Subthalamic nucleus activity dissociates proactive and reactive inhibition in patients with Parkinson's disease
Introduction
The ability to inhibit unwanted actions is a hallmark of executive control. Current models of action suppression distinguish between reactive stopping, when subjects are required to inhibit an action in response to infrequent stimuli (Aron and Poldrack, 2006, Li et al., 2008) and proactive stopping, when subjects stand ready to inhibit forthcoming actions when necessary (Ballanger et al., 2009, Chikazoe et al., 2009). Functional magnetic resonance imaging (fMRI) studies have identified a prefrontal–subcortical network supporting both types of executive control processes, and established within this network the important involvement of the subthalamic nucleus (STN, Aron and Poldrack, 2006, Ballanger et al., 2009, Chikazoe et al., 2009).
Although proactive and reactive inhibitory processes are related, different predictions about their respective time course of STN activity can be made. One model postulates that during the stop signal task (SST), the STN participates in reactive stopping by relaying a “STOP” signal to quickly inhibit thalamo-cortical activation through the direct glutamatergic excitation of the globus pallidus pars interna, which in turn sends GABAergic inhibitory signals to the thalamus (Aron and Poldrack, 2006). In particular, this model predicts a phasic increase of STN activity triggered by a STOP cue at a latency that should precede the stop signal reaction time (SSRT, which provides an estimation of the amount of time necessary to process the stop signal and to inhibit the prepared response, Logan et al., 1984, Schall and Godlove, 2012, Verbruggen and Logan, 2008). Another model predicts that the STN is also involved during proactive inhibitory control by mediating a “hold your horse” signal to dynamically delay response execution, when subjects tend to favor accuracy over speed (Aron, 2011, Ballanger et al., 2009, Frank et al., 2007, van Maanen et al., 2011). Thus, the proactive inhibition hypothesis predicts a tonic change of STN activity, for example when subjects prepare to eventually stop their response in a near future. This type of behavior is typical during the stop signal task because subjects slightly slow-down their responses (compared to their maximal response speed) due to the unpredictability of stop cues. This type of proactive inhibitory control may reflect a modulation of a decision threshold in the STN (Aron, 2011, Ballanger et al., 2009, Chikazoe et al., 2009, Frank et al., 2007, Jahfari et al., 2010, van Maanen et al., 2011).
This study aimed at clarifying the temporal dynamics of the STN during proactive and reactive inhibition using local field potential (LFP) recordings of the STN performed in patients suffering from Parkinson's disease (PD). We tested the abovementioned model predictions using a version of the stop signal task (SST) modified to dissociate reactive inhibition and proactive inhibition, while controlling for a potential attentional confound (Chikazoe et al., 2009, Logan et al., 1984, Sharp et al., 2010). We determined the amount of time required to dissociate using LFPs the trials involving reactive stopping from motor execution trials. We also quantified STN involvement during proactive inhibitory control by contrasting trials involving a “preparation to stop” component and trials during which this component was absent. We focused the analysis of LFP signals on beta band activity (βA, [13–35 Hz]) because (i) this neural marker is a signature of the sensorimotor STN (Chen et al., 2006, Miyagi et al., 2009), (ii) βA is correlated to the akineto-rigidity syndrome (Zaidel et al., 2010), (iii) βA supports inhibitory processes (Alegre et al., 2012, Kuhn et al., 2004, Ray et al., 2011, Swann et al., 2009) and (iv) βA is correlated with irregular bursting activity in PD (Kuhn et al., 2005).
Section snippets
Participants
Data were obtained from 12 PD patients (5 males and 7 females, mean ± SEM age: 59 ± 2 y.o.), selected for their capacity and readiness of collaborating in a demanding cognitive task. They had no psychiatric comorbidity, and were undergoing bilateral STN implantation of deep brain stimulation (DBS) electrodes using standard surgical procedure and inclusion criteria for PD (Benabid et al., 2009). Table 1 summarizes the clinical characteristics of the patients. They had been suffering from idiopathic
Behavioral results
Patients performed the task accurately (mean success rate was 92 ± 2% in GO trials and 50 ± 3% in STOP trials.). As expected, the accuracy during STOP trials was significantly lower than during GO trials (Friedman's test performed on success rates with factor trial Type [GO, GF, GC, Unsuccessful STOP]; F(11,3) = 30.28, p < 0.0001; Dunn's post-hoc test: p = 0.0124). The accuracy observed during GC trials (85 ± 5%) and GF trials (96 ± 1%) was not significantly different from the accuracy observed during GO
Discussion
The main finding of this study is that the temporal dynamics of STN activity dissociates two forms of executive processes, namely reactive and proactive inhibitory signals. We showed that the decrease of βA in the STN, observed during GO trials, is prematurely interrupted during reactive stopping (Successful STOP trials) at a latency that precedes the SSRT by ~ 100–150 ms. This result is in agreement with the hypothesis that the STN plays a central role in the stopping network to quickly suppress
Conclusion
To sum up, our results demonstrate the existence of at least two different temporal dynamics in the STN, implementing two forms of executive control functions: whereas reactive stopping signals are supported by early STN responses, proactive stopping signals are mediated by a more sustained STN activity that also predicts subjects' inhibitory performances during the SST.
Acknowledgments
The authors would like to thank all patients for their participation; we thank Alexandre Krainik, Andrea Kistner, Nathalie Chivoret and Manik Bhattacharjee for their invaluable help. This study was funded by the Grenoble University Hospital (project DRCI 2010 IP-ESCP). JPL was supported by MLA ANR and by the LABEX CORTEX (ANR-11-LABX-0042) of Université de Lyon, within the program “Investissements d'Avenir” (ANR-11-IDEX-0007).
Conflict of interest
The authors declare no competing financial
References (51)
From reactive to proactive and selective control: developing a richer model for stopping inappropriate responses
Biol. Psychiatry
(2011)- et al.
Direct recordings in human cortex reveal the dynamics of gamma-band [50–150 Hz] activity during pursuit eye movement control
Neuroimage
(2012) - et al.
Therapeutic electrical stimulation of the central nervous system
C. R. Biol.
(2005) - et al.
Deep brain stimulation of the subthalamic nucleus for the treatment of Parkinson's disease
Lancet Neurol.
(2009) - et al.
Basal ganglia local field potential activity: character and functional significance in the human
Clin. Neurophysiol.
(2005) - et al.
Intra-operative recordings of local field potentials can help localize the subthalamic nucleus in Parkinson's disease surgery
Exp. Neurol.
(2006) - et al.
Thresholding of statistical maps in functional neuroimaging using the false discovery rate
Neuroimage
(2002) - et al.
The relationship between local field potential and neuronal discharge in the subthalamic nucleus of patients with Parkinson's disease
Exp. Neurol.
(2005) - et al.
The role of the subthalamic nucleus in response inhibition: evidence from deep brain stimulation for Parkinson's disease
Neuropsychologia
(2009) - et al.
Current advances and pressing problems in studies of stopping
Curr. Opin. Neurobiol.
(2012)