Old dog, new tricks: the attentional set-shifting test as a novel cognitive behavioral task after controlled cortical impact injury

Abstract

Cognitive impairment associated with prefrontal cortical dysfunction is a major component of disability in traumatic brain injury (TBI) survivors. Specifically, deficits of cognitive flexibility and attentional set-shifting are present across all levels of injury severity. Though alterations in spatial learning have been extensively described in experimental models of TBI, studies investigating more complex cognitive deficits are relatively scarce. Hence, the aim of this preclinical study was to expand on this important issue by evaluating the effect of three injury levels on executive function and behavioral flexibility performance as assessed using an attentional set-shifting test (AST). Isoflurane-anesthetized male rats received a controlled cortical impact (CCI) injury (2.6, 2.8, and 3.0 mm cortical depth at 4 m/sec) or sham injury, whereas an additional group had no surgical manipulation (naïve). Four weeks postsurgery, rats were tested on the AST, which involved a series of discriminative tasks of increasing difficulty, such as simple and compound discriminations, stimulus reversals, and intra- and extradimensional (ED) shifts. TBI produced accompanying impact depth-dependent increases in cortical lesion volumes, with the 3.0-mm cortical depth group displaying significantly larger injury volumes than the 2.6-mm group (p=0.05). Further, injury severity-induced deficits in ED set-shifting and stimulus reversals, as well as increases in total response error rates and total set loss errors, were observed. These novel findings demonstrate executive function and behavioral flexibility deficits in our animal model of CCI injury and provide the impetus to integrate the AST in the standard neurotrauma behavioral battery to further evaluate cognitive dysfunction after TBI. Ongoing experiments in our laboratory are assessing AST performance after pharmacological and rehabilitative therapies post-TBI, as well as elucidating possible mechanisms underlying the observed neuropsychological deficits.

FIG. 1.

(A) The attentional set-shifting Plexiglas apparatus, fitted with a removable divider separating the start box, which serves as a holding area between trials, and the digging terracotta pots defined by a pair of cues along two stimulus dimensions (an odor and the digging medium filling the pot; pictured in the arena are pots for the simple discrimination training session for medium: paper strips and felt pieces). (B) Digging terracotta pots illustrating stimulus pairs used throughout the attentional set-shifting test stages and the Honey Nut Cheerios reward (General Mills Cereals, Minneapolis, MN).

FIG. 2.

Mean (±standard error of the mean [SEM]) trials to reach criterion on simple discriminations tasks on the training day preceding testing. There were no significant differences among groups, whether subjected to TBI injury (TBI [2.6 mm], TBI [2.8 mm], and TBI [3.0 mm]) or controls (naïve and sham groups), on ability to complete training procedures with relevant dimension being either odor or medium. Data are expressed as mean trials to criterion±SEM (n=6–9/group; p>0.05).

FIG. 3.

Progression through the attentional set-shifting protocol stages renders comparable cognitive performance patterns regardless of the initial discriminative dimension (odor or medium), suggesting no injury-related bias against a sensory modality necessary for contingency rule acquisition. Data are expressed as mean trials to criterion±standard error of the mean (n=19–20/group). SD, simple discriminations; CD, compound discrimination; ID, intradimensional; ED, extradimensional.

FIG. 4.

Impact depth-dependent effect of controlled cortical impact (CCI) injury, ranging from 2.6 to 3.0 mm cortical deformation depth, on cognitive performance on the AST. Significantly more trials were required to reach criterion on the R1, ED, and R3 tasks (main effect of stage, ^#p<0.05, compared to the other stages, collapsed across surgical groups). CCI injury produced impairments in cognitive performance on the AST, manifested as an overall significantly higher number of trials to criterion across AST stages for both the TBI (2.8 mm) and TBI (3.0 mm) groups, compared to naïve, sham, and TBI (2.6 mm) deformation depth groups (all p<0.05). Newman-Keuls' post-hoc analyses for individual test stages determined that the CCI injury also induced significant performance deficits in the moderate and severe injury groups (2.8 and 3.0 mm, respectively), relative to naïve, sham (*p<0.05), and TBI (2.6 mm) groups (^{^}p<0.05) on the ED stage, as well as on the third reversal stage in the TBI (2.8 mm) group, compared to sham controls (*p<0.05). Data are expressed as mean trials to criterion±standard error of the mean (n=6–9/group). TBI, traumatic brain injury; SD, simple discriminations; CD, compound discrimination; ID, intradimensional; ED, extradimensional; AST, attentional set-shifting test; R1, first reversal; R2, second reversal; R3, third reversal.

FIG. 5.

Effect of controlled cortical impact (CCI) injury (2.6, 2.8, and 3.0 mm cortical deformation depth) on total response error rates in the AST. CCI produced a significant overall injury group effect (p=0.05), as well as significant individual stage group effect on R1, with TBI rats displaying higher rates of total response errors while learning the contingency rule. Specifically, the TBI (2.8 mm) rats displayed significantly higher total error rates on R1, compared to naïve rats (*p<0.05, Newman-Keuls' post-hoc tests). Data are expressed as mean trials to criterion±standard error of the mean (n=6–9/group). TBI, traumatic brain injury; SD, simple discriminations; CD, compound discrimination; ID, intradimensional; ED, extradimensional; AST, attentional set-shifting test.

FIG. 6.

(A) Panels are representative depictions of the lesion within each group at the level of the dorsal hippocampus. (B) Mean (±standard error of the mean) cortical lesion volume (mm³) 5 weeks after controlled cortical impact (CCI) injury at three levels of injury (2.6, 2.8, and 3.0 mm cortical deformation depth, left to right, respectively). CCI injury produced a significantly larger cortical lesion volume in the TBI (3.0 mm) group, in comparison to the TBI (2.6 mm), deformation depth group (*p=0.05). TBI, traumatic brain injury.