- Email: [email protected]
Inhibitory Control Differences Following Mild Head Injury
Brain and Cognition 41, 411–416 (1999) Article ID brcg.1999.1141, available online at http://www.idealibrary.com on
NOTES AND DISCUSSION
Inhibitory Control Differences Following Mild Head Injury Jo-Ann L. Stewart and Rosemary Tannock The Hospital for Sick Children Toronto, Ontario, Canada Complex inhibitory control, defined as the ability to inhibit a planned or ongoing action, was assessed in a sample of individuals with a history of mild head injury, case-matched with normal control subjects for age and gender. This central act of control was assessed using a modification of the stop-signal paradigm. The group with mild head injury took longer to inhibit their on going action and reported more accidents than the normal control subjects. The group that reported having had a mild head injury did not differ in terms of their go reaction time, number of correct responses, handedness, education level, or reported learning disabilities. Limitations of this design and directions for future research are discussed. 1999 Academic Press
Every year, many individuals experience a head injury that involves a loss of consciousness. Many of these individuals experience a traumatic brain injury (TBI) that involves a loss of consciousness and serious, long-term cognitive and behavioral changes (Levin et al., 1988; Robertson et al., 1997). However, for the majority of these individuals, the head injury is mild, associated with only a brief loss of consciousness, with few noticeable long-term effects of the injury (Jennett, 1989). In fact, some of these injuries may go unreported to health-care professionals. Recent research suggests that, although mild head injury (MHI) is inadequately defined and there is little research on deficits following a MHI compared to those of TBI, there seem to be some cognitive effects of the injury that may go unnoticed by many people affected by a MHI (Stewart & TanThis research was supported in part by NIH Grant HD31714 to R. Tannock. The authors are grateful for the contributions of Drs. Russell Schachar and Gordon Logan and thank the Staff at the Ontario Science Centre for their support. Correspondence should be addressed to Dr. R. Tannock, Brain and Behaviour Research Programme, The Research Institute of The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada. Fax: (416) 813-6565. E-mail: [email protected]. 411 0278-2626/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
412
NOTES AND DISCUSSION
nock, 1999). The information gained by the authors in this original study suggests that individuals who had experienced a MHI had a more difficult time inhibiting an ongoing action than matched control subjects when a stopsignal paradigm was used to assess inhibitory control. This finding contrasts many studies in which few, if any long-term effects are noted following a MHI (Jennet, 1989; Satz, Zaucha, McCleary et al., 1997). The complex stop-signal paradigm is a well-established method for assessing inhibitory control (Logan, 1994). This task has subjects stop a planned or ongoing action when the action is suddenly rendered inappropriate. Subjects are engaged in a go task (a choice reaction time task in which they have to press buttons on a computer in response to visual signals) and occasionally and unpredictably they are presented with two auditory signals (a high tone and a low tone). They are told to withhold their response to the go task on a trial when one of the tones is presented, but to ignore the other tone. This method is advantageous because it provides a clear definition of the conditions that trigger the act of control (i.e., the presentation of the stop signal) and the changes that result from executing the act (i.e., inhibition of the response) (Logan, 1994). Also, the model provides a way of estimating the latency of the internally generated act of control (stop-signal reaction time, SSRT), even though successful inhibition produces no overt behavior. Individuals who fail to stop when required to do so appear to be impulsive (Logan, Schachar, & Tannock, 1997). Moreover, failure or slowness in stopping may have a fatal outcome (e.g., failing to stop one’s flight into an oncoming vehicle while running into the street). Based on previous research using the stop-signal paradigm (Stewart & Tannock, 1999), it was predicted that individuals who reported having had a MHI at some point in their lives would also exhibit inhibitory control deficits (a slower SSRT than matched control subjects) and that these same individuals would report more accidents than matched control subjects. METHOD
Participants The current study examined data from a subset of 385 individuals (ages 5 to 82 years) who participated in a study of developmental change in inhibitory control across the life span that was conducted at the Ontario Science Centre (OSC). Participants were selected on the basis of their response to a question about head injury on a questionnaire. Subjects were classified as having a MHI if they responded ‘‘once’’ or ‘‘more than once’’ to the question ‘‘have you ever had a head injury that included a loss of consciousness?’’ These subjects were matched by sex and age using a case–control design with individuals who did not report a head injury. It has been found that individuals with attention deficit hyperactivity disorder (ADHD) show a different pattern of response on the Stop task (Schachar & Logan, 1990); therefore, subjects who reported having been diagnosed with ADHD were not selected for the following analyses. None of the participants had major physical or sensory impairments. Forty-two pairs of subjects, ages 8.38 to 72.12 years (mean age ⫽ 33.85, SD ⫽ 16.76) were included. Four pairs (9.5%) were under 13 years of age; 5 pairs (11.9%) ranged between
NOTES AND DISCUSSION
413
13.0 and 19.9 years; 20 pairs (47.6%) were between 20.0 and 39.9 years; and 13 pairs (30.9%) were 40 years of age or older. Twenty-six pairs (61.9% of the sample) were male and 18 pairs (38.1%) were female.
Questionnaire Data Two questionnaires were used to gain information about age, gender, educational history, handedness, health, current medication, and accident history. An Accident Index was calculated by summing scores across six items (number of admissions to hospital emergency departments, vehicle accidents, head injuries, burns, poisonings, and broken bones in the past 6 months). For subjects in this sample the question about head injury was excluded from the Index. The Accident Index score ranged from 0 (no accidents) to 10 (all endorsed as ‘‘more than once’’).
Complex Stop Signal Paradigm This computerized paradigm involves a go task and a stop task (Logan, 1994). The go task is a choice reaction time task which requires subjects to discriminate between two go signals, an X and an O. The frequencies of X and O presentations were even across the task. The stop task requires subjects to inhibit their response to the go task on 25% of go-signal trials. Six blocks of trials with 24 go-signal and 8 stop-signal trials were administered. The first block of trials was considered a practice block. Therefore, the data from the first block is not included in the analyses that follow. The presentation of stop signals was distributed randomly for each block. The stop signal was a tone (1000 or 500 Hz) delivered through headphones. Subjects were instructed to stop responding when presented with one of these tones and to ignore the other tone. This instruction was counterbalanced across subjects so that half of them were told to inhibit their response to the high tone and half were told to inhibit their response to the low tone. This task, which lasted approximately 15 min was administered to each subject individually by a member of the research team.
RESULTS
Subjects were matched on age, sex, and stop-signal tone (high or low). Paired t-tests were conducted for continuous data and χ 2 tests were used for categorical data. The MHI subjects showed a slower stop-signal reaction time (SSRT) than the control (C) subjects (t ⫽ 2.46; p ⬍ .05). Figure 1 displays the SSRT data for each matched pair in the data set, along with the line of best fit for each of the groups. Although there is an outlier in the MHI group (defined as a subject with an SSRT greater than three standard deviations above the mean), the group differences remain when the data are analyzed without this subject (t ⫽ 2.23; p ⬍ .05). There were no differences between the groups on the remaining stop signal variables (see Table 1). The MHI subjects differed from the C subjects on their Accident score, with the MHI subjects reporting significantly more accidents than the C subjects (p ⬍ .05). There was no difference between subjects on handedness, level of education, or presence of learning disabilities (see Table 1). Thirty-one of the subjects responded to the question inquiring about the length of time of unconsciousness. Subjects were divided into three groups
414
NOTES AND DISCUSSION
FIG. 1. SSRT data for each matched pair in the data set, along with the line of best fit for each of the groups.
depending on their answers to this question. Group 1 consisted of subjects who had been unconscious for ‘‘minutes,’’ Group 2 consisted of subjects who had been unconscious for ‘‘hours,’’ and Group 3 consisted of subjects who had been unconscious for ‘‘days.’’ An analysis of variance (ANOVA) was conducted on these data and it was found that there was no significant difference in SSRT across these groups. An analysis of the Pearson r correlation between the number of accidents reported and stop-signal reaction time was conducted for all subjects. These variables were not significantly correlated (r ⫽ ⫺.11, p ⬎ .05). DISCUSSION
The major finding of the current study was that individuals who reported having had a MHI had a slower SSRT and a higher Accident Index than those individuals with no MHI. It is important to note that the MHI group did not differ from the C group on reaction time to the ‘‘go’’ signal. This suggests that they were not simply slow in responding to all signals. Moreover, the differences that were found cannot be accounted for by other variables such as handedness, reported learning disabilities, or level of education. These results replicate the findings of earlier research that was completed using a similar paradigm (Stewart & Tannock, 1999). Notwithstanding the imposed limitations of this study (i.e., retrospective design and lack of information about preinjury status), these replicated results are provocative and demonstrate a difference in cognitive function, specifically in inhibitory control, in individuals who report having had a MHI. It is interesting to note, once again, that these subjects had presumably recovered from their injuries
415
NOTES AND DISCUSSION
TABLE 1 Comparison of Group Means (n ⫽ 29) Variable/Group
Mean
SD
SSRT MHI
260.93
131.03
C Go RT MHI
208.86
72.34
459.04
128.66
C % Correct HI
452.80
106.67
97.62
3.68
C Accident Index HI
97.92
4.10
1.26
1.27
C Education level HI
0.79
1.22
6.02
2.95
C Handedness (Proportion of sample) HI
6.82
2.72
C Learning Difficulties (Proportion of sample) HI
0.22
t Value(χ 2)
2.46*
0.30
0.36
2.08*
1.84
0.26 (χ 2) 2.29
0.17 (χ 2) 0.42
C
0.05
* p ⬍ .05.
and did not report any long-lasting effects of their injuries to the examiners. Unfortunately, the design of this research project does not allow speculation about whether or not the MHI subjects displayed difficulties with inhibition before their injuries or whether the differences in inhibitory control were a result of the MHI. Further research that uses a prospective design, a more precise definition of MHI, and ascertains details about the accidents, the nature of the injury, and recovery phases in order to answer these questions, is indicated. REFERENCES Jennett, B. 1989. Some international comparisons. In H. S. Levine, H. M. Eisenberg, & A. L. Benton (Eds.), Mild head injury (Pp. 23–34). Oxford Univ. Press: New York.
416
NOTES AND DISCUSSION
Levin, H. S., High, W., Ewing-Cobbs, L., et al. 1988. Memory functioning during the first year after closed head injury in children and adolescents. Neurosurgery, 22, 1043–1052. Logan, G. D. 1994. On the ability to inhibit thought or action: A user’s guide to the stop signal paradigm. In D. Dagenbach & T. H. Carr (Eds.), Inhibitory processes in attention, memory, and language Pp. 189–239. San Diego: Academic Press. Logan, G., Schachar, R., & Tannock, R. 1997. Impulsivity and inhibitory control. Psychological Science, 8, 60–64. Robertson, I. H., Manly, T., Andrade, J., et al., 1997. ‘Oops!’: Performance correlates of everyday attentional failures in traumatic brain injured and normal subjects. Neuropsychologia, 35, 747–758. Satz, P., Zaucha, K., McCleary, C., et al., 1997. Mild head injury in children and adolescents: A review of studies 1970–1995. Psychological Bulletin, 122, 107–131. Schachar, R., & Logan, G. G. 1990. Impulsivity and inhibitory control in normal development and childhood psychopathology. Developmental Psychology, 26, 710–720. Stewart, J. L., & Tannock, R. 1999. Mild head injury and inhibitory control. Brain and Cognition, 40, 1–284 (abstract).
NOTES AND DISCUSSION
Inhibitory Control Differences Following Mild Head Injury Jo-Ann L. Stewart and Rosemary Tannock The Hospital for Sick Children Toronto, Ontario, Canada Complex inhibitory control, defined as the ability to inhibit a planned or ongoing action, was assessed in a sample of individuals with a history of mild head injury, case-matched with normal control subjects for age and gender. This central act of control was assessed using a modification of the stop-signal paradigm. The group with mild head injury took longer to inhibit their on going action and reported more accidents than the normal control subjects. The group that reported having had a mild head injury did not differ in terms of their go reaction time, number of correct responses, handedness, education level, or reported learning disabilities. Limitations of this design and directions for future research are discussed. 1999 Academic Press
Every year, many individuals experience a head injury that involves a loss of consciousness. Many of these individuals experience a traumatic brain injury (TBI) that involves a loss of consciousness and serious, long-term cognitive and behavioral changes (Levin et al., 1988; Robertson et al., 1997). However, for the majority of these individuals, the head injury is mild, associated with only a brief loss of consciousness, with few noticeable long-term effects of the injury (Jennett, 1989). In fact, some of these injuries may go unreported to health-care professionals. Recent research suggests that, although mild head injury (MHI) is inadequately defined and there is little research on deficits following a MHI compared to those of TBI, there seem to be some cognitive effects of the injury that may go unnoticed by many people affected by a MHI (Stewart & TanThis research was supported in part by NIH Grant HD31714 to R. Tannock. The authors are grateful for the contributions of Drs. Russell Schachar and Gordon Logan and thank the Staff at the Ontario Science Centre for their support. Correspondence should be addressed to Dr. R. Tannock, Brain and Behaviour Research Programme, The Research Institute of The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada. Fax: (416) 813-6565. E-mail: [email protected]. 411 0278-2626/99 $30.00 Copyright 1999 by Academic Press All rights of reproduction in any form reserved.
412
NOTES AND DISCUSSION
nock, 1999). The information gained by the authors in this original study suggests that individuals who had experienced a MHI had a more difficult time inhibiting an ongoing action than matched control subjects when a stopsignal paradigm was used to assess inhibitory control. This finding contrasts many studies in which few, if any long-term effects are noted following a MHI (Jennet, 1989; Satz, Zaucha, McCleary et al., 1997). The complex stop-signal paradigm is a well-established method for assessing inhibitory control (Logan, 1994). This task has subjects stop a planned or ongoing action when the action is suddenly rendered inappropriate. Subjects are engaged in a go task (a choice reaction time task in which they have to press buttons on a computer in response to visual signals) and occasionally and unpredictably they are presented with two auditory signals (a high tone and a low tone). They are told to withhold their response to the go task on a trial when one of the tones is presented, but to ignore the other tone. This method is advantageous because it provides a clear definition of the conditions that trigger the act of control (i.e., the presentation of the stop signal) and the changes that result from executing the act (i.e., inhibition of the response) (Logan, 1994). Also, the model provides a way of estimating the latency of the internally generated act of control (stop-signal reaction time, SSRT), even though successful inhibition produces no overt behavior. Individuals who fail to stop when required to do so appear to be impulsive (Logan, Schachar, & Tannock, 1997). Moreover, failure or slowness in stopping may have a fatal outcome (e.g., failing to stop one’s flight into an oncoming vehicle while running into the street). Based on previous research using the stop-signal paradigm (Stewart & Tannock, 1999), it was predicted that individuals who reported having had a MHI at some point in their lives would also exhibit inhibitory control deficits (a slower SSRT than matched control subjects) and that these same individuals would report more accidents than matched control subjects. METHOD
Participants The current study examined data from a subset of 385 individuals (ages 5 to 82 years) who participated in a study of developmental change in inhibitory control across the life span that was conducted at the Ontario Science Centre (OSC). Participants were selected on the basis of their response to a question about head injury on a questionnaire. Subjects were classified as having a MHI if they responded ‘‘once’’ or ‘‘more than once’’ to the question ‘‘have you ever had a head injury that included a loss of consciousness?’’ These subjects were matched by sex and age using a case–control design with individuals who did not report a head injury. It has been found that individuals with attention deficit hyperactivity disorder (ADHD) show a different pattern of response on the Stop task (Schachar & Logan, 1990); therefore, subjects who reported having been diagnosed with ADHD were not selected for the following analyses. None of the participants had major physical or sensory impairments. Forty-two pairs of subjects, ages 8.38 to 72.12 years (mean age ⫽ 33.85, SD ⫽ 16.76) were included. Four pairs (9.5%) were under 13 years of age; 5 pairs (11.9%) ranged between
NOTES AND DISCUSSION
413
13.0 and 19.9 years; 20 pairs (47.6%) were between 20.0 and 39.9 years; and 13 pairs (30.9%) were 40 years of age or older. Twenty-six pairs (61.9% of the sample) were male and 18 pairs (38.1%) were female.
Questionnaire Data Two questionnaires were used to gain information about age, gender, educational history, handedness, health, current medication, and accident history. An Accident Index was calculated by summing scores across six items (number of admissions to hospital emergency departments, vehicle accidents, head injuries, burns, poisonings, and broken bones in the past 6 months). For subjects in this sample the question about head injury was excluded from the Index. The Accident Index score ranged from 0 (no accidents) to 10 (all endorsed as ‘‘more than once’’).
Complex Stop Signal Paradigm This computerized paradigm involves a go task and a stop task (Logan, 1994). The go task is a choice reaction time task which requires subjects to discriminate between two go signals, an X and an O. The frequencies of X and O presentations were even across the task. The stop task requires subjects to inhibit their response to the go task on 25% of go-signal trials. Six blocks of trials with 24 go-signal and 8 stop-signal trials were administered. The first block of trials was considered a practice block. Therefore, the data from the first block is not included in the analyses that follow. The presentation of stop signals was distributed randomly for each block. The stop signal was a tone (1000 or 500 Hz) delivered through headphones. Subjects were instructed to stop responding when presented with one of these tones and to ignore the other tone. This instruction was counterbalanced across subjects so that half of them were told to inhibit their response to the high tone and half were told to inhibit their response to the low tone. This task, which lasted approximately 15 min was administered to each subject individually by a member of the research team.
RESULTS
Subjects were matched on age, sex, and stop-signal tone (high or low). Paired t-tests were conducted for continuous data and χ 2 tests were used for categorical data. The MHI subjects showed a slower stop-signal reaction time (SSRT) than the control (C) subjects (t ⫽ 2.46; p ⬍ .05). Figure 1 displays the SSRT data for each matched pair in the data set, along with the line of best fit for each of the groups. Although there is an outlier in the MHI group (defined as a subject with an SSRT greater than three standard deviations above the mean), the group differences remain when the data are analyzed without this subject (t ⫽ 2.23; p ⬍ .05). There were no differences between the groups on the remaining stop signal variables (see Table 1). The MHI subjects differed from the C subjects on their Accident score, with the MHI subjects reporting significantly more accidents than the C subjects (p ⬍ .05). There was no difference between subjects on handedness, level of education, or presence of learning disabilities (see Table 1). Thirty-one of the subjects responded to the question inquiring about the length of time of unconsciousness. Subjects were divided into three groups
414
NOTES AND DISCUSSION
FIG. 1. SSRT data for each matched pair in the data set, along with the line of best fit for each of the groups.
depending on their answers to this question. Group 1 consisted of subjects who had been unconscious for ‘‘minutes,’’ Group 2 consisted of subjects who had been unconscious for ‘‘hours,’’ and Group 3 consisted of subjects who had been unconscious for ‘‘days.’’ An analysis of variance (ANOVA) was conducted on these data and it was found that there was no significant difference in SSRT across these groups. An analysis of the Pearson r correlation between the number of accidents reported and stop-signal reaction time was conducted for all subjects. These variables were not significantly correlated (r ⫽ ⫺.11, p ⬎ .05). DISCUSSION
The major finding of the current study was that individuals who reported having had a MHI had a slower SSRT and a higher Accident Index than those individuals with no MHI. It is important to note that the MHI group did not differ from the C group on reaction time to the ‘‘go’’ signal. This suggests that they were not simply slow in responding to all signals. Moreover, the differences that were found cannot be accounted for by other variables such as handedness, reported learning disabilities, or level of education. These results replicate the findings of earlier research that was completed using a similar paradigm (Stewart & Tannock, 1999). Notwithstanding the imposed limitations of this study (i.e., retrospective design and lack of information about preinjury status), these replicated results are provocative and demonstrate a difference in cognitive function, specifically in inhibitory control, in individuals who report having had a MHI. It is interesting to note, once again, that these subjects had presumably recovered from their injuries
415
NOTES AND DISCUSSION
TABLE 1 Comparison of Group Means (n ⫽ 29) Variable/Group
Mean
SD
SSRT MHI
260.93
131.03
C Go RT MHI
208.86
72.34
459.04
128.66
C % Correct HI
452.80
106.67
97.62
3.68
C Accident Index HI
97.92
4.10
1.26
1.27
C Education level HI
0.79
1.22
6.02
2.95
C Handedness (Proportion of sample) HI
6.82
2.72
C Learning Difficulties (Proportion of sample) HI
0.22
t Value(χ 2)
2.46*
0.30
0.36
2.08*
1.84
0.26 (χ 2) 2.29
0.17 (χ 2) 0.42
C
0.05
* p ⬍ .05.
and did not report any long-lasting effects of their injuries to the examiners. Unfortunately, the design of this research project does not allow speculation about whether or not the MHI subjects displayed difficulties with inhibition before their injuries or whether the differences in inhibitory control were a result of the MHI. Further research that uses a prospective design, a more precise definition of MHI, and ascertains details about the accidents, the nature of the injury, and recovery phases in order to answer these questions, is indicated. REFERENCES Jennett, B. 1989. Some international comparisons. In H. S. Levine, H. M. Eisenberg, & A. L. Benton (Eds.), Mild head injury (Pp. 23–34). Oxford Univ. Press: New York.
416
NOTES AND DISCUSSION
Levin, H. S., High, W., Ewing-Cobbs, L., et al. 1988. Memory functioning during the first year after closed head injury in children and adolescents. Neurosurgery, 22, 1043–1052. Logan, G. D. 1994. On the ability to inhibit thought or action: A user’s guide to the stop signal paradigm. In D. Dagenbach & T. H. Carr (Eds.), Inhibitory processes in attention, memory, and language Pp. 189–239. San Diego: Academic Press. Logan, G., Schachar, R., & Tannock, R. 1997. Impulsivity and inhibitory control. Psychological Science, 8, 60–64. Robertson, I. H., Manly, T., Andrade, J., et al., 1997. ‘Oops!’: Performance correlates of everyday attentional failures in traumatic brain injured and normal subjects. Neuropsychologia, 35, 747–758. Satz, P., Zaucha, K., McCleary, C., et al., 1997. Mild head injury in children and adolescents: A review of studies 1970–1995. Psychological Bulletin, 122, 107–131. Schachar, R., & Logan, G. G. 1990. Impulsivity and inhibitory control in normal development and childhood psychopathology. Developmental Psychology, 26, 710–720. Stewart, J. L., & Tannock, R. 1999. Mild head injury and inhibitory control. Brain and Cognition, 40, 1–284 (abstract).