Automated symbolic orienting is not modulated by explicit temporal attention

Acta Psychologica 171 (2016) 93–98

Contents lists available at ScienceDirect

Acta Psychologica journal homepage: www.elsevier.com/locate/actpsy

Automated symbolic orienting is not modulated by explicit temporal attention Dana A. Hayward a,⁎, Jelena Ristic b,⁎ a b

Department of Psychology, Concordia University, 7141 Sherbrooke Ave West, Montreal, QC, H4B 1R6, Canada Department of Psychology, McGill University, 1205 Dr. Penfield Ave, Montreal, QC, H3A 1B1, Canada

a r t i c l e

i n f o

Article history: Received 1 March 2016 Received in revised form 29 September 2016 Accepted 6 October 2016 Available online 13 October 2016 Keywords: Automated symbolic orienting Temporal attention Spatial attention

a b s t r a c t Recent studies show that spatial attention is uniquely engaged by the selection history of a stimulus. One example of this process is Automated Symbolic Orienting, which is thought to reflect overlearned spatial links between a behaviorally relevant stimulus and a target event. However, since automated symbolic effects have been found to vary with temporal expectancies about when a target might occur, it is possible that this spatial effect may also depend on processing resources associated with voluntary temporal attention. To test this idea, here we elicited automated symbolic orienting and voluntary temporal attention in isolation and in combination. Across all conditions, both types of orienting remained typical without interacting. Thus, typical automated symbolic orienting is not modulated by participants' explicit utilization of temporal information; however, and as we have shown previously, typical ASO does appear to require the presence of an implicit temporal structure within a task. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Recent studies (Awh, Belopolsky, & Theeuwes, 2012; Hayward & Ristic, 2013b; Ristic & Kingstone, 2012) indicate that a prominent dichotomy between bottom-up or reflexive attention and top-down or voluntary attention (Jonides, 1981; Posner, Snyder, & Davidson, 1980) does not fully capture the richness of attentional processes occurring in real life. In addition to stimulus properties and individual goals, attention has been found to be independently driven by events that hold motivational value (e.g., Anderson, Laurent, & Yantis, 2011) and the history of stimulus selection (Awh et al., 2012). One example of this novel type of attentional control is Automated Symbolic Orienting (ASO; Ristic & Kingstone, 2012), in which overlearning the meaning of useful everyday symbols such as arrows leads to faster responses for targets that are congruent with the cue's direction relative to events that occur elsewhere (Ristic & Kingstone, 2012; Ristic, Landry, & Kingstone, 2012). ASO effects emerge quickly by 100 ms post cue and persist until about 1000 ms (Brignani, Guzzon, Marzi, & Miniussi, 2009; Ristic, Friesen, & Kingstone, 2002; Ristic & Kingstone, 2006, 2012). ASO effects occur when the cue is spatially uninformative (i.e., points equally often towards and away from the target, Ristic et al., 2002; Tipples, 2002), proceed in parallel with reflexive and voluntary spatial orienting (Ristic & Kingstone, 2012; Ristic et al., 2012), and facilitate both the speed of target processing, as reflected by target detection measures, as well as its ⁎ Corresponding authors. E-mail addresses: [email protected] (D.A. Hayward), [email protected] (J. Ristic).

http://dx.doi.org/10.1016/j.actpsy.2016.10.004 0001-6918/© 2016 Elsevier B.V. All rights reserved.

perceptual analysis, as reflected by target discrimination measures (Ristic et al., 2012). It was recently proposed however that these typical ASO effects may depend on participants anticipating when in time a response target might occur (Hayward & Ristic, 2015). We (Hayward & Ristic, 2015) measured ASO under conditions in which the implicit structure of cue-target events within the task did and did not provide temporal information about the target's occurrence. We found typical early and prolonged ASO when the cue-target temporal sequence was implicitly predictable, however when the implicit temporal link between the cue and the target was unpredictable, automated symbolic orienting was delayed in its onset until 900 ms. This opens up the possibility that while ASO may occur in parallel with reflexive and voluntary spatial orienting (Ristic & Kingstone, 2012; Ristic et al., 2012), it may nevertheless share processing resources with temporal attention, which modulates expectancies about the timing of target events (e.g., Nobre, 2001). To address this question, here we assessed whether spatial automated symbolic orienting was affected by voluntary temporal attention. A large number of investigations (Griffin, Miniussi, & Nobre, 2002; Los, 2004; MacKay & Juola, 2007; Milliken, Lupiáñez, Roberts, & Stevanovski, 2003; Miniussi, Wilding, Coull, & Nobre, 1999; Weinbach, Shofty, Gabay, & Henik, 2015) suggest that while spatial and temporal orienting may share some processing resources, the two types of attention serve different purposes, and involve several different underlying neural mechanisms (cf. Doherty, Rao, Mesulam, & Nobre, 2005; Rohenkohl, Gould, Pessoa, & Nobre, 2014). This dissociation was first demonstrated by Coull and Nobre (1998; see also Miniussi et al.,

94

D.A. Hayward, J. Ristic / Acta Psychologica 171 (2016) 93–98

1999; Nobre, 2001; Nobre, Correa, & Coull, 2007). In their study, the authors used a modified cuing paradigm in which participants were presented with a spatially predictive cue, a temporally predictive cue, or a spatially and temporally predictive cue. Behavioral data indicated reliable orienting for both spatial and temporal cues in isolation and in conjunction with no interactions. That is, like the informative spatial cue, participants were also able to use an informative temporal cue to respond faster when the target appeared at the expected time point as compared to when the target appeared at an unexpected time point. The analyses of brain activity further revealed that while spatial orienting preferentially engaged the right inferior parietal lobe, temporal orienting preferentially engaged the left inferior parietal lobe and left premotor areas. Based on these results, it was argued that orienting in the spatial and temporal domains was additive because the two attentional systems differentially affected the underlying sensory and cognitive processes. Since this pioneering study, the relationship between spatial and temporal orienting has been described as both additive and synergistic, with spatial and temporal attention each linked with facilitating the target's sensory analysis (Correa, Lupiàñez, & Tudela, 2005; Cravo, Rohenkohl, Wyart, & Nobre, 2013; Rohenkohl et al., 2014). However, temporal attention has been specifically linked with furnishing the preparation and timing of responses (Nobre, 2001) and more recently with accelerating the perceptual processing of the target (Seibold & Rolke, 2014a, 2014b; Vangkilde, Coull, & Bundesen, 2012). These ideas were supported by EEG investigations. Miniussi et al. (1999) reported that modulations in spatial target expectancies affected the early event-related components associated with early sensory processing (e.g., P1, N1; Hillyard, Hink, Schwent, & Picton, 1973; Mangun, 1995; Van Voorhis & Hillyard, 1977), while modulations in temporal target expectancies affected the later event-related components associated with response execution and expectancy formation (e.g., the CNV and P300). More recently, Seibold and Rolke (2014b) found that latencies of the early sensory event-related components (i.e., the N2pc) also became shorter when the task included a predictable temporal sequence. Thus, one distinctive function of temporal attention relates to the temporal fine-tuning of responses for expected targets, which in turn contributes to the acceleration of a target's sensory processing (Griffin, Miniussi, & Nobre, 2001; Griffin et al., 2002; Miniussi et al., 1999; Seibold & Rolke, 2014a, 2014b; cf. Correa, Lupiàñez, Milliken, & Tudela, 2004; Vangkilde et al., 2012). Our recent result showing delays in ASO's onset under conditions in which the implicit temporal links between the cue and the target were reduced dovetails with this proposed role of temporal attention. That is, the absence of early ASO under reduced temporal predictability (Hayward & Ristic, 2015) suggests that ASO's rapid onset may be contingent on the presence of a reliable temporal structure between cues and targets. Generally, temporal expectancies can be either explicit or implicit (Coull & Nobre, 2008). Explicit temporal expectancies involve the deliberate use of temporal information. Experimentally, this is achieved by presenting cues that indicate a likely time at which a target may appear. Participants are instructed to utilize this contingency to maximize their performance. Implicit temporal expectancies, on the other hand, involve participants' non-deliberate utilization of the more subtle temporal regularities present within the task sequence. Experimentally, this is often achieved by presenting the cues and targets within a rhythmical sequence of cue-target intervals, in which the probability of target occurrence within a trial increases with the lengthening of cue-target time (i.e., an aging distribution of trials; Gabay & Henik, 2008; Näätänen, 1970). And although participants are typically not informed about this regularity, their performance is facilitated by such task conditions. A classic example of implicit temporal performance facilitation is the foreperiod effect (e.g., Bertelson, 1967; Gabay & Henik, 2008; Hayward & Ristic, 2013a), which is indexed by an overall speeding up of responses with increases in the time delay between the cue

and the target, when the cue-target intervals are presented in an intermixed fashion. In our previous study (Hayward & Ristic, 2015), we manipulated implicit temporal expectancies within a cuing task by modulating the presence and absence of a foreperiod effect (i.e., by using aging and nonaging distributions of cue-target intervals; see also Gabay & Henik, 2008; Hayward & Ristic, 2013a). In contrast to the aging distribution, in which the presentation of an equal number of targets at each cue-target interval contributes to the increased probability of target occurrence at longer cue-target times, the non-aging distribution keeps the probability of target occurrence equal by halving the number of targets presented at each successive cue-target time (Gabay & Henik, 2008; Näätänen, 1970). Our data (e.g., Hayward & Ristic, 2015) indicated that the rapid ASO effects were contingent on the presence of the tasks' implicit temporal structure, as early ASO emerged only when the task involved an aging distribution of trials, and not when the task involved a non-aging distribution of trials. While in general this implicates temporal predictability in the development of early spatial symbolic orienting, it does not specifically reveal whether participants utilized the task's temporal structure in a deliberate manner. In other words, it remains unknown if typical ASO depends on the implicit or explicit utilization of available temporal information. To address this question, here we assessed if ASO was influenced by the formation of explicit temporal expectancies about when a target might occur. To do so, we manipulated and measured spatial automated symbolic orienting and explicit voluntary temporal orienting alone in isolation, and simultaneously in combination. If ASO did not require explicit voluntary temporal attention, we expected to observe the two processes in their typical forms across all conditions (Coull & Nobre, 1998; Miniussi et al., 1999). More specifically, we expected to find no interactions between spatial and temporal orienting when the two processes were elicited simultaneously. 2. Methods 2.1. Participants Thirty-two (32) undergraduate students (5 males, mean age 19.9 ± 1.4 years) participated in the experiment. 2.2. Stimuli & Apparatus The stimuli are shown in Fig. 1A. They were rendered in black and presented against a white background. Stimuli included a large circle (8.0°), a small circle (0.3°), and an arrow, which was comprised of a horizontal line (2.9° long), an arrowhead (0.6°), and a vertical stop line (0.8°). A capital letter ‘X’ (1°) served as the response target, appearing with an eccentricity of 7.1° from central fixation along the horizontal meridian. The stimuli were presented on a 16-in CRT monitor at an approximate viewing distance of 57 cm. 2.3. Design Four possible Trial types were distributed across 552 target-present trials: (i) No Cue trials, in which neither a Space nor a Time cue occurred; (ii) Space only trials, in which an arrow pointing to the left or to the right served as a spatial cue; (iii) Time only trials, in which the brightening (from 1 to 5 points) of either the large or the small circle served as a temporal cue; and (iv) Space and Time trials, in which both the Space cue and the Time cue occurred simultaneously. In this last condition, Space and Time cues could converge onto the same target (i.e., Both valid; Both invalid) or be committed in a divergent fashion to the processing of two different targets (i.e., Space valid/Time invalid or Time

D.A. Hayward, J. Ristic / Acta Psychologica 171 (2016) 93–98

95

Fig. 1. Stimuli and Procedure. A depicts the four different Trial types (not drawn to scale): No Cue, Space Only, Time Only, Space & Time. B illustrates an example stimulus presentation sequence. Each trial started with a fixation display. After 600 ms, one of the four cue types was presented (Time only cue shown here). Following a variable cue-target interval of 100 or 1200 ms, a target (‘X’) appeared to the left or the right of the cue. For the No Cue trials, the target appeared 100 or 1200 ms after the presentation of the initial fixation screen. The cue(s) and target remained on the screen until response or until 2000 ms had elapsed. For the no target trials, the cue remained on the screen for 1250 ms before the trial timed out. The intertrial interval was 650 ms.

valid/Space invalid)1, 2. Table 1 shows all trial counts as a function of cue condition. Trial type, cue-target interval, and cue validity were manipulated within participants. The arrow cue was spatially uninformative of where the target would appear (p = 0.5). The circle cue was temporally informative of when the target would appear (p = 0.88). In all conditions, the targets were presented using an aging distribution of trials, in which a target is equally likely to be presented following each cuetarget interval, a practice that creates an implicitly probable temporal sequence (Gabay & Henik, 2008; Hayward & Ristic, 2013a; Näätänen, 1972). The cue-target intervals of 100 and 1200 ms were used to capture the time courses of ASO and voluntary temporal attention, both of which normally exert strong effects at early cue-target times. In all of these respects, the procedures for eliciting ASO and explicit voluntary temporal orienting mirrored previous work (e.g., Coull & Nobre, 1998; Ristic & Kingstone, 2012). The assignment of temporal cue type (i.e., brightening of small vs. large circle) in predicting the target at either the early or late target interval (i.e., 100 vs. 1200 ms) was counterbalanced between participants. Cue validity reflected the spatial and/or temporal congruency between the cue and the target. For the Isolated Space and Time conditions, trials could either be Valid (i.e., the target was congruent with the cue) or Invalid (i.e., the target was incongruent with the cue). For the conditions in which both Space and Time cues appeared, trials could be Both cues valid, One cue (either Space or Time) valid, or Both cues invalid. About 6% of trials (36 trials) contained no target, which along with 552 target-present trials formed a total of 588 testing trials. All conditions were presented in an intermixed random fashion, with trials distributed equally across two testing blocks.

left or right side of fixation and participants were asked to detect its onset by pressing the spacebar as fast and as accurately as possible. For the No Cue trials, the target appeared 100 or 1200 ms after the presentation of the initial fixation screen. The target remained on the screen until participants responded or until 2000 ms had elapsed. For the no-target trials the cue was presented for 1250 ms. The intertrial interval was 650 ms. Participants were informed about the cues' spatial and temporal contingencies. RT was measured from target onset, and five practice trials were run at the start. 3. Results Response errors, which included false alarms (i.e., responding on a no-target trial; 1.1%), timed-out responses (i.e., trials with no response; 0.03%), and trials with incorrect key presses (i.e., trials where another key besides the spacebar was pressed; 0.05%) accounted for b1.2% of all data and were removed from all analyses. Interparticipant mean correct median RTs were calculated for each Trial type as a function of cue validity and cue-target interval. The data were analyzed using repeated measures ANOVAs. We conducted two sets of analyses to examine our main question regarding the relationship between ASO and voluntary temporal attention. First, we examined if typical spatial and temporal effects emerged for the isolated Space and Time conditions. Then, we analyzed the condition in which Space and Time cues were presented simultaneously. These analyses were performed separately due to a different number of cue validity levels for the Isolated and Simultaneous cue conditions (i.e., two levels of cue validity for the isolated conditions vs. four levels of cue validity for the simultaneous condition). 3.1. Isolated spatial and temporal effects

2.4. Procedure Fig. 1B shows an example stimulus presentation sequence. Each trial began with the presentation of a fixation display for 600 ms. Then a Space Only cue, Time Only cue, Space and Time cue or No Cue was presented. After 100 or 1200 ms a response target appeared on either the 1 Please note that the use of the ‘convergent/divergent’ terminology corresponds to the existing language in the field. In the present case, we use this terminology to capture the idea of two processes (i.e., spatial and temporal attention) being committed to the processing of the same target event (convergent) versus the processing of two target events (divergent). 2 As the No Cue condition did not contain an alerting signal (e.g., a nonspecific temporal cue, as per Coull & Nobre, 1998), it was examined in separate analyses. A comparison of RTs for No Cue trials with conditions in which attentional cues were presented indicated that overall RTs were reliably slower in the No Cue case [Trial Type main effect; F(3,93) = 44.1, p b 0.001], replicating Coull and Nobre's (1998) original results. More specifically, No Cue trials (RT: 368 ms, SD: 38 ms) were slower than Space Only trials (RT: 347 ms, SD: 35 ms; t(31) = − 5.2, p b 0.001), Time Only (RT: 341 ms, SD: 37 ms; t(31) = − 6.3, p b 0.001), and Space and Time trials (RT: 334 ms, SD: 34 ms; t(31) = −8.9, p b 0.001). No Cue trials were also reliably slower than all invalid trials, relative to Isolated invalid Space [F(1,31) = 17.2, p b 0.001], Isolated invalid Time [F(1,31) = 16.4, p b 0.001] and Both invalid trials [F(1,31) = 32.9, p b 0.001]. All t-tests reflect two-tailed paired comparisons with Bonferroni-corrected p-values.

Fig. 2A and B show that when the Space and Time cues were presented in isolation typical orienting effects emerged for both automated symbolic orienting (2 A, e.g., Ristic & Kingstone, 2012) and voluntary temporal attention (2B, e.g., Coull & Nobre, 1998). An omnibus analysis with Trial Type (Space only, Time only), Cue Validity (valid, invalid) and Cue-target Interval (100 ms, 1200 ms) confirmed this observation. Main effects of Trial Type [F(1,31) = 6.3, p b 0.05], showing overall faster RTs for Time cues (e.g., Coull & Nobre, 1998), Cue Validity [F(1,31) = 15.7, p b 0.001], showing cuing effects for each Space and Time cues, and Cue-target Interval [F(1,31) = 37.1, p b 0.001] indicating a typical foreperiod effect (Bertelson, 1967) were reliable. No interactions Table 1 Target-present trial distribution counts as a function of condition. Valid (V); Invalid (IV); Space (S); Time (T). No cue

Count

48 48

Space

Time

V

IV

V

IV

Both V

Space & time SV/T IV

TV/S IV

Both IV

Sum

48 96

48

120 136

16

120 272

16

120

160 552

96

D.A. Hayward, J. Ristic / Acta Psychologica 171 (2016) 93–98

Fig. 2. Results. Mean correct median RTs for the Isolated Space (A), Isolated Time (B) and Simultaneous Space and Time cues (C). For the Isolated Space and Time cues, the RTs are depicted as a function of Cue validity (valid; invalid) and Cue-target Interval (100 ms; 1200 ms). For the simultaneous cues, the RTs are depicted as a function of Cue validity (Both cues valid, Space valid/Time invalid, Time valid/Space invalid, and Both cues invalid) and Cue-target Interval (100 ms; 1200 ms). Error bars depict the standard error of the difference between the means (Loftus & Masson, 1994).

reached significance (Trial Type × Cue-target Interval, F b 1; Trial Type × Cue Validity F(1,31) = 1.9, p = 0.17; Cue-target Interval × Cue Validity F(1,31) = 2.6, p = 0.11; Trial Type × Cue-target Interval × Cue Validity F(1,31) = 1.5, p = 0.23). The same results emerged when the Space and Time cues were examined in separate analyses, each run as a function of Cue Validity and Cue-target Interval. Both analyses indicated reliable cuing effects for ASO [Space only Cue Validity: F(1,31) = 9.8, p b 0.01] and temporal attention [Time only Cue Validity: F(1,31) = 8.7, p b 0.01]. ASO effects did not vary with cue-target time [Cue validity × Cue-target Interval, F b 1] while a marginally significant interaction between Cue validity and Cue-target Interval for the temporal attention condition suggested diminishing effects with the lengthening of cue-target time [F(1,31) = 3.8, p = 0.06]. Thus, robust ASO effects emerged at both early and late cue-target intervals. Likewise, typical voluntary temporal orienting, which was strong at the early cue-target time and trended towards reduction at the long cue-target time emerged, closely replicating previous reports (Correa, Lupiàñez, & Tudela, 2006; Coull & Nobre, 1998). 3.2. Simultaneous spatial and temporal effects Similar results indicating typical spatial and temporal effects were found when we examined the condition in which Space and Time cues occurred together, as illustrated in Fig. 2C. To remind, here both Space and Time cues occurred on each trial. The two cues could either both indicate the target location and target time (i.e., convergent cues), or only one cue, either Space or Time could be congruent with the target location or target time (i.e., divergent cues). To examine if ASO was modulated across Time cue validity, we analyzed the data as a function of Space Validity (Valid, Invalid), Time Validity (Valid, Invalid) and Cue-target Interval (100 ms, 1200 ms). This ANOVA returned reliable main effects of Space Validity [F(1,31) = 24.0, p b 0.001], Time Validity [F(1,31) = 9.0, p b 0.01] and Cue-target Interval [F(1,31) = 19.7, p b 0.001], with no interactions between Space and Time cue validity and/or as a function of Cue-target Interval (Space validity × Time validity, F b 1; Space validity × Time validity × Cue-target Interval, F(1,31) = 2.6, p = 0.11). Thus, orienting in the spatial domain did not interact with orienting in the temporal domain. Finally, to control for the diminishing magnitude of voluntary temporal orienting at the later cue-target time, we also examined ASO and temporal attention at each Cue-target interval separately, as a function of Space Validity and Time Validity. At the early cue-target time, both Space Validity [F(1,31) = 20.7, p b 0.001] and Time Validity [F(1,31) = 7.0, p b 0.05] were reliable, with no interactions [F(1,31) = 3.1, p = 0.09]. At the late cue-target time, only Space Validity [F(1,31) = 4.3, p b 0.05] was significant, with no other reliable effects [Time Validity, F b 1; Space Validity × Time Validity, F b 1]. As such, these analyses further

show that spatial and temporal orienting did not interact even when we accounted for the different time courses of the two effects. That is, ASO continued to show robust early and late effects, while voluntary temporal orienting was maximal at the short cue-target interval. Taken together, these results show that ASO was reliable both when it occurred in isolation and when it occurred in the presence of an explicit temporal cue.3 No delays in the spatial ASO emerged. Likewise, voluntary temporal orienting was typical both when it occurred in isolation and when it occurred in the presence of ASO. Thus, when the Space and Time cues were presented in isolation, typical effects for each cue were observed. Similarly, when the Space and Time cues were presented simultaneously reliable orienting effects with no interactions were observed. 4. Discussion In this study we examined whether spatial automated symbolic orienting depended on the engagement of voluntary temporal attention. While ASO was previously found to be independent from both reflexive and voluntary spatial attention, our recent work (Hayward & Ristic, 2015) suggested that it may share resources with temporal attention, as ASO became delayed when the temporal expectation about when in time a target would appear within the task was reduced. Here we elicited spatial automated symbolic orienting and voluntary temporal orienting alone in isolation, and simultaneously in combination. Across all cue conditions, both the nonpredictive spatial arrow cue and the predictive temporal cue produced their typical effects. ASO emerged quickly by 100 ms and persisted until 1200 ms, both when it was elicited in isolation and when it was elicited along with a Time cue. Voluntary temporal orienting was robust at 100 ms and showed a typical decline at the longest cue-target interval of 1200 ms. No interactions between spatial and temporal orienting were found. Note that the short cue-target interval of 100 ms used in the present study was considerably shorter than the cue-target intervals that were employed in previous investigations of voluntary temporal attention 3 Further support for the resilience of ASO to voluntary temporal orienting is provided by the comparison of performance for valid trials across the Isolated and Simultaneous divergent conditions. Namely, we found no evidence that valid cues in the Isolated conditions facilitated performance more than valid cues in the Simultaneous divergent conditions. This was supported by two one-way ANOVAs, which contrasted (i) valid trials from the Isolated Space condition with the Space valid/Time invalid trials from the Simultaneous cue condition, and (ii) valid trials from the Isolated Time condition with the Time valid/Space invalid trials from the Simultaneous cue condition. The first analysis indicated that valid Space trials from the Simultaneous divergent condition (i.e., Valid Space/Invalid Time) were faster than valid trials from the Isolated Space condition [F(1,31) = 17.8, p b 0.001]. The second ANOVA returned no reliable difference between the valid Time trials from the Simultaneous divergent condition (i.e., Valid Time/Invalid Space) versus valid trials from the Isolated Time condition (F b 1, p N 0.8). Thus, both ASO and voluntary temporal orienting remained typical even under conditions when the other cue was invalid.

D.A. Hayward, J. Ristic / Acta Psychologica 171 (2016) 93–98

(i.e., 400 ms; Correa et al., 2004; Coull & Nobre, 1998). This suggests that, similarly to voluntary spatial attention (e.g., Ristic & Kingstone, 2006, 2009), voluntary temporal effects may also be present earlier than originally thought. Thus, our data show that automated symbolic orienting is not modulated by participants' explicit temporal knowledge about when the target might occur. As such, these data suggest that ASO and voluntary temporal attention may be independent. One method that is used to examine process independence is additive factors logic (Sternberg, 1969). This technique is based on examining the independence of successive processing stages that are associated with a particular cognitive operation. If, as we have hypothesized in the Introduction, ASO required the utilization of voluntary temporal processing resources, one would predict that spatial and temporal orienting would share a common processing stage related to the extraction of the target's temporal predictability. This in turn predicts the emergence of interactions between spatial and temporal orienting, especially when the two processes are engaged in opposition, like in the divergent Space and Time condition. One way to test for independence using additive factors method is to compare the magnitudes of spatial and temporal orienting in conditions in which the two processes converged onto the processing of the same target (i.e., convergent cues) with an additive sum of the magnitudes of the two processes when they diverged towards the processing of two targets (i.e., divergent cues; Hayward & Ristic, 2013b; Ristic & Kingstone, 2012). The equivalence between these magnitudes would suggest that ASO and voluntary temporal attention did not share processing stages either when they converged towards the processing of the same target or when they diverged towards processing of two different targets (i.e., they were elicited in opposition). The results from our analyses supported this notion. One-way ANOVAs compared the magnitudes of orienting for the convergent Space and Time cues (i.e., RT Both cues invalid – RT Both cues valid) with the sum of orienting magnitudes obtained for the divergent Space and Time cues (i.e., [RT Both cues invalid – RT Space valid/Time invalid] + [RT Both cues invalid – RT Time valid/Space invalid]). Comparisons were performed for each cue-target interval separately. Neither contrast returned a statistically reliable difference between the magnitudes of orienting (100 ms: F(1,31) = 3.1, p = 0.09; 1200 ms: F b 1).4 In summary, our data indicated reliable spatial and temporal effects, with no interactions regardless of whether the processes were elicited alone or simultaneously. These results held both in omnibus analyses as well as when the effects were examined at each cue-target interval separately. Examinations of spatial and temporal orienting magnitude additivity further indicated that the magnitudes of convergent spatial and temporal effect differed neither from the sum of the magnitudes obtained from the divergent spatial and temporal orienting nor from the sum of isolated spatial and temporal effects. As such, these data show that spatial automated orienting and voluntary temporal attention do not interact and likely do not share common processing stages, i.e., they are independent. It was suggested to us that despite the nonsignificant differences, an inspection of Fig. 2C suggests a numerical trend in which the magnitude of orienting in the convergent space and time condition appears to be smaller than the sum of the orienting magnitudes from the divergent space and time conditions. We can speculate that such numerical variation may potentially suggest that partial foreknowledge about an upcoming event (i.e., one cue valid) can be almost as effective as full 4 The same result suggesting process independence, (i.e., no reliable difference) emerged when the data were examined using an omnibus analysis [main effect: F b 1; interaction with Cue-target interval: F(1,31) = 2.6, p = 0.11]. Further, the same result also emerged when we contrasted the magnitude of orienting obtained in the convergent Space and Time condition (i.e., RT Both cues invalid − RT Both cues valid) with the additive sum of the effect magnitudes obtained from the Isolated Space and Time conditions (i.e., [RT Isolated Space invalid − RT Isolated Space valid] + [RT Isolated Time invalid − RT Isolated Time invalid]), 100 ms Analysis: F b 1; 1200 ms Analysis: F b 1. Omnibus analysis [main effect: F b 1; interaction with Cue-target interval: F b 1].

97

foreknowledge about an event (i.e., both cues valid; see Ristic & Landry, 2015 for a similar finding within the spatial domain). Overall then, these results replicate the original Coull and Nobre's (1998) finding and extend it in three ways. First, we observed equivalent temporal orienting effects when all Trial types were presented in an intermixed rather than in a blocked fashion. Second, the additivity between spatial and temporal attention held when instead of a spatially predictive cue we utilized a spatially nonpredictive cue. This confirms that voluntary temporal attention is not only independent from voluntary spatial orienting, but that it is also independent from the less cognitively demanding automated spatial orienting. Finally, our design also included trials in which only one cue was valid in the space and time condition, which allowed for a more stringent test of the independence between the two processes (see also a recent paper by Weinbach et al., 2015 for a similar design). Across all of these design changes, the original result suggesting independence between spatial and temporal attention held. With respect to the theoretical advances, our results provide two novel insights regarding automated symbolic orienting, i.e., spatial attentional orienting in response to overlearned behaviorally relevant symbols. First, they show that in addition to being independent from classic reflexive and voluntary spatial orienting, automated symbolic orienting also does not interact with the classic explicit voluntary temporal attention, and as such does not depend on the explicit utilization of temporal expectancies. However, and as we have previously demonstrated (Hayward & Ristic, 2015), ASO is delayed by the reduction in the implicit temporal contingencies between the cue and the target. Along with the explicit temporal links between the cues and targets, the present manipulation included the implicit temporal structure of cue-target intervals (i.e., by using an aging distribution of trials). The effectiveness of this manipulation was indexed by consistently reliable foreperiod effects (Gabay & Henik, 2008; Näätänen, 1972). Consequently, ASO remained typical in its time course, without delays. Unlike explicit temporal attention, which reflects the deliberate utilization of temporal contingencies, implicit temporal effects are thought to reflect the participants' passive entrainment to the task's temporal structure. It is known that an organized temporal structure aids attentional orienting, especially when sensory processing or motor responses depend on the regular occurrence of events in the environment (Coull & Nobre, 2008). ASO appears to benefit from such regular temporal structure, in line with recent findings (Seibold & Rolke, 2014a, 2014b) indicating that implicit temporal contingencies accelerate the spatial sensory analysis of the target. It is however worth noting here that it remains possible that ASO may be modulated by voluntary temporal contingencies under conditions in which the degree of temporal uncertainty is reduced rather than fully eliminated. Specifically, the manipulation of explicit temporal cues like the ones employed in the present study provided high temporal certainty, while the manipulation of implicit temporal cues like the ones manipulated in the Hayward and Ristic (2015) study provided little or no temporal certainty, due to the utilization of a non-aging distribution of trials. In other words, it is possible that a quantitative difference reflecting the degree of temporal certainty rather than a qualitative difference reflecting the utilization of implicit versus explicit temporal contingencies may be responsible for the temporal fine-tuning that contributes to rapid ASO effects. Future investigations that manipulate the degree of temporal certainty as well as those that examine the time course of processing associated with automated symbolic orienting in relation to the known behavioral and neural markers of spatial and temporal attention will shed more light onto the specific contribution of temporal attention to automated symbolic orienting. Second, when our current findings are considered together with those from Hayward and Ristic (2015), it appears that behaviorally relevant cues like arrows drive human attention uniquely because their effects require regularities from both spatial and temporal domains. While the implicit temporal structure in the environment facilitates the

98

D.A. Hayward, J. Ristic / Acta Psychologica 171 (2016) 93–98

target's early sensory analysis, the overlearned spatial associations ensure the continuation of spatial orienting towards the target. Both of these findings dovetail well with the everyday relevance of symbolic cues like arrows, which are hypothesized to engage attention because of their utility for current behavior. This coupling between past learning and present context so far appears to be unique to the spatial effects elicited by behaviorally relevant cues like arrows. Social cues like eye gaze, in contrast, elicit spatial attentional effects that are largely unaffected by the task's structure (Hayward & Ristic, 2015) and the spatial predictability of the cue (Friesen, Ristic, & Kingstone, 2004; Hayward & Ristic, 2013b). Thus, human attention is sensitive to both prior experience and the present environmental structure, a finding that supports an expanded view of attention beyond the traditional reflexive and voluntary attentional control (Ristic & Enns, 2015). Acknowledgements Supported by a graduate fellowship from the Natural Sciences and Engineering Research Council of Canada (NSERC) to DAH, and by grant support from NSERC, the Social Sciences and Humanities Research Council of Canada (SSHRC), G.W. Stairs, and William Dawson funds to JR. Thanks to C. Dick for help with data collection, and to Sander Los and an anonymous reviewer for their insightful comments on a previous version of this manuscript. References Anderson, B. A., Laurent, P. A., & Yantis, S. (2011). Value-driven attentional capture. Proceedings of the National Academy of Sciences of the United States of America, 108(25), 10367–10371. Awh, E., Belopolsky, A. V., & Theeuwes, J. (2012). Top-down versus bottom-up attentional control: A failed theoretical dichotomy. Trends in Cognitive Sciences, 16(8), 437–443. Bertelson, P. (1967). Time course of preparation. The Quarterly Journal of Experimental Psychology, 19(3), 272–279. Brignani, D., Guzzon, D., Marzi, C. A., & Miniussi, C. (2009). Attentional orienting induced by arrows and eye-gaze compared with an endogenous cue. Neuropsychologia, 47(2), 370–381. Correa, A., Lupiàñez, J., Milliken, B., & Tudela, P. (2004). Endogenous temporal orienting of attention in detection and discrimination tasks. Perception & Psychophysics, 66(2), 264–278. Correa, A., Lupiàñez, J., & Tudela, P. (2005). Attentional preparation based on temporal expectancy modulates processing at the perceptual level. Psychonomic Bulletin & Review, 12(2), 328–334. Correa, A., Lupiàñez, J., & Tudela, P. (2006). The attentional mechanism of temporal orienting: Determinants and attributes. Experimental Brain Research, 169, 58–68. Coull, J. T., & Nobre, A. C. (1998). Where and when to pay attention: The neural systems for directing attention to spatial locations and to time intervals as revealed by both PET and fMRI. The Journal of Neuroscience, 18(18), 7426–7435. Coull, J. T., & Nobre, A. C. (2008). Dissociating explicit timing from temporal expectation with fMRI. Current Opinion in Neurobiology, 18, 137–144. Cravo, A. M., Rohenkohl, G., Wyart, V., & Nobre, A. C. (2013). Temporal expectation enhances contrast sensitivity by phase entrainment of low-frequency oscillations in visual cortex. The Journal of Neuroscience, 33(9), 4002–4010. Doherty, J. R., Rao, A., Mesulam, M. M., & Nobre, A. C. (2005). Synergistic effect of combined temporal and spatial expectations on visual attention. The Journal of Neuroscience, 25(36), 8259–8266. Friesen, C. K., Ristic, J., & Kingstone, A. (2004). Attentional effects of counterpredictive gaze and arrow cues. Journal of Experimental Psychology: Human Perception and Performance, 30(2), 319–329. Gabay, S., & Henik, A. (2008). The effects of expectancy on inhibition of return. Cognition, 106, 1478–1486. Griffin, I. C., Miniussi, C., & Nobre, A. C. (2001). Orienting attention in time. Frontiers in Bioscience, 6, d660–d671. Griffin, I. C., Miniussi, C., & Nobre, A. C. (2002). Multiple mechanisms of selective attention: Differential modulation of stimulus processing by attention to space or time. Neuropsychologia, 40, 2325–2340.

Hayward, D. A., & Ristic, J. (2013a). Measuring attention using the Posner cuing paradigm: The role of across and within trial target probabilities. Frontiers in Human Neuroscience, 7(205). http://dx.doi.org/10.3389/fnhum.2013.00205. Hayward, D. A., & Ristic, J. (2013b). The uniqueness of social attention revisited: Working memory load interferes with endogenous but not social orienting. Experimental Brain Research, 231(4), 405–414. Hayward, D. A., & Ristic, J. (2015). Exposing the cuing task: The case of gaze and arrow cues. Attention, Perception, & Psychophysics, 77(4), 1088–1104. Hillyard, S. A., Hink, R. F., Schwent, V. L., & Picton, T. W. (1973). Electrical signs of selective attention in the human brain. Science, 182, 177–180. Jonides, J. (1981). Voluntary versus automatic control over the mind's eye's movement. In J. B. Long, & A. D. Baddeley (Eds.), Attention and performance. vol. IX. (pp. 187–203). NJ: Erlbaum: Hillsdale. Loftus, G. R., & Masson, M. E. J. (1994). Using confidence intervals in within-subject designs. Psychonomic Bulletin & Review, 1(4), 476–490. Los, S. A. (2004). Inhibition of return and nonspecific preparation: Separable inhibitory control mechanisms in space and time. Perception & Psychophysics, 66(1), 119–130. MacKay, A., & Juola, J. F. (2007). Are spatial and temporal attention independent? Perception & Psychophysics, 69(6), 972–979. Mangun, G. R. (1995). Neural mechanisms of visual selective attention. Psychophysiology, 32, 4–18. Milliken, B., Lupiáñez, J., Roberts, M., & Stevanovski, B. (2003). Orienting in space and time: Joint contributions to exogenous spatial cuing effects. Psychonomic Bulletin & Review, 10(4), 877–883. Miniussi, C., Wilding, E. L., Coull, J. T., & Nobre, A. C. (1999). Orienting attention in time: Modulation of brain potentials. Brain, 122, 1507–1518. Näätänen, R. (1970). The diminishing time-uncertainty with the lapse of time after the warning signal in reaction-time experiments with varying fore-periods. Acta Psychologica, 34, 399–419. Näätänen, R. (1972). Time uncertainty and occurrence uncertainty of the stimulus in a simple reaction time task. Acta Psychologica, 36, 492–503. Nobre, A. C. (2001). Orienting attention to instants in time. Neuropsychologia, 39, 1317–1328. Nobre, A. C., Correa, A., & Coull, J. T. (2007). The hazards of time. Current Opinion in Neurobiology, 17(4), 465–470. Posner, M. I., Snyder, C. R. R., & Davidson, B. J. (1980). Attention and the detection of signals. Journal of Experimental Psychology: General, 109(2), 160–174. Ristic, J., & Enns, J. T. (2015). The changing face of attentional development. Current Directions in Psychological Science, 24(1), 24–31. Ristic, J., & Kingstone, A. (2006). Attention to arrows: Pointing to a new direction. The Quarterly Journal of Experimental Psychology, 59(11), 1921–1930. Ristic, J., & Kingstone, A. (2009). Rethinking attentional development: Reflexive and volitional orienting in children and adults. Developmental Science, 12(2), 289–296. Ristic, J., & Kingstone, A. (2012). A new form of human spatial attention: Automated symbolic orienting. Visual Cognition, 20(3), 244–264. Ristic, J., & Landry, M. (2015). Combining attention: A novel way of conceptualizing the links between attention, sensory processing, and behavior. Attention, Perception, & Psychophysics, 77(1), 36–49. Ristic, J., Friesen, C. K., & Kingstone, A. (2002). Are eyes special? It depends on how you look at it. Psychonomic Bulletin & Review, 9(3), 507–513. Ristic, J., Landry, M., & Kingstone, A. (2012). Automated symbolic orienting: The missing link. Frontiers in Psychology, 3(560). http://dx.doi.org/10.3389/fpsyg.2012.00560. Rohenkohl, G., Gould, I. C., Pessoa, J., & Nobre, A. C. (2014). Combining spatial and temporal expectations to improve visual perception. Journal of Vision, 14(4), 1–13 8. Seibold, V. C., & Rolke, B. (2014a). Does temporal preparation facilitate visual processing in a selective manner? Evidence from attentional capture. Acta Psychologica, 151, 51–61. Seibold, V. C., & Rolke, B. (2014b). Does temporal preparation speed up visual processing? Evidence from the N2pc. Psychophysiology, 51, 529–538. Sternberg, S. (1969). Memory-scanning: Mental processes revealed by reaction-time experiments. American Scientist, 57(4), 421–457. Tipples, J. (2002). Eye gaze is not unique: Automatic orienting in response to uninformative arrows. Psychonomic Bulletin & Review, 9(2), 314–318. Van Voorhis, S., & Hillyard, S. A. (1977). Visual evoked potentials and selective attention to points in space. Perception & Psychophysics, 22, 54–62. Vangkilde, S., Coull, J. T., & Bundesen, C. (2012). Great expectations: Temporal expectation modulates perceptual processing speed. Journal of Experimental Psychology: Human Perception and Performance, 38(5), 1183–1191. Weinbach, N., Shofty, I., Gabay, S., & Henik, A. (2015). Endogenous temporal and spatial orienting: Evidence for two distinct attentional mechanisms. Psychonomic Bulletin & Review, 22, 967–973.