Rehabilitation of endogenous task shift processes in closed head injury patients

Abstract

This paper reports a study that was aimed to rehabilitate executive functions in CHI patients. When a subject is engaged in two speeded tasks, not simultaneously but with some form of alternation, the response is slower to an item of task A if it was preceded by an item of task B, than when it was preceded by an item of task A. This shift cost is small when subjects can prepare in advance for the new task (endogenous task shift), whereas the cost is much greater when preparation is not possible (exogenous task shift). The groups tested comprised 20 severe closed head injury (CHI) patients (10 who underwent treatment and 10 controls), 8 mild CHI patients, and 18 non-brain damaged (NBD) controls. In the present study, the shift cost was greater for severe CHI patients than for NBD controls. Treatment consisted of five sessions, in which an endogenous task shift paradigm was used. A significant reduction of the endogenous shift cost from assessment to retest was found. The reduction remained stable at the 4-month follow-up session. These results are not simply due to retesting, as the control patients did not show any improvement at retest. Interestingly, no reduction of exogenous task shift cost was found. The results showed also that the beneficial effect of the treatment generalises to other executive functions.

Portions of these data were presented at the Nineteenth European Workshop on Cognitive Neuropsychology, Bressanone (Italy), 21–26 January 2001, and at the Second International Vipiteno/Sterzing Conference on Neuropsychological Rehabilitation after Brain Injury, Vipiteno (Italy), 9–12 June 2004.

Acknowledgments

This research was supported by a Univeristy of Padova Grant to Franca Stablum (Progetti di Ateneo Padova 2003; CPDA035783). Sandro Bettella is recognised for the technical support he provided. We would like to thank the people working at the Servizio di Psicologia, Ospedale di Bolzano (Italy), and Servizio di Neuropsicologia, Ospedale San Bortolo di Vicenza (Italy), for allowing us to test their patients. We are also indebted with the NBD controls and patients that agreed to participate in our study. Also, we thank John Whyte for suggesting many improvements to the original paper.

Notes

Portions of these data were presented at the Nineteenth European Workshop on Cognitive Neuropsychology, Bressanone (Italy), 21–26 January 2001, and at the Second International Vipiteno/Sterzing Conference on Neuropsychological Rehabilitation after Brain Injury, Vipiteno (Italy), 9–12 June 2004.

¹Although the effect size of the differences on time since injury was rather large, it should be noted that the treated patients were the most distant to the time of injury, and therefore the probability of a bias in the results that confounds treatment effects with spontaneous recovery seems rather low (see Table 1).

²Stablum et al. (2000) found that the dual task cost correlated with the Wisconsin Card Sorting Test (number of category achieved: r = .627, p < .005; number of perseverative errors: r = .548, p < .01) and PASAT (r = .591, p < .005).

³The reason for these missing data is that these subjects' native language was German and the PASAT German version, at the time of testing, was not available to us. We believe the likelihood of bias in the data is not high.

⁴These subjects were not tested on BADS for time constraints in the testing sessions.

⁵We report only the analyses with position as a two level factor. That is because the analyses conducted on all the positions (positions 1 to 4 for the 4-item sequences and positions 1 to 10 for the 10-item sequences) do not add further relevant information and do not invalidate the results of the analyses reported.

⁶The CHI control group was not tested on dual task or BADS because of time constraints in the testing sessions.

⁷A note of caution should be added. The improvement found on dual task and BADS may still be due to practice: Untreated subjects were not administered these tasks.