Should I run to catch the bus, or wait for the next one? Deciding whether or not to move may be a common part of our life, but many decisions we make every day do not involve any physical activity, such as deciding whether something is important or not. But does the brain distinguish between decisions involving movement and those without? Previous research has shown that a cluster of neurons called the basal ganglia and a region within these clusters called the subthalamic nucleus play a critical role in both movement and decision-making that involves activity.

The basal ganglia are widely connected to other regions in the brain such as the prefrontal cortex, which is important for decision making and forming associated memories. Therefore, scientists think that the basal ganglia may also play a role in making decisions that do not involve movement, such as deciding whether to pay attention and form relevant memories. Indeed, in patients with Parkinson’s disease – a condition that damages parts of the brain and affects their movement – the degree of physical impairment correlates with their memory deficits.

To test this, Zavala et al. recorded the activity of the subthalamic nucleus and the prefrontal cortex of patients with Parkinson's disease while they performed a task that required them to decide whether to remember something or not. The recordings were taken as the patients underwent ‘deep brain stimulation surgery’, which is used to treat the symptoms of Parkinson’s.

The results showed that brain activity in both regions decreased during the task. This is what normally happens when people make movements, but in this case, these decreases occurred even when individuals were not moving and instead just formed memories. In addition, individuals were presented with specific items to remember and others to forget. When they were presented with items to forget, the activity rebounded back to its original levels. Similar activity patterns have also been observed when individuals decided to stop a movement.

This confirms that the subthalamic nucleus plays a role in decision-making and shows that this area is involved in decisions, even when they do not involve a movement. A full understanding of the purpose of the subthalamic nucleus and basal ganglia will help us understand why changes in the activity of basal ganglia leads to the memory and movement deficits seen in Parkinson’s disease.

The basal ganglia have traditionally been implicated in facilitating or inhibiting motor movements (Albin et al., 1989; DeLong, 1990). As such, studies tying the basal ganglia to decision making have focused on decisions in which participants either make or withhold a motor command (Zavala et al., 2015). Yet, many of the choices that we make in our daily lives involve decisions that do not explicitly require an immediate motor action. For instance, when meeting someone at a dinner party, one may decide that their name is worth trying to remember. Whether and how the basal ganglia may also participate in such non-motor decisions remains poorly understood.

Given their widespread connections to cortical regions, we hypothesized that the circuits and structures that enable the basal ganglia to mediate motor decisions may also be used in an analogous manner to participate in non-motor decisions. We were specifically interested in the non-motor decision to attend to and encode items into working memory. Our ability to dynamically control access to working memory is critical as only a small portion of the large number of stimuli we are exposed to each day are relevant for our behavioral goals. Most studies of working memory have implicated several regions in the prefrontal cortex (PFC) in regulating whether and how memories are successfully encoded (McNab and Klingberg, 2008; Spitzer et al., 2010; Voytek and Knight, 2010; Heinrichs-Graham and Wilson, 2015; Wiesman et al., 2016). We hypothesized, however, that gating access to working memory, like gating movement, also involves the basal ganglia.

Several lines of evidence suggest this possibility. The decision to execute or withhold a movement involves a race between competing basal ganglia pathways that facilitate or inhibit motor commands (Schmidt et al., 2013). At the center of this race lies the subthalamic nucleus (STN), a structure within the basal ganglia previously implicated in motor decisions (Kühn et al., 2004; Frank et al., 2007; Cavanagh et al., 2011; Zaghloul et al., 2012; Alegre et al., 2013; Zavala et al., 2014). The ability to gate movement may also grant the basal ganglia, and the STN, the ability to regulate non-motor cognitive processes such as memory. In addition, the basal ganglia exhibit extensive connections with regions of the PFC that have been implicated in working memory (Alexander et al., 1986; Nakano et al., 2000; Chudasama et al., 2003), and imaging studies reveal increased basal ganglia activity during tasks that involve working memory (Chang et al., 2007; McNab and Klingberg, 2008; Murty et al., 2011; Chatham et al., 2014). Empiric support for the basal ganglia’s role in gating memory also arises from studies of patients with Parkinson’s disease who have disruptions of normal basal ganglia circuitry (Hammond et al., 2007). These patients often experience working memory deficits that correlate with the degree of motor and executive function impairment (Owen et al., 1997; Lewis et al., 2003; Moustafa et al., 2008; Chiaravalloti et al., 2014; Trujillo et al., 2015; Wiesman et al., 2016). Moreover, patients receiving deep brain stimulation (DBS), in which high-frequency electrical stimulation is delivered directly to the STN, may experience changes in working memory performance, although studies examining whether DBS improves or worsens working memory have not been conclusive (Hälbig et al., 2004; Hershey et al., 2008; Mayer et al., 2016; Ventre-Dominey et al., 2014; Mollion et al., 2011).

Collectively, these studies suggest a link between the basal ganglia and non-motor decisions such as those used to regulate and access working memory. Despite such empiric support, however, few studies have directly examined neural activity of the human basal ganglia, and specifically the STN, during a working memory task. We address this here by examining local field potential and neuronal spiking activity in the human STN in patients undergoing DBS surgery for Parkinson’s disease as they participate in a task requiring them to make non-motor decisions to selectively attend to and encode items into working memory. Because of the known involvement of the PFC in working memory, and because of the known connections between the PFC and the basal ganglia, we also simultaneously capture intracranial EEG (iEEG) activity from the PFC in order to ask how these structures communicate during this process of making non-motor decisions. As such, our work builds upon previous studies establishing the role of the human STN in motor response inhibition and decision making (Zavala et al., 2015) to demonstrate the STN’s involvement in non-motor decisions used to dynamically modulate working memory.

Eighteen participants (16 males; 58.4 $\pm$ 1.64 (mean $\pm$ SEM) years old) undergoing DBS surgery for Parkinson’s disease performed a novel working memory task in which we sequentially presented eight single-digit numbers (Figure 1A; see Materials and methods). We instructed the participants to attend to and memorize only numbers presented within a target shape (square or octagon), which was pseudo-randomly chosen and presented before each block of eight numbers. During each block, four of the numbers appeared within the target shape (target trials). The target numbers were randomly interleaved with four numbers appearing within a distractor shape (distractor trials). Correct performance in each block required the participant to vocalize only the four target numbers at the end of each block. Participants vocalized the correct response on 71.2 $\pm$ 2.6% of the blocks they were shown. In the remaining 28.8 $\pm$ 2.6% blocks, the participants made one of two types of errors: they could fail to vocalize all four target numbers (96.4 $\pm$ 1.2% of error blocks) or they could erroneously include a distractor number in the vocalized sequence (63 $\pm$ 4.5% of error blocks). On 62.1 $\pm$ 4.4% of the error blocks, they made both types of errors.

Figure 1 with 1 supplement see all

Download asset Open asset

We simultaneously captured micro- and macroelectrode recordings from the subthalamic nucleus (STN), and intracranial EEG recordings from subdural electrodes temporarily placed over the anterior and lateral prefrontal cortex (PFC) as participants performed the task (Figure 1B and Figure 1—figure supplement 1). We first investigated overall changes in power during all trials in each of three recorded brain regions - the anterior PFC, the lateral PFC, and the STN - by computing the average power across all trials during each experimental session and across all electrodes in each brain region (Figure 2A,B). Across sessions, all three brain regions demonstrated significant decreases in beta band (15–30 Hz) power across all trials (p < 0.005, permutation test; see Materials and methods). We also found significant increases in theta band (2–8 Hz) power in the anterior and lateral PFC during the task (p < 0.005 for anterior PFC and lateral PFC, permutation test, Figure 3—figure supplement 1A). These changes were not specific to either left- or right-sided brain regions, as we found no significant differences in spectral power between experimental sessions captured in the separate hemispheres (left anterior PFC vs right anterior PFC $p=0.37$, left lateral PFC vs right lateral PFC $p=0.55$, left STN vs right STN $p=0.60$, permutation test; see Materials and methods).

Figure 2 with 1 supplement see all

Download asset Open asset

Although we observed these decreases in beta oscillatory power in both the anterior and lateral PFCs by averaging the spectral power across all electrode contacts, we were interested in whether the observed changes represented a focal process or a diffuse cortical phenomenon. We therefore examined changes in spectral power within each electrode contact. In 19 out of the 24 sessions with lateral PFC electrodes, at least one electrode contact exhibited a significant decrease in beta power during the task (p < 0.05, permutation test; see Materials and methods). This effect was not present in all electrode contacts, however. Indeed, within these 19 sessions, we found that only 2.6 $\pm$ 0.29 of seven lateral PFC electrode contacts (36.8 $\pm$ 4.1%) showed a significant decrease in beta power relative to baseline (p < 0.05, permutation test; Figure 2—figure supplement 1). In contrast, 14 out of 29 sessions with anterior PFC electrodes demonstrated at least one electrode contact exhibiting a significant decrease in beta power during the task (p < 0.05, permutation test). Within these sessions, we found that 2.1 $\pm$ 0.31 of five electrode contacts (42.9 $\pm$ 6.2%) showed the change in beta power relative to baseline (p < 0.05, permutation test). Although we did not have access to intra-operative computed tomography (CT), we estimated all anterior and lateral PFC electrode locations on each participant’s brain (see Supplementary Information) to determine whether the observed effects were localized to a particular brain region. We found that despite the fact that the anterior PFC, the lateral PFC, and the STN all showed a similar drop in beta power during the task, the beta power within individual participants was focal in nature, although the location of these focal drops was not consistent across participants (Figure 2—figure supplement 1).

We were interested in understanding how the observed changes in beta oscillatory power evolved over the course of successive trials in each individual block. Although each individual trial exhibited a relative transient decrease in beta power, overall beta power exhibited a systematic decrease from the first through the last trial of each block in all three brain regions (Figure 2C). The observed decreases across trials were not related to movement as participants were instructed to withhold all movements until receiving an instruction to vocalize their response one second after the final number was cleared from the screen. We quantified this overall decrease by examining beta oscillatory power averaged across the entire trial (0 to 1000 ms following stimulus onset). A repeated-measures ANOVA revealed a significant effect of trial number on the changes in average beta oscillatory power in all three brain regions (anterior PFC $F=4.88$, $df=7$, p < 0.001; lateral PFC $F=6.46$, $df=7$, p < 0.001; STN $F=4.07$, $df=7$, p < 0.001). We also performed a linear regression of beta power against trial number and found a significant relation across all experimental sessions in all three brain regions (anterior PFC $\rho =-0.31\pm 0.1$, $t(27)=3.4$, p < 0.002; lateral $\rho =-0.31\pm 0.1$, $t(23)=3.1$, p < 0.005; STN $\rho =-0.35\pm 0.1$, $t(28)=3.35$, p < 0.002; best fit line in Figure 2C). See Supplementary Data and Figure 3—figure supplement 3 for analysis of progressive decrease in beta oscillatory power separately for target and distractor trials.

Our data suggest that participation in this non-motor task evokes decreases in beta oscillatory power in both the STN and PFC following the presentation of each number. The observed decreases in beta oscillatory power during each trial are similar to decreases previously observed with motor movements (Kühn et al., 2004; Ray et al., 2012; Alegre et al., 2013). However, we were specifically interested in whether these changes in beta oscillatory power depend on whether the presented stimulus was a target or distractor item. We therefore investigated whether there were significant differences in oscillatory power between the target and distractor conditions. We computed the average power separately across all target and distractor trials during each experimental session. Across sessions, all three brain regions exhibited initial decreases in beta power following the presentation of both target and distractor. However, during the distractor trials, both the STN and lateral PFC demonstrated an early termination of the beta band decrease. This resulted in significantly higher beta band power relative to the target trials (p = 0.020 for STN and p = 0.025 for lateral PFC, permutation test; Figure 3; see Materials and methods; for changes in theta band power, see Figure 3—figure supplement 1B).

Figure 3 with 4 supplements see all

Download asset Open asset

As with the changes in overall beta power during the task, we were also interested in whether the observed differences in beta power between target and distractor trials were specific to individual electrode contacts. We found that 13 out of the 24 experimental sessions with lateral PFC electrodes demonstrated a significant difference between target and distractor trials in at least one contact (p < 0.05, permutation test). Indeed, the observed effect averaged across the entire brain region was only mediated by a subset of electrodes. Within these 13 sessions, only 2.1 $\pm$ 0.2 of seven electrode contacts (29.7 $\pm$ 3.4%) showed a significant difference in beta power between target and distractor trials (p < 0.05, permutation test; Figure 3—figure supplement 2). Conversely, in the anterior PFC, only 6 out of 29 sessions demonstrated at least one contact exhibiting a significant difference in average beta power between target and distractor trials (p < 0.05, permutation test).

Although our data demonstrate that target and distractor stimuli involved differences in beta oscillatory power during each trial, we were interested in whether these differences were directly related to the behavioral performance on each trial. As such, we next examined the relatively few error trials in which participants failed to encode the target number. If participants failed to encode these target stimuli because they treated them as distractors, then physiologically the observed beta decreases should be more similar to the distractor rather than the target trials. In the lateral PFC electrodes, we found that the successfully encoded target trials exhibited significantly greater decreases in beta power than those target trials that participants failed to encode (p = 0.04, permutation test; Figure 3—figure supplement 4). Indeed, the changes in beta power during the target trials that were not encoded did not differ from the distractor trials, suggesting that the failure to encode those target trials were related to the same early rebounds in beta power that were present when participants intentionally chose not to encode the distractor numbers. We did not, however, find similar differences between the error trials and the successfully encoded target trials in the anterior PFC (p = 0.17, permutation test) or the STN (p = 0.20, permutation test, Figure 3—figure supplement 4). We also did not find significant differences between the erroneously encoded distractor trials and the correct distractor trials in any brain region (p > 0.05, permutation test, data not shown).

As the lateral PFC and the STN both demonstrated significant differences in oscillatory power between target and distractor trials, we were next interested in asking whether these conditions also exhibited significant differences in connectivity between the two regions. Within each participant, we first identified individual lateral and anterior PFC contacts exhibiting significant differences in power between target and distractor trials that mediated the overall changes in these brain regions (Figure 3—figure supplement 2). We focused our analysis of connectivity only on the lateral PFC electrodes that exhibited a significant difference in power within participants (p < 0.05, permutation test; 12 experimental sessions with simultaneously LFP recordings within the borders of the STN; see Materials and methods; an insufficient number of individual anterior PFC electrodes exhibited significant differences in power within participants, consistent with the lack of an overall effect of trial type in the anterior PFC). We examined connectivity by comparing the spectral coherence between the lateral PFC and STN to the coherence observed during a baseline period (see Materials and methods). As with the analysis of power, we first calculated the average coherence difference between target and distractor trials for the relevant STN-lateral PFC pairs within a session before testing for a significant trial-type related difference across all sessions. Distractor trials exhibited significantly higher coherence in the beta band between the lateral PFC and the STN than target trials (p = 0.015, permutation procedure; Figure 4), suggesting that the elevated levels of beta power in both brain regions during distractor trials were accompanied by elevated levels of beta band coherence. We subsequently tested whether the higher levels of coherence were due to increased synchrony of phase or increased correlation of power (Cohen, 2014). The increased coherence we observed during distractor trials was primarily driven by higher levels of phase synchrony (p = 0.025, permutation test; Figure 4—figure supplement 1). Distractor trials also had higher levels of correlation of beta power, but this difference did not survive multiple comparison testing.

Figure 4 with 1 supplement see all

Download asset Open asset

Finally, we extracted spiking activity from microelectrode recordings in the STN while participants were engaged in the task. We identified spiking activity from 48 neuronal firing clusters across 21 STN recording sessions (35.9 $\pm$ 3.7 spikes/s (mean $\pm$ SEM) per cluster). On average, across all neurons, we found a significant decrease in overall spiking activity during each trial compared to baseline (p = 0.02, permutation test; see Materials and methods; Figure 5A). However, the responses of individual neuronal clusters were heterogenous. Some individual clusters demonstrated decreases in spiking activity during each trial, whereas others demonstrated increases (Figure 5B). Within individual recordings, we found that 47.9% (23 out of 48 clusters) of the recorded clusters demonstrated a significant change in firing rate (p < 0.05, permutation test) during the task (across all trials) when compared to baseline (Figure 5C). Of these neuronal clusters that showed a change in firing, most (65%, 15 out of 23 clusters) exhibited significant decreases in firing during the task, but some (35%, 8 out of 23 clusters) exhibited significant increases (average responses of both populations in Figure 5D). Relative to the dorsal and ventral borders of the STN, there were no significant differences in depth between the upward (mean depth 70.4 $\pm$ 7.8%; see Materials and methods) and the downward firing clusters (mean depth 56.2 $\pm$ 5.4%; p = 0.14, unpaired t-test).

Figure 5

Download asset Open asset

We separately examined the neuronal clusters that individually demonstrated decreases in spiking activity during the task, and found that the average response of these clusters across an entire block exhibited similar transient dynamics during each trial to those observed in beta oscillatory power (Figure 5E). However, unlike the changes in beta oscillatory power, we did not find that there was a progressive change in overall spiking activity with subsequent trials across each block ($\rho =0.13\pm 0.20$, $t(14)=0.61$, p = 0.55). Nevertheless, we examined the relationship between individual spike events and the ongoing LFP oscillations recorded from the STN macroelectrodes, and found that these neuronal clusters were significantly entrained to ongoing beta oscillations during the task (p = 0.02, permutation test; Figure 5F; see Materials and methods). Notably, the neuronal clusters that individually demonstrated increases in spiking activity during the task did not exhibit a significant interaction between LFP phase and spiking events (p = 0.16, permutation test; Figure 5F).

The observed changes in STN spiking activity suggest that indeed the STN modulates its spiking output during this non-motor task. However, as with the changes in beta oscillatory power, we were also interested in whether the spiking responses of these neurons were differentially modulated by target and distractor trials. Individual recordings accounting for approximately 20% of the recorded neuronal clusters exhibited significant differences in spiking activity between target and distractor trials (p < 0.05, permutation test; Figure 5G). However, we found no significant overall differences in spiking activity between target and distractor trials across either decreasing (p = 0.34, permutation test; Figure 5H) or increasing neurons (p = 0.36, permutation test). Moreover, although neuronal clusters exhibiting decreased spiking activity during the task were locked to the LFP beta oscillation, we found no significant differences in beta band spike phase locking between the target and distractor trials (p > 0.05, permutation test). This was the case when we examined the full 1000 ms window of all trials, as well as when we used smaller time windows (250 ms) centered at various points within the task (data not shown).

Our data demonstrate that the human STN modulates both oscillatory and spiking activity as participants make decisions regarding whether to attend to and remember individual items. Our data therefore build upon previous empiric findings suggesting that the basal ganglia may play a role in regulating working memory, and provide direct evidence that the STN exhibits changes in activity during a non-motor decision. Moreover, by simultaneously recording activity from the lateral PFC, our data also demonstrate changes in oscillatory coherence between the PFC and STN, thereby tying cortical regions known to be involved in working memory with the STN.

In the novel working memory task used here, participants make internal decisions to attend to and encode numbers if they appear within a target shape, or to ignore them if they appear within a distractor shape. Importantly, both choices involve no immediate motor movements, enabling us to directly ask whether the STN participates in such non-motor decisions. As when inhibiting motor movements (Zavala et al., 2015), participants should optimally perform this task by immediately recognizing the indicator shape on each trial, and in the case of distractors, either ignore the presented item completely or prevent any processes that actively encode it. Although our data cannot distinguish these possibilities, the observed differences in neural activity suggest that participants are able to successfully make an internal distinction between the two trial types. Of note, this strategy would spare working memory capacity from storing distracting items that are unnecessary for correct completion of the block. It is possible, however, that the participants encode every number presented in each trial, along with the shape within which each number was presented. In this latter scenario, the participants would make a distinction between the target and distractor items only at the time of retrieval when vocalizing their response. Although we do not explicitly test their memory for the distractor items, it is less likely that participants are adopting this latter more challenging approach given how well they performed this task, and given the differences in neural activity observed at the time of presentation between target and distractor items.

The most robust changes we observed during our task in both the STN and in the lateral PFC were related to beta oscillatory activity. Changes in STN beta oscillatory power have been extensively studied in the context of movement, decreasing during the execution of a movement, and increasing when a movement is inhibited. The relation between beta oscillatory activity and movement is most evident during stop inhibition tasks when participants are asked to specifically begin or stop a motor movement (Kühn et al., 2004; Ray et al., 2012; Alegre et al., 2013). In the task described here, participants make a similar decision, albeit without moving. Remarkably, both the STN and the lateral PFC exhibit similar decreases in beta oscillatory power during this non-motor decision, with significantly earlier termination of these decreases when the presented item should be ignored.

The absence of movement here therefore allows us to specifically examine changes in STN activity during non-motor decisions and show that these changes are similar to those seen during movement inhibition. However, this also makes interpreting the timing of when these changes in beta power occur more difficult since there is no explicit behavioral measure of when each individual is internally processing the numbers and distractor signals, or when they are deciding to encode or ignore each stimulus. Nevertheless, we find that the timing of beta power differences between trial types is similar between our task and the tasks that required an inhibition of an actual movement, suggesting that the changes observed here are analogous to the previously identified stop signal (Kühn et al., 2004; Ray et al., 2012; Alegre et al., 2013). Moreover, in stop inhibition tasks, differences in beta power can still alter the behavior of the failed stop trials (in which a subject failed to properly inhibit an action) even when occurring after the erroneous response has already begun. These differences in beta power persist after that behavior has been initiated, and in the case of movement inhibition, this manifests itself as a suppression of the effort and duration of an erroneous response (Ruiz et al., 2011; Cohen and van Gaal, 2014). In the case of our current task then, increased STN beta activity may prevent the encoding of an item into working memory, or it may impair the maintenance of that item in memory. Although rebounds in beta power have also been observed following the execution of a movement, because these rebounds occur much later in the task than the timing observed here (Alegre et al., 2013), we think our data are more consistent with an internal stop signal rather than a post-action rebound in beta power.

Given the relative timing of the observed changes in beta power and the behavioral and physiological evidence that participants are able to successfully distinguish trial types, our data are therefore consistent with previous studies suggesting that beta band activity may in fact represent the brain’s mechanism for globally suppressing all actions. The concept of global suppression has been well described in the context of movement, but more recently has been extended to include even non-motor processes such as working memory (Wessel et al., 2016a). Increases in STN beta oscillatory power have been observed when participants actively refrain from updating working memory (Oswal et al., 2012), and following the presentation of a surprise auditory stimulus that impaired the maintenance of items in working memory (Wessel et al., 2016b). Moreover, inhibiting movements can result in an unintentional inhibition of memory (Chiu and Egner, 2015), suggesting that the suppression of actions related to movement and to memory may be linked.

Indeed, the suggestion that beta band oscillatory activity could serve as a global suppression signal for both motor and non-motor functions throughout the brain finds support both in our observation that the STN and the lateral PFC exhibit similar decreases and rebounds in beta oscillatory power during this non-motor decision, and in previous studies of the PFC. In particular, beta oscillations in the lateral PFC decrease during the encoding phase of working memory (Spitzer et al., 2010; Brookes et al., 2011; Heinrichs-Graham and Wilson, 2015; Kornblith et al., 2016; Lundqvist et al., 2016; Wiesman et al., 2016), and exhibit similar changes as found in the STN when movements are inhibited (Swann et al., 2009, 2012). In patients with Parkinson’s disease, therefore, excessive beta oscillatory activity may not only impact movement, but could also play a role in the observed deficits in working memory due to the global nature of this activity (Owen et al., 1997; Lewis et al., 2003; Chiaravalloti et al., 2014; Trujillo et al., 2015; Wiesman et al., 2016). These effects may be related to the loss of dopaminergic innervation to the lateral PFC that result in increased beta oscillations (Puig and Miller, 2012; Puig et al., 2014; Puig and Miller, 2015). Importantly, here we demonstrate that the changes in beta oscillatory power are accompanied by significant changes in beta oscillatory coherence between the lateral PFC and the STN. Although the lack of intraoperative CT imaging limited our ability to determine precisely what areas of the lateral PFC were interacting with the STN, our data provide direct evidence linking these structures and their respective roles as participants make these non-motor decisions related to memory.

Interestingly, the progressive decreases in overall beta oscillatory power over successive trials observed here suggest a link with working memory load. Previous studies have demonstrated progressive increases in gamma band power in the PFC and hippocampus with working memory load (van Vugt et al., 2010; Roux et al., 2012), and in alpha band power in the parietal lobe (Tuladhar et al., 2007). These progressive changes are not consistent, however, as progressive decreases in oscillatory activity with working memory load have also been observed across several cortical regions (Meltzer et al., 2007), similar to our data. Of note, we observed such progressive decreases separately for both target and distractor trials, with no progression in the differences in beta oscillatory power between them. It is also possible, therefore, that these progressive decreases are not related to memory load, and instead emerge as each trial brings the participant closer to the forthcoming motor movement at the end of each list. Conversely, if these progressive decreases are indeed linked with memory load, then participants may be retaining some memory of the distractor items even though, during individual trials, their neural data suggest that they treat these items differently. Alternatively, the progressive decreases could be related only to increasing memory load of the target items, in which case the distractor trials may not add to the decreases, but also may not reverse them.

Ultimately, the STN exerts its effects on the rest of the basal ganglia through its spiking activity. Our data demonstrate that the firing rates of a significant fraction of individual neurons within the STN were also modulated during this working memory task that did not involve an explicit motor movement. Overall firing rates averaged across all neurons decreased during each trial, consistent with the suggestion that decreases in overall STN spiking activity are facilitatory. However, just as movement elicits variability in the spiking responses of individual neurons (Zavala et al., 2017), the decision to attend to and encode items was accompanied by heterogeneous spiking responses. Indeed, the observed overall decreases in spiking activity was mediated by only a subset of neurons. Interestingly, these neurons also exhibited significant phase locking to the ongoing beta oscillations, suggesting a link between spiking activity in the STN and LFP oscillations.

Despite the overall decreases in spiking activity, however, only a subset of neurons demonstrated a significant difference in spiking activity between target and distractor trials. Moreover, we did not find evidence that there was a significant difference in spike phase locking between the two conditions. Hence, while our oscillatory data would implicate the basal ganglia at large in such non-motor decisions that distinguish target and distractor items, the spiking data would suggest that only a subset of STN neurons are involved in this process, and that these neurons do not alter the extent to which they are locked to the ongoing beta oscillations accordingly.

Finally, our data demonstrate that the changes in cortical beta oscillatory power were also accompanied by changes in cortical theta oscillatory power. Unlike beta oscillations, the precise roles played by theta oscillations in the cortex and STN are less clear. The elevated levels of cortical theta oscillatory power during the distractor trials are consistent with the role cortical theta oscillations may play during conflict related response inhibition (Cavanagh et al., 2011, 2012). Given our previous work demonstrating increases in STN theta power and cortico-STN theta coherence during conflict (Zavala et al., 2014, 2016b, 2016a), however, we were surprised to find no trial-type related differences in theta power in the STN during this task. One possibility, however, is that STN theta increases are specific to conflict induced slowing of response times rather than complete response inhibition. Indeed, previous studies have demonstrated that motor response inhibition as tested using Go-NoGo paradigms fails to elicit differences in theta oscillatory activity despite showing robust changes in the beta band power similar to those reported here (Kühn et al., 2004; Ray et al., 2012; Alegre et al., 2013).

Together, our data demonstrate that many of the circuit mechanisms of the basal ganglia that have been previously described in the context of movement are used in an analogous fashion during non-motor cognitive decisions. Although our data were collected from patients with Parkinson’s disease in an intraoperative environment, we primarily contrast beta band activity when participants were making decisions between different trial types, suggesting that these differences may exist or even be enhanced (Oswal et al., 2012) in the healthy brain. Here, we were specifically interested in the non-motor decision to attend to and encode items into working memory. Our data raise the possibility, however, that similar neural mechanisms and interactions between the basal ganglia and the cortex could be used for any non-motor decision that involves proceeding with or aborting a cognitive process.

1. 1. Zavala BA
   2. Jang A
   3. Zaghloul KA

   (2018) Data from: Human subthalamic nucleus activity during non-motor decision making

   Available at Dryad Digital Repository under a CC0 Public Domain Dedication.

   http://dx.doi.org/10.5061/dryad.9t740

1. 1. Albin RL
   2. Young AB
   3. Penney JB

   (1989) The functional anatomy of basal ganglia disorders

   Trends in Neurosciences 12:366–375.

   https://doi.org/10.1016/0166-2236(89)90074-X

   • PubMed
   • Google Scholar
2. 1. Alegre M
   2. Lopez-Azcarate J
   3. Obeso I
   4. Wilkinson L
   5. Rodriguez-Oroz MC
   6. Valencia M
   7. Garcia-Garcia D
   8. Guridi J
   9. Artieda J
   10. Jahanshahi M
   11. Obeso JA

   (2013) The subthalamic nucleus is involved in successful inhibition in the stop-signal task: a local field potential study in Parkinson's disease

   Experimental Neurology 239:1–12.

   https://doi.org/10.1016/j.expneurol.2012.08.027

   • PubMed
   • Google Scholar
3. 1. Alexander GE
   2. DeLong MR
   3. Strick PL

   (1986) Parallel organization of functionally segregated circuits linking basal ganglia and cortex

   Annual Review of Neuroscience 9:357–381.

   https://doi.org/10.1146/annurev.ne.09.030186.002041

   • Google Scholar
4. 1. Brittain JS
   2. Watkins KE
   3. Joundi RA
   4. Ray NJ
   5. Holland P
   6. Green AL
   7. Aziz TZ
   8. Jenkinson N

   (2012) A role for the subthalamic nucleus in response inhibition during conflict

   Journal of Neuroscience 32:13396–13401.

   https://doi.org/10.1523/JNEUROSCI.2259-12.2012

   • PubMed
   • Google Scholar
5. 1. Brookes MJ
   2. Wood JR
   3. Stevenson CM
   4. Zumer JM
   5. White TP
   6. Liddle PF
   7. Morris PG

   (2011) Changes in brain network activity during working memory tasks: a magnetoencephalography study

   NeuroImage 55:1804–1815.

   https://doi.org/10.1016/j.neuroimage.2010.10.074

   • PubMed
   • Google Scholar
6. 1. Cavanagh JF
   2. Wiecki TV
   3. Cohen MX
   4. Figueroa CM
   5. Samanta J
   6. Sherman SJ
   7. Frank MJ

   (2011) Subthalamic nucleus stimulation reverses mediofrontal influence over decision threshold

   Nature Neuroscience 14:1462–1467.

   https://doi.org/10.1038/nn.2925

   • PubMed
   • Google Scholar
7. 1. Cavanagh JF
   2. Zambrano-Vazquez L
   3. Allen JJ

   (2012) Theta lingua franca: a common mid-frontal substrate for action monitoring processes

   Psychophysiology 49:220–238.

   https://doi.org/10.1111/j.1469-8986.2011.01293.x

   • PubMed
   • Google Scholar
8. 1. Chang C
   2. Crottaz-Herbette S
   3. Menon V

   (2007) Temporal dynamics of basal ganglia response and connectivity during verbal working memory

   NeuroImage 34:1253–1269.

   https://doi.org/10.1016/j.neuroimage.2006.08.056

   • PubMed
   • Google Scholar
9. 1. Chatham CH
   2. Frank MJ
   3. Badre D

   (2014) Corticostriatal output gating during selection from working memory

   Neuron 81:930–942.

   https://doi.org/10.1016/j.neuron.2014.01.002

   • PubMed
   • Google Scholar
10. 1. Chiaravalloti ND
    2. Ibarretxe-Bilbao N
    3. DeLuca J
    4. Rusu O
    5. Pena J
    6. García-Gorostiaga I
    7. Ojeda N

    (2014) The source of the memory impairment in Parkinson's disease: acquisition versus retrieval

    Movement Disorders 29:765–771.

    https://doi.org/10.1002/mds.25842

    • PubMed
    • Google Scholar
11. 1. Chiu YC
    2. Egner T

    (2015) Inhibition-induced forgetting: when more control leads to less memory

    Psychological Science 26:27–38.

    https://doi.org/10.1177/0956797614553945

    • PubMed
    • Google Scholar
12. 1. Chudasama Y
    2. Baunez C
    3. Robbins TW

    (2003)

    Functional disconnection of the medial prefrontal cortex and subthalamic nucleus in attentional performance: evidence for corticosubthalamic interaction

    Journal of Neuroscience 23:5477–5485.

    • PubMed
    • Google Scholar
13. 1. Cohen MX
    2. Cavanagh JF

    (2011) Single-trial regression elucidates the role of prefrontal theta oscillations in response conflict

    Frontiers in Psychology 2:30.

    https://doi.org/10.3389/fpsyg.2011.00030

    • PubMed
    • Google Scholar
14. 1. Cohen MX
    2. van Gaal S

    (2014) Subthreshold muscle twitches dissociate oscillatory neural signatures of conflicts from errors

    NeuroImage 86:503–513.

    https://doi.org/10.1016/j.neuroimage.2013.10.033

    • PubMed
    • Google Scholar
15. Book

    1. Cohen MX

    (2014)

    Analyzing Neural Time Series Data: Theory and Practice

    Cambridge, MA: MIT Press.

    • Google Scholar
16. 1. Cox RW

    (1996) AFNI: software for analysis and visualization of functional magnetic resonance neuroimages

    Computers and Biomedical Research 29:162–173.

    https://doi.org/10.1006/cbmr.1996.0014

    • PubMed
    • Google Scholar
17. 1. DeLong MR

    (1990) Primate models of movement disorders of basal ganglia origin

    Trends in Neurosciences 13:281–285.

    https://doi.org/10.1016/0166-2236(90)90110-V

    • PubMed
    • Google Scholar
18. 1. Frank MJ
    2. Samanta J
    3. Moustafa AA
    4. Sherman SJ

    (2007) Hold your horses: impulsivity, deep brain stimulation, and medication in parkinsonism

    Science 318:1309–1312.

    https://doi.org/10.1126/science.1146157

    • PubMed
    • Google Scholar
19. 1. Fries P

    (2005) A mechanism for cognitive dynamics: neuronal communication through neuronal coherence

    Trends in Cognitive Sciences 9:474–480.

    https://doi.org/10.1016/j.tics.2005.08.011

    • PubMed
    • Google Scholar
20. 1. Hammond C
    2. Bergman H
    3. Brown P

    (2007) Pathological synchronization in Parkinson's disease: networks, models and treatments

    Trends in Neurosciences 30:357–364.

    https://doi.org/10.1016/j.tins.2007.05.004

    • PubMed
    • Google Scholar
21. 1. Heinrichs-Graham E
    2. Wilson TW

    (2015) Spatiotemporal oscillatory dynamics during the encoding and maintenance phases of a visual working memory task

    Cortex 69:121–130.

    https://doi.org/10.1016/j.cortex.2015.04.022

    • PubMed
    • Google Scholar
22. 1. Hershey T
    2. Wu J
    3. Weaver PM
    4. Perantie DC
    5. Karimi M
    6. Tabbal SD
    7. Perlmutter JS

    (2008) Unilateral vs. bilateral STN DBS effects on working memory and motor function in Parkinson disease

    Experimental Neurology 210:402–408.

    https://doi.org/10.1016/j.expneurol.2007.11.011

    • PubMed
    • Google Scholar
23. 1. Holmes CJ
    2. Hoge R
    3. Collins L
    4. Woods R
    5. Toga AW
    6. Evans AC

    (1998) Enhancement of MR images using registration for signal averaging

    Journal of Computer Assisted Tomography 22:324–333.

    https://doi.org/10.1097/00004728-199803000-00032

    • PubMed
    • Google Scholar
24. 1. Hälbig TD
    2. Gruber D
    3. Kopp UA
    4. Scherer P
    5. Schneider GH
    6. Trottenberg T
    7. Arnold G
    8. Kupsch A

    (2004) Subthalamic stimulation differentially modulates declarative and nondeclarative memory

    NeuroReport 15:539–543.

    https://doi.org/10.1097/00001756-200403010-00031

    • PubMed
    • Google Scholar
25. 1. Kornblith S
    2. Buschman TJ
    3. Miller EK

    (2016) Stimulus load and oscillatory activity in higher cortex

    Cerebral Cortex 26:3772–3784.

    https://doi.org/10.1093/cercor/bhv182

    • PubMed
    • Google Scholar
26. 1. Kühn AA
    2. Williams D
    3. Kupsch A
    4. Limousin P
    5. Hariz M
    6. Schneider GH
    7. Yarrow K
    8. Brown P

    (2004) Event-related beta desynchronization in human subthalamic nucleus correlates with motor performance

    Brain 127:735–746.

    https://doi.org/10.1093/brain/awh106

    • PubMed
    • Google Scholar
27. 1. Lachaux JP
    2. Lutz A
    3. Rudrauf D
    4. Cosmelli D
    5. Le Van Quyen M
    6. Martinerie J
    7. Varela F

    (2002) Estimating the time-course of coherence between single-trial brain signals: an introduction to wavelet coherence

    Neurophysiologie Clinique/Clinical Neurophysiology 32:157–174.

    https://doi.org/10.1016/S0987-7053(02)00301-5

    • PubMed
    • Google Scholar
28. 1. Lewis SJ
    2. Cools R
    3. Robbins TW
    4. Dove A
    5. Barker RA
    6. Owen AM

    (2003) Using executive heterogeneity to explore the nature of working memory deficits in Parkinson's disease

    Neuropsychologia 41:645–654.

    https://doi.org/10.1016/S0028-3932(02)00257-9

    • PubMed
    • Google Scholar
29. 1. Lundqvist M
    2. Rose J
    3. Herman P
    4. Brincat SL
    5. Buschman TJ
    6. Miller EK

    (2016) Gamma and beta bursts underlie working memory

    Neuron 90:152–164.

    https://doi.org/10.1016/j.neuron.2016.02.028

    • PubMed
    • Google Scholar
30. 1. Maris E
    2. Oostenveld R

    (2007) Nonparametric statistical testing of EEG- and MEG-data

    Journal of Neuroscience Methods 164:177–190.

    https://doi.org/10.1016/j.jneumeth.2007.03.024

    • PubMed
    • Google Scholar
31. 1. Mayer JS
    2. Neimat J
    3. Folley BS
    4. Bourne SK
    5. Konrad PE
    6. Charles D
    7. Park S

    (2016) Deep brain stimulation of the subthalamic nucleus alters frontal activity during spatial working memory maintenance of patients with Parkinson's disease

    Neurocase 22:369–378.

    https://doi.org/10.1080/13554794.2016.1197951

    • PubMed
    • Google Scholar
32. 1. McNab F
    2. Klingberg T

    (2008) Prefrontal cortex and basal ganglia control access to working memory

    Nature Neuroscience 11:103–107.

    https://doi.org/10.1038/nn2024

    • PubMed
    • Google Scholar
33. 1. Meltzer JA
    2. Negishi M
    3. Mayes LC
    4. Constable RT

    (2007) Individual differences in EEG theta and alpha dynamics during working memory correlate with fMRI responses across subjects

    Clinical Neurophysiology 118:2419–2436.

    https://doi.org/10.1016/j.clinph.2007.07.023

    • PubMed
    • Google Scholar
34. 1. Mollion H
    2. Dominey PF
    3. Broussolle E
    4. Ventre-Dominey J

    (2011) Subthalamic nucleus stimulation selectively improves motor and visual memory performance in Parkinson's disease

    Movement Disorders 26:2019–2025.

    https://doi.org/10.1002/mds.23769

    • PubMed
    • Google Scholar
35. 1. Moustafa AA
    2. Sherman SJ
    3. Frank MJ

    (2008) A dopaminergic basis for working memory, learning and attentional shifting in Parkinsonism

    Neuropsychologia 46:3144–3156.

    https://doi.org/10.1016/j.neuropsychologia.2008.07.011

    • PubMed
    • Google Scholar
36. 1. Murty VP
    2. Sambataro F
    3. Radulescu E
    4. Altamura M
    5. Iudicello J
    6. Zoltick B
    7. Weinberger DR
    8. Goldberg TE
    9. Mattay VS

    (2011) Selective updating of working memory content modulates meso-cortico-striatal activity

    NeuroImage 57:1264–1272.

    https://doi.org/10.1016/j.neuroimage.2011.05.006

    • PubMed
    • Google Scholar
37. 1. Nakano K
    2. Kayahara T
    3. Tsutsumi T
    4. Ushiro H

    (2000) Neural circuits and functional organization of the striatum

    Journal of Neurology 247:V1–V15.

    https://doi.org/10.1007/PL00007778

    • PubMed
    • Google Scholar
38. 1. Oswal A
    2. Litvak V
    3. Sauleau P
    4. Brown P

    (2012) Beta reactivity, prospective facilitation of executive processing, and its dependence on dopaminergic therapy in Parkinson's disease

    Journal of Neuroscience 32:9909–9916.

    https://doi.org/10.1523/JNEUROSCI.0275-12.2012

    • PubMed
    • Google Scholar
39. 1. Owen AM
    2. Iddon JL
    3. Hodges JR
    4. Summers BA
    5. Robbins TW

    (1997) Spatial and non-spatial working memory at different stages of Parkinson's disease

    Neuropsychologia 35:519–532.

    https://doi.org/10.1016/S0028-3932(96)00101-7

    • PubMed
    • Google Scholar
40. 1. Puig MV
    2. Antzoulatos EG
    3. Miller EK

    (2014) Prefrontal dopamine in associative learning and memory

    Neuroscience 282:217–229.

    https://doi.org/10.1016/j.neuroscience.2014.09.026

    • PubMed
    • Google Scholar
41. 1. Puig MV
    2. Miller EK

    (2012) The role of prefrontal dopamine D1 receptors in the neural mechanisms of associative learning

    Neuron 74:874–886.

    https://doi.org/10.1016/j.neuron.2012.04.018

    • PubMed
    • Google Scholar
42. 1. Puig MV
    2. Miller EK

    (2015) Neural substrates of dopamine D2 receptor modulated executive functions in the monkey prefrontal cortex

    Cerebral Cortex 25:2980–2987.

    https://doi.org/10.1093/cercor/bhu096

    • PubMed
    • Google Scholar
43. 1. Ray NJ
    2. Brittain JS
    3. Holland P
    4. Joundi RA
    5. Stein JF
    6. Aziz TZ
    7. Jenkinson N

    (2012) The role of the subthalamic nucleus in response inhibition: evidence from local field potential recordings in the human subthalamic nucleus

    NeuroImage 60:271–278.

    https://doi.org/10.1016/j.neuroimage.2011.12.035

    • PubMed
    • Google Scholar
44. 1. Roux F
    2. Wibral M
    3. Mohr HM
    4. Singer W
    5. Uhlhaas PJ

    (2012) Gamma-band activity in human prefrontal cortex codes for the number of relevant items maintained in working memory

    Journal of Neuroscience 32:12411–12420.

    https://doi.org/10.1523/JNEUROSCI.0421-12.2012

    • PubMed
    • Google Scholar
45. 1. Ruiz MH
    2. Strübing F
    3. Jabusch HC
    4. Altenmüller E

    (2011) EEG oscillatory patterns are associated with error prediction during music performance and are altered in musician's dystonia

    NeuroImage 55:1791–1803.

    https://doi.org/10.1016/j.neuroimage.2010.12.050

    • PubMed
    • Google Scholar
46. 1. Saad ZS
    2. Reynolds RC

    (2012) SUMA

    NeuroImage 62:768–773.

    https://doi.org/10.1016/j.neuroimage.2011.09.016

    • PubMed
    • Google Scholar
47. 1. Schmidt R
    2. Leventhal DK
    3. Mallet N
    4. Chen F
    5. Berke JD

    (2013) Canceling actions involves a race between basal ganglia pathways

    Nature Neuroscience 16:1118–1124.

    https://doi.org/10.1038/nn.3456

    • PubMed
    • Google Scholar
48. 1. Sharott A
    2. Gulberti A
    3. Zittel S
    4. Tudor Jones AA
    5. Fickel U
    6. Münchau A
    7. Köppen JA
    8. Gerloff C
    9. Westphal M
    10. Buhmann C
    11. Hamel W
    12. Engel AK
    13. Moll CK

    (2014) Activity parameters of subthalamic nucleus neurons selectively predict motor symptom severity in Parkinson's disease

    Journal of Neuroscience 34:6273–6285.

    https://doi.org/10.1523/JNEUROSCI.1803-13.2014

    • PubMed
    • Google Scholar
49. 1. Spitzer B
    2. Wacker E
    3. Blankenburg F

    (2010) Oscillatory correlates of vibrotactile frequency processing in human working memory

    Journal of Neuroscience 30:4496–4502.

    https://doi.org/10.1523/JNEUROSCI.6041-09.2010

    • PubMed
    • Google Scholar
50. 1. Swann N
    2. Tandon N
    3. Canolty R
    4. Ellmore TM
    5. McEvoy LK
    6. Dreyer S
    7. DiSano M
    8. Aron AR

    (2009) Intracranial EEG reveals a time- and frequency-specific role for the right inferior frontal gyrus and primary motor cortex in stopping initiated responses

    Journal of Neuroscience 29:12675–12685.

    https://doi.org/10.1523/JNEUROSCI.3359-09.2009

    • PubMed
    • Google Scholar
51. 1. Swann NC
    2. Cai W
    3. Conner CR
    4. Pieters TA
    5. Claffey MP
    6. George JS
    7. Aron AR
    8. Tandon N

    (2012) Roles for the pre-supplementary motor area and the right inferior frontal gyrus in stopping action: electrophysiological responses and functional and structural connectivity

    NeuroImage 59:2860–2870.

    https://doi.org/10.1016/j.neuroimage.2011.09.049

    • PubMed
    • Google Scholar
52. 1. Trujillo JP
    2. Gerrits NJ
    3. Veltman DJ
    4. Berendse HW
    5. van der Werf YD
    6. van den Heuvel OA

    (2015) Reduced neural connectivity but increased task-related activity during working memory in de novo Parkinson patients

    Human Brain Mapping 36:1554–1566.

    https://doi.org/10.1002/hbm.22723

    • PubMed
    • Google Scholar
53. 1. Tuladhar AM
    2. ter Huurne N
    3. Schoffelen JM
    4. Maris E
    5. Oostenveld R
    6. Jensen O

    (2007) Parieto-occipital sources account for the increase in alpha activity with working memory load

    Human Brain Mapping 28:785–792.

    https://doi.org/10.1002/hbm.20306

    • PubMed
    • Google Scholar
54. 1. van Vugt MK
    2. Schulze-Bonhage A
    3. Litt B
    4. Brandt A
    5. Kahana MJ

    (2010) Hippocampal gamma oscillations increase with memory load

    Journal of Neuroscience 30:2694–2699.

    https://doi.org/10.1523/JNEUROSCI.0567-09.2010

    • PubMed
    • Google Scholar
55. 1. Ventre-Dominey J
    2. Bourret S
    3. Mollion H
    4. Broussolle E
    5. Dominey PF

    (2014) Dissociable dorsal and ventral frontostriatal working memory circuits: evidence from subthalamic stimulation in Parkinson's disease

    Human Brain Mapping 35:552–566.

    https://doi.org/10.1002/hbm.22205

    • PubMed
    • Google Scholar
56. 1. Voytek B
    2. Knight RT

    (2010) Prefrontal cortex and basal ganglia contributions to visual working memory

    PNAS 107:18167–18172.

    https://doi.org/10.1073/pnas.1007277107

    • PubMed
    • Google Scholar
57. 1. Weinberger M
    2. Mahant N
    3. Hutchison WD
    4. Lozano AM
    5. Moro E
    6. Hodaie M
    7. Lang AE
    8. Dostrovsky JO

    (2006) Beta oscillatory activity in the subthalamic nucleus and its relation to dopaminergic response in Parkinson's disease

    Journal of Neurophysiology 96:3248–3256.

    https://doi.org/10.1152/jn.00697.2006

    • PubMed
    • Google Scholar
58. 1. Wessel JR
    2. Ghahremani A
    3. Udupa K
    4. Saha U
    5. Kalia SK
    6. Hodaie M
    7. Lozano AM
    8. Aron AR
    9. Chen R

    (2016a) Stop-related subthalamic beta activity indexes global motor suppression in Parkinson's disease

    Movement Disorders 31:1846–1853.

    https://doi.org/10.1002/mds.26732

    • PubMed
    • Google Scholar
59. 1. Wessel JR
    2. Jenkinson N
    3. Brittain JS
    4. Voets SH
    5. Aziz TZ
    6. Aron AR

    (2016b) Surprise disrupts cognition via a fronto-basal ganglia suppressive mechanism

    Nature Communications 7:11195.

    https://doi.org/10.1038/ncomms11195

    • PubMed
    • Google Scholar
60. 1. Wiesman AI
    2. Heinrichs-Graham E
    3. McDermott TJ
    4. Santamaria PM
    5. Gendelman HE
    6. Wilson TW

    (2016) Quiet connections: Reduced fronto-temporal connectivity in nondemented Parkinson's Disease during working memory encoding

    Human Brain Mapping 37:3224–3235.

    https://doi.org/10.1002/hbm.23237

    • PubMed
    • Google Scholar
61. 1. Zaghloul KA
    2. Weidemann CT
    3. Lega BC
    4. Jaggi JL
    5. Baltuch GH
    6. Kahana MJ

    (2012) Neuronal activity in the human subthalamic nucleus encodes decision conflict during action selection

    Journal of Neuroscience 32:2453–2460.

    https://doi.org/10.1523/JNEUROSCI.5815-11.2012

    • PubMed
    • Google Scholar
62. 1. Zavala B
    2. Brittain JS
    3. Jenkinson N
    4. Ashkan K
    5. Foltynie T
    6. Limousin P
    7. Zrinzo L
    8. Green AL
    9. Aziz T
    10. Zaghloul K
    11. Brown P

    (2013) Subthalamic nucleus local field potential activity during the Eriksen flanker task reveals a novel role for theta phase during conflict monitoring

    Journal of Neuroscience 33:14758–14766.

    https://doi.org/10.1523/JNEUROSCI.1036-13.2013

    • PubMed
    • Google Scholar
63. 1. Zavala B
    2. Damera S
    3. Dong JW
    4. Lungu C
    5. Brown P
    6. Zaghloul KA

    (2017) Human subthalamic nucleus theta and beta oscillations entrain neuronal firing during sensorimotor conflict

    Cerebral Cortex 27:496–508.

    https://doi.org/10.1093/cercor/bhv244

    • PubMed
    • Google Scholar
64. Data

    1. Zavala B
    2. Jang A
    3. Zaghloul K

    (authors) (2018) Data from: Human subthalamic nucleus activity during non-motor decision making

    Dryad Digital Repository.

    https://doi.org/10.5061/dryad.9t740
65. 1. Zavala B
    2. Tan H
    3. Ashkan K
    4. Foltynie T
    5. Limousin P
    6. Zrinzo L
    7. Zaghloul K
    8. Brown P

    (2016a) Human subthalamic nucleus-medial frontal cortex theta phase coherence is involved in conflict and error related cortical monitoring

    NeuroImage 137:178–187.

    https://doi.org/10.1016/j.neuroimage.2016.05.031

    • PubMed
    • Google Scholar
66. 1. Zavala B
    2. Tan H
    3. Little S
    4. Ashkan K
    5. Green AL
    6. Aziz T
    7. Foltynie T
    8. Zrinzo L
    9. Zaghloul K
    10. Brown P

    (2016b) Decisions made with less evidence involve higher levels of corticosubthalamic nucleus theta band synchrony

    Journal of Cognitive Neuroscience 28:811–825.

    https://doi.org/10.1162/jocn_a_00934

    • PubMed
    • Google Scholar
67. 1. Zavala B
    2. Zaghloul K
    3. Brown P

    (2015) The subthalamic nucleus, oscillations, and conflict

    Movement Disorders 30:328–338.

    https://doi.org/10.1002/mds.26072

    • PubMed
    • Google Scholar
68. 1. Zavala BA
    2. Tan H
    3. Little S
    4. Ashkan K
    5. Hariz M
    6. Foltynie T
    7. Zrinzo L
    8. Zaghloul KA
    9. Brown P

    (2014) Midline frontal cortex low-frequency activity drives subthalamic nucleus oscillations during conflict

    Journal of Neuroscience 34:7322–7333.

    https://doi.org/10.1523/JNEUROSCI.1169-14.2014

    • PubMed
    • Google Scholar

Author details

1. Baltazar A Zavala

   Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, Bethesda, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—original draft

   Competing interests

   No competing interests declared

2. Anthony I Jang

   Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, Bethesda, United States

   Contribution

   Data curation, Formal analysis, Writing—review and editing

   Competing interests

   No competing interests declared

3. Kareem A Zaghloul

   Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, Bethesda, United States

   Contribution

   Conceptualization, Data curation, Supervision, Funding acquisition, Methodology, Project administration, Writing—review and editing

   For correspondence

   kareem.zaghloul@nih.gov

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-8575-3578

• 2,529

  views

• 435

  downloads

• 41

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.