Advancements in neuroimaging techniques over the course of the last 20 years have driven clinico-anatomical correlation studies in relation to aphasia resulting from lesions of the subcortical region. Consequently, subcortical aphasia represents a prevailing clinical entity which has served to challenge established localisationist theory. A number of models of subcortical participation in language were developed on the basis of these findings, largely established upon a corpus of data relating to populations with subcortical lesions of vascular origin. The attestation of these paradigms, however, has been fundamentally encumbered by an incapacity to strictly define spontaneous vascular neuropathology, and the acknowledged limitations of conventional neuroimaging techniques in reliably identifying the true extent of cerebral damage.
A recent resurgence in stereotactic neurosurgical procedures, involving the inducement of
circumscribed subcortical lesions, to treat the motor sequelae associated with Parkinson's disease (PD), now provides an opportunity to subject these theoretical constructs to rigorous hypothesis testing. Henceforth, the aims of the current thesis were threefold: 1) to investigate the impact of surgically-induced functional inhibition of the globus pallidus, thalamus and subthalamic nucleus (STN) on language function, utilising a comprehensive linguistic test battery; 2) to discuss post-operative language profiles in relation to operative theories of subcortical participation in language; and 3) to ratify and/ or extend available schemas relative to proposed roles for the basal ganglia and thalamus in language.
A total of 15 stereotactic neurosurgical PD, together with 16 matched nonsurgical PD (NSPD) and 16 non-neurologically impaired (NC) subjects participated in this research. All subjects were administered a series of off-line language assessments,
comprised of general and high-level linguistic indices, utilising repeated measures experimental designs. The first of 7 investigations (Chapter 2) reported the effects of bilateral posteroventral pallidotomy (PVP) (N = 6) on language function, as compared to the group of NSPD and NC subjects. The results obtained revealed significant postoperative declines in performance on 3 specific measures, namely phonemic fluency (p < .05), semantic ambiguity resolution (p < .05) and confrontation naming (p < .01). The exclusive nature of the significant changes observed was interpreted with caution, by reason of inherent performance heterogeneity within Parkinsonian cohorts, prompting supplemental data analysis. Reliable change indices (RCI's) were subsequently utilised as a means of investigating clinically consequential change over time, respecting individual bilateral PVP and NSPD subject
profiles, relative to demonstrated variance within the NC group. Within this context, a significantly higher proportion of reliable declines in performance were demonstrated by the bilateral PVP group, in comparison to the NSPD group, on the Boston Naming Test (BNT) (p < .05) and tool fluency (p < .05). Furthermore, case by case analyses relative to performance on the above tasks revealed a higher incidence of reliable change within the bilateral PVP group, dominated by declines, in contrast to a matched subcohort (n = 6) of NSPD subjects, who consistently demonstrated improvements in performance over time. Despite these trends, stark intersubject variability regarding magnitudes and directions of reliable change was identified both within and between groups, necessitating the future employment of case study designs relevant to this population.
In response to this directorate. Chapter 3
presented the plenary linguistic profiles of 2 bilateral PVP subjects, prior to and at 3 and 12 months following surgery, affording a detailed investigation of the short and long term effects of pallidal lesions on language function. The findings proffered indicated that bilateral PVP has the potential to exert long term effects on linguistic functioning, up to 1 year post-operatively. Furthermore, heterogeneous profiles with respect to middle point magnitudes and directions of reliable change were demonstrated between subjects, and primary symptom complex appeared to carry weight respecting performance stability, relative to linguistic outcomes over time. Discernible changes (i.e., improvements and decrements) were, however, typically restricted to tasks demanding complex linguistic operations. A propos of dyskinetic symptoms, bilateral PVP effected an initial overall improvement in complex language function, which diminished over time, possibly indicative of a decrescendo effect
relative to surgical beneficence. In contrast, akinetic symptoms were associated with a more stable performance profile, albeit defined by higher proportions of reliable decline versus improvement in post-operative scores across testing phases.
Progressing along the procession of basal ganglia-thalamocortical circuitry components. Chapter 4 reported the effects of dominant and non-dominant hemisphere ventral intermediate (Vim) thalamotomy (THAL) on language function in 2 cases. The findings summarised indicated a discernable laterality effect with respect to middle point magnitudes and directional proportions of change, with left Vim THAL evoking greater antagonistic outcomes corresponding to post-operative high-level linguistic abilities. Thalamotomy within the non-dominant hemisphere also influenced complex language function, but to a lesser extent, symptomatic of potential interhemispheric effects relative to thalamocortical interactions.
The experiments conducted in Chapters 6 through 8 aimed to empirically validate extensions to operative models of subcortical participation in language proffered in Chapter 5, which proposed a role for the STN in the mediation of linguistic processes. Chapter 6 investigated the impact of bilateral STN deep brain stimulation (DBS) (N = 5) on language function. Non-parametric comparisons failed to identify significant post-operative changes in general or high-level language abilities, as compared to the NSPD and NC groups. When RCI's were employed as a measure of clinically consequential change, however, a significantly (p < .05) higher proportion of reliable improvement on the TWT-R composite score was demonstrated by the bilateral STNDBS group, as compared to NSPD controls. When case by case analyses relative to individual performance on the TWT-R were examined, the bilateral STNDBS subjects demonstrated a trend towards
reliable improvement across subtests, in contrast to a matched subcohort (n = 5) of NSPD subjects who generally demonstrated no reliable change in performance over time. Despite these trends, stark performance heterogeneity was again evident both within and between groups, mandating the need for detailed case study reports.
Chapter 7 subsequently reported the plenary linguistic profiles of 2 cases prior to and at 3 and 12 months following bilateral STNDBS, also affording consideration of the long term effects of DBS on language. Overall, a cumulative increase in the proportion of reliable post-operative improvements in performance demonstrated, again typically confined to high-level linguistic variables, was exhibited by both subjects across testing phases. These results suggested that bilateral STNDBS may serve to enhance the proficiency of basal ganglia-thalamocortical linguistic circuits over time.
reported in Chapter 8 specifically tested the hypothesis of STN laterality, by way of investigating the impact of unilateral STNDBS/ ablation within the dominant and non-dominant hemispheres on language function. The 2 cases presented demonstrated similar fluctuations in post-operative high-level linguistic performance, characterised by an overall median decline in function associated with a higher proportion of reliable declines versus improvements on relevant variables. These findings suggested that the STN may represent a component of an interhemispheric high-level language processing system.
Overall, the findings presented in this thesis served to validate as well as to extrapolate many of the precepts proffered within operative theories of subcortical participation in language, specifically those constructs proposing integration within frontal lobe language systems. The results provided evidence to suggest that the concatenation of subcortical nuclei
comprising basal ganglia-thalamocortical circuits play a fundamental role in high-level linguistic processes, potentially underpinning the recruitment and directed interplay of frontal and temporoparietal cortical regions. Intersubject variability was largely attributed to potential additive and subtractive postlesional effects within the context of neuropathologically/ functionally distinct basal ganglia circuitry. The marriage of multi-modal neuroimaging techniques to the aforementioned postulates was recognised as an imperative direction for future research.