Sensory testing as a predictor of short vs. long term trajectory of recovery from concussion

Abstract

There have been numerous reports of neurological assessments of post-concussed athletes.  However, the majority of the methods commonly deployed are either qualitative assessments that are simply symptom based or are psycho-social questionnaires.  The information provided from those studies does not provide insight into the neural mechanisms impacted by concussion, and more importantly, does not contribute to a prognostic view of overall brain health that would facilitate or predict the recovery of the concussed individual. Cortical metrics are measures that were designed to probe brain function via the somatosensory system (i.e., with high fidelity tactile inputs) and have been demonstrated to be both objective, quantifiable and physiologically based.  The methods have also been recently reported to parallel findings in a neurophysiological animal model of brain injury (Challener et al, 2020) that support the concept that these metrics parallel alterations in specific neural mechanisms post-injury.  In this report, the battery of tactile based measures are reaction time (RT), reaction time variability (RTvar), sequential and simultaneous amplitude discrimination, temporal order judgement (TOJ) and duration discrimination (DD).  These methods  are administered with a computer mouse sized tactile stimulator (the Brain Gauge) that delivers sinusoidal stimuli to digits 2 and 3 with precision control of both amplitude and frequency. The results obtained during the first week of observation post-injury predict the recovery trajectory of the concussed individual.  Interestingly, some of the metrics of the individuals who take longer to recover from injury obtained during the first week outperform the metrics of individuals that recover quickly, and these findings parallel the findings from the animal model.

Introduction

There have been an overwhelming number of studies of concussion or mild traumatic brain injury (mTBI) over the past several decades. However, few, if any, of these studies have successfully described methods for assessing the brain health of concussed individuals that successfully predict recovery trajectory. One reason for this is that the majority of widely deployed methods are not physiologically based and do not address the neural mechanisms of information processing that are impacted by traumatic insult. For example, balance testing is widely used in sports-concussion studies (despite never demonstrating significant accuracy for assessing concussion) only addresses a symptom of concussion and does not target information processing in the central nervous system (CNS). Online questionnaires that test cognitive functions such as memory and learning do not address base level neural mechanisms that contribute to those cognitive functions that are difficult to objectively quantify. Cortical metrics – methods designed to probe CNS functional mechanism via the somatosensory system - on the other hand, target changes in specific neural mechanisms, and for this reason, are sensitive to physiological alterations [1-3](Challener et al, 2020). If neural mechanisms of information processing are altered in a manner that predicts trajectory of recovery to concussion, then hypothetically, the cortical metrics influenced by those mechanisms would also be predictive of the trajectory of recovery. Based on a recent report by Challener et al, 2020, cortical metrics parallel these neural mechanisms.

A significant issue with concussion, or mTBI, is determination of return-to-duty status for the military or return to-play status for athletes at multiple levels of competition (secondary school, college/university, and professional level). Because injury from secondary concussions can be much more serious, if not fatal, during the critical post-concussion recovery period, it is imperative that methods for this determination be developed. In this report, we describe a method for utilization of the somatosensory system to probe the CNS to ascertain its potential as a biomarker for post-concussion assessments. In particular, we analyze the data of concussed individuals to determine if a recovery trajectory can be predicted. In other words, can the cortical metrics of individuals that recover relatively quickly from concussion (1-2 weeks) be differentiated from the cortical metrics of individuals that take longer to recover? In this report, we examine differences between individuals who recover relatively quickly from concussion with individuals who take much longer to recover.

Methods

Subject Enrollment.

Data were collected from 380 healthy subjects (278 males, 102 females, mean age 19.6 years, SD 0.75 years) who were participating in a university athletic program. A survey about medication and medical history was filled out by each subject before experimental tests to exclude subjects with a history of neurological impairment. All participating subjects were naïve of the study design and issue under investigation. The study was performed in accordance with the Declaration of Helsinki, all subjects gave their written informed consent, and the experimental procedures were reviewed and approved in advance by an institutional review board.

Of the 380 enrolled subjects, 246 were diagnosed with mTBI during the duration of the study. All athletes were diagnosed with mild traumatic brain injury (mTBI) in the form of a concussion by a certified athletic trainer and the team physician with the help of the Sport Concussion Assessment Tool 2 (SCAT-2) and had no prior history of concussion or any other diagnosed medical conditions.

Testing was completed for healthy subjects prior to the beginning of the collegiate athletic season. Testing of subjects who were diagnosed with concussions varied based on the schedule of the athlete and the team physician but all were tested within the first 7 days after the initial insult.

Sensory Assessment.

A two-point vibro-tactile stimulator, pictured in Figure 1 (the Brain Gauge by Cortical Metrics, Carrboro, NC), was used to deliver stimuli to the tips of digits 2 and 3. All of the protocols used were originally conducted on a four-site mechanical stimulator (CM4; Cortical Metrics Model #4) that is functionally identical for 2 digits and was previously described in [4], and was utilized to assess multiple sensory information processing characteristics in a number of subject populations [5-13](Francisco et al., 2008, 2012a). The prominent feature of these protocols, which have demonstrated significant sensitivity to alterations in CNS processing, is that they are independent of detection thresholds or skin sensitivity [13,14].

Figure 1.Figure 1

During the evaluation session, subjects were seated comfortably in a chair with their hand positioned on the stimulator. The independent, computer-controlled probe tips can deliver a wide range of sinusoidal vibrotactile stimulation of varying amplitudes and frequencies. In this study, vibrotactile flutter stimulation (25 or 40Hz) was delivered via 5mm Delrin probes to the glabrous tips of either, or both, the second (index, D2) and/or the third (middle, D3) digits of the left hand. These digits were chosen as test sites for convenience and comfort and also because of the wealth of neurophysiological data that supports the evaluation of the associated somatotopic regions in the non-human primate cerebral cortex. The right hand was placed on a two-button response device and subjects were instructed to press the left or right button when the correct stimulus was perceived on the middle or index finger, respectively.

A computer monitor provided visual cueing during each of the experimental runs. The cues indicated when the experimental stimuli would be delivered and when subjects were to respond. Training trials conducted prior to each task familiarized subjects with the test; correct responses on three consecutive training trials were required before the start of each assessment. Subjects were not provided with performance feedback or knowledge of the results during data acquisition. Stimulus parameters were specified interactively by test algorithms based on specific protocols and the responses of the subjects during those protocols. Delivering these tactile based neurosensory tasks as a set of tasks, or battery of tests, has been previously documented in multiple reports [1,3,15-18].

A series of sensory perceptual measures were employed to assess tactile information processing ability. In sum, these tests lasted approximately thirty minutes and consisted of evaluations of reaction time (RT), amplitude discrimination (AD), temporal order judgment (TOJ), and duration discrimination (DD). The individual tests, all of which are described in previous reports [5-13]( Francisco et al., 2008, 2012a), are described below.

Reaction Time.

For the simple reaction time (RT) task, a single tap (300µm, 40ms) was delivered to D2 and subjects were instructed to respond by clicking the response device as soon as the tap was perceived. A randomized delay ranging from 2 to 7s separated the trials. Response times were recorded for each of the 10 trials. This method was first reported in Zhang et al., 2011b [13] and more recently reported in Pearce et al, 2019 [18] and Favorov et al, 2019 [3]. The standard deviation of the results from the 10 trials of reaction time testing was used as a measure of reaction time variability. Timeline with randomized inter-trial interval is displayed in Figure 2.

Figure 2. Reaction Time (RT) task Subject responds as soon as they perceive the stimulus.

Amplitude Discrimination.

Amplitude discriminative capacity is defined as the minimal difference in amplitudes of two mechanical sinusoidal vibratory stimuli for which an individual can successfully identify the stimulus of larger magnitude. For the amplitude discrimination (AD) task, the device delivered simultaneous sinusoidal vibrotactile stimuli (initial stimulus parameters: 400µm peak-to-peak amplitude test, 200µm standard, 25Hz, 500ms, 20µm step size) to D2 and D3 over 20 trials. Discrimination capacity was assessed using a 2-alternative forced choice (2AFC) tracking protocol that has been described and implemented in a number of previous studies [5-13]( Francisco et al., 2008, 2012a). The magnitude of the test stimulus was always greater than that of the standard stimulus, but the loci of the stimuli were randomly varied on a trial-by-trial basis. Subjects were questioned as to which of the two digits received the higher magnitude stimulus. The difference between the amplitudes of the test and standard amplitudes was adjusted on the basis of the response such that correct responses decreased, while incorrect responses increased, the test amplitude on subsequent trials. Three conditions of amplitude discrimination were tested. In the first condition, stimuli were delivered sequentially, and the subject felt one stimulus at a time Figure 3.

Figure 3.Sequential Amplitude Discrimination taskSubject selects which of two stimuli (test vs. standard) are larger.

The second condition of amplitude discrimination delivered the two stimuli simultaneously (i.e., the subject received both stimuli at the same time). This condition is displayed in Figure 4.

Figure 4. Simultaneous amplitude discrimination task Subject selects which of two stimuli (test vs. standard) are larger.

A third condition of amplitude discrimination was obtained after an illusory conditioning task and is most often referenced as “single site adaptation”. The conditioning stimulus is delivered to the digit site prior to it receiving the test stimulus of the amplitude discrimination task. The theoretical impetus for this is that the conditioning stimulus will reduce the perception of the magnitude of the test stimulus (which is always larger than the standard) and make it more difficult for the subject to differentiate between test and standard. The conditioning stimulus has been found to have significant impact on healthy individuals ability to discriminate between the amplitudes of two stimuli, but little or no impact on an individual’s ability for amplitude discrimination if they have compromised neurological status [1,8]. Additionally, the measure has been correlated with GABA levels via MRS [19]. First described in Tannan et al [20], these methods have been described in a significant number of reports [3,18,21,22].

Figure 5.Single site adaptation taskSubject selects which of two stimuli (test vs. standard) are larger

Temporal Order Judgement.

For the temporal order judgment (TOJ) task, two sequential taps (200µm, 40ms) were delivered, one to each digit tip. These were initially temporally separated by an interstimulus interval (ISI) of 150ms. The stimulus location that received the first of the two pulses was randomized on a trial-by-trial basis. Subjects were queried to select the digit that received the first stimulus. The temporal separation between the two pulses was adjusted on the basis of the previous response through employment of percentage tracking (15% step size) such that correct responses resulted in shorter ISIs while incorrect responses increased the ISIs. Each task consisted of 20 trials. Schematic of TOJ timeline is displayed in Figure 6. These methods have been described previously [1-3,9,22-24]).

Figure 6. Temporal order judgement (TOJ) task Subject selects which of two stimuli is presented first

Duration Discrimination.

Duration discriminative capacity is defined as the minimal difference in durations of two stimuli for which an individual can successfully identify the stimulus of longer duration. For the duration discrimination (DD) task, sequential stimuli were delivered to D2 and D3 in 20 trials (initial stimulus parameters: 750ms test, 500ms standard, 300µm, 25Hz, 25ms step size). Discrimination capacity was assessed using a 2AFC tracking protocol (in a manner similar to that described for amplitude discrimination capacity). The duration of the test stimulus was always greater than that of the standard stimulus, but the location of the stimulus of longer duration was randomly selected on a trial-by-trial basis. Subjects were asked to determine which of the two digits received the longer stimulus duration. The difference between the duration of the test and standard amplitudes was adjusted on the basis of subject response; correct responses resulted in shortening the test duration in subsequent trials while incorrect responses resulted in increasing the test duration in subsequent trials (total of 20 trials). This method has been previously described [1,3,17,18,25] and the timeline schematic of the method is shown in Figure 7.

Figure 7.Duration discrimination taskSubject selects which of two stimuli are perceived to last longer in duration.

Data Analysis

Two-sample t-tests were used to evaluate the difference of the subject’s performance across different groups. Data are presented as means and SE. A probability of less than 0.05 was considered statistically significant.

Results

Of the 246 subjects diagnosed with mTBI, 174 were cleared by a team physician within 2 weeks of the initial diagnosis. These 174 subjects were included in the “Acute” group for data analysis. The remaining 72 subjects continued to experience concussion symptoms for longer than 14 days (some for up to 48 days beyond the initial insult). These 72 subjects were included in the “Chronic” group for data analysis.

Reaction Time and Variability

The average reaction times and reaction time variabilities obtained during the reaction time task for healthy controls, individuals from our “acute” concussion sub-group, and individuals from our “chronic” concussion sub-group are summarized in Figure 8. The “acute” and “chronic” concussion groups both include data collected within the first 7 days after a subject was diagnosed with an mTBI. Two-sample t-tests were employed to compare the reaction time and reaction time variabilities and p-values are stored in Figure 8.

The average reaction times and reaction time variabilities collected from both of concussion group are significantly longer than those collected from the healthy controls. The average reaction time from the control group was 223 ± 3 msec, 283 ± 15 msec for the “acute” concussion group, and 264 ± 16 msec for the “chronic” concussion group; a 29% increase for subjects with acute concussion and a 19% increase for subjects with chronic concussion. The reaction time variabilities show a similar trend with the controls have a variability of 19 ± 1 msec, the “acute” concussion group have a variability of 34 ± 4 msec, and the “chronic” concussion group have a variability of 27.5 ± 3 msec; a 79% increase for the “acute” group and a 45% increase for the chronic group. A statistically significant difference was found between the controls and both of the concussion sub-groups in both reaction time means and variabilities. The stastistical difference between the two concussion sub-groups for reaction time means and variability was not significant, but in the case of reaction time variability a strong difference did exist (p-value = 0.60).

Figure 8. Reaction time and reaction time variability. Note that although both concussed cohorts performed worse than controls during the initial first week of testing, the long-term cohort was not outperformed by the acute cohort.

Amplitude Discrimination

The average difference limen (DL) obtained during the sequential and simultaneous amplitude discrimination task for healthy controls, individuals from our “acute” concussion sub-group, and individuals from our “chronic” concussion sub-group are summarized in Figure 9. Two-sample t-tests were employed to compare the difference limens from the amplitude discrimination tasks and p-values are stored in Figure 9.

The average DL collected from both of concussion group are significantly larger than those collected from the healthy controls. The average DL from sequential amplitude discrimination task for the control group was 17.5 ± 0.5 microns, 49 ± 4 microns for the “acute” concussion group, and 41 ± 3 microns for the “chronic” concussion group; a 280% increase for subjects with acute concussion and a 234% increase for subjects with chronic concussion. The average DL from the simultaneous amplitude discrimination task show a similar trend: the controls have a DL of 21 ± 1 msec, the “acute” concussion group have a variability of 65 ± 5 msec, and the “chronic” concussion group have a variability of 62 ± 6 msec; a 310% increase for the “acute” group and a 295% increase for the chronic group. A statistically significant difference was found between the controls and both of the concussion sub-groups in both simultaneous and sequential amplitude discrimination. The statistical difference between the two concussion sub-groups for simultaneous amplitude discrimination was not significant, but in the case of sequential amplitude discrimination a statistically significant difference did exist (p-value = 0.30).

Figure 9. Amplitude discrimination. Healthy controls significantly outperformed both concussed cohorts in both amplitude discrimination tasks. Although there was no difference in performance between the two concussion groups for simultaneous amplitude discrimination, the long-term cohort outperformed the acute cohort on the sequential amplitude discrimination task.

TOJ and Duration Discrimination

The average Inter-stimulus interval (ISI) obtained during the temporal order judgment (TOJ) task and the average DL from the duration discrimination task for healthy controls, individuals from our “acute” concussion sub-group, and individuals from our “chronic” concussion sub-group are summarized in the first section of Figure 10. The difference limen for the duration discrimination task was calculated as a percentage of the standard. Two-sample t-tests were employed to compare both the ISI’s from the TOJ task and the average DL for the duration discrimination task; p-values are stored in Figure 10.

The average ISI collected from both of concussion group are significantly larger than those collected from the healthy controls on the TOJ task. The average ISI for the control group was 24 ± 0.8 msec, 37 ± 2 microns for the “acute” concussion group, and 36.8 ± 3 microns for the “chronic” concussion group; a 54% increase for both subjects with acute concussion and those with chronic concussion. The average DL from the duration discrimination task show a similar trend: the controls have a DL of 8.5 ± 0.5 %, the “acute” concussion group have a variability of 15.8 ± 1 %, and the “chronic” concussion group have a variability of 14.4 ± 1.1 %; a increase of 85% for the “acute” group and an increase of 69% for the chronic group. A statistically significant difference was found between the controls and both of the concussion sub-groups in both simultaneous and sequential amplitude discrimination.

Figure 10. TOJ and duration discrimination. On these tasks, controls outperformed both concussed cohorts, and there does not appear to be any significant difference in performance of the two concussed groups.

Since some of the LONG-TERM subjects had varying time courses of recovery, a Support Vector Machine (SVM) was trained to discriminate between cortical metrics performance of healthy control subjects and performance of LONG-TERM subjects more than 2 weeks post-concussion. SVM identified 72 subjects as being distinctly different from healthy control subjects more than 2 weeks post-concussion. These subjects were tentatively classified as CHRONIC subjects.

A statistical comparison was made between the performance of ACUTE vs. CHRONIC subjects on 6 cortical metrics tests during the first 10 days post-concussion. The means, standard error of the mean, and the statistical significance of the differences of the means (tested using Welch’s T-test) are listed in the Table. The Table also shows means and SEM for the healthy control population.

Figure 11. Summary plot of differences in cortical metrics between 3 populations. In this 3-D rendering, a comparison of 3 measures significantly distinguishes the 3 cohorts. Note that the values used are sequential amplitude discrimination, reaction time variability and simultaneous amplitude discrimination , and for each of these, lower values indicate better performance. In other words, individuals tested during the first week post-injury that took longer to recover outperformed the acute group significantly on these metrics. Also note that the size of the ellipsoid indicates standard deviation and that healthy controls significantly outperformed both concussed cohorts.

Discussion

Several tactile based neurosensory tasks were performed on both healthy control subjects and at multiple time points on concussed individuals post-injury. The results demonstrated that there were significant differences not only between concussed and non-concussed individuals on the neurosensory performance tasks, but there were significant differences between the results on some of the tasks of two different cohorts within the group of concussed individuals when comparison was made between the performance evaluated the first week post-concussion versus the performance evaluated from healthy controls. Interestingly, the cohort of individuals that took longer to recover outperformed the cohort of individuals that recovered more quickly.

Why would some measures be sensitive to the trajectory of recovery and other measures not? To address the question, let’s first examine the differences between the metrics themselves. Although many (incorrectly) assume that any metric derived from a tactile based task such as were administered in this report is testing only the parietal cortex (somatosensory cortex resides in the anterior parietal cortex), there have been numerous demonstrations that this is not the case. Validation studies point to activity in somatosensory cortex as playing a predominant role in vibrotactile amplitude discrimination [26](Francisco et al, 2008) and some studies point to the metric as being significantly impacted by systemic alterations in GABA-mediated inhibition [19,20,24,27,28]. However, other factors clearly play significant roles in timing perception (duration discrimination) and temporal order judgement (TOJ). Blocking cortical activity via TMS in the parietal lobe has little or no effect on TOJ [29] and other cortical areas, such as prefrontal cortices, could play an integral role in TOJ processing [30]. Similarly, blocking activity in cerebellar regions has been demonstrated to have a pronounced impact on timing perception regardless of the sensory modality – visual, auditory or somatosensory (for review, see [31]). Additional evidence for cerebellar involvement in the timing perception task is that individuals with Parkinson’s perform very poorly on the task (Kursun, et al, 2013). While it would be difficult to correlate specific performance on a task to a specific mechanism or cortical locus, the tasks address questions that can be differentiated enough to make the argument that when combined, give a more complete profile of an individual’s brain function than a simple assay of somatosensory function. Additionally, since these metrics appear to be impacted differentially, it stands to reason that they could be impacted differently by such factors as brain injury history, environmental exposures and recovery rate. Thus, variability in the measures of each individual would be expected and outcomes from different cohorts in terms of trajectory recovery would not be anticipated to be the same.

One noteworthy finding are the differences observed in reaction and reaction time variability of the three cohorts. Reaction time has been shown be impacted with concussion or TBI for a number of decades [32-44] although there appears to be some ambivalence in several studies as to whether or not it has been a consistent measure. Holden and colleagues [45] reviewed the literature and found that some reports described little or no difference between the reaction time of concussed and non-concussed individuals (Naesheim et al, 2005; Iverson et al, 2004), and a few have even reported that concussed individuals have faster reaction times than non-concussed individuals (Iverson, et al, 2019; Norman, et al, 2019; Lynall et al, 2018). One explanation for these discrepancies with the reaction time error could simply be the utilization of inaccurate methods for collecting reaction time. Plant and Turner (2009) pointed out the problem with utilization of consumer grade technology (i.e. computer systems) for collecting reaction time. The report by Holden et al [45] investigated the systems that most online computerized assessment tools utilize to routinely to collect reaction times and found that there were significant timing inaccuracies. In many cases, it is impossible to make specific statements regarding the accuracy of the RT or RT Var measurements because the methods presented in these papers related to these measurements are vague, incomplete, and nonspecific. Nevertheless, reaction time in this report is demonstrated to be significantly impacted by concussion and appears to show a difference, though not statistically different, between the acute and long-term recovery cohorts.

Reaction time variability reflects an individual’s consistency at attending to the reaction time task and appears to be a sensitive metric for differentiating between concussed and non-concussed individuals as well as differentiating the trajectory recovery within the concussed individuals. Recently, investigators have recognized that reaction time variability could be a better indicator for assessing concussion than reaction time alone [46] and the demonstration of reaction time variability as a potentially good biomarker for concussion in both this report and prior reports utilizing the method [3,18]Tommerdahl et al, 2018, 2020) provides evidence for that. However, the above described resolution problems for collecting reaction time accurately are magnified tremendously for the accurate collection of reaction time variability, and this calls into question the contrary findings of some reports (for review see [45]). For example, in a recent report, Pearce et al (2019b) compared healthy controls with individuals with persistent PCS and made reaction time observations with both the Brain Gauge (as described in this report) and a visual reaction time task. Results collected with the Brain Gauge found that individuals with persistent PCS had a reaction time variability of 25.6 msec which was distinctly different from the control population average obtained of 14.7 msec. However, the visual reaction time task showed little or no difference; reaction time variability for both the symptomatic and asymptomatic group were 98 msec. The reaction time metric itself demonstrated a 6% difference between symptomatic and asymptomatic populations with the visual reaction time task but a 30% difference with the tactile based task. Pearce et al [47] interpretation of that finding is that the latency errors described in Holden et al [45]accounted for the lack of resolution in the visual reaction time task. Thus, it appears that reaction time and reaction time variability are potentially very good biomarkers for concussion if accurate measures are made. Using inaccurate methods, especially those that could not even theoretically resolve the differences in reaction times and reaction time variability, renders the utility of reaction time testing as equivalent to that of many of the methods currently routinely in use for concussion assessments that have been repeatedly demonstrated to be inadequately sensitive, and therefore of no clinical value.

Why was reaction time variability for the acute group worse than the reaction time variability for the long-term recovery group? Some insight for this can come from the animal model described in Challener et al, 2020. Although the initial response to mild brain injury does not result in spontaneous neural activity changing significantly, longer term injury (induced by a second mild trauma) does lead to a reduction in inter-neuron response diversity. In other words, if an injury is worse, there is less variability in the stimulus evoked neural response. It is possible that a system reduction in neuronal variability could contribute to an improvement in reaction time variability, but more research will have to be conducted over a longer-term window in the animal model to determine if this is, in fact, the case.

Another challenging question to address is the differences observed in the recovery trajectory within the concussed individuals in sequential amplitude discrimination. The better performance in the long-term injury cohort for sequential amplitude discrimination could be attributed to a couple of possibilities. First (from Challener et al, 2020), more severe brain injury appears to result in a decrease in stimulus response amplitude. Based on previous studies, that would imply that simultaneous amplitude discrimination would potentially degrade because of the lower stimulus evoked response and a reduction in lateral inhibition (Simons et al, 2005) that would degrade discrimination capacity. Simultaneous amplitude discrimination capacity is dependent on contrast enhancement between the neural activity of two adjacent cortical regions corresponding to the two adjacent digits (D2 and D3) where the stimulus is delivered. Reduction in overall stimulus response would lead to a reduction in the signal to noise ratio and the subsequent reduction in contrast between the two zones would be a deviation from the Weber’s fraction normally observed by this task (Francisco et al, 2008). Thus, the degradation in simultaneous amplitude discrimination with concussed individuals could be attributed to a degradition in lateral inhibition and/or reduction in stimulus evoked response to adjacent sensory inputs. Sequential amplitude discrimination, on the other hand, does not depend on lateral inhibition in the same manner as simultaneous amplitude discrimination [2]. Both concussed cohorts degraded in sequential amplitude discrimination capacity, but not to the degree that was observed with simultaneous amplitude discrimination capacity, suggesting that other mechanisms of information processing are involved with the sequential task and were disrupted differentially within the groups of concussed individuals. The question that remains to be addressed is why the presumably more injured cohort would outperform the less injured group on the sequential task. One possibility would be to consider that with the worse injury led to more of an inflammatory response and consequently more of a suppression of the glial response [48](. With the sequential task, the glial response that follows the neural response could interfere with the discrimination task via prolonged relative time course. Shortened time course of glial response (as described in [48]) could result in a relative improvement in the sequential discriminative capacity between the two concussed cohorts with different recovery trajectories. The authors view these explanations as working hypotheses, and further investigations of neurophysiological outcomes in the animal model could provide beneficial insights into recovery trajectories of concussed individuals.

References

1. Cole Wesley R., Gregory Emma, Arrieux Jacques P., Haran F. Jay. Intraindividual Cognitive Variability: An Examination of ANAM4 TBI-MIL Simple Reaction Time Data from Service Members with and without Mild Traumatic Brain Injury. Journal of the International Neuropsychological Society. 2017; 24(2)DOI
2. Collins Michael W., Iverson Grant L., Lovell Mark R., McKeag Douglas B., Norwig John, Maroon Joseph. On-Field Predictors of Neuropsychological and Symptom Deficit Following Sports-related Concussion. Clinical Journal of Sport Medicine. 2003; 13(4)DOI
3. Collins Kassondra L., McKean Danielle L., Huff Katherine, Tommerdahl Mark, Favorov Oleg Vyacheslavovich, Waters Robert S., Tsao Jack W.. Hand-to-Face Remapping But No Differences in Temporal Discrimination Observed on the Intact Hand Following Unilateral Upper Limb Amputation. Frontiers in Neurology. 2017; 8DOI
4. Eckner James T., Richardson James K., Kim Hogene, Joshi Monica S., Oh Youkeun K., Ashton-Miller James A.. Reliability and Criterion Validity of a Novel Clinical Test of Simple and Complex Reaction Time in Athletes. Perceptual and Motor Skills. 2015; 120(3)DOI
5. Favorov Oleg V, Francisco Eric, Holden Jameson, Kursun Olcay, Zai Laila, Tommerdahl Mark. Quantification of Mild Traumatic Brain Injury via Cortical Metrics: Analytical Methods. Military Medicine. 2019; 184(Supplement_1)DOI
6. Folger Stephen E, Tannan Vinay, Zhang Zheng, Holden Jameson K, Tommerdahl Mark. Effects of the N-methyl-D-Aspartate receptor antagonist dextromethorphan on vibrotactile adaptation. BMC Neuroscience. 2008; 9(1)DOI
7. Francisco E., Holden J., Zhang Z., Favorov O., Tommerdahl M.. Rate dependency of vibrotactile stimulus modulation. Brain Research. 2011; 1415DOI
8. Francisco Eric M., Holden Jameson K., Nguyen Richard H., Favorov Oleg V., Tommerdahl Mark. Percept of the duration of a vibrotactile stimulus is altered by changing its amplitude. Frontiers in Systems Neuroscience. 2015; 9DOI
9. Francisco Eric M., Holden Jameson K., Nguyen Richard H., Favorov Oleg V., Tommerdahl Mark. Percept of the duration of a vibrotactile stimulus is altered by changing its amplitude. Frontiers in Systems Neuroscience. 2015; 9DOI
10. Gould Jennifer A., Ciuffreda Kenneth J., Yadav Naveen K., Thiagarajan Preethi, Arthur Benjamin. The effect of retinal defocus on simple eye-hand and eye-foot reaction time in traumatic brain injury (TBI). Brain Injury. 2013; 27(13-14)DOI
11. Güçlü Burak, Tanıdır Canan, Çanayaz Emre, Güner Bora, İpek Toz Hamiyet, Üneri Özden Ş., Tommerdahl Mark. Tactile processing in children and adolescents with obsessive–compulsive disorder. Somatosensory & Motor Research. 2015; 32(3)DOI
12. Hanley Claire J., Tommerdahl Mark, McGonigle David J.. Stimulating somatosensory psychophysics: a double-blind, sham-controlled study of the neurobiological mechanisms of tDCS. Frontiers in Cellular Neuroscience. 2015; 9DOI
13. Hanley Claire J., Burianová Hana, Tommerdahl Mark. Towards Establishing Age-Related Cortical Plasticity on the Basis of Somatosensation. Neuroscience. 2019; 404DOI
14. HETHERINGTON C. R., STUSS D. T., FINLAYSON M. A. J.. Reaction time and variability 5 and 10 years after traumatic brain injury. Brain Injury. 1996; 10(7)DOI
15. Holden Jameson K., Nguyen Richard H., Francisco Eric M., Zhang Zheng, Dennis Robert G., Tommerdahl Mark. A novel device for the study of somatosensory information processing. Journal of Neuroscience Methods. 2012; 204(2)DOI
16. Holden Jameson, Francisco Eric, Tommerdahl Anna, Lensch Rachel, Kirsch Bryan, Zai Laila, Pearce Alan J., Favorov Oleg V., Dennis Robert G., Tommerdahl Mark. Methodological Problems With Online Concussion Testing. Frontiers in Human Neuroscience. 2020; 14DOI
17. Jones Christina B., Lulic Tea, Bailey Aaron Z., Mackenzie Tanner N., Mi Yi Qun, Tommerdahl Mark, Nelson Aimee J.. Metaplasticity in human primary somatosensory cortex: effects on physiology and tactile perception. Journal of Neurophysiology. 2016; 115(5)DOI
18. Khan Sheraz, Michmizos Konstantinos, Tommerdahl Mark, Ganesan Santosh, Kitzbichler Manfred G., Zetino Manuel, Garel Keri-Lee A., Herbert Martha R., Hämäläinen Matti S., Kenet Tal. Somatosensory cortex functional connectivity abnormalities in autism show opposite trends, depending on direction and spatial scale. Brain. 2015; 138(5)DOI
19. Koch Giacomo, Oliveri Massimiliano, Caltagirone Carlo. Neural networks engaged in milliseconds and seconds time processing: evidence from transcranial magnetic stimulation and patients with cortical or subcortical dysfunction. Philosophical Transactions of the Royal Society B: Biological Sciences. 2009; 364(1525)DOI
20. Lee Kevin GH, Jacobs Mark F, Asmussen Michael J, Zapallow Christopher M, Tommerdahl Mark, Nelson Aimee J. Continuous theta-burst stimulation modulates tactile synchronization. BMC Neuroscience. 2013; 14(1)DOI
21. Lynall Robert C., Blackburn J. Troy, Guskiewicz Kevin M., Marshall Stephen W., Plummer Prudence, Mihalik Jason P.. Reaction Time and Joint Kinematics During Functional Movement in Recently Concussed Individuals. Archives of Physical Medicine and Rehabilitation. 2018; 99(5)DOI
22. MacFlynn G, Montgomery E A, Fenton G W, Rutherford W. Measurement of reaction time following minor head injury.. Journal of Neurology, Neurosurgery & Psychiatry. 1984; 47(12)DOI
23. Maeda Yumi, Kettner Norman, Holden Jameson, Lee Jeungchan, Kim Jieun, Cina Stephen, Malatesta Cristina, Gerber Jessica, McManus Claire, Im Jaehyun, Libby Alexandra, Mezzacappa Pia, Morse Leslie R., Park Kyungmo, Audette Joseph, Tommerdahl Mark, Napadow Vitaly. Functional deficits in carpal tunnel syndrome reflect reorganization of primary somatosensory cortex. Brain. 2014; 137(6)DOI
24. Nelson Aimee J., Premji Azra, Rai Navjot, Hoque Tasnuva, Tommerdahl Mark, Chen Robert. Dopamine Alters Tactile Perception in Parkinson's Disease. Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques. 2012; 39(1)DOI
25. Nguyen R.H., Ford S., Calhoun A.H., Holden J.K., Gracely R.H., Tommerdahl M.. Neurosensory assessments of migraine. Brain Research. 2013; 1498DOI
26. Nguyen Richard H., Gillen Cody, Garbutt J.C., Kampov-Polevoi Alexei, Holden Jameson K., Francisco Eric M., Tommerdahl Mark. Centrally-mediated sensory information processing is impacted with increased alcohol consumption in college-aged individuals. Brain Research. 2013; 1492DOI
27. Nguyen Richard H., Forshey Theresa M., Holden Jameson K., Francisco Eric M., Kirsch Bryan, Favorov Oleg, Tommerdahl Mark. Vibrotactile discriminative capacity is impacted in a digit-specific manner with concurrent unattended hand stimulation. Experimental Brain Research. 2014; 232(11)DOI
28. Niogi S.N., Mukherjee P., Ghajar J., Johnson C., Kolster R.A., Sarkar R., Lee H., Meeker M., Zimmerman R.D., Manley G.T., McCandliss B.D.. Extent of Microstructural White Matter Injury in Postconcussive Syndrome Correlates with Impaired Cognitive Reaction Time: A 3T Diffusion Tensor Imaging Study of Mild Traumatic Brain Injury. American Journal of Neuroradiology. 2008; 29(5)DOI
29. Pearce Alan J., Tommerdahl Mark, King Doug A.. Neurophysiological abnormalities in individuals with persistent post-concussion symptoms. Neuroscience. 2019; 408DOI
30. Pearce Alan J., Kidgell Dawson J., Frazer Ashlyn K., King Doug A., Buckland Michael E., Tommerdahl Mark. Corticomotor correlates of somatosensory reaction time and variability in individuals with post concussion symptoms. Somatosensory & Motor Research. 2019; 37(1)DOI
31. Plant Richard R., Turner Garry. Millisecond precision psychological research in a world of commodity computers: New hardware, new problems?. Behavior Research Methods. 2009; 41(3)DOI
32. Ponsford Jennie, Kinsella Glynda. Attentional deficits following closed-head injury. Journal of Clinical and Experimental Neuropsychology. 1992; 14(5)DOI
33. Benham Alex, Powell Dylan, Tommerdahl Mark, King Doug, Godfrey Alan, Stuart Samuel. Exploratory Analysis of the effects of University Rugby on Brain Health monitored via a Somatosensory Device. Unpublished. 2018. DOI
34. Puts Nicolaas A.J., Wodka Ericka L., Harris Ashley D., Crocetti Deana, Tommerdahl Mark, Mostofsky Stewart H., Edden Richard A.E.. Reduced GABA and altered somatosensory function in children with autism spectrum disorder. Autism Research. 2016; 10(4)DOI
35. Puts Nicolaas A. J., Harris Ashley D., Mikkelsen Mark, Tommerdahl Mark, Edden Richard A. E., Mostofsky Stewart H.. Altered tactile sensitivity in children with attention-deficit hyperactivity disorder. Journal of Neurophysiology. 2017; 118(5)DOI
36. Puts Nicolaas A. J., Wodka Ericka L., Tommerdahl Mark, Mostofsky Stewart H., Edden Richard A. E.. Impaired tactile processing in children with autism spectrum disorder. Journal of Neurophysiology. 2014; 111(9)DOI
37. Puts Nicolaas A.J., Edden Richard A.E., Wodka Ericka L., Mostofsky Stewart H., Tommerdahl Mark. A vibrotactile behavioral battery for investigating somatosensory processing in children and adults. Journal of Neuroscience Methods. 2013; 218(1)DOI
38. Puts Nicolaas A., Wodka Ericka L., Tommerdahl Mark, Mostofsky Stewart H., Edden Richard A.. Reply to Dickinson and Milne. Journal of Neurophysiology. 2014; 112(6)DOI
39. Puts Nicolaas A.J., Wodka Ericka L., Harris Ashley D., Crocetti Deana, Tommerdahl Mark, Mostofsky Stewart H., Edden Richard A.E.. Reduced GABA and altered somatosensory function in children with autism spectrum disorder. Autism Research. 2016; 10(4)DOI
40. Rai N., Premji A., Tommerdahl M., Nelson A.J.. Continuous theta-burst rTMS over primary somatosensory cortex modulates tactile perception on the hand. Clinical Neurophysiology. 2012; 123(6)DOI
41. Ruesch Jurgen. Dark Adaptation, Negative After Images, Tachistoscopic Examinations and Reaction Time in Head Injuries. Journal of Neurosurgery. 1944; 1(4)DOI
42. Stuss D T, Stethem L L, Hugenholtz H, Picton T, Pivik J, Richard M T. Reaction time after head injury: fatigue, divided and focused attention, and consistency of performance.. Journal of Neurology, Neurosurgery & Psychiatry. 1989; 52(6)DOI
43. Takahashi T., Kansaku K., Wada M., Shibuya S., Kitazawa S.. Neural Correlates of Tactile Temporal-Order Judgment in Humans: an fMRI Study. Cerebral Cortex. 2012; 23(8)DOI
44. Tannan V., Simons S., Dennis R.G., Tommerdahl M.. Effects of adaptation on the capacity to differentiate simultaneously delivered dual-site vibrotactile stimuli. Brain Research. 2007; 1186DOI
45. Tannan V., Dennis R.G., Zhang Z., Tommerdahl M.. A portable tactile sensory diagnostic device. Journal of Neuroscience Methods. 2007; 164(1)DOI
46. Tannan Vinay, Whitsel Barry L., Tommerdahl Mark A.. Vibrotactile adaptation enhances spatial localization. Brain Research. 2006; 1102(1)DOI
47. Tannan V., Dennis R., Tommerdahl M.. A novel device for delivering two-site vibrotactile stimuli to the skin. Journal of Neuroscience Methods. 2005; 147(2)DOI
48. Tannan Vinay, Holden Jameson K., Zhang Zheng, Baranek Grace T., Tommerdahl Mark A.. Perceptual metrics of individuals with autism provide evidence for disinhibition. Autism Research. 2008; 1(4)DOI
49. Tavassoli Teresa, Bellesheim Katherine, Tommerdahl Mark, Holden Jameson M., Kolevzon Alexander, Buxbaum Joseph D.. Altered tactile processing in children with autism spectrum disorder. Autism Research. 2015; 9(6)DOI
50. Tommerdahl Mark, Tannan Vinay, Zachek Matt, Holden Jameson K, Favorov Oleg V. Effects of stimulus-driven synchronization on sensory perception. Behavioral and Brain Functions. 2007; 3(1)DOI
51. Tommerdahl M., Tannan V., Cascio C.J., Baranek G.T., Whitsel B.L.. Vibrotactile adaptation fails to enhance spatial localization in adults with autism. Brain Research. 2007; 1154DOI
52. Tommerdahl Mark, Tannan Vinay, Holden Jameson K, Baranek Grace T. Absence of stimulus-driven synchronization effects on sensory perception in autism: Evidence for local underconnectivity?. Behavioral and Brain Functions. 2008; 4(1)DOI
53. Tommerdahl Mark, Favorov Oleg V., Whitsel Barry L.. Dynamic representations of the somatosensory cortex. Neuroscience & Biobehavioral Reviews. 2010; 34(2)DOI
54. Tommerdahl Mark, Dennis Robert G., Francisco Eric M., Holden Jameson K., Nguyen Richard, Favorov Oleg V.. Neurosensory Assessments of Concussion. Military Medicine. 2016; 181(5S)DOI
55. Tommerdahl Mark, Lensch Rachel, Francisco Eric, Holden Jameson, Favorov Oleg. The Brain Gauge: a novel tool for assessing brain health. The Journal of Science and Medicine. 2019; 1(1)DOI
56. Tommerdahl Mark, Francisco Eric, Holden Jameson, Lensch Rachel, Tommerdahl Anna, Kirsch Bryan, Dennis Robert, Favorov Oleg. An Accurate Measure of Reaction Time can Provide Objective Metrics of Concussion. The Journal of Science and Medicine. 2020; 2(2)DOI
57. Turco Claudia, Locke Mitchell, El-Sayes Jenin, Tommerdahl Mark, Nelson Aimee. Exploring Behavioral Correlates of Afferent Inhibition. Brain Sciences. 2018; 8(4)DOI
58. van Zomeren A.H., Deelman B.G.. Differential effects of simple and choice reaction after closed head injury. Clinical Neurology and Neurosurgery. 1976; 79(2)DOI
59. Warden D. L., Bleiberg J., Cameron K. L., Ecklund J., Walter J., Sparling M. B., Reeves D., Reynolds K. Y., Arciero R.. Persistent prolongation of simple reaction time in sports concussion. Neurology. 2001; 57(3)DOI
60. WILLISON J, TOMBAUGH T. Detecting simulation of attention deficits using reaction time tests☆. Archives of Clinical Neuropsychology. 2006; 21(1)DOI
61. Wodka Ericka L., Puts Nicolaas A. J., Mahone E. Mark, Edden Richard A. E., Tommerdahl Mark, Mostofsky Stewart H.. The Role of Attention in Somatosensory Processing: A Multi-trait, Multi-method Analysis. Journal of Autism and Developmental Disorders. 2016; 46(10)DOI
62. Zahn Theodore P., Mirsky Allan F.. Reaction Time Indicators of Attention Deficits in Closed Head Injury. Journal of Clinical and Experimental Neuropsychology. 1999; 21(3)DOI
63. Zhang Zheng, Francisco Eric M., Holden Jameson K., Dennis Robert G., Tommerdahl Mark. The impact of non-noxious heat on tactile information processing. Brain Research. 2009; 1302DOI
64. Zhang Zheng, Zolnoun Denniz A., Francisco Eric M., Holden Jameson K., Dennis Robert G., Tommerdahl Mark. Altered Central Sensitization in Subgroups of Women With Vulvodynia. The Clinical Journal of Pain. 2011; 27(9)DOI
65. Zhang Zheng, Francisco Eric M., Holden Jameson K., Dennis Robert G., Tommerdahl Mark. Somatosensory Information Processing in the Aging Population. Frontiers in Aging Neuroscience. 2011; 3DOI
66. Zhang Zheng, Tannan Vinay, Holden Jameson K, Dennis Robert G, Tommerdahl Mark. A quantitative method for determining spatial discriminative capacity. BioMedical Engineering OnLine. 2008; 7(1)DOI
67. Tannan V, Dennis RG, Tommerdahl M. Stimulus-dependent effects on tactile spatial acuity.. Behav Brain Funct. 2005; 10:1-18. DOI
68. Tommerdahl M. Non-invasive CNS diagnostics.. 2010.
69. Tommerdahl M. Commentary On: Absence of stimulus-driven synchronization effects on sensory perception in autism: evidence for local under-connectivity. J Neurology Neuromedicine. 2016; 1(1):24-27.
70. Lowe S, Tommerdahl A, Lensch R, Francisco E, Holden J, Tommerdahl M. An Undergraduate Laboratory Exercise that Demonstrates the Difference Between Peripherally and Centrally Mediated Measures.. JUNE. 2016.
71. King DA, Hume PA, Cummins C, Tommerdahl M, Pearce A, Clark T, McGeown J. Use of the King-Devick test and Brain Gauge for the management of concussion. Journal of Science and Medicine in Sport. 2018; 21:38-39.
72. King DA, Hume PA, Tommerdahl M. Use of the Brain Gauge Somatosensory Assessment for monitoring recovery from concussion: A case study. Journal of Physiotherapy Research. 2018; 2(13)
73. Holden JK, Francisco EM, Zhang Z, Baric C, Tommerdahl M. An Undergraduate exercise to study Weber’s Law. J Undergrad Neurosci Educ. 2011; 9(2):71-74.
74. Favorov O, Kursun O, Tommerdahl M. Role of Feed-Forward Inhibition in Neocortical Information processing: Implications for Neurological Disorders.. 2017.
75. Favorov O, Whitsel B, Tommerdahl M. Discrete, Place-Define Macrocolumns in Somatosensory Cortex: Lessons for Modular Organization of the Cerebral Cortex.. 2015.