Reduced locus coeruleus integrity linked to response inhibition deficits in parkinsonian disorders

Parkinson's disease and progressive supranuclear palsy (PSP) both impair response inhibition, exacerbating impulsivity. Inhibitory control deficits vary across individuals, and have been linked with worse prognosis and lack of improvement on dopaminergic therapy. Motor and cognitive control are associated with noradrenergic innervation of the cortex, arising from the locus coeruleus noradrenergic system. Here we test the hypothesis that loss of structural integrity of the locus coeruleus explains response inhibition deficits in progressive supranuclear palsy and Parkinson's disease. This cross-sectional observational study recruited 24 people with idiopathic Parkinson's disease, 14 with PSP-Richardson's syndrome, and 24 age- and sex-matched controls. All participants undertook a stop-signal task and ultrahigh field 7T-magnetic transfer weighted imaging of the locus coeruleus. Hierarchical Bayesian estimation of the parameters of 'race models' of go- versus stop-decisions was used to quantify the cognitive processes of response inhibition. We tested the multivariate relationship between locus coeruleus integrity and model parameters using partial least squares. Both disorders impaired response inhibition at the group level. Progressive supranuclear palsy caused a distinct pattern of abnormalities in inhibitory control, relative to Parkinson's disease and healthy controls, with a paradoxically reduced threshold for go responses, but longer non-decision times, and more lapses of attention. The variation in response inhibition correlated with variation in the integrity of the locus coeruleus, across participants in both clinical groups. Structural imaging of the locus coeruleus, coupled with behavioural modelling in parkinsonian disorders, confirms that locus coeruleus integrity is associated with response inhibition and its degeneration contributes to neurobehavioural changes. The noradrenergic system is therefore a promising target to treat impulsivity in these conditions. The optimisation of noradrenergic treatment is likely to benefit from stratification according to locus coeruleus integrity.


Introduction
Progressive supranuclear palsy (PSP) and Parkinson's disease (PD) have distinct neuropathology and clinical features 1,2 , but they can both cause cognitive and behavioural problems 3 . One of the consequences is an impairment of response inhibition, contributing to disinhibited behaviour and impulsivity. These are associated with poor clinical outcomes [4][5][6] .
Impulsivity is a multi-faceted behavioural construct, including abnormal sensitivity to reward, intolerance to delayed reward and a failure to inhibit inappropriate responses 7,8 . These deficits are mediated by abnormalities in complementary frontostriatal neural circuits and neurochemical systems 9 . Here we focus on response inhibition, as the execution of responses represents a point of convergence for upstream changes in cognition and behavioural decisions. Deficits in inhibitory control arise from pathology in frontostriatal circuits, including the ventrolateral prefrontal cortex, pre-motor cortex, caudate and sub-thalamic nuclei, and their monoaminergic regulation. Noradrenergic manipulations influence response inhibition, especially the cancellation of a response, in preclinical models [10][11][12] , healthy humans 13 , attention deficit disorders 14,15 and Parkinson's disease [16][17][18][19][20] . They are therefore a promising route to ameliorate response inhibition deficits in diverse neurological and psychiatric disorders.
The locus coeruleus is the main source of noradrenaline in the brain and an early site of pathology in progressive supranuclear palsy and Parkinson's disease [21][22][23][24] . Locus coeruleus degeneration is severe in symptomatic progressive supranuclear palsy and Parkinson's disease, although post mortem studies indicate a high degree of variability in locus coeruleus cell loss 24,25 . The heterogeneity in locus coeruleus damage has been linked to variability in the response to drugs that increase noradrenergic transmission such as atomoxetine 26 . To facilitate more targeted treatment of impulsivity in progressive supranuclear palsy and Parkinson's disease, it is necessary to quantify locus coeruleus structural integrity in vivo and determine its relationship to response inhibition. Specialist magnetic resonance imaging (MRI) sequences for ultrahigh field scanners (7T) have enabled sensitive and well-tolerated quantification of locus coeruleus pathology [25][26][27][28] . The resolution is sufficient to examine regional effects of pathology within the locus coeruleus 29,30 .
Noradrenergic deficits may influence response inhibition at both motor and cognitive (decisional) levels, given their widespread cortical projections 31 . Response inhibition can be . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint measured using performance on a stop signal task, but singular parameters of performance may obscure the complexity of underlying decision mechanisms 9,16,18 . A multivariate model of computational parameters of response inhibition overcomes this limitation, to distinguish motor, attentional and decisional components of inhibition 26,32,33 .
In this study, we used complementary parametric analyses of the stop signal task, estimated with hierarchical Bayesian models, to advance our understanding of the response inhibition deficits in progressive supranuclear palsy and Parkinson's disease. The first model explains response accuracy and reaction times as function of a race between three processes: a stop process, a go process for the response that matches the choice stimulus, and a go process for the response that mismatches the choice stimulus [34][35][36] . To decompose the cognitive mechanisms of response inhibition, we also parameterised accumulation-to-threshold models in the second type of model, in which evidence accumulates stochastically until a go response threshold is reached. The mean time required to trigger the stop process served as the estimate of the stop signal reaction time (SSRT). The models also estimate attentional failures to trigger the stop and go processes.
We tested the hypothesis that these parameters of response inhibition relate to locus coeruleus structural integrity at an individual patient level, as measured in vivo using 7T MRI. We predicted that the response inhibition deficits characterising progressive supranuclear palsy and Parkinson's disease would be associated with reduced locus coeruleus integrity. Given the multivariate nature of the model parameters and the topographic organisation of the locus coeruleus, we investigated the multivariate relationship between locus coeruleus integrity and response inhibition.

Methods and Materials Participants
Fourteen patients with probable PSP-Richardson's syndrome (MDS 2017 criteria), 24 with idiopathic Parkinson's disease (UK Parkinson's disease Brain Bank criteria), and 24 age-and sex-matched healthy controls were included in the study. Controls did not use psychoactive medications and exclusions criteria for all participants included history of stroke, severe . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint medical co-morbidity, and any contraindications to 7T MRI. None of the patients met criteria for impulse control disorders, based on clinical impression and/or the Questionnaire for Impulsive-Compulsive Disorders. Participants were not demented, based on a Mini-Mental State Examination score >26 and clinical impression. All patients with Parkinson's disease and 10/14 progressive supranuclear palsy patients were on dopaminergic medications (Table   1). Eighteen patients with Parkinson's disease were part of a single-dose, placebo-controlled, crossover drug study where the relationship between locus coeruleus integrity and drug responsiveness was investigated 26 . The behavioural performance of these patients was examined for the placebo session, and potential placebo or practice effects were explicitly modelled (see Statistical Analyses). All participants underwent a structured clinical, cognitive, and behavioural assessment (

Stop-signal task
Response inhibition was measured using a stop-signal paradigm where a two-choice reaction time (RT) 'go' task was occasionally interrupted by a 'stop signal'. For the go task, participants were instructed to indicate the direction of a black arrow presented at the centre of the screen by pressing a left or right button. On stop trials, the arrow changed to a red colour in conjunction with a tone (i.e., the stop signal) and participants were instructed to withhold their initiated response (Fig 1, top-left). The stop signal occurred after a variable delay (i.e., the stop-signal delay, SSD), the length of which was determined by an adaptive staircase method. The SSD ranged from 50 ms to 1500 ms and increased or decreased by 50 ms after a successful or failed stop trial, respectively. Further details of the task design are provided in O'Callaghan et al. 26 .

Modelling of response inhibition
Parametric race models We applied two complementary models to the stop-signal task data. The first model assumed that the finish time distributions of the stop and go processes followed ex-Gaussian distributions 37 , with mean µ and standard deviation σ of the Gaussian component, and mean τ of the exponential component. The mean finish time of each process was estimated as the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint mean of the corresponding ex-Gaussian distribution, which is given by µ + τ . The mean finish times of the stop process and matching go process were taken as the stop signal reaction time (SSRT) and go RT, respectively.
The second model was similar, except that the finish time distributions of the go processes were assumed to follow shifted Wald distributions 36 . The Wald distribution describes the first-passage time distribution of a single-boundary diffusion process, where evidence accumulates stochastically at a positive mean rate (cf. the drift rate) until a threshold is reached. A subject-specific constant non-decision time shifts the lower bound of the distribution to account for peripheral processes, such as stimulus encoding and motor output.
The shifted Wald distribution therefore enables a process model that explains how the go RT distributions were generated.
Both models additionally included the probabilities of attentional failures related to the stop process (trigger failure) and go processes (go failure). Thus, the first model featured 11 free parameters: three ex-Gaussian parameters for each of the three processes, and the trigger and go failure probabilities. The second model featured 9 free parameters: separate drift rates for the matching and mismatching go processes, a threshold and non-decision time that were shared across the go processes, three ex-Gaussian parameters for the stop process, and the trigger and go failure probabilities.
Hierarchical Bayesian modelling was used to fit each model to the observed task data, separately for the progressive supranuclear palsy, Parkinson's disease, and control groups.
Markov Chain Monte Carlo (MCMC) sampling methods were used to estimate the posterior distributions of all group-and participant-level parameters. We assigned broad ("weakly informative") priors on the group-level means and standard deviations of the model parameters (Table S1). Sampling convergence was confirmed by visual inspection of the time series plots of the MCMC samples, and by the potential scale reduction statistic R-hat (< 1.1 for all parameters). The absolute goodness of fit was assessed by visually comparing the observed data to simulated data generated from the model's posterior predictive distribution ( Fig.S5-7).
Model fitting was performed using the Dynamic Models of Choice (DMC) toolbox 38 implemented in R (version 3.6.1). The number of sampling chains was set to three times the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint number of free parameters. Automated procedures were used to continue sampling until convergence was reached (h.run.unstuck.dmc and h.run.converge.dmc functions in the DMC toolbox). After this, an additional 500 iterations were obtained for each chain to create a final posterior distribution for each parameter, which was used for statistical analyses.

MRI acquisition and processing
Participants were scanned on a 7T Magnetom Terra (Siemens, Erlangen, Germany) with a 32-channel receive head coil (Nova Medical, Wilmington, USA). Following the acquisition and processing protocol described previously 26,27 , the locus coeruleus was imaged using a individual T1 to T1 group template and finally T1 group template to the ICBM152 template.
The co-registered MT images were converted to contrast-to-noise (CNR) maps by subtracting the mean and dividing by the standard deviation of the signal in a central pontine reference region 26,27 . A probabilistic locus coeruleus atlas was applied on the CNR maps with a conservative threshold (25%) to extract voxel-wise locus coeruleus CNRs for later statistical tests and slice-wise means for group comparisons. Structural T1-weighted images were subjected to FreeSurfer (v 6.0) recon-all pipeline with -highres and -brainstem options. The . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint resulting total intracranial and brainstem volumes were used for the estimation of global and local atrophy, respectively.

Group differences in response inhibition parameters
The go error rate was defined as the proportion of go trials with an incorrect response, including commission errors (responses that mismatched the choice stimulus) and omission errors (missing responses). The stop accuracy rate was defined as the proportion of stop trials with successfully inhibited (i.e. missing) responses. Each of these measures served as dependent variables in an ANOVA with group as the between-subjects factor. Differences between specific groups were then examined with post-hoc Tukey's tests.
We focused on group differences by examining the posterior distributions of the group-level means of the race model parameters. For each posterior distribution, we took the median as the posterior estimate, and the 95% quantile interval (QI) as the range of plausible values. We derived posterior distributions for group contrasts by subtracting the set of MCMC samples of the two groups under consideration. For each group contrast, we computed the probability of direction (P dir ) as an index of the presence of an effect 41 . This measure indicates the proportion of the contrast's posterior distribution that is strictly positive or negative (whichever is the most probable). Note that P dir can be directly interpreted as the probability that a group difference is non-zero, and is therefore not subject to a particular significance threshold. To examine individual differences in mechanisms of response inhibition, we extracted the medians of all participant-level posterior distributions.

The relationship between locus coeruleus integrity and response inhibition
We studied the relationship between locus coeruleus integrity and response inhibition deficits with a two-level analytical strategy 42,43 . First, the multidimensional relationship was examined using partial least squares (PLS) 42,44 with in-house Matlab (R2018b) scripts where pairs of latent variables were computed from all response inhibition parameters (LV inhibition ) and voxelwise locus coeruleus contrast (LV LC ). The PLS multivariate method is particularly . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint suitable given the ratio of variables to participants 45 . The significant pair of latent variables was identified using a permutation test (10,000 iterations, P < 0.05).
Second, individual loadings on LVs were subjected to linear regression models to confirm the group-wise relationship between locus coeruleus integrity and response inhibition ability.
Subsequent regression models examined the effects of nuisance covariates including age, global and local atrophy, disease duration and motor severity. To mitigate the potential placebo and/or practice effects in the 18/24 of the patients with Parkinson's disease who had completed the task as part of a drug study 26 , we adopted two approaches using (i) a categorical variable indicating session (0 = first / only session; 1 = second session) and (ii) a categorical indicator for drug order (0 = one session without placebo; 1 = first session on placebo; 2 = second session on placebo). Adding these categorical indicators as a covariate of no interest in the regression models did not meaningfully change the relationship between locus coeruleus integrity and response inhibition (see results). Prior to analysis, continuous variables were z-scored and categorical variables (including nuisance covariates) were assigned sum-to-zero contrasts.

Basic task performance
Participant characteristics and clinical summary data are presented in Table 1. Groups were similar by age, sex and education, while the patient groups were similar by disease duration and motor severity, although cognitive function was lower in progressive supranuclear palsy.
There were expected significant main effects of group on both the stop accuracy rate ( Fig.2A; . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Locus coeruleus integrity and individual differences
The locus coeruleus integrity was reduced in the caudal region for both patient groups ( Fig.S1-S4; Fig.5C). Partial least squares analysis (PLS) of the response inhibition parameters and voxel-wise locus coeruleus contrast identified a single significant pair of latent variables (r = 0.75, P = .040, 10,000 permutations).
The locus coeruleus latent variable (LV LC ) expressed negative loadings throughout the structure, to a variable degree across sub-regions (Fig.5A). The response inhibition latent variable (LV inhibition ) expressed positive loadings on the SSRT, go and trigger failure probabilities, and non-decision time, and expressed negative loadings on the drift rate and response threshold of the go process (Fig.5B). Thus, higher scores on this behavioural latent variable reflected impaired response inhibition with prolonged SSRT, greater probability of attentional failures, reduced response threshold, lower drift rate, and longer non-decision time.
Mean go RTs had negligible loading on the LV inhibition , confirming that go RTs were not related to the cognitive mechanisms of response inhibition.
A regression analysis examined the relationship between response inhibition and locus coeruleus integrity, and potential group differences. The LV inhibition participant score was the dependent variable, and the LV LC participant score, group, and their interaction were independent variables. There was a significant overall relationship between the LV LC scores and the LV inhibition scores ( Fig.5D; ß = -0.24, F(1, 56) = 7.23, P = .009; BF = 6.95). This suggests that individuals with reduced locus coeruleus integrity have more severe deficits in response inhibition. There was a significant main effect of group on the LV inhibition scores (F (2, 56) = 29.88, P < .001; BF = 1.98 10 7 ), reflecting impaired response inhibition in the progressive supranuclear palsy group compared to the control group (t (56) = 7.46, P < .001) and PD group (t (56) = 6.33, P < .001), regardless of locus coeruleus CNR. There was no significant interaction effect between the LV LC scores and group (F (2, 56) = 0.31, P = .734; BF = 0.15), suggesting that the slope between the LV LC and LV inhibition scores is similar across groups.
We confirmed the robustness of these results with four additional regression analyses. First, adding age, gender, and years of education to the regression model as covariates of no interest did not meaningfully change the relationship between the LV LC and LV inhibition scores . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint (ß = -0.23, F (1, 53) = 6.31, P = .015; BF = 6.99), nor did the inclusion of additional covariates, including disease duration, motor severity, and local and global brain atrophy. Model selection procedures consistently identified a relatively sparse model as the optimal account of the data, retaining only the LV LC scores and group as predictors of LV inhibition (Table S2-5).
Third, re-running the regression analysis with only the Parkinson's disease and progressive supranuclear palsy groups yielded a similar relationship between the LV LC and LV inhibition scores (ß = -0.26, F (1, 34) = 4.90, P = .034; BF = 2.07; Table S6). Fourth, accounting for potential placebo and / or practice effects in the Parkinson's disease group did not meaningfully change the results (Table S7-8).
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Discussion
This study confirms the hypothesis that response inhibition deficits in Parkinson's disease and progressive supranuclear palsy are linked to reduced structural integrity of the locus coeruleus, the principal source of cerebral noradrenaline. Diminished response inhibition (i.e., multivariate disinhibition-related parameters) was associated with reduced locus coeruleus integrity (i.e., multivariate voxelwise loadings) across all groups, in keeping with psychopharmacological and pre-clinical studies 10,11,[16][17][18][19][20]46 .
The race models identified latent variables to explain the behavioural performance. Despite similar mean reaction times in the go responses, there were disease-specific patterns underlying abnormal inhibitory control. Patients with progressive supranuclear palsy had a reduced response threshold, consistent with a paradoxical bias towards committing go responses as proposed by Zhang et al. 32 , noting that the current study was manual not oculomotor. The progressive supranuclear palsy group also had a slower non-decision time, suggesting they required more time for sensory encoding and execution of motor outputs.
There was evidence for a lower drift rate among patients with progressive supranuclear palsy relative to patients with Parkinson's disease and controls, confirming the slower accumulation of evidence to reach a decision.
Task performance might also have been influenced by attentional problems, which are a common cognitive feature in parkinsonian disorders, especially in progressive supranuclear palsy 3 . To account for this possibility, we included attentional failures related to the stop and go processes (i.e., trigger failure and go failure) in the race models of the stop signal task.
These indices revealed that patients with progressive supranuclear palsy had greater attentional deficits than Parkinson's disease and control groups, albeit not to the extent that prevented them from correctly executing the task. The attentional impairments during the Stop/Go paradigm were also captured within the main PLS-derived LV inhibition variable that related to the LV LC component reflecting locus coeruleus structural integrity.
The results support the hypothesis that the decision threshold is reduced in progressive supranuclear palsy to compensate for the impairment in evidence accumulation (reduced drift rate) and execution (slower non-decision time). This threshold compensation is liable to change the quality of decisions, and thereby promote impulsivity. This relationship is disease . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint agnostic and relates to the level of pathology with the locus coeruleus which is more severely affected in progressive supranuclear palsy 25 .
The variability of locus coeruleus degeneration across patients in our study is also evident post mortem 21,22,24 , consistent with what is observed in other neurodegenerative disorders such as Alzheimer's disease, corticobasal degeneration and Lewy Body dementia [47][48][49][50] . This highlights the potential of locus coeruleus imaging as a trans-diagnostic marker, to understand individual differences in cognition beyond the classic nosological borders and diagnostic criteria 51 .
On average, locus coeruleus degeneration is more severe in progressive supranuclear palsy than Parkinson's disease, and our in vivo imaging data confirms this, particularly in the caudal portions 25 . This distribution is consistent with neuropathological studies reporting greater degeneration in the caudal locus coeruleus 52 . However, the response inhibition deficits spanning progressive supranuclear palsy and Parkinson's disease topographically map to the mid-caudal and rostral locus coeruleus, that innervate the forebrain regions associated with response inhibition and impulsive behaviour 30 .
The locus coeruleus-noradrenergic system's influence on response inhibition and impulsivity may be non-linear (e.g. a U-inverted shape function), and involve multiple brain networks 10,11,53 . The locus coeruleus diffusely projects to many sites within the brain where noradrenaline has a state-dependant effect on the neuronal input-output gain function 54 .
Although we did not directly measure noradrenaline transmission or noradrenergic receptor density, locus coeruleus structural integrity is a proxy index of noradrenergic function 51,55 .
Our study focused on noradrenergic contributions to response inhibition deficits in parkinsonian disorders. However, we recognise that noradrenergic projections from the locus coeruleus have secondary pharmacological interactions with other neurotransmitter systems, including dopamine and GABA. For example, dopamine and noradrenaline can be coreleased from the same LC-noradrenergic terminals 56-58 . Locus coeruleus activity can alter midbrain dopamine cell firing 59 and directly participate in the regulation of dopamine release in hippocampus 60 . Nevertheless, pharmacological studies using selective dopamine manipulations have found no effects on response inhibition [61][62][63] , and there is no clear evidence for a relationship between response inhibition and the levodopa equivalent daily dose or the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint on/off state of dopaminergic medication in Parkinson's disease 64,65 . Taken together, our results are consistent with a growing body of work that indicates a robust link between the locus-coeruleus noradrenaline system and response inhibition, although further work is needed to elucidate the potential contribution of dopaminergic mechanisms.
However, we anticipate that any future use of noradrenergic treatments that target response inhibition deficits would be adjunctive to standard dopaminergic therapy, and not an alternative. Therefore, all patients were tested on their usual clinically optimised dopaminergic medication. We acknowledge that patients' task performance may have differed if they were taken off their dopaminergic medication, which consequently could have affected the relationship between response inhibition and locus coeruleus integrity.
Our study has several limitations. We acknowledge that all the patients were diagnosed with clinical criteria, without pathological confirmation. Misclassification of progressive supranuclear palsy subtypes and other atypical parkinsonian syndromes can occur for PSP-Richardson's syndrome diagnosis 66 and future studies including post-mortem confirmation would be critical to enhance the diagnostic accuracy. Data from some patients with Parkinson's disease are drawn from a placebo-controlled drug study 26 . Only the placebo data were analysed, but we acknowledge that this might have resulted in heterogeneity due to effects of placebo expectancy and/or task practice. To mitigate these issues, we used two approaches to explicitly model the impact of placebo/practice confounds in the statistical analyses. Reassuringly, neither of these steps meaningfully altered the results.
To conclude, our study further elucidates the role of the locus coeruleus noradrenergic contribution in response inhibition and its impairment in Parkinson's disease and Progressive Supranuclear Palsy. We propose that locus coeruleus imaging could be used as a heuristic stratification marker in clinical trials, targeting response inhibition deficits with noradrenergic drugs in those most likely to benefit. Individual differences in response to drug are marked 26 , due to variation in disease severity, allelic variations in the noradrenaline transporter, polymorphisms in the CYP2D6 liver enzyme catabolising noradrenergic drugs, baseline brain networks structure or function, and the integrity of the locus coeruleus 16,17,20,26,46,67,68 .
Optimisation of noradrenergic treatments will benefit from better understanding the mechanisms of response inhibition and their relationship to the integrity of the locus coeruleus.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Disclosures
The authors declared no conflict of interest.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021.  is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint Figure 1 | Schematic representation of data analysis pipeline. The trial-by-trial stop signal task performance was subjected to a two parametric race model following ex-Gaussian and shifted Wald distributions. An array of behavioural parameters were estimated hierarchically from the models for both stop and go response, including stop-signal reaction time (SSRT), go reaction time (go RT), trigger failure, go failure, drift rate (v), response threshold (B) and non-decision time (t0). These parameters altogether provided more mechanistic understanding of response inhibition. The locus coeruleus (LC) integrity was assessed by computing voxel-wise contrast-to-noise ratio (CNR) and extracted using an independent LC probability atlas. The multivariate relationship between LC integrity and response inhibition was then examined using partial least squares using resulted behavioural and imaging matrices from previous data processing steps. Significant pairs of latent variables were identified with the permutation test. The contribution of LC in response inhibition was finally confirmed in linear regression models with individual subject loading scores on the inhibition latent variable as dependent variable, loading scores on the LC latent variable, group and nuisance covariates as predictors.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint Figure 5 | The relationship of locus coeruleus integrity and response inhibition. As confirmed using partial least squares, a significant pair of latent variables was identified between voxel-wise LC contrast and response inhibition parameters estimated from cognitive models. The spatial distribution of voxelwise LC loadings (A) and individual loadings on response inhibition latent variable (B) were presented. Negative LC loadings were associated with positive loadings on stop-signal reaction time (SSRT), go failure (gf), trigger failure (tf) and non-decision time (t0), and negative loadings on drift rate (v) and response threshold (B). This suggested that impaired response inhibition is linked with reduced LC integrity seen in both PD and PSP patients (C). An overall relationship of LC integrity and response inhibition was further confirmed in the linear regression model consistent across all groups as supported by a significant main effect of LC when including the group predictor in the model (D). Individual fitted lines for each group were presented with curved areas indicating 95% confidence intervals.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted October 16, 2021. ; https://doi.org/10.1101/2021.10.14.21264996 doi: medRxiv preprint