjects were pseudo-randomly assigned to one of two groups, vision or no vision. Within each group, subjects experienced two conditions, vibrotactile feedback, or no feedback. The order of these conditions was alternated between subjects to negate order effects.
Experimental Measures
Motion Capture
Reliable assessment of kinematics requires the ability to track objects in 3-dimensional space locked to a fourth dimension -- time. The trakSTAR\texttrademark{} system (Ascension Technology Corp., Shelburne, VT, USA) incorporates four electromagnetic sensors along with an electromagnetic transmitter that allows motion capture at 100\,Hz within a one meter sphreical envelope. Each sensor reports 6-degrees of information: $X,Y,Z$ as well as yaw, pitch, and roll. Motion Monitor software (Innsport, Chicago, IL, USA) was used for data collection. Analysis was conducted in MATL ABhe(T Mathworks, Inc., Natick, Mass. USA). Sensors were placed as follows: over the subject's right scapula ,on the fixed jaw of the prosthesis and on the moveable jaw of the prosthesis. Subjects who used tongs rather than the prosthesis were set up with a sensor over the right scapula, and one sensor on each jaw of the tongs.
Electroencephalography (EEG)
Populations of cortical neurons firingi n synchrony create dipoles whose electrical fields are measurable at the scalp. Participants will be fitted with a 58-channel EEG cap (Electrocap, Eaton, OH, USA). Neural activity is sampled at a rate of 1000\,Hz using a Syanmps 2 (Neuroscan, Charlotte, NC, USA) amplifier and Neuroscan software. Event-related spectral power (ERSP) will be analyzed for selected experiment conditions using EEGLAB \citep{Delorme2004} and custom MATLAB code.
Experimental Design
Subjects in the occluded group performed the paradigm while using a device for occluding vision (see \cref{fig:occluder}) during the reach-to-grasp and return phases of the paradigm (see \cref{fig:task}).
On their right arm, subjects donned a prosthetic limb (see \cref{fig:fams}) designed to be worn by intact subjects. Control of the terminal device was via a cable and loop around the left shoulder so that horizontal adduction of the shoulder resulted in opening the terminal device. An elastic band closed the terminal device when not held open by the subject. Subjects moved (in order) each of three discs (medium, small and large) from the right target area to the left target area, then back, pressing the home button before each movement (see \cref{fig:task}), and after the final movement. These six disc movements constituted one trial. Each trial was followed by a 20\,s rest. In either the first or last 15 trials, the vibrotactile feedback device (see \cref{fig:vffeedback}) was enabled. Fifteen trials were followed by a 3\,min rest during which the vibrotactile feedback was enabled or disabled, after which the remaining 15 trials were executed. Execution of the task was self-paced. If subjects asked about the pace they were told, ''a normal pace, as if you were doing a task in the kitchen.''
A second paradigm was created based on the first paradmig. In this experiment, subjects used a pair of tongs (see \cref{fig:tongs}) to carry out the same disc movement task. Subjects were placed into either vision or occluded groups. Vibrotactile feedback was not used in this paradigm due to the availability of proprioceptvie and tactile feedback from the subjects' hand while using the tongs.