Elsevier

Neuropsychologia

Volume 47, Issue 12, October 2009, Pages 2615-2624
Neuropsychologia

Neural mechanisms of attentional shifts due to irrelevant spatial and numerical cues

https://doi.org/10.1016/j.neuropsychologia.2009.05.011Get rights and content

Abstract

Studies of endogenous (cue-directed) attention have traditionally assumed that such shifts must be volitional. However, recent behavioural experiments have shown that participants make automatic endogenous shifts of attention when presented with symbolic cues that are systematically associated with particular spatial directions, such as arrows and numerals, even when such cues were not behaviourally relevant. Here we used event-related potentials (ERPs) to test whether these automatic shifts of attention use the same mechanisms as volitional shifts of attention. We presented participants with non-predictive (50% valid) task-irrelevant arrow and numeral cues while measuring cue- and target-locked ERPs. Although the cues were task-irrelevant, they elicited attention-related ERP components previously found in studies that used informative and/or task-relevant cues. These findings further substantiate the dissociation between endogenous and volitional attentional control, and suggest that the same fronto-parietal networks involved in volitional shifts of attention are also involved in reflexive endogenous shifts of attention.

Introduction

The ability to selectively attend to spatial locations allows the brain to process the vast amount of information that it receives. Numerous studies have demonstrated the existence of two attentional systems in which attention can be controlled either exogenously (reflexively), by external stimuli in the environment, or endogenously (volitionally), by changes in an individual's goals and intentions (Posner, 1980, Posner and Dehaene, 1994). Studies of voluntary attention have traditionally used the spatial cueing paradigm, in which participants respond as quickly as possible to a peripheral target that is expected to appear in the direction indicated by a central symbolic orienting cue, such as an arrow (Posner, 1980). Targets appearing at the cued locations (valid condition) are generally detected faster than targets appearing away from the cued location (invalid condition). However, these studies traditionally use more valid than invalid cues and instruct participants to direct their attention to the side indicated by the cue, on the assumption that symbolic cues can only give rise to volitional shifts of attention.

Recently, Hommel, Pratt, Colzato, and Godijn (2001) questioned this association between endogenous and voluntary attention using a cueing paradigm with non-predictive (50% valid) symbolic cues that have a consistent association with one side of space, such as arrows (leftward or rightward) and direction words (“left” or “right”). They found that these non-predictive cues automatically orient attention in the direction indicated. Subsequent studies have suggested that cuing effects traditionally measured with informative arrows involve a super-additive combination of automatic shifts due to the arrows and controlled shifts due to volitional orientation of attention (Ristic and Kingstone, 2006, Tipples, 2002).

Arabic numerals can also elicit automatic shifts of attention, further demonstrating that symbolic cues can cause reflexive shifts of visuospatial attention, and highlighting the dissociation between endogenous and voluntary attentional mechanisms (Casarotti et al., 2007, Fischer et al., 2003, Galfano et al., 2006, Ristic et al., 2006). Western participants seem to represent numbers on a left to right oriented mental number line (for reviews, see Fias and Fischer, 2005, Hubbard et al., 2005). For example, in the SNARC effect (Spatial Numerical Association of Response Codes effect; Dehaene, Bossini, & Giraux, 1993), reaction times in a parity judgment (odd/even) task are faster to smaller numbers with left-sided responses, and to larger numbers with right-sided responses. Fischer et al. (2003) found that subjects responded faster to targets presented in the left visual field when cued by smaller numerals (1/2), and to targets in the right visual field when cued by larger numerals (8/9), even though the cues were non-predictive and task-irrelevant. Similarly, Casarotti et al. (2007) showed that task-irrelevant numerals biased temporal order judgments, such that targets appearing in the hemifield congruent with the mental number line (small-left, large-right) were judged to appear earlier than targets in the incongruent hemifield. At a cortical level, the parietal lobe contains multiple regions involved in calculation and visuospatial processes (Simon, Mangin, Cohen, Le Bihan, & Dehaene, 2002), leading us to suggest that automatic associations between numbers and space emerge as the result of interactions between parietal regions involved in numerical and spatial processes (Dehaene et al., 2003, Hubbard et al., 2005).

The neural mechanisms that underlie these reflexive endogenous shifts of attention have not been extensively explored (although see Eimer, 1997, Hietanen et al., 2008, Sato et al., 2009). One important method for investigating visuospatial attention is the use of event-related potentials (ERPs), which permit the observation of the time course of cognitive processes (for a review, see Luck, Woodman, & Vogel, 2000). Studies using the classic spatial cueing paradigm described above have reported three early cue-locked ERP components involved in voluntary shifts of attention (Harter et al., 1989, Hopf and Mangun, 2000, Nobre et al., 2000, Talsma et al., 2005, Van der Stigchel et al., 2006). The “early directing attention negativity” (EDAN), a relative negativity appearing 200–400 ms after cue onset over contralateral posterior scalp sites, is thought to originate in parietal cortex and to reflect mechanisms of spatial orienting (but see Van Velzen & Eimer, 2003). The “anterior directing attention negativity” (ADAN), which appears over contralateral anterior scalp sites 300–500 ms after cue onset, is thought to reflect supramodal attentional control processes originating in frontal structures (e.g., Eimer, Van Velzen, & Driver, 2002), although recent research has suggested the presence of more than one generator (Green et al., 2008, Praamstra et al., 2005). Finally, the “late directing attention positivity” (LDAP), a positive waveform arising over 500 ms after cue onset and following the EDAN and ADAN components, seems to reflect the biasing of excitability in visual sensory areas by supramodal top-down attentional control (Eimer et al., 2002, Harter et al., 1989), and is thought to be generated by occipital areas involved in endogenous spatial attention (Praamstra et al., 2005). As such, the EDAN, ADAN and LDAP are thought to reflect the actions of a fronto-parietal attentional network engaged in the control of spatial attention (Corbetta and Shulman, 2002, Posner and Petersen, 1990).

In the current study, we examined the EDAN and ADAN components described in previous studies of voluntary attention while participants performed a cueing task with non-predictive, task-irrelevant symbolic stimuli (arrows and numerals) to address two issues. First, we investigated whether these cues elicited similar attentional mechanisms as task-relevant cues by looking for the presence of similar electrophysiological signatures. Second, we explored whether numerical cues induce involuntary activation of spatial representations, by testing whether the EDAN and ADAN were elicited for both arrows and numerals. In particular, we expected that our non-informative cues would elicit the same attentional modulations that have been found for informative cues in the hemisphere contralateral to the direction associated with the cue, indicating that small numbers and left arrows directed attention to the left, and that large numbers and right arrows directed attention to the right. Based on previous behavioural studies that have suggested that interactions between numbers and space begin only after numerical magnitude has been processed (e.g., Galfano et al., 2006), we also predicted that interactions between numerical magnitude and hemisphere would begin only after the P2p component, a positive component appearing between 200 and 260 ms over parietal sites, which has been shown to be linked to processing of numerical magnitude (Dehaene, 1996). Finally, we predicted that early ERP responses (P1 and N1 components) to the target stimuli would be enhanced when they appeared at a location congruent with that indicated by the cue, as seen in studies that have used informative cues (Hillyard et al., 1998, Luck et al., 2000). As predicted, we found that non-informative task-irrelevant arrows and numerical cues both elicited the EDAN and ADAN, and that these non-informative cues modulated ERP responses to the targets. The results of the current study also suggest that symbolic cues do not need to be task-relevant in order to elicit shifts of attention, and that similar visuospatial processes are elicited by both arrows and numbers, due to their overlearned associations with space.

Section snippets

Participants

We tested 20 French volunteers but excluded 5 from analysis because of excessive noise or artefacts (blinks, movement, etc.) during the recording (final group: 15 subjects, 4 females; mean age = 22.2, range = 20–29). All gave informed consent, were right handed and had normal or corrected to normal vision.

Experimental design and procedure

Participants were instructed to respond as quickly as possible to a peripheral target, which appeared in the left or right visual field with equal probability (see Fig. 1). Each trial started with

Behavioural results

Misses (less than 1% of trials) and reaction times faster than 100 ms or longer than 1000 ms were excluded from analyses, as well as all catch trials (false alarm rate less than 2% of trials). Repeated measures ANOVAs were then computed on reaction times with Cue Type (arrow/number), Cue Direction/Magnitude (for arrows: left/right; for numbers: small/large), and Target Position (left, right) as factors. Neither the Cue Direction/Magnitude × Target Position interaction, nor the three-way interaction

Discussion and conclusions

The present study aimed to investigate the time course of the orienting shifts induced by non-informative symbolic cues with spatial meaning such as arrows and numbers by exploring electrophysiological activity during a simple detection task. Our results are consistent with previous behavioural studies that find symbolic cues with spatial meaning can elicit automatic endogenous shifts of attention. Consistent with these previous studies reporting spatial attentional biases for non-predictive

Acknowledgements

This research was supported by the Institut National de la Santé et de la Recherche Médicale (SD and MP) a James S. McDonnell Centennial Fellowship (SD), and a Marie-Curie Numeracy and Brain Development (NUMBRA) postdoctoral fellowship (EMH). We thank T. Bekinschtein, G. Dehaene-Lambertz, A. Del Cul, G. Hesselmann, A. Jobert and A. Knops for help in data collection and analyses; D. Basso, L. Girelli, and T. Vecchi for assistance during pilot behavioural testing, and M. Bonato, L.E. Williams and

References (39)

  • O. Simon et al.

    Topographical layout of hand, eye, calculation, and language-related areas in the human parietal lobe

    Neuron

    (2002)
  • D. Talsma et al.

    The orienting of visuo-spatial attention: An event-related brain potential study

    Cognitive Brain Research

    (2005)
  • D. Talsma et al.

    Faster, more intense! The relation between electrophysiological reflections of attentional orienting, sensory gain control, and speed of responding

    Brain Research

    (2007)
  • S. Van der Stigchel et al.

    An ERP study of preparatory and inhibitory mechanisms in a cued saccade task

    Brain Research

    (2006)
  • M. Bonato et al.

    Normal and impaired reflexive orienting of attention following central non-predictive cues

    Journal of Cognitive Neuroscience

    (2009)
  • M. Corbetta et al.

    Control of goal-directed and stimulus-driven attention in the brain

    Nature Reviews Neuroscience

    (2002)
  • S. Dehaene

    The organization of brain activations in number comparison: Event-related potentials and the additive-factors method

    Journal of Cognitive Neuroscience

    (1996)
  • S. Dehaene et al.

    The mental representation of parity and number magnitude

    Journal of Experimental Psychology: General

    (1993)
  • S. Dehaene et al.

    Three parietal circuits for number processing

    Cognitive Neuropsychology

    (2003)
  • Cited by (77)

    View all citing articles on Scopus

    Institution where the study was carried out: INSERM, U562, Cognitive Neuroimaging Unit, CEA/SAC/DSV/I2BM/NeuroSpin, Bât 145, Point Courrier 156, F-91191 Gif/Yvette, France.

    View full text