Skip to main content
NCBI home page
Journal of Neurophysiology logoLink to Journal of Neurophysiology
. 2017 Feb 22;117(6):2152–2162. doi: 10.1152/jn.00908.2016

Fixation target representation in prefrontal cortex during the antisaccade task

Xin Zhou 1,2, Christos Constantinidis 2,✉
PMCID: PMC5454469  PMID: 28228585

Properties of neurons that discharge during eye fixation and go silent before saccade initiation have been described in subcortical structures involved in eye movement generation, but their role in the dorsolateral prefrontal cortex presents a puzzle. Our results demonstrate the role of fixation neurons in the prefrontal cortex during tasks requiring precise timing of appropriate eye movement and inhibition of inappropriate actions.

Keywords: monkey, neurophysiology, fixation, saccade, inhibition

Abstract

Neurons that discharge strongly during the time period of fixation of a visual target and cease to discharge before saccade initiation have been described in the brain stem, superior colliculus, and cortical areas. In subcortical structures, fixation neurons play a reciprocal role with saccadic neurons during the generation of eye movements. Their role in the dorsolateral prefrontal cortex is less obvious, and it is not known if they are activated by fixation, inhibit saccade generation, or play a role in more complex functions such as the inhibition of inappropriate responses. We examined the properties of prefrontal fixation neurons in the context of an antisaccade task, which requires an eye movement directed away from a prepotent visual stimulus. We tested monkeys with variants of the task, allowing us to dissociate activity synchronized on the fixation offset, presentation of the visual stimulus, and saccadic onset. Fixation neuron activity latency was most strongly tied to the offset of the fixation point across task variants. It was not well predicted by the appearance of the visual stimulus, which is essential for planning of the correct eye movement and inhibiting inappropriate ones. Activity of fixation neurons was generally negatively correlated with that of saccade neurons; however, critical differences in timing make it unlikely that they provide precisely timed signals for the generation of eye movements. These results demonstrate the role of fixation neurons in the prefrontal cortex during tasks requiring timing of appropriate eye movement and inhibition of inappropriate actions.

NEW & NOTEWORTHY Properties of neurons that discharge during eye fixation and go silent before saccade initiation have been described in subcortical structures involved in eye movement generation, but their role in the dorsolateral prefrontal cortex presents a puzzle. Our results demonstrate the role of fixation neurons in the prefrontal cortex during tasks requiring precise timing of appropriate eye movement and inhibition of inappropriate actions.

neuronal activity elicited by a fixation target in oculomotor tasks has been described in brain areas involved in the control of eye movements, including the brain stem (Keller 1974) superior colliculus (Munoz et al. 1991; Munoz and Wurtz 1993), the posterior parietal cortex (Mountcastle et al. 1975), the frontal eye fields (Bizzi 1968; Suzuki and Azuma 1977), and the dorsolateral prefrontal cortex (Tinsley and Everling 2002). Fixation neurons in the brain stem and superior colliculus resemble the mirror image of saccadic neurons; they are active during stable fixation and are suppressed before the onset of eye movements (Munoz et al. 1991; Munoz and Wurtz 1993). Fixation and movement neurons are thought to form a mutually inhibitory circuit in the superior colliculus, regulating the generation of eye movements (Meredith and Ramoa 1998; Munoz and Istvan 1998).

The functional role of fixation neurons that have been described in the dorsolateral prefrontal cortex represents more of a puzzle. In principle, it is possible that they replicate oculomotor control circuits described in the superior colliculus and brain stem, which involve mutual inhibition of fixation and saccade neurons (Everling et al. 1998). This possibility seems uncertain, however, considering that the dorsolateral prefrontal cortex, unlike the superior colliculus or frontal eye field, is only indirectly involved in the control of eye movements. Even frontal eye field neurons projecting to the superior colliculus are activated by cognitive processes such as working memory and attention and do not convey information only limited to eye movements (Sommer and Wurtz 2001). A wide range of information about stimuli and internal variables are represented in the activity of dorsolateral prefrontal neurons (Katsuki and Constantinidis 2012; Miller and Cohen 2001), and prefrontal responses are plastic, based on learning of the demands of various tasks (Constantinidis and Klingberg 2016; Qi et al. 2011). It is possible, therefore, that fixation neurons in the dorsolateral prefrontal cortex are involved in more complex roles. These include the inhibition of inappropriate actions, a critical function of the prefrontal cortex (Bari and Robbins 2013), and the representation of the task significance of the fixation point, whose onset and offset can cue the animals about generating (or withholding) a response in the context of trained behavioral tasks.

We were motivated, therefore, to investigate the response properties of neurons active during fixation in variants of an antisaccade task. This task requires the ability to inhibit inappropriate eye movements toward a prepotent stimulus (Munoz and Everling 2004) and for this reason was well suited to examine the role of fixation neurons. We employed different variants of the task, differing in their relative timing of the fixation point offset, visual stimulus presentation, and saccadic onset. Our analysis focused on three questions: whether prefrontal fixation neurons encode the presence of the fixation target; whether they could provide inhibition of inappropriate movements toward the visual stimulus in the anti-saccade task; and whether they could provide inhibition of saccade neurons before the generation of saccades.

METHODS

Four male, rhesus monkeys (Macaca mulatta) were used in these experiments. The animals were young adults, 6–8 yr of age and weighing 7–11 kg at the time these recordings were obtained. These animals had previously undergone neurophysiological recordings from the prefrontal cortex at the time of adolescence. More details about the monkeys, and the properties of neurons with responses to the visual stimuli and eye movements, have been reported elsewhere (Zhou et al. 2016b; Zhou et al. 2016c). All surgical and animal-use procedures in this study followed guidelines by the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals and the National Research Council’s Guide for the Care and Use of Laboratory Animals, and were reviewed and approved by the Wake Forest University Institutional Animal Care and Use Committee.

Behavioral tasks.

The animals were trained to perform an oculomotor delayed response (ODR) task (Funahashi et al. 1989) and variants of the antisaccade task (Hallett 1978). The ODR task is a spatial working memory task that requires the subjects to remember the location of a cue stimulus flashed on a screen for 0.5 s. The monkey had to maintain fixation throughout the trial on a white fixation point and remember the location of a visual cue. Deviations greater than 2° from the fixation point resulted in errors, and the trial was aborted immediately. The cue was a 1° white square stimulus that could appear at one of eight locations arranged on a circle of 10° eccentricity. After a 1.5-s delay period, the fixation point was extinguished and the monkey was trained to make an eye movement to the remembered location of the cue within 0.6 s. The saccade needed to terminate in a 5°–6° radius window centered on the stimulus (within 3°–4° from the edge of the stimulus, slightly different for each monkey), and the monkey was required to hold fixation within this window for at least 0.1 s (to prevent reward for an eye movement that swept past the correct location). In this study, we relied on the ODR task mainly to determine the effect of saccadic eye movements on fixation neurons without being influenced by the response to the stimulus, since the saccade is temporally separated from the stimulus presentation in the ODR task.

In the antisaccade task, each trial started with the monkey fixating a central green point on the screen, also having to remain within a 2° window. After 1-s fixation, the cue was presented, consisting of a 1° white square stimulus that could appear at one of eight locations arranged on a circle of 10° eccentricity for 0.1 s. These were the same locations used in the ODR task, although the duration of the stimulus presentation was shorter to encourage as fast a response as possible. The monkey was then required to make a saccade to the location diametric to the cue. The saccade needed to end within a 5°–6° radius window around this location, and the monkey needed to maintain fixation for at least 0.1 s, as in the ODR task. In all tasks, monkeys typically maintained stable gaze on the saccade target for 200–300 ms. Animals were rewarded with fruit juice for successful completion of a trial. Eye position was monitored with an infrared eye tracking system (RK-716; ISCAN, Burlington, MA). Breaking fixation at any point before the offset of the fixation point aborted the trial and resulted in no reward. The stimulus presentation and online behavioral control was achieved by in-house software (Meyer and Constantinidis 2005).

We used three different variants for the antisaccade task: overlap, zero gap, and gap, differing in the sequence of the cue onset relative to the fixation point offset (Fig. 1). In the overlap condition, the cue appears first, and then both fixation point and cue are simultaneously extinguished. In the zero-gap condition, the fixation offset and the cue onset occur at the same time. In the gap condition, the fixation turns off and a 100-ms blank screen is inserted before the cue onset. Two of the monkeys were additionally tested with a 200-ms gap variant of the task. This task provided a longer interval over which the monkey was maintaining fixation, without a physical fixation target present, allowing us to better dissociate the foveal response to the fixation point from the oculomotor act of maintaining stable gaze. This condition was tested after recordings with the 100-ms gap variant were complete and at that time, it was not practical for the two other animals to be trained in this new task. The location of the stimulus and the task variant were presented pseudorandomly across trials.

Fig. 1.

Fig. 1.

Open in a new tab

Behavioral tasks. A: sequence of events in the overlap variant of the antisaccade task. The cue presentation and fixation point overlap for 100 ms before they both turn off and signal the requirement for a saccade away from the cue. B: zero-gap variant. The fixation point turns off simultaneously with the cue onset. C: 100-ms gap variant of the antisaccade task. The fixation point turns off, and after a 100-ms gap, the cue appears. D: oculomotor delayed response (ODR) task. A visual cue is presented, requiring an eye movement toward it, after a delay period. E: possible stimulus locations in the antisaccade and ODR tasks.

Monkeys performed blocks of ODR trials and antisaccade trials on a daily basis, although the two tasks were not mixed on a trial-by-trial basis, to minimize confusion about the task. The color of the fixation point served as an additional indication of the task to be performed in each trial. Eight to 10 correct trials were typically collected from each location in each antisaccade task variant. Correct performance in the antisaccade task averaged ~75% across tasks variants and individual monkeys, as we have reported in detail elsewhere (Zhou et al. 2016b). A negligible percentage of trials were aborted in this task, because responses tended to be immediate after the presentation of the stimuli.

Surgery and neurophysiology.

Recordings were performed from a 20-mm-diameter recording cylinder implanted over the dorsolateral prefrontal cortex, accessing areas 8a and 46. More details of surgical and recording procedures have been reported previously (Zhou et al. 2016b). The localization of the recording cylinder was based on magnetic resonance (MR) imaging, processed with the BrainSight system (Rogue Research, Montreal, ON, Canada), and this was further verified based on the sulcal pattern encountered by our electrodes. Precise histological localization was not available, but we distinguished between area 46 (encompassing the 2 banks of the principal sulcus) and area 8a (involving the area between the principal and arcuate sulcus). Recordings were collected with Epoxylite-coated tungsten electrodes with a diameter of 250 μm and an impedance of 4 MΩ at 1 kHz (FHC, Bowdoin, ME). Electrical signals recorded from the brain were amplified, bandpass filtered between 0.5 and 8 kHz, and stored through a modular data acquisition system at 25-μs resolution (APM System; FHC).

Neural data analysis.

Recorded spike waveforms were sorted into separate units using an automated cluster analysis method referred to as the KlustaKwik algorithm (Harris et al. 2000). This relies on principal component analysis of the waveforms. Mean firing rate of each unit was then determined in several task epochs. We identified units as prefrontal fixation neurons if they exhibited significant elevation of firing rate (evaluated with a paired t-test, pooling all available trials, and evaluated at the 0.001 significance level) in the 1,000-ms fixation period compared with a 1,000-ms period during the intertrial interval, beginning 400 ms after the disappearance of the stimulus in the antisaccade task. The monkeys were free to move their eyes during the intertrial interval, and the period analyzed contained a series of eye movements and fixations (in the dark). We excluded from the sample of fixation neurons those that had significant responses to visual stimuli evidenced by significant elevation of firing rate (paired t-test, P < 0.05) during the 200 ms following the stimulus presentation period, as well as those with responses to the saccade in the 200 ms following the onset of the saccade, for any of the stimuli used, over the fixation period. These selection criteria for identifying task selective neurons were used in prior studies (Zhou et al. 2016a; Zhou et al. 2016b). We also excluded neurons with a transient increase in activity after the cue onset of 5 spikes/s or more, which was consistent for the same stimulus location across task variants, even if these neurons otherwise failed to meet the criteria of responsiveness in the antisaccade task. Finally, we excluded neurons with “inverted tuning” (Zhou et al. 2012), displaying selectivity for the stimulus location in the cue or delay period (1-way ANOVA, P < 0.05) by virtue of suppressed firing rate for some locations. The same method for the identification of fixation neurons could have been based in the ODR task, as well, but we sometimes observed differences in firing rate of the same neuron between tasks and an overall higher level of fixation activity in the antisaccade task compared with the ODR (Zhou et al. 2016a; Zhou et al. 2016b). For this reason, we limited all of our comparisons within the antisaccade task. Details about the properties of other types of neurons obtained in this experiment have been reported elsewhere (Zhou et al. 2016a; Zhou et al. 2016b).

To determine the relative timing of different task intervals, activity of prefrontal fixation neurons in the ODR and antisaccade tasks was aligned to three task time points: the offset of the fixation point, the onset of the saccade, and the onset of the stimulus. To determine the onset and offset of saccades, we first determined the eye position corresponding to the time point of maximum eye velocity following the offset of fixation point, based on the maximum eye position distance between successive samples during the saccade period, smoothed with an 8-ms triangular filter and resampled at 4 ms (Zhou et al. 2013). The onset and offset of the saccade was determined as the eye position at the first and last point of monotonic deviation away from the fixation point, preceding or following the point of maximum velocity, respectively.

Baseline activity was determined in a 200-ms interval before the offset of the fixation point. A second baseline activity was determined in a 200-ms interval after the end of the saccade. Spike trains were convolved with a Gaussian kernel of σ = 10 ms, to yield a continuous spike density function, and were averaged across trials pooled from all stimulus locations. The first time point in which activity decreased below the midpoint between the two baseline levels was used to determine the timing of the activity change. The statistical significance of the difference in timing between two tasks was evaluated by a permutation test. We simulated population activity across pseudovariants by randomly assigning trials from each neuron to different task variants. We then evaluated the midpoint of firing rate decrease in each condition and determined the difference in midpoint timing between conditions. This process was repeated 1,000 times, allowing us to compare the empirical difference with the simulated distribution of differences.

Firing rate differences between tasks were compared with a repeated-measures ANOVA test (comparing firing rates of the same neurons across tasks). For this test, the mean firing rate was calculated in 200-ms bins aligned on the onset of the cue, on the onset of the fixation point, and preceding the onset of the saccade. These rates were based on the raw spike counts of neurons, without any smoothing.

RESULTS

Neurophysiological data were collected from four monkeys while they performed variants of the antisaccade task, which requires an eye movement to a location diametrically opposed to a visual stimulus (Fig. 1, A–C). We used three temporal variants of the antisaccade task: one in which the visual cue and fixation point overlapped for 100 ms before both turn off (overlap; Fig. 1A), one in which onset of the visual cue and offset of the fixation point occurred simultaneously (zero gap; Fig. 1B), and one in which the visual cue was presented 100 ms after the fixation point was extinguished (gap; Fig. 1C). Saccades were initiated faster in the gap variant; however, performance was also lowest in this variant, because it was most difficult to suppress an eye movement toward the stimulus without the fixation point being present (Zhou et al. 2016b). The oculomotor delayed response (ODR) task (Fig. 1D) was also used. This task requires monkeys to make an eye movement to the location of a remembered visual stimulus, which could be at one of eight locations (Fig. 1E).

Database.

A total of 1,076 neurons were recorded in areas 8a and 46 of the dorsolateral prefrontal cortex (Fig. 2) while the monkeys executed the antisaccade and ODR tasks. Of those, 715 neurons responded significantly to a stimulus presentation or to a saccade in the antisaccade task compared with baseline activity (paired t-test, P < 0.05). These neurons were initially omitted from analysis. From the remaining 361 neurons, we identified those with significantly elevated firing rate in the fixation period compared with the intertrial period, evaluated with a paired t-test pooling all available trials at the α = 0.001 significance level. A total of 34 neurons met the criteria and were selected for further analysis. These represented 9.4% of the neurons that did not otherwise respond in the antisaccade task (compared with an expected false positive rate of 0.1% of the statistical test used) and 3.2% of the total number of neurons recorded (11/204 = 5.4%, 5/235 = 2.1%, 17/597 = 2.8%, and 1/40 = 2.5% for the 4 monkeys, respectively). Penetrations where these neurons were recorded were identified in both areas 8a and 46. The locations of penetrations relative to the principal and arcuate sulcus are shown for the two animals with the most recordings in Fig. 2, B and C.

Fig. 2.

Fig. 2.

Open in a new tab

Recording areas. A: schematic diagram of the monkey brain with areas 8a and 46 of the dorsolateral prefrontal cortex indicated. AS, arcuate sulcus; PS, principal sulcus. B: electrode penetrations where prefrontal fixation neurons were recorded in one animal. C: electrode penetrations in a second animal.

Overview of activity and selectivity of prefrontal fixation neurons.

Responses from a typical prefrontal fixation neuron in the anti-saccade task are shown in Fig. 3. The neuron discharged tonically during the fixation interval and the onset of the stimulus, without regard to the location of the stimulus (Fig. 3, A–D). Synchronized on the saccade, discharge rate declined rapidly, regardless of the direction of the eye movement (Fig. 3, A–D). The timing of discharge varied subtly but consistently for the different task variants, with activity decaying faster in the gap variant and most slowly in the overlap variant (Fig. 3E). Responses from all neurons in the antisaccade task are shown in Fig. 4, confirming this timing difference across the population. Activity of the same neurons in the ODR task is shown in Fig. 5 for comparison. In this task, too, activity started to rise within 50 ms after the onset of the fixation point; it remained elevated through the entire interval period of the stimulus presentation and delay period and declined abruptly following the saccade.

Fig. 3.

Fig. 3.

Open in a new tab

Activity of one prefrontal fixation neuron during the antisaccade task. A: eye position traces for horizontal (X) and vertical position (Y), rasters, and peristimulus time histograms (PSTHs) of a single prefrontal fixation neuron are shown during the antisaccade task. Gray square at top indicates that the cue appeared on the right of the screen, requiring a leftward saccade. Plot limits in eye position traces extend from −15° to 15°. The vertical lines in all plots indicate the time of initiation of the saccade. Top plots were recorded from the overlap variant of the antisaccade task, middle plots from the zero-gap variant, and bottom plots from the 100-ms gap variant. B–D: activity of the same neuron as in A, for 3 different stimulus and saccade locations (sequence of screen displays indicated by gray squares at top). E: average activity for the same neuron, collapsed across all stimulus/saccade locations, plotted for the 3 variants of the task. Activity is synchronized to the onset of the saccade (indicated as a vertical line).

Fig. 4.

Fig. 4.

Open in a new tab

Population activity of fixation neurons in the antisaccade task. A: average PSTH for all trials of prefrontal fixation neurons during the antisaccade task (n = 34). Activity is plotted in 1 s of the intertrial interval; shaded area represents SE. Gray square at top represents blank screen. B: activity recorded from the same neurons as in A, relative to the onset of the fixation point. C: activity from the 3 variants of the antisaccade task shown separately (overlap, zero gap, 100-ms gap). Gray squares at top illustrate sequence of events in each task variant. Activity is synchronized to the offset of the fixation point (indicated by the dashed vertical line). D: same data as in C, with activity synchronized to the onset of the saccade (indicated by the vertical line). E: same data as in C, with activity synchronized to the stimulus presentation (indicated as a gray bar).

Fig. 5.

Fig. 5.

Open in a new tab

Population activity of fixation neurons in the ODR task. A: average population PSTH for all trials of prefrontal fixation neurons during the ODR task (n = 34; same neurons as in Fig. 4). Activity is shown before the onset of the fixation point, synchronized to fixation point onset (Fix on; vertical line), and synchronized on the cue presentation (gray bar) and on the fixation offset (Fix off; dashed vertical line). Gray area around line represents SE. Sac, saccade.

The general lack of selectivity for stimulus location and saccade direction was confirmed when we compared the firing rate of prefrontal fixation neurons following the presentation of the visual stimulus at each of the eight locations in the ODR task. A 1-way ANOVA indicated a significant (P < 0.05) preference for stimulus location in only 3/34 fixation neurons. Similarly, significant selectivity in firing rate after the fixation offset for the direction of the upcoming saccade was present in 4/34 neurons and in the postsaccadic interval in 3/34 neurons (ANOVA test, P < 0.05). These findings were not substantially different when we pooled data from all ipsilateral and contralateral locations. A total of 3/34, 5/34, and 3/34 neurons exhibited significant selectivity for ipsilateral vs. contralateral locations in the cue, presaccadic, and postsaccadic intervals, respectively (ANOVA test, P < 0.05). We conclude that prefrontal fixation neurons were generally not sensitive to the location of the peripheral visual stimulus or saccade.

Activity relative to fixation point offset and cue onset.

An important question about the role of prefrontal fixation neurons was the timing of their activation with respect to the offset of the fixation point. The activity of prefrontal fixation neurons may decline after the fixation target is extinguished or may continue to be elevated while the monkeys maintain fixation after the fixation target turns off, indicating that gaze maintenance primarily activates them. To distinguish between these two alternatives, we compared activity in the variants of the antisaccade task. The gap variant of the task, in particular, requires sustained gaze after the fixation point has been extinguished.

The offset of the fixation point predicted best the change in activity across all task conditions (Fig. 4C). We quantified the difference in timing between task conditions by determining the time point at which firing rate decreased below the midpoint between the stable firing rate before fixation offset and the new baseline after saccade execution. Aligned on the offset of the fixation point, the midpoints of firing rate decreases were tightly clustered between tasks: gap, 126 ms; zero gap, 139 ms; and overlap, 107 ms (Fig. 4C). Comparison between the gap and overlap task variants revealed that the timing of neural activity differed by 19 ms, although the period of stable gaze differed by 200 ms in these conditions. The decrease in activity also preceded the onset of the saccade and the shift in gaze that this signified (Fig. 4D). The difference between task conditions was most evident when we synchronized activity on the cue (Fig. 4E). Activity in the gap condition began to decline well before the onset of the cue, after the fixation point was extinguished, even though gaze was maintained stable for another ~200 ms, a time necessary for the monkey to identify the appropriate target of the saccade. Indeed, the midpoints of firing rate decreases relative to the cue onset now spanned 177 ms (task latency: gap, 20 ms; zero gap, 139 ms; and overlap, 197 ms in Fig. 4E). This difference in timing between the gap and overlap tasks was significantly different from chance (permutation test, P < 0.001).

A subset of 8 prefrontal fixation neurons was additionally tested with a variant of the anti-saccade task imposing a 200-ms gap between the fixation offset and cue onset, allowing us to better dissociate foveal responses to the fixation point from maintaining gaze (Fig. 6). The offset of the fixation point relative to the cue for the 200-ms gap and overlap conditions differed by 300 ms. The midpoints of firing rate decreases were tightly clustered relative to the fixation offset across task variants: 200-ms gap, 151 ms; 100-ms gap, 132 ms; zero gap, 138 ms; overlap, 116 ms (Fig. 6A). On the other hand, the decrease of activity occurred at discrete intervals relative to the onset of the cue (Fig. 6C). In the 200-ms gap variant, activity began to decline well before the onset of the cue and had essentially returned to baseline by the time the cue had appeared and the saccade direction could be planned. The midpoints of firing rate decreases relative to the cue onset were separated accordingly (200-ms gap, −57 ms; 100-ms gap, 19 ms; zero gap, 108 ms; and overlap, 212 ms in Fig. 6C). The difference in midpoint between firing rates recorded in the 200-ms gap task variant and the overlap task variant spanned 269 ms, an interval very close to the 300-ms difference in the offset of the fixation point relative to the cue for these tasks. This difference between task variants in midpoint of firing rate decline was also significant compared with chance (permutation test, P < 0.001). These results indicate that in the context of the antisaccade, task prefrontal fixation neurons are modulated primarily by the presence or absence of a fixation point.

Fig. 6.

Fig. 6.

Open in a new tab

Population activity in the 200-ms gap variant of the antisaccade task. A: average population PSTH for all trials of prefrontal fixation neurons tested with a 200-ms gap variant of the antisaccade task, in addition to the other variants (n = 8). Activity is synchronized to the offset of the fixation point (indicated by the dashed vertical line). B: same data as in A, with activity synchronized to the onset of the saccade (indicated by the vertical line). C: same data as in A, with activity synchronized to the cue stimulus presentation (indicated as a gray bar).

Prefrontal fixation neurons may in principle be modulated by the timing of the visual cue, in addition to the fixation point. We tested this possibility by comparing firing rates synchronized to the cue onset, across antisaccade task conditions. The results did not reveal a consistent modulation of activity by the onset of the cue. Whereas activity of fixation neurons decayed after the cue appearance in the overlap condition, it decayed before the onset of the cue in the 100-ms gap condition (Fig. 4E). An ANOVA test comparing mean firing rates in a 500-ms period following the onset of the cue revealed a highly significant difference between task conditions (1-way ANOVA, F2,66 = 17.48, P < 10−5). In contrast, firing rate after the fixation offset was consistent across all task conditions, and an ANOVA test comparing mean rates in a 500-ms period following the fixation point offset showed that rates did not differ significantly between tasks (1-way ANOVA, F2,66 = 0.92, P = 0.4). The result indicates that the offset of the fixation point modulates activity of fixation neurons across tasks. Because firing rate began at the same level during the fixation interval across all tasks, decayed with the same rate, and returned to the same baseline after the execution of the saccade, the result also argues against a consistent output of fixation neurons that could play a role in the inhibition of the appropriate responses across task conditions. The cue onset, only after which the monkey is informed about the direction of saccade to be executed, was not a good predictor of activity change among prefrontal fixation neurons.

Activity related to saccade preparation.

Fixation neurons in the superior colliculus are thought to actively inhibit saccadic neurons; the release of saccadic neurons from inhibition after the offset of the fixation point is partially responsible for the faster reaction times in the gap condition (Everling et al. 1998). The activity of fixation neurons in cortical areas such as the frontal eye field also exhibits an inverted pattern of activation compared with that of saccade neurons (Izawa and Suzuki 2014; Izawa et al. 2009). We therefore sought to examine the activity of prefrontal fixation neurons relative to saccade preparation. Aligned to the saccade, the activity decrease of neurons in our sample preceded the onset of the eye movement (Fig. 4D).

To more directly test the role of prefrontal fixation neurons in saccade generation, we compared their activity with that of saccade neurons (Fig. 7). A total of 68 neurons with motor but no visual activity were recorded during the course of the same experiments, using the same task (Zhou et al. 2016b). We thus plotted the difference between the mean firing rates of fixation neurons, minus the activity of saccade neurons. Positive values of this variable represent the (putative) inhibitory drive of fixation neurons onto saccade ones, and negative values are indicative of release from inhibition. The time course of activation appeared very similar between fixation and saccade neurons, and with little difference between task conditions (Fig. 8A). Prefrontal fixation neurons therefore could be contributing to the generation of saccades by releasing saccade neurons from inhibition. Some differences between the two groups of neurons were also evident, however. The activity of prefrontal saccade neurons best synchronized with the onset of the saccade (Fig. 7B), and a >100-ms difference in latency between task conditions was present in saccade neurons (but not in fixation neurons) when we aligned responses to the offset of the fixation point (Fig. 7A). Only a partial separation between task conditions was also evident in saccade neuron responses aligned on the cue, which also deviated from the timing of fixation neuron responses (Fig. 7C). As a result, the putative inhibitory drive differed considerably between task variants when aligned to the cue (Fig. 8B). In the 100-ms gap task, firing rate of fixation neurons decays essentially simultaneously with the increase in firing rate of saccade neurons, so the difference in firing rate declines abruptly. In the overlap task, however, activity of saccade neurons begins to rise before the activity of fixation neurons starts to decline. As a consequence, the inhibitory drive appears to lag the gap-variant curve, by ~100 ms. This suggests that the level of activation of saccade neurons is not tightly determined by the activity of fixation neurons.

Fig. 7.

Fig. 7.

Open in a new tab

Population activity of saccade neurons in the anti-saccade task. A: average population PSTH for all trials of saccade neurons during the antisaccade task (thick lines; n = 68). Three variants of the antisaccade task are shown (overlap, zero gap, 100-ms gap). Activity is synchronized to the offset of the fixation point (indicated by the dashed vertical line). Activity of fixation neurons is plotted on the same axes, for comparison (thin lines), duplicating the set of traces in Fig. 4. B: average activity of saccade neurons, with activity synchronized to the onset of the saccade (indicated by the vertical line). C: average activity of saccade neurons, with activity synchronized to the stimulus presentation (indicated as a gray bar).

Fig. 8.

Fig. 8.

Open in a new tab

Difference in activation between fixation and saccade neurons. A: difference in firing rate between the mean firing rate of the population of fixation neurons (n = 34) and the mean firing rate of the population of saccade neurons (n = 68), plotted as a function of time. Results are plotted separately for each variant of the antisaccade task, synchronized on the onset of the saccade. B: difference in firing rate in the same groups of neurons as in A, now synchronized to the onset of the cue. C: difference in activity of saccade neurons pooled from trials with slow and fast reaction times (n = 68). Activity from all task variants has been pooled together in this illustration. D: difference in activity of fixation neurons, pooled from trials with slow and fast reaction times (n = 34).

To further investigate the potential role of fixation neurons on saccade generation, we compared activity in trials with slower and faster reaction times, defined by a median split of reaction times, which was performed separately for each recording session, task variant, and stimulus location, so as to produce balanced fast- and slow-reaction-time groups. The activity of saccade neurons was distinguishing of the two groups. Firing rate aligned on the saccade onset reached a peak in a shorter interval in trials with faster reaction times (shifted to the right in Fig. 8C). As a result, mean firing rate in the 200 ms preceding the saccade onset was higher in the faster than in the slower trials, by an average of 1.3 spikes/s. This difference was highly significant in a three-way ANOVA test using as factors neuron; fast/slow reaction time group; and antisaccade task type (main fast/slow group effect: F1,6417 = 21.5, P = 3.6 × 10−6). In fixation neurons, activity also decayed in a shorter interval for faster trials, synchronized to the saccade (Fig. 8D). The difference in firing rate between slow and fast conditions was smaller for fixation neurons (0.8 spikes/s), although the effect still reached significance (F1,3088 = 6.57, P = 0.01). This analysis, too, suggested that the activation of fixation neurons correlated with motor behavior, although it was likely not the primary determinant of the saccade neuron pattern of activation.

Fixation modulation by neurons with task activity.

Our analysis so far focused on neurons that were not significantly activated by the visual stimulus or saccade. However, it is possible that a subset of these neurons is also modulated by the fixation point and represents fixation information, which could be used to guide behavior in the task. For this reason, we repeated our comparison of the fixation interval with the intertrial interval for all 715 neurons responsive to the cue presentation or saccade, during the antisaccade task. Indeed, a total of 143 neurons (20%) exhibited significantly elevated activation during the fixation interval relative to that during the intertrial interval (paired t-test, P < 0.001).

These task-related neurons responded strongly to the onset of the fixation point, as well as the presentation of the cue, in the context of the ODR task (average response across 8 peripheral locations is shown in Fig. 9A). In the antisaccade task, task-related neurons exhibited strong anticipatory activity before the onset of the stimulus (Fig. 9B), as we have described elsewhere (Zhou et al. 2016a). These neurons also responded strongly to the presentation of the stimulus itself (Fig. 9A).

Fig. 9.

Fig. 9.

Open in a new tab

Fixation activity of neurons responsive to the task. A: average population PSTH for all trials of prefrontal neurons exhibiting fixation activity and responding during the task (n = 143). Activity in the ODR task is shown before the onset of the fixation point, synchronized to fixation point onset, and synchronized on the cue presentation and fixation offset. Gray area around discharge rate trace represents SE. B: average population activity for the same neurons as in A, in the antisaccade task, with responses synchronized to the onset of the cue. C: average population activity for the same neurons as in A, in the antisaccade task, with responses synchronized to the fixation offset.

A notable difference between task-related and fixation neurons had to do with their pattern of activity after the offset of the fixation point. Task-related neurons continued to be active after the offset of the fixation point in the antisaccade variants in which the fixation point turned off before or simultaneously with the cue, in contrast with the fixation neurons, whose rapid decline in activity signaled reliably the offset of the fixation point. Whereas activity of task-related neurons rapidly declined in the overlap task after the fixation offset, a strong visual transient was observed precisely at the time after the fixation offset occurred in the zero-gap and gap conditions (Fig. 9C). Additionally, many of the task-related neurons were responsive to saccade generation toward the receptive field, which followed closely the offset of the fixation point. This was most clearly evident in the ODR task, with a strong transient response being present at the time of the fixation point offset (Fig. 9A). In this case, too, the activity of task-related neurons could not provide a signal for the offset of the fixation point across conditions, as the fixation neurons did.

DISCUSSION

Our study determined the properties of prefrontal neurons representing the fixation target in the context of the antisaccade task. Activity of the dorsolateral prefrontal cortex is thought to influence and control presaccadic activation in the superior colliculus, exerting a net excitatory effect on the ipsilateral side (Johnston et al. 2014). However, frontal neurons represent a variety of information and transmit signals not only related to eye movements (Sommer and Wurtz 2001). It has been unclear, therefore, what information modulates the activity of prefrontal fixation neurons and what their functional output might be. We identified neurons with elevated activity during fixation relative to the intertrial interval. Variants of the antisaccade task allowed us to determine the properties of these neurons relative to the presentation of visual stimuli, the presence or absence of the fixation point, and the onset of saccades. We also contrasted their activity with those visual neurons responsive to the fixation point and to motor neurons activated by the saccade. Our findings illustrate that prefrontal fixation neurons are primarily modulated by the offset of the fixation point rather than the onset of a saccade, that their activity cannot guide inhibition of specific eye movements toward inappropriate targets, and that their activation profile is generally reciprocally related but distinct from those of saccade neurons, whose activation is more tightly tied to the onset of saccades.

Properties of fixation neurons in cortical and subcortical structures.

Fixation neurons have been described extensively in the brain stem and superior colliculus (Munoz et al. 1991; Munoz and Wurtz 1993). Their properties in many respects represent the mirror image of movement (saccadic) neurons; they are active during stable fixation and are suppressed before the onset of eye movements. Fixation and movement neurons are thought to form a mutually inhibitory circuit in the superior colliculus, regulating the generation of eye movements (Meredith and Ramoa 1998; Munoz and Istvan 1998). When a gap intervenes between the fixation point offset and the appearance of a target that instructs a saccade toward it, neurons in the superior colliculus attenuate their firing, releasing saccadic neurons from inhibition and presumably facilitating faster reaction times compared with the condition involving the fixation point and target appearing simultaneously (Dorris and Munoz 1995). Our current results mirror this finding in the dorsolateral prefrontal cortex for the antisaccade task (Fig. 4).

Neurons active during the fixation in the superior colliculus were further distinguished in two categories, based on their behavior during a period of constant gaze, when the fixation point briefly disappeared (Munoz et al. 1991; Munoz and Wurtz 1993): neurons that continued to discharge and neurons whose firing diminished during the “blink” period (and were termed foveal visual neurons). The pattern of activity we describe now in the prefrontal cortex more closely matches that of the foveal visual neurons in the superior colliculus.

Fixation neurons have been described in the frontal eye field (FEF) and dorsolateral prefrontal cortex (Bizzi 1968; Suzuki and Azuma 1977; Suzuki et al. 1979; Tinsley and Everling 2002). Fixation neurons in the FEF were shown to consist of corticofugal neurons projecting to the superior colliculus (Segraves and Goldberg 1987). Although they can be found across the entire FEF, they are concentrated in a region of the ventral limb of the arcuate sulcus, electrical microstimulation of which suppresses visually guided saccades and smooth pursuit movements (Izawa and Suzuki 2014; Izawa et al. 2009). It may be tempting to assume that fixation neurons of the frontal lobe transmit signals relevant to eye position onto the fixation of the superior colliculus, but FEF fixation neurons do not target predominantly collicular fixation neurons, which reside rostrally in the superior colliculus (Sommer and Wurtz 2000). Information transmitted to the superior colliculus by the frontal cortex has also been shown to originate from a variety of neurons with activation related to cognitive processes such as memory and attention, and not only those signaling information about eye movements (Sommer and Wurtz 2001). These findings rule out a pure point-to-point transmission of fixation signals from the frontal lobe into the superior colliculus.

Even less of such a link would be expected for the dorsolateral prefrontal cortex, whose neurons are modulated by a variety of cognitive processes, other than the control of movement (Miller et al. 2002; Riley and Constantinidis 2016). Fixation neurons in the frontal cortex exhibit reciprocal patterns of activation with neurons active during saccades or smooth pursuit (Izawa and Suzuki 2014; Izawa et al. 2009). Their activity is related to active fixation and differs when the monkey maintains its gaze stable without a fixation target compared with when it is fixating one (Izawa and Suzuki 2014; Suzuki et al. 1979). It is thought therefore that they are part of a mutually inhibitory circuit that suppresses all types of eye movements, at least in the contralateral hemifield (Izawa and Suzuki 2014). Activity of frontal fixation neurons is also critical in canceling the generation of an impending eye movement, e.g., after a stop signal has been issued in a countermanding paradigm (Hanes et al. 1998; Lo et al. 2009).

Role of prefrontal fixation neurons.

The dominant representation of cognitive signals in the prefrontal cortex raises the possibility of a role of fixation neurons in cognitive tasks, including the recall of the remembered stimulus in the ODR task and the inhibition of an inappropriate response toward the stimulus in the antisaccade task. Such a role, based on abstract stimulus-response associations, may be acquired after a particular task is learned, given that responses of prefrontal neurons have been shown to be particularly plastic (Asaad et al. 1998; Qi et al. 2011). Neurons tested in the ODR task confirmed prior findings about fixation neurons in the dorsolateral prefrontal cortex (Fig. 5). A population of neurons became active after the onset of fixation and remained active during the stimulus presentation, regardless of its location. These neurons also remained active during the delay period of the task while the monkey continued to maintain fixation. Their activity decayed rapidly after the offset of the fixation point, which required the monkey to make an eye movement toward the remembered location of the cue, regardless of the direction of the saccade.

To dissociate the effect of stimulus offset, stimulus representation, and saccadic preparation, we analyzed activity in three variants of the antisaccade task in which the fixation point appeared either before, simultaneously, or after a visual stimulus that instructed a saccade away from the stimulus (Fig. 4). In such a task, inhibition is presumably required not only to suppress the generation of a saccade while the fixation point is still on, but also to prevent a saccade toward the stimulus. Fixation neurons appeared unlikely to be involved in the latter, because the majority of prefrontal fixation neurons did not exhibit selectivity for the location of the stimulus. Additionally, in the gap variant of the antisaccade task, the stimulus appeared 100 ms after the offset of the fixation point, only after which the monkey knew where the saccade could be directed. At that time activity had already started to decay before the onset of the cue, when the fixation point was not visible but before the monkey knew which direction of saccade to plan (and which to inhibit). The results suggest that response inhibition is mediated by a different neuronal circuit in the prefrontal cortex.

We also considered a role of prefrontal fixation neurons in inhibiting saccade generation in analogy to oculomotor circuits in subcortical structures, which involve mutual inhibition of fixation and saccade neurons (Everling et al. 1998). We found that responses of both fixation and saccade neurons in the prefrontal cortex were generally negatively correlated, similarly timed relative to the onset of the saccade, and differing in groups of slow and fast trials. Some subtle differences in timing were also present, which we revealed when we compared timing in different variants of the antisaccade task, synchronized to the onset of the cue. As a result, fixation neurons are unlikely to dominate the activation of saccade neurons or provide spatially selective signals, although they may facilitate the generation of saccades.

In addition to neurons that were activated solely by the fixation point, we identified a larger subgroup of neurons that also were activated by the stimulus presentation. Prefrontal visual neurons exhibit large, often bilateral, receptive fields that often include the fovea (Funahashi et al. 1990; Riley et al., in press; Suzuki and Azuma 1983). It is not surprising, therefore, that the fixation point was sufficient to provide activation for a subset of such visual neurons in our sample. In principle, their activity could provide information for the execution of the task, superimposed to signals representing directional information about the stimulus and saccadic target. However, these neurons remained active after the offset of the fixation point in some task conditions. In the zero-gap variant, a strong visual response to the stimulus was evident precisely at the time the fixation target was turned off. Similarly, in the ODR task, strong (saccade related) activity was present after the fixation offset.

Prefrontal fixation neurons represented faithfully the visual fixation point. This was the most reliable predictor of activity across task variants and event timings. This activity could not directly guide the generation of saccades or aid in the selection of appropriate over inappropriate responses, but it was essential in the context of the task. A saccade was only permitted after the fixation point turned off. In the overlap task, a saccade could be planned already during the stimulus presentation, yet the task rule required knowledge of whether the fixation point was present or not for the decision of whether to saccade or not. Even though the activity of fixation neurons appears to represent low-level sensory information, our experiment represents an instance where this was woven in the context of a cognitive operation. In the ODR task, the offset of the fixation point triggers recall of the information maintained in memory. If persistent activity during the delay period represents the neural correlate of working memory, as advocated by a class of models (Constantinidis and Klingberg 2016; Wimmer et al. 2014), interactions between fixation neurons and neurons with persistent activity are likely to transmit this signal. Another class of models propose short-term synaptic changes between neurons as the mechanism of working memory representation, which requires reactivation of the same neurons at the time of recall (Mongillo et al. 2008; Stokes 2015). It is less clear if these models can account for performance in a recall task such as the ODR, given the patterns of activity observed. In any case, our current results illustrate that the simultaneous representation of a diversity of sensory and abstract information is essential for the integrative role of the dorsolateral prefrontal cortex (Katsuki and Constantinidis 2012; Miller and Cohen 2001).

GRANTS

Research reported in this article was supported by National Eye Institute Grants R01 EY017077 and R01 EY016773 (to C. Constantinidis) and the Tab Williams Family Endowment.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

X.Z. and C.C. conceived and designed research; X.Z. and C.C. performed experiments; X.Z. and C.C. analyzed data; X.Z. and C.C. interpreted results of experiments; X.Z. and C.C. prepared figures; X.Z. and C.C. drafted manuscript; X.Z. and C.C. edited and revised manuscript; X.Z. and C.C. approved final version of manuscript.

ACKNOWLEDGMENTS

We acknowledge the contributions of Terry Stanford and Emilio Salinas in the design of the original study that generated the results analyzed for this report, Dantong Zhu and Samson King for participation in some experiments, Joshua Seideman for assistance in the analysis of eye movements, and Kathini Palaninathan for technical help.

REFERENCES

  1. Asaad WF, Rainer G, Miller EK. Neural activity in the primate prefrontal cortex during associative learning. Neuron 21: 1399–1407, 1998. doi: 10.1016/S0896-6273(00)80658-3. [DOI] [PubMed] [Google Scholar]
  2. Bari A, Robbins TW. Inhibition and impulsivity: behavioral and neural basis of response control. Prog Neurobiol 108: 44–79, 2013. doi: 10.1016/j.pneurobio.2013.06.005. [DOI] [PubMed] [Google Scholar]
  3. Bizzi E. Discharge of frontal eye field neurons during saccadic and following eye movements in unanesthetized monkeys. Exp Brain Res 6: 69–80, 1968. doi: 10.1007/BF00235447. [DOI] [PubMed] [Google Scholar]
  4. Constantinidis C, Klingberg T. The neuroscience of working memory capacity and training. Nat Rev Neurosci 17: 438–449, 2016. doi: 10.1038/nrn.2016.43. [DOI] [PubMed] [Google Scholar]
  5. Dorris MC, Munoz DP. A neural correlate for the gap effect on saccadic reaction times in monkey. J Neurophysiol 73: 2558–2562, 1995. [DOI] [PubMed] [Google Scholar]
  6. Everling S, Paré M, Dorris MC, Munoz DP. Comparison of the discharge characteristics of brain stem omnipause neurons and superior colliculus fixation neurons in monkey: implications for control of fixation and saccade behavior. J Neurophysiol 79: 511–528, 1998. [DOI] [PubMed] [Google Scholar]
  7. Funahashi S, Bruce CJ, Goldman-Rakic PS. Mnemonic coding of visual space in the monkey’s dorsolateral prefrontal cortex. J Neurophysiol 61: 331–349, 1989. [DOI] [PubMed] [Google Scholar]
  8. Funahashi S, Bruce CJ, Goldman-Rakic PS. Visuospatial coding in primate prefrontal neurons revealed by oculomotor paradigms. J Neurophysiol 63: 814–831, 1990. [DOI] [PubMed] [Google Scholar]
  9. Hallett PE. Primary and secondary saccades to goals defined by instructions. Vision Res 18: 1279–1296, 1978. doi: 10.1016/0042-6989(78)90218-3. [DOI] [PubMed] [Google Scholar]
  10. Hanes DP, Patterson WF 2nd, Schall JD. Role of frontal eye fields in countermanding saccades: visual, movement, and fixation activity. J Neurophysiol 79: 817–834, 1998. [DOI] [PubMed] [Google Scholar]
  11. Harris KD, Henze DA, Csicsvari J, Hirase H, Buzsáki G. Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements. J Neurophysiol 84: 401–414, 2000. [DOI] [PubMed] [Google Scholar]
  12. Izawa Y, Suzuki H. Activity of fixation neurons in the monkey frontal eye field during smooth pursuit eye movements. J Neurophysiol 112: 249–262, 2014. doi: 10.1152/jn.00816.2013. [DOI] [PubMed] [Google Scholar]
  13. Izawa Y, Suzuki H, Shinoda Y. Response properties of fixation neurons and their location in the frontal eye field in the monkey. J Neurophysiol 102: 2410–2422, 2009. doi: 10.1152/jn.00234.2009. [DOI] [PubMed] [Google Scholar]
  14. Johnston K, Koval MJ, Lomber SG, Everling S. Macaque dorsolateral prefrontal cortex does not suppress saccade-related activity in the superior colliculus. Cereb Cortex 24: 1373–1388, 2014. doi: 10.1093/cercor/bhs424. [DOI] [PubMed] [Google Scholar]
  15. Katsuki F, Constantinidis C. Unique and shared roles of the posterior parietal and dorsolateral prefrontal cortex in cognitive functions. Front Integr Neurosci 6: 17, 2012. doi: 10.3389/fnint.2012.00017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Keller EL. Participation of medial pontine reticular formation in eye movement generation in monkey. J Neurophysiol 37: 316–332, 1974. [DOI] [PubMed] [Google Scholar]
  17. Lo CC, Boucher L, Paré M, Schall JD, Wang XJ. Proactive inhibitory control and attractor dynamics in countermanding action: a spiking neural circuit model. J Neurosci 29: 9059–9071, 2009. doi: 10.1523/JNEUROSCI.6164-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Meredith MA, Ramoa AS. Intrinsic circuitry of the superior colliculus: pharmacophysiological identification of horizontally oriented inhibitory interneurons. J Neurophysiol 79: 1597–1602, 1998. [DOI] [PubMed] [Google Scholar]
  19. Meyer T, Constantinidis C. A software solution for the control of visual behavioral experimentation. J Neurosci Methods 142: 27–34, 2005. doi: 10.1016/j.jneumeth.2004.07.009. [DOI] [PubMed] [Google Scholar]
  20. Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci 24: 167–202, 2001. doi: 10.1146/annurev.neuro.24.1.167. [DOI] [PubMed] [Google Scholar]
  21. Miller EK, Freedman DJ, Wallis JD. The prefrontal cortex: categories, concepts and cognition. Philos Trans R Soc Lond B Biol Sci 357: 1123–1136, 2002. doi: 10.1098/rstb.2002.1099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Mongillo G, Barak O, Tsodyks M. Synaptic theory of working memory. Science 319: 1543–1546, 2008. doi: 10.1126/science.1150769. [DOI] [PubMed] [Google Scholar]
  23. Mountcastle VB, Lynch JC, Georgopoulos A, Sakata H, Acuna C. Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space. J Neurophysiol 38: 871–908, 1975. [DOI] [PubMed] [Google Scholar]
  24. Munoz DP, Everling S. Look away: the anti-saccade task and the voluntary control of eye movement. Nat Rev Neurosci 5: 218–228, 2004. doi: 10.1038/nrn1345. [DOI] [PubMed] [Google Scholar]
  25. Munoz DP, Guitton D, Pélisson D. Control of orienting gaze shifts by the tectoreticulospinal system in the head-free cat. III. Spatiotemporal characteristics of phasic motor discharges. J Neurophysiol 66: 1642–1666, 1991. [DOI] [PubMed] [Google Scholar]
  26. Munoz DP, Istvan PJ. Lateral inhibitory interactions in the intermediate layers of the monkey superior colliculus. J Neurophysiol 79: 1193–1209, 1998. [DOI] [PubMed] [Google Scholar]
  27. Munoz DP, Wurtz RH. Fixation cells in monkey superior colliculus. I. Characteristics of cell discharge. J Neurophysiol 70: 559–575, 1993. [DOI] [PubMed] [Google Scholar]
  28. Qi XL, Meyer T, Stanford TR, Constantinidis C. Changes in prefrontal neuronal activity after learning to perform a spatial working memory task. Cereb Cortex 21: 2722–2732, 2011. doi: 10.1093/cercor/bhr058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Riley MR, Constantinidis C. Role of prefrontal persistent activity in working memory. Front Syst Neurosci 9: 181, 2016. doi: 10.3389/fnsys.2015.00181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Riley MR, Qi XL, Constantinidis C. Functional specialization of areas along the anterior-posterior axis of the primate prefrontal cortex. Cereb Cortex. 1–15, 2016. doi: 10.1093/cercor/bhw190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Segraves MA, Goldberg ME. Functional properties of corticotectal neurons in the monkey’s frontal eye field. J Neurophysiol 58: 1387–1419, 1987. [DOI] [PubMed] [Google Scholar]
  32. Sommer MA, Wurtz RH. Composition and topographic organization of signals sent from the frontal eye field to the superior colliculus. J Neurophysiol 83: 1979–2001, 2000. [DOI] [PubMed] [Google Scholar]
  33. Sommer MA, Wurtz RH. Frontal eye field sends delay activity related to movement, memory, and vision to the superior colliculus. J Neurophysiol 85: 1673–1685, 2001. [DOI] [PubMed] [Google Scholar]
  34. Stokes MG. ‘Activity-silent’ working memory in prefrontal cortex: a dynamic coding framework. Trends Cogn Sci 19: 394–405, 2015. doi: 10.1016/j.tics.2015.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Suzuki H, Azuma M. Prefrontal neuronal activity during gazing at a light spot in the monkey. Brain Res 126: 497–508, 1977. doi: 10.1016/0006-8993(77)90600-X. [DOI] [PubMed] [Google Scholar]
  36. Suzuki H, Azuma M. Topographic studies on visual neurons in the dorsolateral prefrontal cortex of the monkey. Exp Brain Res 53: 47–58, 1983. doi: 10.1007/BF00239397. [DOI] [PubMed] [Google Scholar]
  37. Suzuki H, Azuma M, Yumiya H. Stimulus and behavioral factors contributing to the activation of monkey prefrontal neurons during gazing. Jpn J Physiol 29: 471–489, 1979. doi: 10.2170/jjphysiol.29.471. [DOI] [PubMed] [Google Scholar]
  38. Tinsley CJ, Everling S. Contribution of the primate prefrontal cortex to the gap effect. Prog Brain Res 140: 61–72, 2002. doi: 10.1016/S0079-6123(02)40042-8. [DOI] [PubMed] [Google Scholar]
  39. Wimmer K, Nykamp DQ, Constantinidis C, Compte A. Bump attractor dynamics in prefrontal cortex explains behavioral precision in spatial working memory. Nat Neurosci 17: 431–439, 2014. doi: 10.1038/nn.3645. [DOI] [PubMed] [Google Scholar]
  40. Zhou X, Katsuki F, Qi XL, Constantinidis C. Neurons with inverted tuning during the delay periods of working memory tasks in the dorsal prefrontal and posterior parietal cortex. J Neurophysiol 108: 31–38, 2012. doi: 10.1152/jn.01151.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhou X, Qi XL, Constantinidis C. Distinct roles of the prefrontal and posterior parietal cortices in response inhibition. Cell Reports 14: 2765–2773, 2016a. doi: 10.1016/j.celrep.2016.02.072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhou X, Zhu D, King SG, Lees CJ, Bennett AJ, Salinas E, Stanford TR, Constantinidis C. Behavioral response inhibition and maturation of goal representation in prefrontal cortex after puberty. Proc Natl Acad Sci USA 113: 3353–3358, 2016b. doi: 10.1073/pnas.1518147113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhou X, Zhu D, Qi XL, Lees CJ, Bennett AJ, Salinas E, Stanford TR, Constantinidis C. Working memory performance and neural activity in prefrontal cortex of peripubertal monkeys. J Neurophysiol 110: 2648–2660, 2013. doi: 10.1152/jn.00370.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zhou X, Zhu D, Qi XL, Li S, King SG, Salinas E, Stanford TR, Constantinidis C. Neural correlates of working memory development in adolescent primates. Nat Commun 7: 13423, 2016c. doi: 10.1038/ncomms13423. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Neurophysiology are provided here courtesy of American Physiological Society

RESOURCES