Relation Between Event-Related Potential Latency and Saccade Latency in Overt Shifts of Attention

Abstract

Controlled shifts of attention between competing stimuli are crucial for effective everyday visual behaviour. While these typically involve overt shifts of fixation, many past studies used covert attention shifts in which fixation is unchanged, meaning that some response components may result from the inhibition of eye movements. In this study, the neural events in the human brain when making overt shifts of attention are studied through the combination of event-related potential recording with simultaneous eye tracking. Fixation shifts under competition (central target remains visible when a peripheral target appears) were compared with noncompetition (central target disappears). A longer latency for competition compared with noncompetition, which is found in the saccadic response, is already present in the early occipital positivity when a single target is presented for the fixation shift. These results indicate that the requirement to disengage from a current target affects the time course of neural processing at an early level. However, the relation is more complex when the participant is required to choose which of two targets to fixate.

Keywords

eye tracking EEG attention overt attention shifts fixation-shift-paradigm

Everyday visual behaviour requires us to shift the focus of our attention between competing objects and events in our environment. Usually, such shifts are overt, including an eye movement to foveate the new target. However, covert shifts of spatial attention, with fixation unchanged, are also possible.

Although there are many studies of the neural events accompanying fixation shifts in nonhuman primates (Dorris & Munoz, 1995; Goldberg & Wurtz, 1972; Schiller & Stryker, 1972), the majority of work on human neural responses (Electroencephalogram [EEG] or functional magnetic resonance imaging) in attention shifts has studied covert attention shifts (Anllo-Vento & Hillyard, 1996; Eimer et al., 2005, 2002; Martinez et al., 1999; Praamstra & Oostenveld, 2003; Shomstein et al., 2012; Yamaguchi et al., 1994, 1995), with few studies investigating overt attention shifts, mainly using magnetic resonance imaging (see Corbetta et al., 1998, for an overview). Measuring covert attention shifts avoids the need to disentangle attention-related responses from eye movement artefacts. However, maintaining fixation while shifting attention is not the most natural way to behave and requires the participants to inhibit the eye movements they would normally make (Perry & Zeki, 2000). Recently, combined eye tracking and EEG has been implemented, making it possible to measure neural mechanisms of overt attention (Huber-Huber et al., 2016; Kulke, 2015, 2019; Kulke et al., 2016a, 2016b; Weaver et al., 2017). One previous study (Kulke et al., 2016a) tested shifts between competing stimuli, with participants instructed either to make an eye movement or to make a manual response while maintaining fixation on the original target. That study focused on comparing overt and covert attention shifts but did not manipulate the level of attentional (dis-)engagement of participants. Although posterior responses were generally comparable between overt and covert conditions, the latter condition showed additional frontal responses believed to reflect the inhibition of saccades. To attempt to avoid such effects, this study compared the latency and distribution of EEG responses in conditions of a paradigm using overt attention shifts. Another study investigating overt attention shifts by Weaver et al. (2017) found occipital responses to be related to the spatial execution of saccades in a visual search task. They also demonstrated that occipital alpha activity occurring before target onset was related to slower saccade latency. In contrast, the temporal execution (i.e., saccade latency) was found to be independent of a distractor-related positivity (Pd). Note however, that they did not investigate early visual responses to the target (e.g., P1). In summary, neural mechanisms of overt attention shifts may differ from covert ones, and overt attention shifts should therefore be further investigated to complement traditional studies focusing on covert attention.

Our overall approach reflects the “Premotor Theory of Attention” introduced by Rizzolatti et al. (e.g., Eimer et al., 2005; Sheliga et al., 1994). That is, the effect of directing spatial attention is to facilitate actions oriented to that location including saccades. From this perspective, neural responses involved in saccade preparation can be considered as elements of the attention process. We therefore consider the disengagement of gaze from a current fixation target, and the preparation of a saccade towards a newly appearing target, to be components of the shift of attention. However, we recognize that alternative formulations may consider the preparation of a gaze shift to constitute an oculomotor process that may be distinguished from cognitive aspects of an attention shift.

The “Fixation Shift Paradigm” is used here (Atkinson & Braddick, 2012 [review]; Atkinson et al., 1988, 1992; Butcher et al., 2000; Hood & Atkinson, 1993; Kulke et al., 2015), originally designed to study the control of shifts of attention in infants. Shifts of attention (measured in terms of latency of eye movements) between a central target and a target in the periphery to either the right or left of the central target are measured in two conditions. In the noncompetition condition, the central target disappears at the same time as the second target appears in the periphery, whereas in the competition condition, the central target remains visible when the peripheral target appears (i.e., the two targets compete for attention). The competition condition requires the participant to disengage from the competing central target, a situation which leads to an increased latency in the eye movement, most strongly seen in infants in the first months of life (Atkinson et al., 1992; Hood & Atkinson, 1993; Kulke et al., 2015, 2016b; Mercuri et al., 1997a, 1997b). An alternative description of events in the noncompetition condition is that the offset of the central target primes a rapid shift of attention, seen as a rapid eye movement response. A closely related method (Csibra et al., 1997; Kalesnykas & Hallett, 1987; for a comparison, see Kulke, 2017; Reulen, 1984; Rolfs & Vitu, 2007) is the “gap/overlap paradigm,” in which there is a variable delay between offset of the central target and the appearance of the peripheral target; the latency is longest if the gap is absent or negative (“overlap”; Farroni et al., 1999; Matsuzawa & Shimojo, 1997). The noncompetition condition in the Fixation Shift Paradigm is comparable to a zero gap in the gap/overlap paradigm.

Previous studies measuring EEG responses associated with the gap effect demonstrated variations associated with the presence of competing targets. These include a parietal positivity peaking 40 ms after target onset, and a following posterior positivity peaking first in the contralateral hemisphere (and later in the ipsilateral; Cordones et al., 2013; Csibra et al., 1997; Gomez et al., 1996; Kawakubo et al., 2007). The comparison between competition and noncompetition conditions in these studies has been primarily in terms of amplitude differences but not based on latency differences between the conditions, while a previous study investigating latencies of responses did not compare competition and noncompetition conditions (Kulke et al., 2016a).

This study was designed to extend this work in two ways while using stimulus materials as similar as possible to previous research (Kulke et al., 2015, 2016a, 2016b; Mercuri et al., 1997a, 1997b; Ricci et al., 2010). First, the behavioural effect of competition is to increase saccade latencies compared with noncompetition conditions. Comparison of the latencies of EEG responses preceding the saccade can indicate how early in the neural processing the differential timing observed for the saccade is initiated.

An additional advantage of this study was the combination of eye tracking with EEG (Kulke, 2019; Kulke et al., 2016a, 2016b). Saccades occurring during overt attention shifts induce artefacts in EEG data, a problem already identified by Csibra et al. (1997) who averaged data across all trials. The current methodology made it possible to exclude trials in which an early saccade intruded into the EEG analysis window, leading to a cleaner EEG signal. In summary, this study aimed to compare neural and saccadic responses between competition and noncompetition conditions using simultaneous recording of EEG responses and eye tracking.

Methods and Materials

Participants

In total, 27 students from the University College London (UCL) Psychology subject pool (Mean age = 24.9 years, SD = 10.1, range = 18 to 58, 7 males, 24 right-handed) volunteered to participate in the study in exchange for course credit or monetary compensation (£10). One subject was excluded because of technical problems, one because of excessive noise in the EEG data, one because they did not complete the whole experiment, and one additional subject because of a prediagnosed neurological condition. Remaining subjects all had normal or corrected to normal vision and no known history of brain disease. The study was approved by the UCL ethics committee, all methods were performed in accordance with the guidelines and regulations and all subjects gave written informed consent in accordance with the Declaration of Helsinki.

Materials and Stimuli

A DELL computer with Linux operating system (Linux Mint 14), with MATLAB (version 7.11.0 (R2010b)) as the presentation program, was used to generate stimuli and present them on a 21.5-in. (54 cm) LCD monitor (Samsung) that extended over 35.8° × 22.8° of visual angle, running at a frame rate of 60 Hz. Stimuli were similar to those used in the earlier reported study (Kulke et al., 2016a), with the main differences being that only saccadic responses, but no manual responses were required and that, as in the original Fixation Shift Paradigm (Hood & Atkinson, 1993), there was a condition where a persisting central stimulus was competing for attention (competition condition) and a condition where the central stimulus disappeared and so was not competing with a second stimulus for attention (noncompetition condition). The stimuli were presented on a grey background with a luminance of 77 cd/m². The experiment began with the presentation of a first stimulus, a black schematic face, changing expression at a rate of 3 Hz to attract attention (as in the original Fixation Shift Paradigm) and a green fixation dot, appearing in the centre of the screen. The dot had a size of 0.7° of visual angle and the face subtended visual angle of 7.7° × 7.7°. The central dot and face remained visible for a randomized intertrial interval between 0.5 and 2.5 seconds. Subjects were instructed to fixate on the central dot to avoid eye movements between different positions within the face region. In a gaze-contingent design, when a fixation on the dot was registered after the random intertrial interval (defined as the measured gaze position deviating from the dot position by less than 2.6° of visual angle for at least 20 samples [∼330 ms]), target stimuli were randomly presented on the left, right, or on both sides of the screen at an eccentricity of 12.9° of visual angle. The condition in which bars appeared on both sides was an exploratory condition outside the main research question, introduced to control the effect of visual compared with attentional responses by providing identical visual input to both hemispheres for attention shifts in either direction. As it involves a decision as to which side to select a shift towards, as well as the process of shifting attention, additional higher cognitive mechanisms may be involved in this condition, overlapping with the automatic attention shifts. Due to two targets appearing simultaneously, the condition furthermore may be expected to elicit a stronger visual response. Analyses including this condition are presented in Online Supplement A. Target stimuli were bars made up of one black and one white rectangle that reversed colour at a rate of 3 reversals per second. In competition conditions, the central face and the dot remained on the screen, whereas in the noncompetition condition, they both disappeared at the onset of the bars (Figure 1). A remote Tobii X120 eye tracker was used to record the gaze position of subjects at a sampling rate of 60 Hz during the experiment. The algorithm developed by Kulke et al. (2015) was used for eye tracking data processing, involving interpolation of missing samples, exclusion of noisy trials (fixation not on the screen at target onset, more than 20% of samples with artefactual excursions of more than 2.2° of visual angle, fast saccades that were unrelated to the stimulus onset, saccades to the incorrect direction). Saccades were identified as changes in horizontal gaze position of more than 2.2° of visual angle between two successive samples, and the latency of the first saccade was defined as the temporal difference between target onset and the first such change in gaze position.

Figure 1.

Stimulus Display in the Competition (Top Panel) and the Noncompetition (Bottom Panel) Condition. The side on which the peripheral bar appeared (left or right) was counterbalanced.

Electroencephalogram

While the subjects were engaged in the behavioural task, their EEG was recorded at a rate of 250 Hz using Electrical Geodesics Inc. (EGI) NetAmp300 amplifier and 128- channel Ag/AgCl electrode nets (Tucker, 1993), connected to a separate computer (Macintosh) using Net Station 4.2 (© 1994–2006, EGI, Eugene, Oregon, USA). Electrode impedance was adjusted to less than 90 kΩ (Kulke et al., 2016a, 2016b), with the majority of electrodes having an impedance of less than 40 kΩ (Ferree et al., 2001).

The EEG data analysis was programmed in MATLAB. The timing of the EEG system was measured to ensure that triggers were aligned with visual events, and the data timing was corrected by subtracting the detected delay. The data were average-referenced, and Butterworth filters were used first for notch filtering around the line noise frequency (49–51 Hz), second for high-pass filtering (cut off: 0.01 Hz), and third for low-pass filtering (cut off: 25 Hz). The data were segmented into epochs of –200 to 180 ms around target onset. Noisy epochs and electrodes were then determined by using the median absolute deviation about the median (Hampel, 1974), as this is a measure that is fairly robust to noise. They were interpolated using spherical spline interpolation, and only those that were acceptable according to the abovementioned criteria after interpolation were used for further analysis. Finally, the average voltage during the baseline period ([–200; 0] ms before target onset) was used to correct the data from target onset onwards, individually for each trial and electrode. Scalp surface maps were created using spherical spline interpolation. Event-Related Potential (ERP) peak amplitudes and latencies were calculated based on intervals defined by previous research (Kulke et al., 2016a). Note that, as ERPs before saccade onset were investigated, no independent component analysis was conducted, as recommended by previous research (Huber-Huber et al., 2016; Kulke et al., 2016a). Trials with saccades occurring before the end of the extracted EEG time window were detected based on the eye tracking data and excluded from further analysis.

Procedure

After a 5 point eye tracking calibration, subjects were asked to focus on a central green dot and shift their gaze to bars that randomly appeared on one or both sides of the screen in peripheral vision. They were instructed that when bars appeared on both sides of the screen, they should choose on each trial whether to look to the bar on the left or bar on the right. When the peripheral bars disappeared, the subjects were asked to shift their focus back to the green dot. An eye tracker monitored their gaze so that the trial was controlled in a gaze-contingent fashion. Short breaks were given between each set of 100 trials, with a longer break occurring after 400 trials while the experimenter readjusted the electrode impedance. The entire experiment was composed of 800 trials, two thirds of which were single targets (one third left and one third right) and one-third double targets, and lasted for approximately an hour. Conditions were presented in a random order. The current article focuses on single target conditions, as the double target condition requires a decision which side to shift towards as well as an attention shift, therefore posing additional cognitive demands. However, detailed analyses of the double target condition are reported in Online Supplement A.

Design

For the behavioural analysis, a 2 × 2 within-subject design was used to investigate the effects of competition condition (competition or noncompetition) and screen side looked at (left or right) on saccade latency. For the neural analysis, the effects of competition condition (competition or noncompetition), brain hemisphere (ipsi- or contralateral brain hemisphere to the eye movement), and brain side (left or right side of the brain) on ERP amplitude and latency were determined. Note that there are two measures for brain lateralisation: Brain hemisphere describes the lateralisation in relation to the target (hemisphere ipsi- or contralateral to the stimulus that was reacted to) and the factor brain side compares the left and right sides of the brain.

Results

Data Processing

On average, 8.6% (SD = 3.0%) of trials were interpolated across all subjects, with every subject having less than 14.4% interpolated trials. The individually calculated noise threshold for ERP samples was 13.7 µV (SD = 3.94 µV), and the individually calculated range threshold within epochs was on average 53.8 µV (SD = 14.0 µV). The average number of trials subjects successfully completed behaviourally was 766 (SD = 48.6). After exclusion of noisy EEG data, an average of 703 (SD = 55.9, min = 577) trials per subject remained in the analysis. IBM SPSS Statistics (version 25) was used for analyses. This program uses the Satterthwaite approximation to degrees of freedom. Effect sizes were estimated based on the fixed effects. Only significant effects are reported later.

Behavioural Results

The eye tracking data quality was high with 98.0% correct refixations, and only 1% excluded by noise criteria based on Kulke et al. (2015; e.g., fixation not on screen at trial onset, data loss, and rapid changes in fixation position). In single target, trials 1.5% of saccades were directed to the wrong direction, and four fifths of these were subsequently corrected.

Saccade latencies (M = 0.307 seconds, SD = 0.050) were analysed using a mixed linear model including participants as a random intercept and screen side of the target stimulus and competition condition as fixed factors. It showed a significant main effect of competition condition, F(1, 59) = 56.83, p < .001, d ∼ 1.96. Latencies were shorter in the noncompetition condition (M = 0.290 seconds, SD = 0.048) than in the competition condition (M = 0.324 seconds, SD = 0.046).

EEG Results

In line with previous research, topographical plots showed a contralateral positivity in posterior areas that moves to the ipsilateral side (Figures 2 and 3). There was furthermore a posterior negativity, peaking toward the end of the extracted time window. Wave plots of the occipital and frontal responses are displayed in Figure 4.

Figure 2.

Topographical Plots of the Posterior Response for Targets on the Left (Left Plots) and Right (Right Plots) Sides of the Screen for Noncompetition (Top Panel) and Competition Conditions (Bottom Panel). The time window (displayed at the top left of each plot) is chosen based on the window during which peak amplitudes appear, as peak amplitudes were analysed further. Scale: –2 µV (blue) to 2 µV (red).

Figure 3.

Topographical Plots of the Posterior Response for Targets on the Left (Left Plots) and Right (Right Plots) Sides of the Screen for Noncompetition (Top Panel) and Competition Conditions (Bottom Panel). The time window is chosen based on the full time window used for extraction of peak amplitude and latency (110–180 ms in both cases). Scale: –2 µV (blue) to 2 µV (red).

Figure 4.

Occipital (Top Panel) and Frontal (Bottom Panel) Wave Plots for Targets in the Left (Left Panel) and the Right Hemisphere (Right Panel) of the Brain Are Displayed.

Linear mixed models including participants as a random intercept were calculated to predict ERP amplitudes and latencies from condition, brain hemisphere (ipsi- or contralateral), and brain side (left or right). In the following, amplitudes are displayed in microvolts, latencies are reported in milliseconds. Identical decisions regarding the significance of effects are obtained without correction for multiple factors as when a Bonferroni–Holm correction (Holm, 1979) is applied.

Occipital Response

Posterior responses were extracted in two electrode clusters around the occipital electrode locations O1 and O2 between 110 and 180 ms (occipital cluster O1 [EGI electrodes 65, 66, 70, 71, 69, 74] and O2 [EGI electrodes 90, 84, 76, 83, 82, 89]). The posterior positivity peaked on average around 134 ms (SD = 22.9) and had an average amplitude of 1.24 µV (SD = 2.53 µV). There were no significant effects on peak amplitude.

Peak latency showed a significant main effect of competition condition, F(1, 139) = 26.37, p < .001, d = ∼ 0.87, with shorter latencies in the noncompetition (M = 127, SD = 18.4) than in the competition condition (M = 142, SD = 24.6). An exploratory analysis using 50% fractional area latency as the outcome variable confirmed shorter latencies in the noncompetition (M = 140, SD = 17.0) than in the competition condition (M = 144, SD = 16.2), F(1, 172) = 3.47, p = .064 (t(172) = –2.21, p = .029).

Frontal Response

Frontal responses were extracted in two lateral fronto-central (FC) electrode clusters around FC3 [EGI 12, 13, 19, 24, 20, 28, 29] and FC4 [EGI 4, 5, 111, 112, 117, 118, 124] between 100 and 180 msec after stimulus onset based on previous research (Kulke et al., 2016a). The frontal negativity peaked at 136 ms (SD = 25.4) and had an average amplitude of –1.44 µV (SD = 8.90 µV). No effects on peak amplitude were significant.

Peak latency showed significant effects of competition condition, F(1, 150) = 21.50, p < .001, d ∼ 0.76, with shorter latencies in the noncompetition (M = 129, SD = 25.0) than in the competition condition (M = 143, SD = 24.1), and hemisphere, F(1, 150) = 13.60, p < .001, d ∼ 0.60, with shorter latencies in the ipsilateral (M = 131, SD = 24.1) than in the contralateral hemisphere (M = 141, SD = 25.7).

Discussion

The aim of this study was to monitor behavioural and neural mechanisms of attention shifts in competition and noncompetition conditions of the Fixation Shift Paradigm in adult subjects. Saccade latencies were significantly longer in the competition than in the noncompetition condition, replicating previous studies comparing saccadic shifts either involving disengagement or no disengagement (Csibra et al., 1997; Fischer, 1986; Hood & Atkinson, 1993; Saslow, 1967). These findings confirm that additional processing time is required to disengage from a stimulus that is reflected in longer latencies. Neural responses in this study involved a posterior positivity and a frontal negativity. The posterior positive response to peripheral stimuli was comparable to previous findings (Csibra et al., 1997). The novel investigation of latency differences between competition and noncompetition conditions showed that peak latencies were significantly shorter in the noncompetition than in the competition condition, in line with the behavioural differences in saccade latencies and therefore suggesting that at an early time after target presentation, brain responses already vary between conditions. Although involvement of occipital areas in attention shifts has been well established in previous research, to our knowledge this is the first study showing latency differences in early ERPs reflecting the same pattern as saccadic attention shift measures in a simple fixation shift paradigm. A previous study by Weaver et al. (2017) found saccade latency to be independent of a later occurring ERP, the distractor-related positivity (Pd). However, they did not investigate early visual responses to the target (e.g., P1). Therefore, latency patterns of ERPs parallel to those of the saccades seem to occur earlier than the responses that have previously been investigated. Similar to our findings, an magnetoencephalography study showed that responses in areas around middle temporal visual area and V4 predicted reaction times a random dot motion detection task (Amano et al., 2006). In a covert attention shift task, larger P3f amplitudes were found to be related to faster button presses (Delorme et al., 2007). Our study suggests that a relation as observed between neural responses and reaction times may also exist between neural responses and eye movements. The study was not originally designed to investigate direct correlations between the two measures on individual trials, as neural responses need to be averaged across several trials to compute reliable peak latencies of the P1. However, future research could further explore direct correlations between both measures.

Peak latencies of the frontal negativity were also significantly shorter in the noncompetition condition than in the competition condition. This is a similar pattern as observed for the posterior positivity in this study, suggesting that these two components show similar variation in response latency variation between conditions. It is possible that they reflect two ends of a dipole. In addition, the frontal negativity peaked in the ipsilateral hemisphere first, followed by the contralateral hemisphere. This seems to be in contrast with previous EEG papers (Csibra et al., 1997; Kulke et al., 2016a; Rugg et al., 1987), suggesting a shift of a posterior neural response from contralateral to ipsilateral sites. However, the laterality change in the opposite direction in frontal compared with posterior areas is in line with the possibility that two ends of one dipole are measured here.

It is possible that interhemispheric connections play a crucial role for this shift of neural responses. In particular, connections are impaired in some subject groups who struggle with the Fixation Shift Paradigm, for example, people with autism (Elison et al., 2013) and neglect patients (Bartolomeo et al., 2007; De Schotten et al., 2005). In line with this, recent findings suggest that the lateralisation of frontal responses changes during infancy (Kulke et al., 2016b), coinciding with an improving ability to shift attention. This highlights the role of interhemispheric connections for overt attention shifts.

In summary, this study replicates previous findings on neural responses involved in overt attention shifts by Kulke et al. (2016a), which only investigated competition, but not noncompetition conditions. It furthermore extends these findings by demonstrating that the latency of neural responses varies as a function of noncompetition and competition conditions, similar to the behavioural pattern of shorter saccade latencies in noncompetition than in competition conditions. This suggests that latency differences already occur on an early neural level within 150 ms after target onset. There are two possible causes of the neural effects. First, the differences might be related to attentional mechanisms and saccade preparation. Second, the differences might be related to visual responses to the different visual inputs. In the noncompetition condition, the central stimulus disappears at target onset while it remains visible in the competition condition. The offset of a visual stimulus can induce neural responses (Cordones et al., 2013; Gomez et al., 1996; Vassilev et al., 1983), which can be similar to early P1 and N1 responses elicited by the appearance of a stimulus (Gomez et al., 1996; Vassilev et al., 1983) and depend in magnitude on stimulus features (Vassilev et al., 1983). These responses may overlap with responses to the onset of peripheral targets and attentional responses. Temporally coinciding ERP components sum together to form the response that can be measured on the scalp (reviews, e.g., Luck, 2005; Regan, 1989). In an additional experiment investigating the effect of the offset (Online Supplement B), we presented participants both with the noncompetition condition and a novel condition in which the face disappeared while no peripheral bars appeared (offset condition). If the offset itself creates an ERP, this response should be visible in the offset condition. Subtracting the offset response from the noncompetition response led to a significant increase in ERP latencies. We conclude that the effect of competition on EEG responses in part reflects the absence of an early offset response when the central target remains on. The effect of summing a visual offset response with signals involved in directing attention, must be distinguished from the possibility that the offset plays an actual role in the attentional processing, for example, by priming the switch to a novel target.

Additional EEG differences between competition and noncompetition may be attributable to the difference in processes mobilising attention and preparing fixation shifts, when the disappearance of the current fixation target removes the need to disengage fixation and/or triggers readiness to shift. For example, theories suggest that occipital areas interact with additional input pathways from other areas, as proposed by previous models (e.g., Atkinson, 1984, 2000; Di Russo et al., 2003; Johnson, 1990, 2002; McAdams & Reid, 2005; Wang et al., 2015). As a direct comparison of the offset response with the competition condition was not possible in this study (see Online Supplement B), future research could further explore how much of the difference can be explained by the offset response and how much by attentional processes.

Different models have suggested that the occipital cortex plays a crucial role for overt attention shifts (Atkinson, 1984; Johnson, 1990; Schiller, 1985). The cortex can activate superior colliculus (SC) (Collins et al., 2005; Schiller & Tehovnik, 2005), which in turn generates saccades (Goldberg & Wurtz, 1972; Schiller & Stryker, 1972) after a specific activation threshold is reached (Neggers et al., 2005). In particular, functional magnetic resonance imaging studies using the gap paradigm in adults show that the offset of a visual stimulus can activate SC (Neggers et al., 2005). The current results show that the offset of a visual stimulus increases activation in occipital areas of the cortex. In line with previous research (Collins et al., 2005; Schiller & Tehovnik, 2005), this activation may lead to higher SC activation, which may cause faster shifts in noncompetition conditions. In this sense, it would be possible that the offset of a visual stimulus could prompt the brain to shift to another location. Such a mechanism would be beneficial in everyday life, as it is crucial for humans to focus on new objects instead of staring at locations where a stimulus once was present. This is in line with findings from the similar gap/overlap paradigms (Csibra et al., 1997; Kalesnykas & Hallett, 1987; Reulen, 1984; Rolfs & Vitu, 2007), showing that very fast saccades can occur if the central stimulus disappears before the peripheral target appears after a temporal gap. Quite similarly, in this condition, the brain may also be prompted to shift attention by the visual offset and as the offset occurs even earlier (due to the temporal gap), very fast saccades can be elicited. Thus we have two possibilities for the temporal advantage of the noncompetition condition that (a) offset of the central target facilitates the gaze shift and (b) the need to disengage from a continuing central target delays the gaze shift. In either case, the evidence of the current experiment is that the advantage is reflected at an early stage of visual cortical processing. Further research may be able to dissociate these two possibilities and explore the details of the mechanisms responsible.

Note that this study focused on bottom-up shifts of attention towards targets appearing in the periphery. If only such bottom-up mechanisms are required, saccade latency patterns and P1 latency patterns seem to be comparable. However, the relationship changes when comparison is made with the condition where an additional target was introduced for the exploratory analyses in Online Supplement A. Interestingly, occipital response latencies were faster in response to double than single targets, while saccade latencies were slower in response to double targets. The double target condition involves two changes compared with the single target condition: First, the visual input appears on both sides of the screen, leading to a stronger visual stimulation. Second, participants can only move their eyes towards one of the two stimuli, requiring them to make a decision where to shift their gaze. On a neural level, occipital responses are known to be strongly affected by stimulus area (Katsumi et al., 1988). Therefore, occipital responses may be faster due to the larger stimulus area in the double target condition. However, after the participant has visually registered the targets but before a saccade can be executed, a decision is required. This decision may involve additional independent neural processes that need to be completed before the saccade can be executed, leading to longer saccade latencies. The findings from Online Supplement A therefore suggest that saccade latencies only reflect neural response latencies under certain conditions but that this pattern is affected by other processes, possibly including but not limited to visual stimulation and cognitive decision processes.

In summary, P1 responses are mainly related to automated shifts of attention to single visual stimuli in bottom-up tasks, while other factors may overshadow effects in cognitively more demanding tasks.

In considering the Fixation Shift Paradigm as a measure of attention processes, this work is reflecting the “Premotor Theory of Attention” (e.g., Eimer et al., 2005; Sheliga et al., 1994) and treating the processes underlying gaze shifts as related to attention shifts (see also Atkinson & Braddick, 2012; Atkinson et al., 1992; Kulke, 2019; Kulke et al., 2016a). However, it may be argued that the control of gaze shifts can be dissociated from an accompanying shift of attention, that is, that the offset of a visual stimulus may prompt a saccade (a visual engagement with the novel stimulus), by processes distinct from an attention shift (i.e., a cognitive form of engagement with the novel stimulus). If this alternative approach is adopted, future research is required to identify distinct neural substrates for the processes of oculomotor shifts and cognitive engagement.

In conclusion, this article investigated neural mechanisms of attention shifts with and without competing targets. It replicates behavioural findings of previous research, showing shorter saccadic latencies in noncompetition conditions than in competition conditions. Crucially, these latency differences were also detected at an early neural level in occipital areas of the brain, suggesting that the timing of early cortical responses may already be related to subsequent shifts of attention. However, this relationship becomes more complex when a decision has to be made on fixating one of multiple targets that appear. The study confirms an involvement of occipital areas in overt attention shifts, as suggested by neural models of attention and previous research, which is however affected by visual and cognitive factors.

Highlights

This study combines simultaneous eye tracking and EEG to monitor responses in overt shifts of attention.

Latencies of early neural responses show parallel variation with saccadic latencies for saccades to single visual stimuli.

ERP latencies are significantly longer if another stimulus is competing for attention.

Supplemental Material

PEC911869 Supplemental Material1 - Supplemental material for Relation Between Event-Related Potential Latency and Saccade Latency in Overt Shifts of Attention

Supplemental material, PEC911869 Supplemental Material1 for Relation Between Event-Related Potential Latency and Saccade Latency in Overt Shifts of Attention by Louisa Kulke Janette Atkinson and Oliver Braddick in Perception

Supplemental Material

PEC911869 Supplemental Material2 - Supplemental material for Relation Between Event-Related Potential Latency and Saccade Latency in Overt Shifts of Attention

Supplemental material, PEC911869 Supplemental Material2 for Relation Between Event-Related Potential Latency and Saccade Latency in Overt Shifts of Attention by Louisa Kulke Janette Atkinson and Oliver Braddick in Perception

Footnotes

Acknowledgements

The EEG analysis was based on programs written by John Wattam-Bell, who was also involved in the planning of this project, but who unfortunately died, unexpectedly, before this article was written. The authors would like to thank Ankita Agharkar and Megan Gawryszewski for their help with data collection and Jyrki Toumainen for helpful feedback on EEG data analysis.

Author Contributions

L. K. was involved in the conception and design of the work, the acquisition, analysis and interpretation of data, drafting the work, and revising it critically for important intellectual content. J. A. and O. B. were involved in the conception and design of the work, interpretation of data, and critical revision of the manuscript for intellectual content. All authors gave final approval of the version to be published and agree to be accountable for all aspects of the work.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was funded by a German Academic Exchange Service stipend to L. K. (Grant Number: 91522396–57044644) and a Leverhulme Trust Emeritus Fellowship to J. A. (Grant Number: EM-2012–053).

ORCID iD

Louisa Kulke

Supplemental Material

Supplemental material for this article is available online.

References

Amano

Goda

Nishida

S. Y.

Ejima

Takeda

Ohtani

(2006). Estimation of the timing of human visual perception from magnetoencephalography. Journal of Neuroscience, 26, 3981–3991.

Anllo-Vento

Hillyard

S. A.

(1996). Selective attention to the color and direction of moving stimuli: Electrophysiological correlates of hierarchical feature selection. Perception & Psychophysics, 58, 191–206.

Atkinson

(1984). Human visual development over the first 6 months of life. A review and a hypothesis. Human Neurobiology, 3, 61–74.

Atkinson

(2000). Newborn vision. In Atkinson

(Ed.), The developing visual brain. Oxford University Press.

Atkinson

Braddick

(2012). Visual attention in the first years: Typical development and developmental disorders. Developmental Medicine & Child Neurology, 54, 589–595.

Atkinson

Hood

Braddick

Wattam-Bell

(1988). Infants' control of fixation shifts with single and competing targets-mechanisms of shifting attention. Perception, 17, 367–368.

Atkinson

Hood

Wattam-Bell

Braddick

(1992). Changes in infants’ ability to switch visual attention in the first three months of life. Perception, 21, 643–643.

Bartolomeo

De Schotten

M. T.

Doricchi

(2007). Left unilateral neglect as a disconnection syndrome. Cerebral Cortex, 17, 2479–2490.

Butcher

P. R.

Kalverboer

A. F.

Geuze

R. H.

(2000). Infants’ shifts of gaze from a central to a peripheral stimulus: A longitudinal study of development between 6 and 26 weeks. Infant Behavior and Development, 23, 3–21.

10.

Collins

C. E.

Lyon

D. C.

Kaas

J. H.

(2005). Distribution across cortical areas of neurons projecting to the superior colliculus in new world monkeys. The Anatomical Record Part A: Discoveries in Molecular, Cellular, and Evolutionary Biology, 285, 619–627.

11.

Corbetta

Akbudak

Conturo

T. E.

Snyder

A. Z.

Ollinger

J. M.

Drury

H. A.

Linenweber

M. R.

Petersen

S. E.

Raihle

M. E.

Shulman

G. L.

Van Essen

D. C.

(1998). A common network of functional areas for attention and eye movements. Neuron, 21, 761–773.

12.

Cordones

Gómez

C. M.

Escudero

(2013). Cortical dynamics during the preparation of antisaccadic and prosaccadic eye movements in humans in a gap paradigm. PLoS One, 8, e63751.

13.

Csibra

Johnson

M. H.

Tucker

L. A.

(1997). Attention and oculomotor control: A high-density ERP study of the gap effect. Neuropsychologia, 35, 855–865.

14.

De Schotten

M. T.

Urbanski

Duffau

Volle

Lévy

Dubois

Bartolomeo

(2005). Direct evidence for a parietal-frontal pathway subserving spatial awareness in humans. Science, 309, 2226–2228.

15.

Delorme

Westerfield

Makeig

(2007). Medial prefrontal theta bursts precede rapid motor responses during visual selective attention. Journal of Neuroscience, 27, 11949–11959.

16.

Di Russo

Martínez

Hillyard

S. A.

(2003). Source analysis of event-related cortical activity during visuo-spatial attention. Cerebral Cortex, 13, 486–499.

17.

Dorris

M. C.

Munoz

D. P.

(1995). A neural correlate for the gap effect on saccadic reaction times in monkey. Journal of Neurophysiology, 73, 2558–2562.

18.

Eimer

Forster

Velzen

J. V.

Prabhu

(2005). Covert manual response preparation triggers attentional shifts: ERP evidence for the premotor theory of attention. Neuropsychologia, 43, 957–966.

19.

Eimer

Van Velzen

Driver

(2002). Cross-modal interactions between audition, touch, and vision in endogenous spatial attention: ERP evidence on preparatory states and sensory modulations. Journal of Cognitive Neuroscience, 14, 254–271.

20.

Elison

J. T.

Paterson

S. J.

Wolff

J. J.

Reznick

J. S.

Sasson

N. J.

Botteron

K. N.

Dager

S. R.

Estes

A. M.

Gerig

Hazlett

H. C.

Schultz

R. T.

Styner

Zwaigenbaum

Piven

Evans

A. C.

, & IBIS Network. (2013). White matter microstructure and atypical visual orienting in 7-month-olds at risk for autism. The American Journal of Psychiatry, 170, 899–908.

21.

Farroni

Simion

Umiltà

Barba

B. D.

(1999). The gap effect in newborns. Developmental Science, 2, 174–186.

22.

Ferree

T. C.

Luu

Russell

G. S.

Tucker

D. M.

(2001). Scalp electrode impedance, infection risk, and EEG data quality. Clinical Neurophysiology, 112, 536–544.

23.

Fischer

(1986). The role of attention in the preparation of visually guided eye movements in monkey and man. Psychological Research, 48, 251–257.

24.

Goldberg

M. E.

Wurtz

R. H.

(1972). Activity of superior colliculus in behaving monkey. I. Visual receptive fields of single neurons. Journal of Neurophysiology, 35, 542–559.

25.

Gomez

Atienza

Gomez

Vazquez

(1996). Response latencies and event-related potentials during the gap paradigm using saccadic responses in human subjects. International Journal of Psychophysiology, 23, 91–99.

26.

Hampel

F. R.

(1974). The influence curve and its role in robust estimation. Journal of the American Statistical Association, 69, 383–393.

27.

Holm

(1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6, 65–70.

28.

Hood

Atkinson

(1993). Disengaging visual attention in the infant and adult. Infant Behavior and Development, 16, 405–422.

29.

Huber-Huber

Ditye

Fernández

M. M.

Ansorge

(2016). Using temporally aligned event-related potentials for the investigation of attention shifts prior to and during saccades. Neuropsychologia, 92, 129–141.

30.

Johnson

M. H.

(1990). Cortical maturation and the development of visual attention in early infancy. Journal of Cognitive Neuroscience, 2, 81–95.

31.

Johnson

M. H.

(2002). The development of visual attention: A cognitive neuroscience perspective. In Johnson

M. H.

Munakata

Gilmore

R. O.

(Eds.), Brain development and cognition: A reader (pp. 134–150). John Wiley & Sons.

32.

Kalesnykas

Hallett

(1987). The differentiation of visually guided and anticipatory saccades in gap and overlap paradigms. Experimental Brain Research, 68, 115–121.

33.

Katsumi

Hirose

Tsukada

(1988). Effect of number of elements and size of stimulus field on recordability of pattern reversal visual evoked response. Investigative Ophthalmology & Visual Science, 29, 922–927.

34.

Kawakubo

Kasai

Okazaki

Hosokawa-Kakurai

Watanabe

K.-I.

Kuwabara

Ishijima

Yamasue

Iwanami

Maekawa

Kato

(2007). Electrophysiological abnormalities of spatial attention in adults with autism during the gap overlap task. Clinical Neurophysiology, 118, 1464–1471.

35.

Kulke

(2015). Cortical mechanisms of visual attention in typically developing infants and adults (PhD thesis). University College London.

36.

Kulke

(2017). The effect of stimulus size and eccentricity on attention shift latencies. Vision, 1, E25.

37.

Kulke

(2019). Neural mechanisms of overt attention shifts to emotional faces. Neuroscience, 418, 59–68.

38.

Kulke

Atkinson

Braddick

(2015). Automatic detection of attention shifts in infancy: Eye tracking in the fixation shift paradigm. PLoS One, 10, e0142505.

39.

Kulke

Atkinson

Braddick

(2016a). Neural differences between covert and overt attention studied using EEG with simultaneous remote eye tracking. Frontiers in Human Neuroscience, 10, 592.

40.

Kulke

Atkinson

Braddick

(2016b). Neural mechanisms of attention become more specialised during infancy: Insights from combined eye tracking and EEG. Developmental Psychobiology, 59, 250–260.

41.

Luck

S. J.

(2005). An introduction to the event-related potential technique. MIT Press.

42.

Martinez

Anllo-Vento

Sereno

M. I.

Frank

L. R.

Buxton

R. B.

Dubowitz

D. J.

Wong

E. C.

Hinrichs

Heinze

H. J.

Hillyard

S. A.

(1999). Involvement of striate and extrastriate visual cortical areas in spatial attention. Nature Neuroscience, 2, 364–369.

43.

Matsuzawa

Shimojo

(1997). Infants’ fast saccades in the gap paradigm and development of visual attention. Infant Behavior and Development, 20, 449–455.

44.

McAdams

C. J.

Reid

R. C.

(2005). Attention modulates the responses of simple cells in monkey primary visual cortex. Journal of Neuroscience, 25, 11023–11033.

45.

Mercuri

Atkinson

Braddick

Anker

Cowan

Rutherford

Pennock

Dubowitz

(1997a). Basal ganglia damage and impaired visual function in the newborn infant. Archives of Disease in Childhood-Fetal and Neonatal Edition, 77, F111–F114.

46.

Mercuri

Atkinson

Braddick

Anker

Cowan

Rutherford

Pennock

Dubowitz

(1997b). Visual function in full-term infants with hypoxic-ischaemic encephalopathy. Neuropediatrics, 28, 155–161.

47.

Neggers

Raemaekers

Lampmann

Postma

Ramsey

(2005). Cortical and subcortical contributions to saccade latency in the human brain. European Journal of Neuroscience, 21, 2853–2863.

48.

Perry

Zeki

(2000). The neurology of saccades and covert shifts in spatial attention: An event-related fMRI study. Brain, 123, 2273–2288.

49.

Praamstra

Oostenveld

(2003). Attention and movement-related motor cortex activation: A high-density EEG study of spatial stimulus–response compatibility. Cognitive Brain Research, 16, 309–322.

50.

Regan

(1989). Human brain electrophysiology: Evoked potentials and evoked magnetic fields in science and medicine. Elsevier.

51.

Reulen

J. P.

(1984). Latency of visually evoked saccadic eye movements. Biological Cybernetics, 50, 251–262.

52.

Ricci

Cesarini

Gallini

Serrao

Leone

Baranello

Vasco

Alfieri

Staccioli

Romeo

D. M.

Tinelli

Molle

Lepore

Baldascino

Ramenghi

L. A.

Torrioli

M. G.

Romagnoli

Cowan

Atkinson

, . . . De Rose

(2010). Cortical visual function in preterm infants in the first year. The Journal of Pediatrics, 156, 550–555.

53.

Rolfs

Vitu

(2007). On the limited role of target onset in the gap task: Support for the motor-preparation hypothesis. Journal of Vision, 7, 7.

54.

Rugg

Milner

Lines

Phalp

(1987). Modulation of visual event-related potentials by spatial and non-spatial visual selective attention. Neuropsychologia, 25, 85–96.

55.

Saslow

(1967). Effects of components of displacement-step stimuli upon latency for saccadic eye movement. Journal of the Optical Society of America, 57, 1024–1029.

56.

Schiller

P. H.

(1985). A model for the generation of visually guided saccadic eye movements. In Rose

Dobson

V. G.

(Eds.), Models of the visual cortex. John Wiley & Sons Ltd.

57.

Schiller

P. H.

Stryker

(1972). Single-unit recording and stimulation in superior colliculus of the alert rhesus monkey. Journal of Neurophysiology, 35, 915–924.

58.

Schiller

P. H.

Tehovnik

E. J.

(2005). Neural mechanisms underlying target selection with saccadic eye movements. Progress in Brain Research, 149, 157–171.

59.

Sheliga

B. M.

Riggio

Rizzolatti

(1994). Orienting of attention and eye movements. Experimental Brain Research, 98, 507–522.

60.

Shomstein

Kravitz

D. J.

Behrmann

(2012). Attentional control: Temporal relationships within the fronto-parietal network. Neuropsychologia, 50, 1202–1210.

61.

Tucker

D. M.

(1993). Spatial sampling of head electrical fields: The geodesic sensor net. Electroencephalography and Clinical Neurophysiology, 87, 154–163.

62.

Vassilev

Manahilov

Mitov

(1983). Spatial frequency and the pattern onset-offset response. Vision Research, 23, 1417–1422.

63.

Wang

Chen

Yan

Zhaoping

(2015). Modulation of neuronal responses by exogenous attention in macaque primary visual cortex. Journal of Neuroscience, 35, 13419–13429.

64.

Weaver

M. D.

van Zoest

Hickey

(2017). A temporal dependency account of attentional inhibition in oculomotor control. NeuroImage, 147, 880–894.

65.

Yamaguchi

Tsuchiya

Kobayashi

(1994). Electroencephalographic activity associated with shifts of visuospatial attention. Brain, 117, 553–562.

66.

Yamaguchi

Tsuchiya

Kobayashi

(1995). Electrophysiologic correlates of visuo-spatial attention shift. Electroencephalography and Clinical Neurophysiology, 94, 450–461.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB

0.11 MB