The human nervous system has two classes of cells: neurons and glia. From the research to date, it is believed that signals within the network of neurons constitute the whole of information processing that results in behavior, while the role of glial cells is to provide physiological support to the neurons. This neural doctrine dominates research in direct monitoring technologies.
Neurons consist of four parts: axon, dendrites, cell body or soma, and presynaptic terminals (see Figure D-1). Electrical information is transmitted to the neuron through the dendrites, proceeds through the cell body, and leaves the cell through the axon at one or more presynaptic terminals. Neurons have one axon and from one to tens of thousands of dendrites. Details of how the action potentials (the electrical signals) travel through the cell or are transmitted across the synapses can be influenced by changes in biochemistry, which may in turn be influenced by either environmental changes or the presence of external (pharmacologic) substances.
Direct Neural Signals
In these bio-electric networks, ions of sodium, potassium, and chlorine move through the cell membranes perpendicular to the propagation of the action potential down the axon. This electrical signaling allows information to be transmitted faster than ions could flow down the axon. The propagation of information is similar to the wave traveling down a length of a rope
FIGURE D-1 A typical vertebrate neuron. “The arrows indicate the direction in which signals are conveyed. The single axon conducts signals away from the cell body, while the multiple dendrites receive signals from the axons of other neurons. The nerve terminals end on the dendrites or cell body of other neurons or on other cell types, such as muscle or gland cells” (Alberts et al., 2002, p. 638). The signal of principal interest for monitoring the electrical activity of a neuron is the axonal firing (travel of action potential from the body to the axon terminals).
SOURCE: Alberts, B. (2002). Molecular Biology of the Cell. New York: Garland Science. Reproduced by permission conveyed through Copyright Clearance Center.
when one end of the rope is moved from side to side quickly with sufficient force. Although the wave travels to the other end of the rope, any part of the rope structure has only moved (nominally) perpendicular to the direction of wave propagation. In a similar fashion, ions flow through channels across the axon’s cell membrane, changing the local membrane potential and thus propagating the electrical signal down the axon.
The signal transmission down the axon of a neuron is an all-or-nothing process. When the cell body is stimulated above its threshold level, the axon transmits the same action potential at the same speed and in the same direction, regardless of the extent above the threshold or the duration of the stimulus.
Action potentials have durations of 1-10 msec. Input signals can result in transmission of multiple action potentials, and thus the frequency and number of neuronal firings do vary with the input. Neurons require some time to reset between firings, which nominally is the duration of the pulse for that axon. A typical maximum firing rate is between 100 Hz and 1
kHz.1 The duration of the action potential and the speed of conduction are properties of the axon diameter and whether the axon is myelinated.
The human brain does not process information as a traditional digital computer does. Information is moved around through pathways, and at certain neurons it is allowed or not allowed to pass down that neuron based on excitory or inhibitory dendritic signals arriving before the triggering of action potentials in that neuron. Local groups of neurons can act nearly coherently, as for example in volition of motor action like a hand movement. Detecting such coherent firing at nodes around the brain is robust both noninvasively and invasively, though noninvasive techniques currently cannot resolve firing sequences of individual neurons within such groups. For example, a surface electroencephalography (EEG) signal requires the coherent firing of tens of thousands of neurons, while electrical detection of a single neuronal firing requires that a measurement probe be placed proximal to the neuron of interest, such that the probe is closer to that neuron than to any adjacent neuron. Obviously this requires opening or mechanical penetration of the skull, and that is outside the parameters of application for widespread assessment. Thus, this limitation of noninvasiveness precludes measurement (detection) of individual neuron firings.
Living neurons in an active tissue are always active at a minimal level, firing even in “resting state.” Changes in the frequency of firing imply that a given neuron is currently involved in the processing of information. Bulk changes in local field-potential oscillations imply that several neurons are active. This is the baseline signal seen in the noninvasive direct measurement techniques of EEG and magnetoencephalography (MEG).
Resting state brain activity is an area of current basic research. Global patterns of activation recorded during these “baseline signal” conditions exhibit coordinated behavior when measured with functional magnetic resonance imaging (fMRI; see below and Barkhof et al., 2014, for details on fMRI). Pathologies or individual traits could eventually be indicated by modified connectivity patterns.
Purposeful brain activity leads to activation patterns different from those of the resting state. Movement as simple as an eye blink involves signal communication through a million neurons. Detecting single firings of individual neurons is a difficult process because the signals are weak to start with and are not isolated from the rest of the electrical activity within the brain. Large groups of coherent neurons, perhaps a few thousand to
1 A hertz is one (firing) cycle per second, so a 100 Hz maximum firing rate, for example, would mean the neuron can fire up to 100 times per second. A firing rate of 1 kHz would mean 1,000 action potentials per second.
tens of thousands all firing at once in relation to an external event, are the most studied of single firing signals.2
The preceding discussion is a greatly simplified version of the electrical dynamics of neuronal firing. For instance, it does not include differences between axon and dendrite signals or the transmission of signals across a synapse. Complete discussions of underlying electrical signals in the nervous system are provided by Huettel and colleagues (2004) and Kandel and colleagues (2000).
Indirect Neuronal Signals—Energy Use
The brain activity mentioned above is a complex chain of ionic motion within the central nervous system. Ion movement within and between cells as ion channels are activated consumes energy. Replenishing the energy supply in brain cells requires the conversion of blood-borne oxyhemoglobin to deoxyhemoglobin. The rate of oxygen consumption in a localized volume varies based on local neural activity. The circulatory system compensates for changes in energy demand by increasing or decreasing both the flow rate and volume of blood, regionally and locally. Local energy demand, expressed in the capillary beds, will alter the rate at which oxygen is metabolized, called the cerebral rate of oxygen metabolism, which is abbreviated as CMRO2. When brain activity increases in a region, the circulatory response, called the hemodynamic response, will be increases in flow and volume, while the local areas increase CMRO2.
The hemodynamic response consistently provides an excess of oxygen over what is required, and this results in some oxyhemoglobin traveling through the capillary bed and local venous structure without being converted into deoxyhemoglobin. Oxyhemoglobin and deoxyhemoglobin have different magnetic susceptibilities and different infrared spectra. The hemodynamic response, by changing the net ratio of oxyhemoglobin to deoxyhemoglobin in the local venous structure, thus changes the local magnetic susceptibility and local infrared resonance spectra around focused brain activity. This complex chain reaction is called the Blood Oxygen Level Dependent (BOLD) effect (Ogawa et al., 1992). The BOLD effect leads to a method to indirectly measure local brain activity by monitoring the hemodynamic response using magnetic resonance imaging (MRI) or near-infrared spectroscopy (NIRS).
The BOLD response is a marker of the energy used locally by the coherent firing of large numbers of neurons. The BOLD response to any event peaks about 4-6 seconds after the event occurs, limiting the applications
2 “Single firing signal” here means a single peak of combined electrical activity relative to an event. This is not necessarily the same as single firings of each neuron contributing to the peak.
for which monitoring these signals and their associated delay may be useful. Furthermore, person-to-person variation in distributed signals shows significant differences in regions activated (Hancock and Szalma, 2008), although there is evidence that these intersubject variations are stable over time for the same subjects (Miller et al., 2002). For use in a selection process, the brain activity signal via BOLD fMRI in an individual would need to be proven to be robust and reliable with respect to the range of environmental conditions (e.g., variations in room temperature) typically encountered during assessment.
There are four noninvasive measurement technologies currently in widespread use in monitoring brain activity. The two direct-measure technologies are EEG, which detects mainly surface currents from relative voltage changes at or just below the scalp, and MEG, which detects near-scalp magnetic fields associated with neural pathway current throughout the brain, but mainly near-surface parallel and perpendicular current flow. The remaining two technologies are indirect measures that monitor the BOLD response either through rapid successions of whole brain MRI scans (using fMRI) or with NIRS.3 The purpose of this section is to explain the capabilities and limitations of currently available technology, thereby demonstrating the feasibility of near-term possibilities to apply neuroscience in enlistment accessions as well as advances necessary for neuroscience to contribute to accurate, efficient, and mass-administrable assessments.
The primary technology used for modeling the electrical activity of the brain is also the oldest. EEG was first described in 1929 (Berger, 1929) and now exists in several derivative forms. Traditional EEG uses electrodes at the surface of the scalp to measure and amplify differences in electrical potential between points above the cortical surface and a fixed reference, such as the average reading from the ear lobes. Neuronal activity is fundamentally ionic motion in solution. Firing neurons produce the primary current, while induced charged-particle motion outside of the neuron is lumped together as volume currents. A noninvasive technology can only measure
3 NIRS is occasionally referred to as functional NIRS or fNIRS. NIRS using multiple sources to produce three-dimensional images of internal changes in blood flow is occasionally called diffuse optical tomography. However, “diffuse optical tomography” is a more general technological term that can also refer to methods such as using visible-light laser excitation of tissue and very high resolution imaging of internal blood vessels (from inside the vessel).
the net effect of primary plus volume currents at the surface of the scalp. In EEG, orientation of the primary currents is not detectable.
Traditional EEG data are analyzed by breaking up the spectrum of combined frequencies into several bands between 0.5 and 100 Hz. A derivative form of EEG developed in the late 1930s is called evoked potentials, or EP. In EP, scalp data is averaged over several electrodes time-locked to a stimulus (Davis, 1939). Similar to EP are event-related potentials, which are measured in a similar fashion but not averaged like EP signals. Both of these methods record summed electrical activity of nominally 50,000 local neurons. Thus, large coherent group spiking activity4 is required to produce appreciable signal.
Current EEG technologies are fast enough to capture signals of interest, making it a viable measure for research on performance as well as for use in direct selections. For example, if future technology such as phased array high impedance antennae makes localization of multiple person unobtrusive EEG recording possible, the measurements are unlikely to be any more precise than the current capabilities of an EEG via scalp electrodes. Therefore, conducting research on ASVAB/TAPAS5 test takers using currently available EEG capabilities would indicate whether investments to develop technology for unobtrusive mass administration should be expected to yield a capability for performance assessment or direct selections.
MEG experiments rely on detection of extremely small magnetic fields produced by the time-varying neuronal currents in brain activity. Some direction information is available from MEG recordings—mainly separating primary currents flowing perpendicular to the scalp from current flowing parallel to the surface.
The signal strengths are measured in hundreds of femtotesla.6 Typical signals are about 100 million times weaker than Earth’s static magnetic field, so measurements are carried out in well-isolated chambers. Only superconducting quantum interference device magnetometers (SQUIDs) can detect such signals, and these devices and therefore the sensory apparatus requires liquid helium cooling. Hence, MEG recordings are not envisioned to be practical outside of the laboratory in the near future. However, the
4Spiking activity is the term used for recognition of action potentials. Spikes are fast and easy to recognize with electronic triggering circuits, while more complex waveforms require additional processing.
5 The ASVAB is the Armed Services Vocational Aptitude Battery; the TAPAS is the Tailored Adaptive Personality Assessment System. Both are discussed in Chapter 1 of this report.
6 A femtotesla (fT) is 10–15 tesla. The tesla (T) is the metric unit of magnetic flux density, equal to one weber of magnetic flux per meter squared.
field frequencies are defined, and theoretically, with future technology to detect femtotesla-scale fields and provide a shield from magnetic-flux noise from the environment, MEG recordings could be possible in assessment settings.
The main promise of MEG, whether in the laboratory for use in basic research or in real-world assessments, is its high temporal resolution and good spatial resolution, especially when combined with EEG information. Multimodal temporal resolution on the order of milliseconds can be combined with a spatial resolution of millimeters or even finer. Of course, detailed methods for combining EEG and MEG measurement are a major challenge; in current research, the acquisitions and analyses are done separately. Analyses are accomplished using either traditional approaches of frequency power analysis or by locking an average signal to the onset of an event cue to search for an event-related localized activity peak (similar to the event-related potentials method used for EEG-only data).
MRI and fMRI
MRI works by a simple excitation and relaxation of the spin state of protons in the nuclei of hydrogen atoms. When molecules containing hydrogen are placed in a strong static magnetic field, a small but detectable number of hydrogen protons align their intrinsic spins along the direction of this external field. An applied radio-frequency (RF) pulse near the resonant frequency of hydrogen protons, 42.6 MHz/tesla or 128 MHz at 3 tesla, knocks the spins perpendicular to the external field, and their relaxation back to ground state releases RF energy in patterns that can be reconstructed to show both the composition and distribution of any hydrogen-rich material. The resonant frequency is a direct function of the local magnetic field, defined by the Larmor relation: ω = γ B, where ω is the frequency of precession, B is the local magnetic field, and γ is a constant of the material (42.6 MHz/tesla for bare protons, as mentioned above).
Small perturbations to the static field will change the resonant frequency. By applying a small gradient to the static field—for example, 20 millitesla/meter along the z-axis—and limiting the bandwidth of the RF excitation signal to dw, one may select a slice of the brain perpendicular to the z-axis for excitation to dz. A change in the gradient field will change the position of the excited slice for the next excitation. Similar gradients in the x- and y-directions can limit the excitation to a single small volume of brain tissue. In current MRI instruments, these gradient fields are produced with electromagnets and the series of time-dependent imaging gradient manipulations is called the scan sequence.
A free hydrogen atom (H) would produce a resonant signal slightly
different from the signal from a bare proton, due to the local field changes induced by its valence electron. Hydrogen gas (H2) would produce a still different frequency since the local field around each proton is altered by the two shared electrons. Water molecules (H2O) contain two hydrogen atoms and an entirely different “electron shield” than either H or H2 and thus possess another, slightly different, resonance frequency. Fats and other lipid molecules, which are important cell-structure building blocks, have long chains of hydrocarbons, and the resulting ensemble of electron screening produces a wide peak that is substantially shifted in frequency from that of water. 7
Brain gray matter and white matter have different macroscopic lipid content and can thus be differentiated in an MRI scan. Different signals also arise in bone, cerebral-spinal fluid, and internal tissue structures of various other organs. Unlike x-ray based technologies, MRI scans can be optimized to contrast any of the many aspects of the physical signal, such as total density of protons, water content, lipid content, and even particle motion in advanced techniques involving diffusion or spin labeling. Using such scan sequences, which take several minutes, one can construct very high resolution images of gray and white matter structure for comparison with, and also mapping onto, a “standard brain” template to detect individual differences. This is important for the assessment of performance potential because different structure sizes have been linked to different behaviors and abilities. For example, larger hippocampal volume has been related to visuospatial memory capability in large-city taxi drivers (Maguire et al., 2006) and reduced medial prefrontal cortex volume has been related to schizophrenia (Mathew et al., 2014). Furthermore, models are under development to explain these differences in brain structure, but for the purposes of this study, the correlations between structure and behavior could be important for selection.
A series of fast scan sequences, typically collecting an entire brain volume at a resolution of 3 mm3 in 2 seconds, that are calibrated to optimize detection of the BOLD signal will show the dynamics of brain functioning under the specific internal or applied conditions at the time of scan; this is known as a functional MRI, or simply fMRI.8 The major advantages of fMRI are unmatched three-dimensional spatial resolution compared
7 Frequency detection sensitivities in MRI are very good, and “substantial” here means about 3 parts per million. The frequency shifts caused by imaging gradients range in the parts per thousand.
8 Specifically this is T2* Echo-Planar Imaging, also called BOLD EPI, Gradient Echo EPI, or BOLD fMRI. This approach is used in well over 90 percent of published functional studies, although there are more advanced techniques that concentrate on smaller portions of the hemodynamic signal. For example, Spin-Echo EPI will provide a higher localization within the gray matter but at the cost of a loss of 90 percent of the signal amplitude.
to other noninvasive imaging methods and complete skull penetration, making it the only imaging modality to unambiguously detect limbic activations important for determining emotionally laden neuropsychological states.
A long-term prospect, likely in the 20-40 year timeframe, is that combined low-field MRI and MEG technology could detect neuronal firing deep in the brain and with high temporal accuracy. Initial experiments indicate some level of feasibility, but there is substantial development work required in room-temperature, low-field magnetic field detection devices, such as atomic magnetometers, and in signal processing algorithms to sift through the substantial electromagnetic background (Kraus et al., 2008; McDermott et al., 2004). It is thought that such future devices, as well as those that might alter the atomic nuclei observed by MRI to nuclei of sodium, calcium, potassium or another element with a nonzero magnetic moment, could be operated by minimally trained technicians, the way Army medics are trained to operate medical imaging equipment for limited applications, or research assistants are trained to acquire EEG data from subjects in a sleep center. Future uses of MRI and fMRI include measurement of the Big Five personality traits and other meta-traits, which could expand upon current research to assess dual and multiple task performance with fMRI. If this research identifies brain patterns indicative of performance capability, then such tests and responses could subsequently be utilized in assessment processes.
NIRS is an additional technology to monitor the BOLD effect noninvasively. The NIRS signal correlates with localized γ activity. (EEG measurements typically divide neuronal firing frequencies into spectra, and the relative power in five bands—0-4 Hz [Δ band], 4-8 Hz [θ band], 8-13 Hz [α band], 13-30 Hz [β band], and 30-100 Hz [γ band]—are calculated. Localized γ activity refers to an increased signal in the EEG γ band.) Studies have shown that NIRS correlates well with the fMRI signal in animal models, although with reduced coverage and lower resolution (Chen et al., 2003). This lowered resolution greatly affects the reproducibility of the technique. A recent study involving reading a preference decision from a subject in single trials only attained 80 percent accuracy (Luu and Chau, 2009). NIRS measures BOLD responses near the surface, so anything that fMRI can measure that occurs in the frontal cortex or other near-scalp regions can be detected currently using NIRS for less expense than is associated with fMRI tests.
Transcranial Doppler Sonography
Transcranial Doppler sonography measures increased blood flow through carbon dioxide–induced vasodilatation. The level of carbon dioxide, which is a byproduct of localized increased metabolism and also an indirect measure of neural activity, has been shown to correlate with the level of a subject’s vigilance (Warm et al., 2008).
Measurements involving eye fixations, dwell time (temporal length of a fixation), and pupillary changes are well-established metrics of workload in visual searching tasks (Backs and Walrath, 1992). Additional measures of ocular changes include blink rate, blink duration, blink latency, and eye movement. Moreover, fixations include small high frequency variation in the eye position that are modified by attention (Steinman et al., 1973). These are recorded using one of various types of either eye-tracking devices or electrodes to measure an electrooculogram. Eye-tracking data include the position of a fixation and the time of each eye movement (or saccade), whereas an electrooculogram only identifies the time that the muscle controlling eye blinks or eye position was activated.
Spontaneous eye blink rate (SBR) is correlated with dopamine activity in the brain (Blin et al., 1990; Dreisbach et al., 2005) and can therefore be used as an indirect objective measure of stress variance. SBR is an ideal biomarker for stress, as changes in dopamine activity can be indirectly tracked by a video recording of the individual’s eyes. Advanced image analysis can perform facial recognition based on naturalistic video captures, and automated eye monitoring can calculate SBR (Jiang et al., 2013). Therefore, it is likely that a robust technique can be developed to determine SBR from naturalistic video recording.
The main human glucocorticoid to be monitored is cortisol. It can be measured in blood, saliva, or urine samples (McWhinney et al., 2010).
Alberts, B., A. Johnson, J. Lewis, M. Raff, K. Roberts, and P. Walter. (2002). Molecular Biology of the Cell. (4th ed.) New York: Garland Science.
Backs, R.W., and L.C. Walrath. (1992). Eye movement and pupillary response indices of mental workload during visual search of symbolic displays. Applied Ergonomics, 23(4): 243–254.
Barkhof, F., S. Haller, and S.A. Rombouts. (2014). Resting-state functional MR imaging: A new window to the brain. Radiology, 272(1):29–49.
Berger, H. (1929). Über das elektrenkephalogramm des menschen. Archiv fur Psychiatrie und Nervenkrankheiten, 87(1):527–580.
Blin, O., G. Masson, J.P. Azulay, J. Fondarai, and G. Serratrice. (1990). Apomorphine-induced blinking and yawning in healthy volunteers. British Journal of Clinical Pharmacology, 30(5):769–773.
Chen, Y., X. Intes, D.R. Tailor, R.R. Regatte, H. Ma, V. Ntziachristos, J.S. Leigh, R. Reddy, and B. Chance. (2003). Probing rat brain oxygenation with near-infrared spectroscopy (NIRS) and magnetic resonance imaging (MRI). Advances in Experimental Medicine and Biology, 510:199–204.
Davis, P. (1939). Effects of acoustic stimuli on the waking human brain. Journal of Neurophysiology, 2:494–499.
Dreisbach, G., J. Muller, T. Goschke, A. Strobel, K. Schulze, K.P. Lesch, and B. Brocke. (2005). Dopamine and cognitive control: The influence of spontaneous eyeblink rate and dopamine gene polymorphisms on perseveration and distractibility. Behavioral Neuroscience, 119(2):483–490.
Hancock, P.A., and J.L. Szalma. (2008). Performance Under Stress. Aldershot, England; Burlington, VT: Ashgate.
Huettel, S.A., A.W. Song, and G. McCarthy. (2004). Functional Magnetic Resonance Imaging. Sunderland, MA: Sinauer Associates.
Jiang, X., G. Tien, D. Huang, B. Zheng, and M.S. Atkins. (2013). Capturing and evaluating blinks from video-based eyetrackers. Behavior Research Methods, 45(3):656–663.
Kandel, E.R., J.H. Schwartz, and T.M. Jessell. (2000). Principles of Neural Science (4th ed.). New York: McGraw-Hill, Health Professions Division.
Kraus, R.H., Jr., P. Volegov, A. Matlachov, and M. Espy. (2008). Toward direct neural current imaging by resonant mechanisms at ultra-low field. Neuroimage, 39(1):310–317.
Luu, S., and T. Chau. (2009). Decoding subjective preference from single-trial near-infrared spectroscopy signals. Journal of Neural Engineering, 6(1): Epub 016003.
Maguire, E.A., K. Wollett, and H.J. Spier. (2006). London taxi drivers and bus drivers: A structural MRI and neuropsychological analysis. Hippocampus, 16:1,091–1,101
Mathew, I., T.M. Gardin, N. Tandon, S. Eack, A.N. Francis, L.J. Seidman, B. Clementz, G.D. Pearlson, J.A. Sweeney, C.A. Tamminga, and M.S. Keshavan. (2014). Medial temporal lobe structures and hippocampal subfields in psychotic disorders: Findings from the bipolar-schizophrenia network on intermediate phenotypes (B-SNIP) study. Journal of the American Medical Association—Psychiatry, 71(7):769–777.
McDermott, R., S. Lee, B. ten Haken, A.H. Trabesinger, A. Pines, and J. Clarke. (2004). Microtesla MRI with a superconducting quantum interference device. Proceedings of the National Academy of Sciences of the United States of America, 101(21):7,857–7,861.
McWhinney, B.C., S.E. Briscoe, J.P. Ungerer, and C.J. Pretorius. (2010). Measurement of cortisol, cortisone, prednisolone, dexamethasone and 11-deoxycortisol with ultra high performance liquid chromatography-tandem mass spectrometry: Application for plasma, plasma ultrafiltrate, urine and saliva in a routine laboratory. Journal of Chromatography B, 878(28):2,863–2,869.
Miller, M.B., J.D. Van Horn, G.L. Wolford, T.C. Handy, M. Valsangkar-Smyth, S. Inati, S. Grafton, and M.S. Gazzaniga. (2002). Extensive individual differences in brain activations associated with episodic retrieval are reliable over time. Journal of Cognitive Neuroscience, 14(8):1,200–1,214.
Ogawa, S., D.W. Tank, R. Menon, J.M. Ellermann, S.G. Kim, H. Merkle, and K. Ugurbil. (1992). Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging. Proceedings of the National Academy of Sciences of the United States of America, 89(13):5,951–5,955.
Steinman, R.M., G.M. Haddad, A.A. Skavenski, and D. Wyman. (1973). Miniature eye movement. Science, 181(4,102):810–819.
Warm, J.S., R. Parasuraman, and G. Matthews. (2008). Vigilance requires hard mental work and is stressful. Human Factors, 50(3):433–441.