Archive for December 2017
Team Robot
As technology improves and increases the likelihood of teams with humans and (semi-)autonomous artificial agents (e.g., virtual or robotic agents), studying potential agent capabilities becomes increasingly meaningful. Studies on organizational science focus on how members of teams communicate effectively. Team members must be able to understand each other’s goals, equipment necessary, and shared information about a given task. Shared mental models (SMMx) have been shown to efficiently expedite this information and allow team members to track each others’ progress.
Matthias Scheutz, Scott DeLoach and Julie A. Adam propose the first formal and computational frameworks for shared organizational mental models for human-robot (H-R) teams by monitoring team members physiological responses. The researchers broke down shared mental models between two key elements; data representations, capturing information and sharing between team members, and computational process, how data representations are shared and maintained. Data representations were broken down into five key component areas; agent capabilities, agent and task states, obligations, activity and equipment types, and functional agent roles. These were all given assigned algorithms to create a formal mathematical model to show how the data representation is maintained. Human Performance Factors (HPF) were also established to potentially provide a means for predicting human behavior and how their performances are affected by various internal, external, organizational and task factors. With the formal framework established, Scheutz, et al. established a computational framework for recording physiological measures to provide what they call “Workload Channel Estimation” that calculates estimates for workloads.
Researchers propose using BIOPAC’s wearable BioHarness monitor that would provide wireless physiological feedback to auditory, tactile, or motor stimuli. These measures would then inform the workload estimation and provide data for the computational framework on how physiological factors maintain the shared mental model. Scheutz, et al’s frameworks can provide new insight into what organizational strategies are most effective in communicating task information and potentially provide a measurement for team members’ most effective workload.
Matthias Scheutz, Scott DeLoach and Julie A. Adam propose the first formal and computational frameworks for shared organizational mental models for human-robot (H-R) teams by monitoring team members physiological responses. The researchers broke down shared mental models between two key elements; data representations, capturing information and sharing between team members, and computational process, how data representations are shared and maintained. Data representations were broken down into five key component areas; agent capabilities, agent and task states, obligations, activity and equipment types, and functional agent roles. These were all given assigned algorithms to create a formal mathematical model to show how the data representation is maintained. Human Performance Factors (HPF) were also established to potentially provide a means for predicting human behavior and how their performances are affected by various internal, external, organizational and task factors. With the formal framework established, Scheutz, et al. established a computational framework for recording physiological measures to provide what they call “Workload Channel Estimation” that calculates estimates for workloads.
Researchers propose using BIOPAC’s wearable BioHarness monitor that would provide wireless physiological feedback to auditory, tactile, or motor stimuli. These measures would then inform the workload estimation and provide data for the computational framework on how physiological factors maintain the shared mental model. Scheutz, et al’s frameworks can provide new insight into what organizational strategies are most effective in communicating task information and potentially provide a measurement for team members’ most effective workload.
Mobita Wireless EEG Takes a Glimpse Into the Future-Monitoring Prestimulus EEG Data
New studies examining physiological causality and electroencephalographic (EEG) correlations have provided insight on how the wearable and wireless Mobita Amplifier from BIOPAC is making an impact. A research team at the University of California, Santa Barbara, working with the American Institute of Physics, gathered data consistent with findings to conclude the psychological existence of retrocausality. The group focused on light and sound stimuli identification rates for frontal, central, and occipital parts of the brain.
The test data was divided into a pre-stimulus identification rate as well as a post-stimulus rate, to examine any correlation between standard physics and their hypothesis. No information about the stimulus should be expected to exist prior to stimulus selection. EEG recordings preceding each stimulus are analyzed to find correlations with the future selected stimulus.
During the case study, experimental subjects were fitted with a wireless EEG head cap (BIOPAC MB-32EEG-CAP-A) with 32 electrodes; the wearable BIOPAC Mobita Amplifier was utilized for its capacity to record 32 channels of high fidelity wireless EEG data. The head cap’s electrodes used paper-cotton strips soaked in water to make electrical connection with the scalp. Accompanied by BIOPAC’s AcqKnowledge research software, researchers implemented a modern and efficient process while exploring the validity of their test results and research findings.
The campus research group sought to measure electro cortical evoked potentials in the general population using random subjects with hopes of creating a basic means of measurement. Furthermore, “real-time analysis of EEG data may allow quantum events to be predicted in advance, which would affect interpretations of quantum mechanics and our notions of causality” (Baumgart et. al.). While conducting their experiment, the research group focused on keeping a furtive introduction of stimuli to test subjects equipped with the Mobita Amplifier. Stimuli were then chosen using a quantum random number generator (qRNG) and introduced to non-selected test samples.
Time durations for both light and sound stimuli vary in lengths to provide a way to differentiate the stimuli in the digital channel recording. Since sound stimuli contain longer durations than light, stimuli were recognized by digital signal length as well as being recorded by the stimulus control program. To quantify the interval of sound stimuli, the digital conduit also evaluated the voltage in parallel across the buzzer circuit to better monitor the reaction to sound stimuli. Furthermore, identification of stimulus type based on post-stimulus reaction was supported by the group’s current data for the frontal and central regions but not the occipital region.
The yielded results have lead to planning of future tests and have implicated that the existence of retrocausality is in fact measurable.