This study examined whether different playing positions of high school football players are associated with impaired memory and auditory comprehension at a sentence level after a concussion. The specific research questions are 1) whether there are significant differences in memory on the Immediate Postconcussion Assessment Cognitive Test (ImPACT) between a speed-positions group and a non-speed positions group, and 2) whether there are significant differences in auditory comprehension on Subtest VIII of the Computerized-Revised Token Test (CRTT) between the speed-positions group and the non-speed positions group.
36 acutely concussed high school football players (Age:
The results revealed that at post-concussion assessment high school athletes who play non-speed positions perform significantly poorer than a matched group of high school athletes who play speed positions on verbal memory and auditory comprehension abilities.
According to a Gallup poll in 2017, football is the most popular sport in United States (Retrieved from
Unfortunately, there is a relative paucity of evidence on the cognitive behaviors (i.e., memory and language) of concussed high school football players, and specifically the impact of the playing positions on memory and language at the high school level is unknown [
There is a growing body of evidence that some playing positions have a higher risk of concussion for professional players [
Non-speed positions are also reported to have significantly more symptoms than other positions on a self-report questionnaire from a group of 730 National Collegiate Athletic Association Division 1 Football Championship Series athletes [
However, there is limited information as to the consequences of positions on the cognitive-communication behavioral changes in high school football players, even though a recent study [
Dysfunctions of memory and language are frequently reported clinical features of CTE [
Considering that a decline in cognitive and language functions are the most representative symptoms in CTE [
A recent study (Montenigro et al. [
Montenigro et al. [
The present study investigated whether playing positions was associated with adverse consequence on a players’ memory and language performance. The specific research questions are: 1) Are there significant differences in memory on the Immediate Post-concussion Assessment Cognitive Test (ImPACT) [
Our hypothesis is that the non-speed position group would demonstrate significantly lower memory and language scores than the scores of the speed-position group, and would exhibit slower processing speed in response time measures than the speed position group. The rationale for the hypothesis is that non-speed positions are more exposed to recurrent sub-concussive hits/impacts than the speed positions. If the sub-concussive hits affect the cognitive and language behaviors, and the non-speed positions players are at risk for more frequent sub-concussive hits we predict they would demonstrate slower and poorer performance on memory and language tasks than the speed positions players.
A total of 36 concussed high school football players (Mean Age: 14.61±1.96) completed both the ImPACT and Subtest VIII of the CRTT, which requires auditory comprehension at the sentence level.
All athletes were referred by their team trainers or physicians to the Concussion Management Clinic in the Department of Rehabilitation Science at the University of Texas at El Paso.
Participants were divided into two groups based on their reported playing positions: speed-position (
The estimation of the number of required subjects for this study was calculated using the sample size software, G-Power analysis. Statistical power was set at 0.80, and alpha level was set at 0.05 for the calculation. The calculated sample size was a total of 32 subjects with the medium effect size Cohen’s
The Verbal Memory Composite, Visual Memory Composite, Visual-Motor Speed (Processing Speed) Composite scores from the ImPACT were initially included as well as the CRTT accuracy and CRTT Efficiency Scores (ES) from the CRTT Subtest VIII. IBM SPSS Statistics 24 was utilized to compare two groups across the ImPACT and CRTT variables. These initial variables were tentative to choose each variable from each task based on the variable selection analysis
All participants met the following selection criteria: (a) they were native American English speakers; (b) reported no history of speech-language therapy; (c) reported no history of learning disability and other special education issues; (d) reported no history of other neurological disease; (e) were referred for post-concussion assessment.
The current study used two computerized neurocognitive test batteries: ImPACT Version 2 which incorporates measures of memory and CRTT Subtest VIII as a language measure.
ImPACT: ImPACT is a computer-administered neuropsychological test for the post-concussed population [
The Verbal Memory Composite Score utilized is comprised of the average of the following scores: Word Memory total percentage correct (immediate+delay)/2 from Module 1, Symbol Match (hidden symbols)/9*100 from Module 4, and Three letters Total letters correct from Module 6.
The Visual Memory Composite Score is average of the following scores: X/s and O’s Total correct (memory)/12/100 from Module 3, and Design memory-total percentage correct (immediate+delay)/2 from Module 2.
The Processing Speed Composite Score is average of the following scores: X’s and O’s total correct (during interference) total/4 from Module 3, and Three letters average accounted correctly*3 from Module 6.
CRTT Subtest VIII: Following the administration of the ImPACT, CRTT Subtest VIII was performed as a language measure, specifically as a measure for the auditory comprehension. The auditory comprehension performance has been highlighted as a predictor variable for the treatment effect and recovery in stroke survivors with aphasia [
The CRTT is the computerized version of the Revised Token Test (McNeil & Prescott [
The CRTT consists of 10 subtests and each subtest consists of 10 imperative sentences. The command in the CRTT Subtest VIII begins with the word “put” and the length of each sentences vary with critical lexical items of color (blue, green, red, white, and black), shape (square and circle), and size (small and big) [
The CRTT software calculates the scores automatically using a multidimensional scoring system. The multidimensional scoring system is specifically designed to capture underlying deficits and behaviors in people with brain damages [
According to the multidimensional scoring system [
The CRTT accuracy and CRTT ES are calculated by the software program for CRTT Subtest VIII and were reported for the current study. CRTT efficiency score was added as a unique feature of CRTT from the original RTT [
Three variables from the ImPACT (Verbal Memory Composite score, Visual Memory Composite score, and Visual-Motor Speed Composite ([Processing Speed, PS] score) and the two CRTT Subtest VIII variables CRTT accuracy, and CRTT ES measures were initially computed for a Pearson correlation coefficient to assess the relationships among the variables. In addition, exploratory PCA was computed to identify the nature and interrelationships among the variables and also to reduce the number of variables based on the nature and the interrelationships. After the PCA, a Between Groups MANOVA was performed to compare the two participant groups across the selected ImPACT and CRTT variables based on the PCA results using IBM SPSS Statistics 24.
The results of Pearson Correlation analyses were summarized in
Five variables were initially included for the data collection, and two variables were chosen based on the PCA. The PCA with varimax rotiation was performed for five variables to identify the variables with the greatest shared variance. Two factors were extracted from the PCA results. Interestingly, all three variables from the ImPACT were highly loaded on one factor, and the two variables from the CRTT were highly loaded on a second factor.
The criteria for choosing the variables based on the PCA results were as follows: 1) The variable with the highest value was chosen as the dependent variable from each factor, 2) The variable that was not highly loaded on the second factor was chosen as the dependent variable. Two variables were chosen from each factor based on these criteria. The Verbal Memory variable from ImPACT was the highest value from the first factor out of two factors, and CRTT Efficiency was the highest value in the second factor out of two factors. Therefore, the Verbal memory Composite Scores and the CRTT Efficiency Scores were choosen as the dependent variables because each variable represents each factor as well as each test, ImPACT and CRTT.
The Multivariate Analysis of Variance (MANOVA) was conducted to assess the difference on the two selected dependent variables, the Verbal Memory Composite scores and the CRTT Efficiency Scores, compared between the speed and non-speed groups. The multivatriate effect was significant by groups,
The followed ANOVAs with post hoc tests showed that the non-speed group composite scores were significantly lower than the speed group on the Verbal Memory Composite scores (Non-speed players:
The number of prior concussions reported by the athletes was also compared between the speed and the non-speed position groups. The analysis revealed that there was no significant difference in the number of prior concussions reported by the groups (Speed players:
The primary goal of the current study was to investigate whether playing positions (Speed vs. Non-speed positions) showed significant differences on high school football athletes’ memory and language performance following a sport-related concussion (SRC).
The results revealed that athletes who played non-speed positions performed slower and poorer on memory and auditory comprehension than those athletes who played speed positions. Specifically, the non-speed players were significantly poorer than the speed players on the Verbal Memory Composite Scores from the ImPACT, and the Efficiency Scores from the CRTT Subtest VIII.
The results of this study show that the non-speed players are more vulnerable in verbal memory and language behaviors, specifically auditory comprehension at a sentence level, than the speed players. This vulnerability may be due to the non-speed players’ increased risk for sub-concussive impacts since the frequency of self-reported concussions was not different between groups [
While the current study found evidence of cognitive behavioral changes by playing positions, Clark et al. [
The current study supports and extends the effects of playing positions on behavior performance findings in collegiate athletes [
The correlation results and PCA results across initial variables provided useful information on the nature and relationships among variables. The two tasks, the ImPACT and CRTT were separated in the results. The ImPACT was included as a memory measure, and the CRTT was included as a language measure, specifically for the auditory comprehension at a sentence level. In the results of the PCA, three variables from the ImPACT were highly loaded in the 1st factor, while the two variables from the CRTT were highly loaded in the 2nd factor. In the correlation results, the variables within each task were positively correlated. Therefore, the variable selection from each task was made based on the most highly loaded in each factor because the highest variable was assumed to be the most representative for each task. These results can suggest an option for the possibility of a variable selection out of many potential variables in concussion research.
The results of this study also suggest several interesting possibilities or speculation regarding the recent CTE literature. First, exhibiting significant differences in auditory comprehension performance, the CRTT ES, might indicate an early symptom of a decline in language function in people with the CTE. According to Mez et al. [
Further investigation of auditory comprehension in a sample of individuals showing potential signs and symptoms of CTE may help validate the above speculation. Based on the findings of the current study, we can assume that playing position is potentially a significant factor in detecting those individuals with an increased risk for developing CTE.
The comparison of the number of prior concussion self-reported by each group was not statistically different between the two groups and therefore cannot explain the differences between the speed position and the non-speed position groups. Furthermore, given that the two groups were matched for age and years of education, these two variables cannot explain the current findings. Therefore, the source of differences between the two speed groups carrying out spoken commands may suggest the possibility that the non-speed position players potentially experienced more sub-concussive hits. The higher rate of sub-concussive hits experienced by high school athletes in non-speed positions may put them at greater risk for developing CTE given the dysfunction of verbal memory and auditory comprehension found in this sample of high school athletes.
This study has limitations, which indicate the need for further investigations. We did not test different domains or levels of language such as syntax, phonology, or word level vs. sentence level. These aspects of language processing require further investigation.
Another limitation of the study was the lack of data on the frequency of sub-concussive hits experienced by the participants. The current study was not designed to investigate the relationship between the recurrent hits and the cognitive performance. Previous studies have provided information on the relationship between playing positions and the number of recurrent hits. In the future, an investigation of the effect of recurrent sub-concussive and concussive hits can be carried out to assess the impact on cognitive communication variables such as auditory comprehension, by playing positions.
This current study used a relatively small number of participants so generalization to a population is limited. However, this study did indicate the young athletes’ cognitive communication performance may be adversely impacted based on the position they play in the game of football. Future studies with a larger dataset are suggested to investigate the cognitive communication differences by playing positions and to replicate the results of the current study.
The current study examined how playing positions may impact a high school athletes’ memory and language performance following a sports-related mild concussion. The findings exhibit that playing positions in high school football athletes can significantly affect cognitive communicative behaviors such as verbal memory and auditory comprehension of sentences.
The research was supported by the grant funded by Department of Health and Human Services (HHS), National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR), Advanced Rehabilitation Research and Training (ARRT) (Program no. 90AR5016-03-00, CFDA No. 93.433).
PCA was performed in order to reduce the number of variables, and also to select the best representative variable of each test based on the nature and the interrelationships among variables.
Demographic information
All athletes | Speed position-group | Non-speed position-group | |
---|---|---|---|
Number of subjects | 36 | 18 | 18 |
Age | 14.61 (1.96) | 14.72 (1.41) | 14.5 (2.2) |
Education | 8.36 (1.92) | 8.61 (1.38) | 8.1 (2.2) |
Reported concussion history | 1.44 (0.69) | 1.56 (0.78) | 1.33 (0.59) |
Correlations results among variables
Variables | Verbal memory composite scores | Visual memory composite scores | Visual-motor speed composite scores | CRTT accuracy scores | CRTT efficiency scores |
---|---|---|---|---|---|
Verbal memory composite scores | 1 | 0.731 |
0.623 |
0.173 | 0.086 |
Visual memory composite scores | 0.731 |
1 | 0.352 |
0.319 | 0.065 |
Visual-motor speed composite scores | 0.623 |
0.352 |
1 | 0.159 | 0.141 |
CRTT accuracy scores | 0.173 | 0.319 | 0.159 | 1 | 0.681 |
CRTT efficiency scores | 0.086 | 0.065 | 0.141 | 0.681 |
1 |
Correlation is significant at the 0.01 level (2-tailed);
Correlation is significant at the 0.05 level (2-tailed).
Coefficients derived from the PCA Rotated Component Matrix
Factor 1 | Factor 2 | |
---|---|---|
Verbal Memory Composite Scores | 0.938 | 0.037 |
Visual Memory Composite Scores | 0.826 | 0.131 |
Visual-Motor Speed Composite Scores | 0.752 | 0.083 |
CRTT Accuracy Scores | 0.181 | 0.902 |
CRTT Efficiency Scores | 0.012 | 0.921 |
Coefficients bolded are those selected to represent each factor.