The Physician and Sportsmedicine
Menubar Home Journal Personal Health Resource Center CME Advertiser Services About Us

Sparring and Cognitive Function in Professional Boxers

Barry D. Jordan, MD; Eric J.T. Matser; Robert D. Zimmerman, MD; and Tania Zazula, PhD


If your browser supports tables click here.

In Brief: A volunteer group of 42 professional boxers provided information about their careers and training practices and underwent neuropsychological testing. Performance on the neuropsychological tests was not associated with age, boxing record (wins, losses, or total number of bouts), length of career, or history of knockout or technical knockout. However, the amount of sparring the boxers did was inversely associated with their performance on several of the tests. Impairments revealed by the tests were in the areas of attention, concentration, and memory.

Most neuropsychological studies of boxers suggest that boxing may lead to impairments in memory, attention, and information processing (1). The risk of cognitive impairment associated with boxing—dementia pugilistica—has been noted for many years. This risk is worthy of physicians' attention, since boxing remains a popular sport with thousands of participants in the United States.

Neuropsychological testing represents the most sensitive technique to detect neurologic dysfunction associated with boxing (1). Despite the sensitivity of such testing, interpretation of the results poses some difficulty. Test score norms for the general population are not representative of the boxing population and therefore are of limited value or should be used with caution (1). As discussed by Levin et al (2), it is hazardous to interpret posttraumatic neuropsychological data without the benefit of appropriate controls matched for age and socioeconomic, sociocultural, and education variables.

Our study, therefore, was designed to determine if selected neuropsychological test scores correlate with selected dimensions of boxing exposure, without reference to the general-population norms.


Forty-two professional boxers who were licensed to box or applying for a license to box in New York State volunteered to participate in the study. Boxers were initially contacted by letter requesting their participation. Boxers who did not respond to the letter were either contacted by phone or visited at local gyms in New York City.

All boxers who participated in the investigation underwent neuropsychological testing and were interviewed the day of the testing regarding neurologic symptoms and boxing history. In addition, the most recent computed tomography (CT) scan for each boxer was reviewed by a neuroradiologist who was blinded to the neuropsychological test performance and clinical status of the boxer. (The CT scans had been obtained previous to the neuropyschological testing, as part of New York's state licensing procedure for boxers.) The scans were classified as normal, borderline, or abnormal according to criteria described elsewhere (3,4).

Boxing exposure. Boxing history was obtained via a questionnaire administered by the examiner at the time of interview. Exposure variables, all self-reported, included amateur boxing record, professional boxing record, duration of boxing career (amateur and professional combined), and prior history of technical knockout or knockout (TKO/KO). In addition to competition exposure, sparring exposure was assessed and defined as follows:

  1. Frequency: sessions per week.
  2. Rounds: rounds per session.
  3. Total rounds: the estimated average number of rounds sparred per week. This was obtained by multiplying the average number of sparring sessions per week by the average number of rounds per session.
  4. Intensity: a subjective estimation by the boxer of the average intensity of his sparring, defined as follows: grade 1, minimal or no contact; grade 2, moderate; grade 3, intense, but not quite as intense as competition; grade 4, as intense as competition.
  5. Exposure index: the total number of rounds per week, multiplied by the intensity.
  6. Cumulative exposure index: the sparring exposure index multiplied by the duration of the boxing career.

Neuropsychological testing. Each boxer underwent a battery of neuropsychological tests at the office of the New York State Athletic Commission. The test battery was selected to assess general areas of cognitive functioning and included measures of general intelligence (5), language ability (6), constructional skills, attention and concentration (7-9), memory (10), visual-motor coordination (11-13), concept formation, cognitive stability and flexibility, and planning (14). The tests are specified in table 1.

Table 1. Neuropsychological Tests Used in a Study of 42 
Professional Boxers
Estimates of General Intelligence
  The Wechsler Adult Intelligence Test-Revised (5) 
    Digital Symbol
  Verbal Fluency (6)
Constructional Skills
  Figure Detection (6)
Attention and Concentration
  Paced Auditory Serial Addition (7)
  Wechsler Memory Scale Digit Span Subtest (8)
  Stroop Color and Word Test (9)
  Wechsler Memory Scale Concentration Subtest (8)
  Wechsler Memory Scale (8)
     Verbal memory (immediate and delayed recall) 
     Visual reproduction (immediate and delayed recall) 
     Associated learning 
     Memory quotient
  Rey Complex Figure (10)
     Immediate recall 
     Delayed recall
Visual Motor Coordination
  Trailmaking A and B (11)
  Puncture (12)
  Bourdon (13)
Concept Formation, Cognitive Stability and Flexibility, 
  Wisconsin Card Sorting (14)

Data analysis. Since accepted norms for neuropsychological test data cannot be applied to the boxing population, the boxers' test scores were not classified according to available normative data. Instead, we examined correlations between the boxers' test performance and their age, boxing record, and exposure variables. Measures of association were calculated using the Pearson product-moment correlation coefficient (15).


The characteristics of the 42 boxers are shown in table 2. The average age was 25.6 years and the mean duration of career was 9.5 years (range, 4 to 20 years). Members of the group apparently were fairly successful in amateur boxing, as evidenced by the relatively high mean win-loss ratio. The boxers had had an average of 10 professional fights. Sparring practices appeared to be quite variable; boxers sparred anywhere from 1 to 7 days per week (mean, 3.6 days), and most engaged in only one sparring bout per day.

Table 2. Characteristics of 42 Professional Boxers Tested for 
Cognitive Function
                                    Mean               Range
Age (yr)                            25.6               (19-31)
Duration of boxing career (yr)       9.5                (4-20)
Amateur boxing record
Wins                                49.5                (2-266)
Losses                               6.2                (1-33)
Total*                              55.6                (4-280)
Professional boxing record
Wins                                 6.9                (0-29)
Losses                               3.1                (0-15)
Draws                                0.2                (0-1)
Total*                              10.0                (0-36)
Technical and actual                1.6                 (0-7) 
  knockouts sustained
Sparring exposure
  Frequency (days/wk)                3.6                (1-7) 
  Rounds per sparring session        6.1                (2-12)
  Total rounds/wk                   22.7                (4-56)
  Sparring intensity                 3.0                (2-4)
    (graded on scale of 1-4**)       
  Sparring exposure index***        67.0               (12-160)
  Cumulative sparring exposure**** 682.2               (60-1440)
*Because of rounding, columns do not add up exactly.
**Grade 1 is minimal contact; grade 4 is as intense as competition.
***The product of total rounds per week and sparring intensity.
****The product of sparring exposure index and duration of boxing career.

Seventeen boxers had CT scans that showed borderline brain atrophy (table 3). Two boxers had frankly abnormal CT scans. One abnormal scan demonstrated a focal hypodense lesion in the left frontal lobe that was consistent with posttraumatic encephalomalacia. The other abnormal scan showed cerebellar atrophy with an enlarged fourth ventricle. Six boxers had a cavum septum pellucidum (CSP).

Table 3. Computed Tomography Findings 
  Among 42 Professional Boxers
Normal         23
Borderline     17
Abnormal        2*
Total          42
*Findings showed one focal hypodense lesion and one 
case of cerebellar atrophy.

Correlations between the boxers' performance on the neuropsychological tests and their age, amateur boxing record, professional boxing record, duration of career, and history of TKO/KO were nonsignificant. However, various indices of increased sparring exposure were inversely associated with test scores. These results mostly reflected dysfunction in the areas of attention, concentration, and memory (table 4).

Table 4. Associations of Neuropsychological Test Results With 
Indices of Sparring Exposure Among 42 Professional Boxers*
                          Rounds    Rounds              Sparring   Cumulative      
               Sparring   per       per     Sparring    Exposure   Sparring 
               Frequency  Session   Week    Intensity   Index      Exposure
Constructional                                          X
Attention and              X        X        X          X          X 
Memory          X          X        X        X          X          X
Visual motor    X
Concept formation, 
cognitive stability 
and flexibility, 
*"X" indicates declining neuropsychological function with increased 
sparring exposure (P<.05).

More frequent sparring was associated with decreased performance on portions of the Rey Complex Figure test, the Wechsler Verbal and Visual Memory tests, the Bourdon test, and the Puncture test (dominant hand). The number of rounds per session (mean, 6.1 rounds; range 2 to 12) was inversely related to performance on the Rey Complex Figure, Wechsler Verbal Memory, Stroop Color and Word, and Paced Auditory Serial Addition tests.

The total sparring rounds per week averaged 22.7 (range, 4 to 56). Greater numbers of rounds per week correlated with poorer performance on the Rey Complex Figure, Wechsler Verbal and Visual Memory, Wechsler Memory Quotient, and Stroop tests. Sparring intensity ranged from moderate (grade 2) to competition level (grade 4). Higher sparring intensity correlated with diminished performance on the Rey Complex Figure, Wechsler Visual Memory, and Stroop tests. Both the sparring exposure index and the cumulative sparring exposure index were inversely associated with performance on the Rey Complex Figure, Wechsler Verbal and Visual Memory, and Stroop tests.

We also compared the neuropsychological test scores of the six boxers who had a CSP to those without this condition. The boxers with a CSP performed significantly less well on the Rey immediate recall (P<.05) and the Stroop D (P<.01) tests. However, boxers with a CSP sparred more rounds per week (P<.05) and were heavier (P<.05) than those without a CSP.


The boxers who volunteered to participate in this investigation were representative of the general professional boxing population of New York State (3). The prevalence of abnormal CT scans in this group was 4.8%, similar to the prevalence of abnormal scans reported among boxers in the state (3,4).

The association between increased sparring exposure and declining performance on the neuropsychological tests was not unexpected. It is widely known among the boxing community that most of a boxer's traumatic exposures occur during sparring. This observation is supported by the findings of a survey of boxing injuries among cadets at the US Military Academy at West Point, New York (16). During a 2-year period, the majority of injuries occurred during instructional boxing rather than competition, even though the injury rate was higher during competition. The greater number of injuries occurring during instruction may reflect greater exposure time during instruction and/or sparring.

The results of this cross-sectional study suggest that sparring, not competition, is associated with poorer performance on neuropsychological tests. The tests that most consistently emerged as sensitive measures of cerebral trauma due to sparring were the Rey Complex Figure, Stroop Color and Word Test, and the Verbal Memory and Visual Reproduction subtests of the Wechsler Memory Scale. These tests assess organization, planning, attention, and concentration, as well as the encoding, consolidation, and retrieval of information.

These significant correlations between cognitive function and sparring exposure are highly compatible with well-documented sequelae of closed cerebral trauma (17-20). Similarly, the lack of correlation between sparring and various Wechsler Adult Intelligence Test-Revised and language subtest scores is compatible with the generally acknowledged insensitivity of these cognitive measures to the effects of closed head injury (20). Thus, our neuropsychological findings in boxers are comparable with data on the effects of closed head injury in the general population.

As noted above, boxers with CSP (who reported higher sparring frequencies) showed a further decline on the Stroop test and the Rey Complex Figure Copy subtest when compared with boxers without CSP. The findings reveal greater deficits in attention/concentration and organization/planning as focal anterior abnormality becomes more evident on CT scan.

In conclusion, sparring, which involves repetitive blows to the head, may be associated with statistically significant reductions in cognitive performance. This association, and the lack of relationship between competition and declining cognitive function, may be explained by the fact that boxers spend far more time sparring than actually competing. Accordingly, measures to prevent brain injury should include supervision of sparring bouts and reassessment by boxers of the amount and intensity of sparring they do.


  1. Levin HS, Jordan BD: Neuropsychological assessment of brain injury in boxing, in Jordan BD: Medical Aspects of Boxing. Boca Raton, FL, CRC Press, 1993, pp 197-206
  2. Levin HS, Gary HE, High WM, et al: Minor head injury and post-concussional syndrome: methodological issues in outcome studies, in Levin HS, Grafman J, Eisenberg HM (eds): Neurobehavioral Recovery From Head Injury. New York City, Oxford University Press, 1987, pp 262-275
  3. Jordan BD, Jahre C, Hauser WA, et al: Serial computed tomography in professional boxers. J Neuroimaging 1992;2:181-185
  4. Jordan BD, Jahre C, Hauser WA, et al: CT of 338 active professional boxers. Radiology 1992;185(2):509-512
  5. Wechsler D: WAIS-R manual. New York City, Psychological Corporation, 1989
  6. Luteyn F: Een nieuwe berkorte Groninger Intelligence Test. Dutch J Psychology 1966;21(10):675-682
  7. Gronwall D, Wrightson P: Delayed recovery of intellectual function after minor head injury. Lancet 1974;2(7881):605-609
  8. Wechsler D: A standardized memory scale for clinical use. J Psych 1945;19:87-95
  9. Stroop JR: Studies of the interference in serial verbal reactions. J Exp Psychology 1935;18:643-662
  10. Rey A: L'Examen psychologique dans les cas d'encephalopathic traumatique. Arch Psych 1942;28: 286-340
  11. Army Individual Test Battery. Manual of Direction and Scoring. Washington, DC, War Department Adjutant General's Office, 1944
  12. Van der Vulught J: Puncture test; Lateralisaties van hersenfuncties; een neuropsychologisch onderzoik naar de relatie tussen handvoorkuer en relatie van taalfunctie. Disseration, Sweys & Zeitlinger, Lisse, 1979
  13. Van de Loo L: Enkele Beschouwigen over de Bourdon-Wiersma Test. Tijikschrift van psychologische kring van Nijmeegse Universiteit. 1956;2:33-46
  14. Berg EA: A simple objective test for measuring flexibility of thinking. J Gen Psychology 1948;39:15-22
  15. Ghiselli EE, Campbell JP, Zedeck S: Measurement Theory for the Behavioral Sciences. San Francisco, W.H. Freeman and Co, 1981
  16. Welch MJ, Sitler M, Kroeten H: Boxing injuries from an instructional program. Phys Sportsmed 1986;14 (9):81-90
  17. Gronwall D, Wrightson P: Memory and information processing capacity after closed head injury. J Neurol Neurosurg Psychiatry 1981;44(10):889-895
  18. Gronwall D: Advances in the assessment of attention and information processing after head injury, in Levin HS, Grafman J, Eisenberg HM (eds): Neurobehavioral Recovery From Head Injury. New York City, Oxford University Press, 1987, pp 355-371
  19. Lezak MD: Neuropsychological Assessment, ed 2. New York City, Oxford University Press, 1983
  20. Newcombe F: Psychometric and behavioral evidence: scope, limitations and ecological validity, in Levin HS, Grafman J, Eisenberg HM (eds): Neurobehavioral Recovery From Head Injury. New York City, Oxford University Press, 1987, pp 129-145

Dr Jordan is a neurologist in the Sports Neurology Program, the Hospital for Special Surgery, New York City; Mr Matser is a neuropyschologist in the Department of Neuropsychology, Hospital St Anna, Geldrop, the Netherlands; Dr Zimmerman is former chief of the Division of Neuroradiology, New York Hospital-Cornell Medical Center, New York City; and Dr Zazula is a neuropsychologist in the Sports Neurology Program, the Hospital for Special Surgery, New York City. Address correspondence to Barry D. Jordan, MD, Sports Neurology Program, Hospital for Special Surgery, 535 E 70th St, New York, NY 10021.