Abstract
Primary Objective:
The current study examined how creative divergent thinking (i.e., the ability to produce varied and original solutions to a problem) is impacted by moderate-to-severe traumatic brain injury (TBI).
Research Design:
Descriptive, observational.
Methods and Procedures:
We administered two tasks of divergent thinking, the Abbreviated Torrance Test for Adults (ATTA) and the Alternative Uses Test (AUT), as well as a battery of neuropsychological tests and psychosocial variables (assessing memory and learning, processing speed, set shifting and psychological distress), to 29 individuals with TBI and 20 demographically-matched healthy comparison participants.
Main Outcomes and Results:
Individuals with TBI performed similarly to healthy individuals on both tests of creative thinking, although they were impaired on the neuropsychological tasks. Moreover, there was no significant correlation between performance on the ATTA and performance on neuropsychological tests, but within the TBI group AUT performance and memory were significantly and positively associated.
Conclusions:
Our findings reveal that divergent thinking, as measured by the ATTA and AUT, might be spared following moderate-to-severe TBI. These findings further our understanding of the higher-level cognitive sequelae of TBI and suggest that divergent thinking might be leveraged during treatment planning.
Keywords: Traumatic brain injury, creativity, divergent thinking, convergent thinking, problem solving
Introduction
Traumatic Brain Injury (TBI) is a leading cause of long-term cognitive sequelae across multiple domains, with over two million new cases every year in the United States alone (1). While a large body of work has examined deficits in traditional neuropsychological domains (e.g., memory, language, attention), less is known about how TBI can influence multi-domain cognitive phenomena, such as creativity.
Creativity can be broadly divided into convergent thinking, which is the ability to identify a specific solution to a problem with deliberate focus, and divergent thinking, which is the ability to generate varied and original open-ended ideas (2–4). These two cognitive constructs are distinct and can be assessed with different types of assessments (4). For instance, convergent thinking can be measured using the Remote Associate Task (RAT) (5), in which participants are given three cue words (e.g., cottage, swiss, cake), and are asked to identify a fourth word that is associated with all three cues (e.g., cheese). Divergent thinking, on the other hand, has traditionally been measured using tasks such as the Alternative Uses Task, in which participants are asked to generate original uses for everyday objects (e.g., using a newspaper to swat flies), or using the Torrance Tests of Creative Thinking, which assesses the ability to imagine unusual situations and describe their possible consequences, and the ability to draw figures based on stylized templates (6).
Creativity is a universal human attribute that plays a crucial role in everyday life (7); divergent thinking in particular is related to psychological health and is important for problem solving, planning and executing original ideas, and adapting to one’s environment by finding alternative solutions to challenges and obstacles (3). Indeed, in the context of populations with TBI, who might need to devise and utilize alternative and compensatory strategies in their daily lives, creative divergent thinking is an especially important skill. To date, one study has investigated creativity in individuals with TBI, and it focused on convergent thinking in individuals with TBI (8). In that study, the researchers found that individuals with TBI were significantly less likely to produce a correct response on the Remote Associate Test than healthy comparison participants, even though the number of attempted responses was not significantly different (i.e., group differences could not be exclusively attributed to differences in processing speed). This suggests that convergent thinking is likely to be impaired following moderate-to-severe TBI. However, much less is known about divergent thinking in TBI.
There have been reports of individuals displaying the emergence of artistic divergent thinking skills de novo following both TBI and damage to the temporal and frontal lobes (9,10). However, there has not been a systematic examination of how TBI can affect divergent thinking performance in those individuals who do not begin to display artistic, or other marked forms of creativity abilities following their injury. Studies examining focal lesions have found that patients with bilateral hippocampal damage performed significantly worse than comparison participants on both the verbal and figural portions of the Torrance Tests of Creative Thinking (11). Moreover, a recent systematic review of neuroimaging findings reported that both brain hemispheres are involved in divergent thinking, and in particular central, parietal, and temporal regions (12). However, relatively little is known about how the type of injuries that usually accompanies TBI can affect divergent thinking. A deeper knowledge of how divergent thinking is affected by TBI can aid clinicians in characterizing the sequelae of brain injury on complex cognitive domains, help identify which domains should be targeted by therapeutic plans, as well as suggest possible skills that might be leveraged during the treatment phase.
The goal of the current study was to investigate divergent thinking in a population of individuals with TBI. To this end, we administered two tests of divergent thinking – the Alternative Uses Task and the Abbreviated Torrance Test for Adults – to a sample of individuals with a history of moderate-to-severe TBI and to demographically-matched healthy comparison participants. Our first aim was to investigate the presence of a group difference in the performance on those two tasks. As no previous study has investigated divergent thinking in TBI, we tested the non-directional hypothesis that the performance of individuals with TBI would differ from that of healthy individuals. Our second, exploratory aim was to investigate the association between performance on divergent thinking tasks and on neuropsychological assessments measuring domains such memory and learning, set shifting, processing speed, and psychological wellbeing.
Methods
Participants
Participants were 29 individuals with moderate-to-severe TBI and 20 healthy comparison (HC) participants. The sample was identical to the one described in a previous paper from our group (8). Individuals with TBI were recruited through the University of Iowa Brain Injury Registry, while healthy participants were recruited from the Iowa City community. Brain injury severity was determined based on the Mayo Classification System (13), with TBI injuries considered moderate-to-severe when at least one of the following criteria were met: (1) Glasgow Coma Scale (GCS) <13 (i.e., moderate (<13) or severe (<8) according to the GCS), (2) positive acute computed tomography findings or lesions visible on magnetic resonance imaging, (3) loss of consciousness (LOC)>30 minutes or (4) post-traumatic amnesia (PTA) >24 hours. Injury-related information was acquired through medical records and semi-structured interviews with the patient. Causes of injury were falls (17), motor vehicle accidents (MVA; 9), assaults (2), and sports injuries (2). One participant had experienced two separate TBIs (a fall and an MVA). Location of damage was available for 14 participants, with several participants having injuries in different lobes, and included frontal (8), temporal (6), parietal (3), occipital (4), and the brainstem (1).
Participants were free of aphasia, as confirmed by an aphasia quotient higher than 93.8 on the Western Aphasia Battery (WAB) (14) or assessment by a speech language pathologist.
The TBI and HC groups were matched for age (t (47) = 1.22, p =.227), sex (X2(1, N = 49) = .65, p = .419) and years of education (t(47) = 1.50, p = .141) (Table 1).
Table 1.
Participant demographics.
| N | AGE (Mean±SD) | SEX (Females) | EDUCATION (Mean±SD) | CHRONICITY (Months, Mean±SD) | |
|---|---|---|---|---|---|
| HC | 20 | 55.63 ± 16.1 | 8 | 15.1 ± 1.9 | N/A |
| TBI | 29 | 49.88 ± 16.2 | 12 | 14.17 ± 2.1 | 91.7 ± 133.1 |
| Group Differences (p) | N/A | .227 | .419 | .141 | N/A |
HC = Healthy comparison participants, TBI = Traumatic brain injury, p = p-value, SD = Standard Deviation, N/A = Not Applicable.
All activities were approved by the University of Iowa Institutional Review Board.
Measures of divergent thinking
Divergent thinking was measured using the Abbreviated Torrance Test for Adults (ATTA) (15) and the Alternative Uses Test (AUT) (16).
ATTA
The ATTA is a shortened version of the Torrance Tests of Creative Thinking. It was chosen over the original version of the test because of its abbreviated format and its ease of administration with clinical populations with fatigue problems. The ATTA is a valid (alpha = .72) and reliable test (r = .78) (15) and it has been used to assess creative divergent thinking in studies on both healthy and clinical populations. It requires approximately 15 minutes to administer; it is a paper and pencil assessment, composed of three activities, each with a 3-minute limit. In Activity 1, participants are asked to imagine that they can walk on air or fly, and to list the problems that they might encounter. In Activity 2, two incomplete figures are presented, and the respondent is asked to draw pictures centered around them, in order to make them as unusual as possible. Each incomplete figure is composed of a few lines (parallel or intersecting), and the two figures are presented side by side. In Activity 3, participants are presented with an array of nine triangles (3 by 3) and asked to draw pictures of one or more objects incorporating them.
The test yields several dependent variables. Those used for the current study were: (1) Scores for the 4 norm-referenced Abilities: Fluency (the ability to produce quantities of ideas that are relevant to the task instructions), Originality (the ability to produce ideas that are new, or uncommon, or unique), Elaboration (the ability to embellish ideas with details), and Flexibility (the ability to process information or objects in different ways given the same stimulus). (2) A cumulative measure, the Creativity Index (CI), the sum of the scaled scores for the four abilities and additional points that participant can obtain for 15 different creativity indicators (e.g., Richness and Colorfulness of Imagery, Emotions/Feelings, Fantasy, Humor, etc). For further information on how the Creativity Index is measured, see: (15). (3) The Creativity Level, a rescaling by reduction of the Creativity Index which ranges from 1 (minimal) to 7 (substantial); this is an ordinal variable. Test forms were sent to the publisher for standardized scoring.
AUT
The AUT is a test designed to measure flexibility of thinking. This task is divided into two parts, each consisting of three items that are presented together with their common uses (e.g., a newspaper, which is used for reading). For each item (shoe, button, key, wooden pencil, automobile tyre, and eye-glasses) participants are asked to list uses for each object (or parts of the object) that is different from the common use (e.g., a newspaper could be used to swat flies). Participants are allowed 4 minutes to complete each part of the task, for a total of 8 minutes (16). The AUT is valid (loading .72 on a divergent-thinking factor) and reliable (r = .67) (16).
Each response was scored by two independent, trained raters who were blinded (inter-rater reliability measured with Cohen’s kappa was .70, for a total of 1060 items). The raters then discussed the items on which the ratings diverged until an agreement was reached. The dependent variable was the total number of correct responses (i.e., number of responses that were decided to represent alternate uses of the item by both raters). In addition, given that the task was timed and individuals with TBI tend to have lower processing speed than HCs, we also investigated the presence of a group difference in the number of attempted responses.
Neuropsychological measures
To further characterize the sample of the current study, as well as to examine the relationship between performance on the ATTA and the AUT and other cognitive domains, participants were administered measures of (1) verbal learning and memory: the California Verbal Learning Test (CVLT-Immediate), short-delay verbal recall (CVLT-Short Delay) and long-delay verbal recall (CVLT-Long Delay) (17); (2) processing speed: the processing speed index of the Wechsler Adult Intelligence Scale – Fourth edition (WAISPSI) (18), computed using the Symbol Search and Coding subtests; (3) set shifting: the Trail Making Test B (Trails B) (19) (both Trails A and B were administered, but A was not examined, as it is a measure of processing speed; the results of Trails B were transformed in Z scores to account for age (20)); and (4) psychological distress and psychiatric disorders: the Brief Symptoms Inventory (BSI) (21). Due to logistical issues during data collection, data for the BSI were not collected for one participant in the HC group.
Statistical analysis
Assumptions of normality and homogeneity of variances were tested. To examine group differences in ATTA performance, we used a mixed design analysis of variance (ANOVA) with group as between-factor (TBI vs. HC) and ATTA ability as within-factor (Fluency, Originality, Elaboration, and Flexibility). We were particularly interested in investigating the presence of a group effect, as well as group-by-Ability interaction. Group differences on the Creativity Index were examined using a two-tailed t-test, while differences in Creativity Level were examined using a Mann-Whitney U test, as Creativity Level is an ordinal variable.
To test group differences on the AUT we carried out a two-tailed independent sample t-test.
Group differences on neuropsychological tasks were determined using one-tailed t-tests, with the hypothesis that individuals with TBI would underperform HCs. Two participants with TBI were unable to complete the Trails B within the allotted test time (300 seconds) and testing was discontinued. Thus, to avoid outlier-driven spurious statistical results, their scores were Winsorized. Multiple comparison correction was carried out using the Benjamini–Hochberg (BH) false discovery rate (FDR) procedure (BH-corrected p values are reported, as well as effect sizes using Cohen’s d). We chose an FDR of .05.
The second, exploratory aim of the current study was to determine the association between neuropsychological performance and performance on the ATTA. To do so, we used within-group Pearson correlations, employing the BH FDR procedure to correct for multiple comparisons. For correlational analyses, the overall Creativity Index was used (as opposed to the four scaled abilities), as we did not have specific hypotheses regarding the relationship between specific aspects of divergent creative thinking and other neuropsychological measures, and were interested in determining whether divergent thinking as a whole is related to specific neuropsychological constructs. We also examined the relationship between AUT and ATTA and demographic variables (age, education, as well as time since injury for the TBI group), and the specific correlation between Flexibility and the Trails B, as the two tasks assess similar underlying cognitive constructs.
We also examined the relationship between the ATTA and the AUT using one-tailed Pearson’s correlations, as we hypothesized that in both groups performance on two tests of divergent thinking would be significantly and positively correlated with each other.
Results
ATTA
A mixed-design ANOVA revealed no significant effect of Group (F1,47 = .82, p = .369, ηp2 = .02), nor a Group-by-Ability interaction (F3,141 = .55, p = .651, ηp2 = .01). In other words, individuals with TBI did not perform differently from HCs on any of the ATTA Abilities scores. There was, however, a significant effect of Ability (F3,141 = 10.93, p < .001, ηp2 = .19), indicating that across both groups the scores differed based on the ability measured. In particular, the highest scores were obtained for Originality (16.45 ± 2.56), followed by Elaboration (15.14 ± 2.7), Fluency (14.71 ± 2.52) and Flexibility (14.55 ± 2.7) (Figure 1). Paired sample t-tests revealed that across groups Fluency scores were significantly lower than Originality scores (t(48) = −2.42, p < .001, d = .73), but did not significantly differ from Elaboration (t(48) = −.43, p = .207, d = .18) and Flexibility (t(48) = .16, p = .519, d = .09). Originality scores were significantly higher than both Elaboration (t(48) = 1.31, p = .002, d = .48) and Flexibility (t(48) = 1.9, p < .001, d = .63). Elaboration did not significantly differ from Flexibility (t(48) = .592, p = .082, d = .25). There were also no significant group differences on the Creativity Index (T(47) = .72, p = .474, d = .22; HC MEAN±SD = 72.6 ± 12.49; TBI MEAN ±SD = 69.86 ± 12.75) (Figure 2) or on the Creativity Level (U = 271, p = .695; HC MEDIAN = 5; TBI MEDIAN = 4, r = .07) (Figure 3).
Figure 1.
Group differences on the Abbreviated Torrance Test for Adults.
Group comparison for the ATTA. A mixed-design ANOVA revealed no significant effect of group, nor a Group-by-Ability interaction. There was, however, a significant effect of Ability. The error bars represent the standard error of the mean.
Figure 2.
Group differences on Creativity Index.
Group comparison for the ATTA’s Creativity Index (no significant group difference).
Figure 3.
Group differences on Creativity Level.
Group comparison for the ATTA’s Creativity Level (no significant group difference).
AUT
A two-tailed t-test revealed no significant difference in performance on the AUT both when examining the number of correct responses (T(47) = 1.56, p = .125, d = .46; HC MEAN±SD = 13.3 ± 4.94; TBI MEAN±SD = 10.59 ± 6.6) and the number of attempted responses (T(47) = −.41, p = .686, d = .04; HC MEAN±SD = 20.4 ± 7.4; TBI MEAN±SD = 21.35 ± 8.33) (Figure 4).
Figure 4.
Group differences on the Alternative Uses Test.
Group comparison for the AUT. There was no significant group difference for the number of attempted responses, nor for the number of correct responses.
Relationship between the ATTA and the AUT
Performance on the AUT (number of correct responses) and the ATTA’s Creativity Index were significantly and positively correlated both in the HC group (r = .67, n = 20, p < .001) and in the TBI group (r = .62, n = 29, p < .001).
Neuropsychological measures
Individuals with TBI significantly underperformed HCs on the following neuropsychological tests: the WAIS-PSI (t(47) = 2.83, BH p = .006, d = .82), the Trails B (t(45.57) = 4.19, BH p < .001, d = 1.16), the CVLT-immediate (t(47) = 2.34, BH p = .021, d = .67), and the CVLT-short delay (t(47) = 2.01, BH p = .034, d = .58). Moreover, individuals with TBI had significantly higher scores on the BSI (t(30.85) = −3.85, BH p < .001, d = 1.17). There were no significant differences in performance on the CVLT-long delay (t(47) = 1.58, BH p = .066, d = .46) (Table 2).
Table 2.
Mean scores and group comparison for the neuropsychological assessments.
| HC (Mean±SD) | TBI (Mean±SD) | Group Comparison (p) | Effect size | |
|---|---|---|---|---|
| WAIS-PSI | 111.1 ± 15.17 | 98.97 ± 14.44 | .006* | d =.82 |
| Trails B (Z score) | .79 ± 1.02 | −.9 ± 1.8 | <.001* | d = 1.16 |
| CVLT-immediate (t score) | 57.6 ± 10.36 | 49.86 ± 12.7 | .021* | d =.67 |
| CVLT-short delay (Z score) | .55 ± 1.02 | −.08 ± 1.17 | .034* | d =.58 |
| CVLT-long delay (Z score) | .35 ± 1.03 | −.16 ± 1.2 | .066 | d =.46 |
| BSI | 43.83 ± 8.9 | 52.52 ± 6.71 | .001* | d = 1.17 |
HC = Healthy comparison participants, TBI = Traumatic brain injury, p = p-value determined using the Benjamini–Hochberg false discovery rate procedure, SD = Standard Deviation, d = Cohen’s d, WAIS-PSI = Processing Speed Index, CVLT = California Verbal Learning Test, BSI = Brief Symptoms Inventory, Trails B = Trail Making Test B.
Correlations with neuropsychological measures and demographic variables
ATTA:
Within both the HC and the TBI groups, Creativity Index was not significantly correlated with any of the neuropsychological measures (all BH p > .05) (Table 3). When we examined the relationship between Flexibility and the Trails B we found that they were significantly correlated within the Comparison group (r = .51, N = 20, p = .011), but not within the TBI group (r = .20, N = 29, p = .148).
Table 3.
Correlations between the ATTA Creativity Index and neuropsychological measures.
| HC r, p | TBI r, p | |
|---|---|---|
| WAIS-PSI | .02,.463 | .43,.053 |
| Trails B (Z score) | .39,.108 | .10,.306 |
| CVLT-immediate (t score) | .47,.108 | .45,.053 |
| CVLT-short delay (Z score) | .37,.108 | .23,.170 |
| CVLT-long delay (Z score) | .22,.261 | .21,.170 |
| BSI | −.06,.463 | −.43,.170 |
ATTA = Abbreviated Torrance Test for Adults; HC = Healthy comparison participants, TBI = Traumatic brain injury, p = p-value determined using the Benjamini–Hochberg false discovery rate procedure, r = Pearson’s r, WAIS-PSI = Processing Speed Index, CVLT = California Verbal Learning Test, BSI = Brief Symptoms Inventory, Trails B = Trail Making Test B. All p-values determined using the Benjamini–Hochberg false discovery rate procedure (.05) were higher than.05 and non-significant.
There was no significant correlation between the Creativity Index and age, education, or time since injury within the TBI or the comparison group (all p > .125, all r < .27).
AUT:
Within the HC group, there were no significant correlations between any of the neuropsychological measures and number of correct responses on the AUT (all BH p > .05) (Table 4). Within the TBI group, AUT performance was significantly correlated with performance on the CVLT-Immediate (r = .57, N = 29, BH p = .006) and on the CVLT-Short (r = .51, N = 29, BH p = .009). All other correlations were non-significant.
Table 4.
Correlations between the AUT and neuropsychological measures.
| HC r, p | TBI r, p | |
|---|---|---|
| WAIS-PSI | −.01,.481 | .36,.051 |
| Trails B (Z score) | .26,.481 | .28,.083 |
| CVLT-immediate (t score) | −.05,.481 | . 57,.006* |
| CVLT-short delay (Z score) | .16,.481 | .51, 009* |
| CVLT-long delay (Z score) | −.08,.481 | .35,.051 |
| BSI | .07,.481 | .16,.204 |
AUT = Alternate Uses Task; HC = Healthy comparison participants, TBI = Traumatic brain injury, p = p-value determined using the Benjamini–Hochberg false discovery rate procedure, r = Pearson’s r, WAIS-PSI = Processing Speed Index, CVLT = California Verbal Learning Test, BSI = Brief Symptoms Inventory, Trails B = Trail Making Test B.
denotes p-values determined using the Benjamini–Hochberg false discovery rate procedure that were lower than.05.
There was no significant correlation between the number of correct response on the AUT and age, or with time since injury within the TBI or the comparison group (all p > .286, all r < .11). There was, however, a significant positive correlation between education and AUT within the TBI group (r = .49, n = 29, p = .003).
Discussion
In the current study, we examined how creative divergent thinking is affected by moderate-to-severe TBI using two established tasks, the AUT and the ATTA. We found that individuals with TBI did not perform differently from healthy comparison participants on either of the tasks. In addition, we examined the relationship between performance on neuropsychological tasks measuring set shifting, memory and learning, processing speed, and divergent thinking, and found that, while the ATTA was not significantly correlated with any of the measures, within the TBI group the AUT was significantly associated with verbal learning and memory. We discuss each finding below.
Although a previous study revealed that individuals with moderate-to-severe TBI perform worse than healthy individuals on the RAT, a test of creative convergent thinking (8), here we found that individuals with TBI were not significantly impaired on two separate tests of divergent thinking. For the ATTA, there was no significant Group effect and no Group-by-Ability interaction: this indicates that individuals with TBI did not perform significantly below comparison participants, and that this was the case across all four Ability scales (Fluency, Originality, Elaboration, and Flexibility). This remained true when additional creativity indicators were taken into consideration (i.e., when Creativity Index and Creativity Level were assessed). Similarly, individuals with TBI did not attempt significantly fewer responses on the AUT, and they did not produce significantly fewer correct responses. This suggests that individuals with moderate-to-severe TBI are not impaired on tasks of divergent thinking.
The lack of group differences constitutes a rare and surprising finding, considering that the literature on moderate-to -severe TBI is dominated by studies that have found impairment across numerous neuropsychological and cognitive domains (22–26). Moreover, it supports the account that creative divergent and convergent thinking are separate and distinguishable constructs, and one can be impaired while the other is spared. However, other considerations should also be made. For instance, the figures for the ATTA and the AUT (Figures 2 and 4) show large degrees of individual variability both within the TBI and the HC sample, with some individuals performing very well and others performing more poorly. Thus, careful observation of the data reveal that, while individuals with TBI might not be significantly impaired on a test of divergent thinking, a larger sample size or more in-depth testing (e.g., using the long version of the Torrance Tests of Creative Thinking) might uncover group differences. Furthermore, given that this is the first study to comprehensively examine divergent thinking in TBI, it is possible that the measures used here (the AUT and the ATTA) may not be sensitive to detecting deficits in this domain in populations with TBI. It is also possible that different results would be obtained using the long version of the Torrance Test of Creative Thinking, as the validity and reliability of the ATTA was calculated exclusively using data collected in healthy populations, and thus we cannot be certain that this test is sensitive enough to measure creativity in a clinical population. Still, it should also be kept in mind that the sample of individuals with TBI described here are representative of the literature, in that they displayed significant deficits in several cognitive domains that are commonly impaired following TBI (memory and learning, set shifting, and processing speed), while performing like healthy comparison participants on the two divergent thinking tasks, suggesting that the latter might be a relative strength for individuals with TBI.
Performance on the ATTA and on the AUT were significantly correlated, with a medium effect size. As both tests measure creative divergent thinking, and one of the activities of the ATTA (number 1) is structured very similarly to the AUT, it is not surprising that scores on these tasks would be correlated. However, it is notable that although in the TBI group performance on the ATTA was not significantly correlated with any of the neuropsychological tasks, the correlations between the ATTA and processing speed, CVLT-immediate, and psychological distress showed medium to large effect sizes (27). This suggests that it is possible that a similar study in a larger sample would reveal significant correlations between these neuropsychological measures (positive in the case of processing speed and CVLT-immediate, negative in the case of psychological distress) and creativity, as assessed by the ATTA. In the TBI group, the AUT was associated with two measures from the CVLT. This might be due to the more verbal nature of the AUT compared to the ATTA (in both activity 2 and 3 participants are asked to draw from incomplete figures or triangles), and to the fact that the CVLT is specifically a measure of verbal learning and memory. It is possible that other neuropsychological measures of memory that do not have a heavy verbal component (e.g., the Rey Complex Figure Test) would show higher correlation with the ATTA and lower correlation with the AUT. Moreover, the significant correlation between CVLT and AUT confirms previous work that has found a relationship between declarative memory and divergent thinking (11,28). While not statistically significant, the correlation between the AUT and the CVLT-long delay showed a moderate effect size, as did the correlation between the AUT and processing speed.
It is also notable that there were negligible correlations between executive functioning (measured by the Trail Making Test B) and divergent thinking, especially considering that two of the scales of the ATTA measure Flexibility and Fluency. However, it should be kept in mind that our executive functioning assessments only measured set shifting, and that tests of letter and semantic fluency might correlate more highly with the ATTA or with one or more of its subscales. Flexibility, in particular, which measures a construct that is very similar to mental shifting, did correlate with the Trails B within the comparison group, but not within the TBI group; other measures of verbal fluency (such as the Controlled Oral Word Association Test) or of set-shifting (such as the Wisconsin Card Sorting Test) might yield different results, and future studies should incorporate a larger battery of neuropsychological tests to better investigate the relationship between divergent thinking and executive functioning. Lastly, there were no significant correlations between processing speed and divergent thinking performance, even though all tasks were timed. This is also supported by the fact that individuals with TBI, who as a group display significantly lower processing speed scores than comparison participants, average the same number of responses on the AUT, and thus do not seem penalized by the time limit. Examination of effect sizes offers further insight: both the AUT and the ATTA are timed tasks, the presence of moderate to large associations between processing speed and performance within the TBI group. but not within the HC group, is consistent with the TBI group’s lower performance on the WAIS-PSI, and suggests that individuals with TBI whose processing speed is more impaired might perform more poorly because of time constraints. It should certainly be kept in mind, when interpreting the various correlations, that the relatively small sample size might have led to low power, and that significant correlations between some neuropsychological measures and divergent thinking might emerge with a larger sample size and a better powered study. Thus, further work is necessary to determine the neuropsychological correlates of creativity in TBI.
In conclusion, the current study found that individuals with TBI are not impaired on two tests of creative divergent thinking when compared to healthy individuals, but also revealed large within-group individual variability in performance. Our analysis of the association between demographic variables revealed that education may explain some of this variability, especially in relation to the AUT, but other demographic variables (such as age, or time since injury) were not significantly associated with creativity. More work, with larger samples and additional assessments of divergent thinking, is necessary to thoroughly investigate how creative thinking is impacted by TBI and to determine if and how multi-domain abilities such as creativity are tied to, or predict, long-term outcomes above and beyond the strength of individual cognitive domains (e.g., memory, attention). Our preliminary findings suggest that divergent thinking might be spared in some individuals with TBI who show impairments in other neuropsychological domains, and that it could be leveraged during rehabilitation and within the therapeutic setting.
Footnotes
Disclosure statement
The authors report no declarations of interest.
References
- 1.Roozenbeek B, Maas AI, Menon DK. Changing patterns in the epidemiology of traumatic brain injury. Nat Rev Neurol. 2013;9 (4):231–36. doi: 10.1038/nrneurol.2013.22. [DOI] [PubMed] [Google Scholar]
- 2.Jaarsveld S, Lachmann T, van Leeuwen C. Creative reasoning across developmental levels: convergence and divergence in problem creation. Intelligence. 2012;40(10):172–88. doi: 10.1016/j.intell.2012.01.002. [DOI] [Google Scholar]
- 3.Runco MA. Children’s divergent thinking and creative ideation. Dev Rev. 1992;12(3):233–64. doi: 10.1016/0273-2297(92)90010-Y. [DOI] [Google Scholar]
- 4.Abraham A. The promises and perils of the neuroscience of creativity. Front Hum Neurosci. 2013;7:246. doi: 10.3389/fnhum.2013.00246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mednick SA. The associative basis of the creative process. Psychol Rev. 1962;69:220–32. doi: 10.1037/h0048850. [DOI] [PubMed] [Google Scholar]
- 6.Torrance EP. Predictive validity of the torrance tests of creative thinking. J Creat Behav. 1972;6(4):236–62. doi: 10.1002/j.21626057.1972.tb00936.x. [DOI] [Google Scholar]
- 7.Runco MA. Creativity. Annu Rev Psychol. 2004;55:657–87. doi: 10.1146/annurev.psych.55.090902.141502. [DOI] [PubMed] [Google Scholar]
- 8.Rigon A, Reber J, Patel NN, Duff MC. Convergent thinking and traumatic brain injury: an investigation of performance on the remote associate test. Brain Inj. 2018;32(9):1110–14. doi: 10.1080/02699052.2018.1483031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Midorikawa A, Kawamura M. The emergence of artistic ability following traumatic brain injury. Neurocase. 2015;21(1):90–94. doi: 10.1080/13554794.2013.873058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pollak TA, Mulvenna CM, Lythgoe MF. De novo artistic behaviour following brain injury. Front Neurol Neurosci. 2007;22:75–88. [DOI] [PubMed] [Google Scholar]
- 11.Duff MC, Kurczek J, Rubin R, Cohen NJ, Tranel D. Hippocampal amnesia disrupts creative thinking. Hippocampus. 2013;23 (12):1143–49. doi: 10.1002/hipo.22208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Runco M, Yoruk S. The neuroscience of divergent thinking. Activitas Nervosa Superior. 2014;56(1–2):1–16. doi: 10.1007/BF03379602. [DOI] [Google Scholar]
- 13.Malec JF, Brown AW, Leibson CL, Flaada JT, Mandrekar JN, Diehl NN, Perkins PK. The mayo classification system for traumatic brain injury severity. J Neurotrauma. 2007;24(9):1417–24. doi: 10.1089/neu.2006.0245. [DOI] [PubMed] [Google Scholar]
- 14.Shewan CM, Kertesz A. Reliability and validity characteristics of the Western Aphasia Battery (WAB). J Speech Hear Disord. 1980;45(3):308–24. doi: 10.1044/jshd.4503.308. [DOI] [PubMed] [Google Scholar]
- 15.Althuizen N, Wierenga B, Rossiter J. The validity of two brief measures of creative ability. Creat Res J. 2010;22(1):53–61. doi: 10.1080/10400410903579577. [DOI] [Google Scholar]
- 16.Guilford JP, Christensen PR, Merrifield PR, Wilson RC. Alternative uses manual. Sheridan Supply Co; 1960. Menlo Park, (CA). [Google Scholar]
- 17.Delis DC, Freeland J, Kramer JH, Kaplan E. Integrating clinical assessment with cognitive neuroscience: construct validation of the California verbal learning test. J Consult Clin Psychol. 1988;56 (1):123–30. doi: 10.1037/0022-006X.56.1.123. [DOI] [PubMed] [Google Scholar]
- 18.Holdnack JA, Xiaobin Z, Larrabee GJ, Millis SR, Salthouse TA. Confirmatory factor analysis of the WAIS-IV/WMS-IV. Assessment. 2011;18(2):178–91. doi: 10.1177/1073191110393106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gordon NG. The Trail Making Test in neuropsychological diagnosis. J Clin Psychol. 1972;28(2):167–69. doi:. [DOI] [PubMed] [Google Scholar]
- 20.Tombaugh TN. Trail Making Test A and B: normative data stratified by age and education. Arch Clin Neuropsychol. 2004;19(2):203–14. doi: 10.1016/S0887-6177(03)00039-8. [DOI] [PubMed] [Google Scholar]
- 21.Derogatis LR, Melisaratos N. The Brief Symptom Inventory: an introductory report. Psychol Med. 1983;13(3):595–605. doi: 10.1017/S0033291700048017. [DOI] [PubMed] [Google Scholar]
- 22.Azouvi P, Arnould A, Dromer E, Vallat-Azouvi C. Neuropsychology of traumatic brain injury: an expert overview. Rev Neurol (Paris). 2017;173(7–8):461–72. doi: 10.1016/j.neurol.2017.07.006. [DOI] [PubMed] [Google Scholar]
- 23.Stierwalt JA, Murray LL. Attention impairment following traumatic brain injury. Semin Speech Lang. 2002;23(2):129–38. doi: 10.1055/s-2002-24989. [DOI] [PubMed] [Google Scholar]
- 24.Whiting MD, Baranova AI, Hamm RJ. Cognitive impairment following traumatic brain injury In: Levin ED, Buccafusco JJ, editors. Animal models of cognitive impairment. Boca Raton (FL); 2006. https://www.ncbi.nlm.nih.gov/pubmed/21204363 [PubMed] [Google Scholar]
- 25.Arciniegas DB, Held K, Wagner P. Cognitive impairment following traumatic brain injury. Curr Treat Options Neurol. 2002;4 (1):43–57. doi: 10.1007/s11940-002-0004-6. [DOI] [PubMed] [Google Scholar]
- 26.McDowell S, Whyte J, D’Esposito M. Working memory impairments in traumatic brain injury: evidence from a dual-task paradigm. Neuropsychologia. 1997;35(10):1341–53. doi: 10.1016/S0028-3932(97)00082-1. [DOI] [PubMed] [Google Scholar]
- 27.Cohen J. Statistical power analysis for the behavioral sciences. New York: Routledge Academic; 1988. [Google Scholar]
- 28.Addis DR, Pan L, Musicaro R, Schacter DL. Divergent thinking and constructing episodic simulations. Memory. 2016;24 (1):89–97. doi: 10.1080/09658211.2014.985591. [DOI] [PMC free article] [PubMed] [Google Scholar]