The attentional selection of spatial and non-spatial attributes in touch: ERP evidence for parallel and independent processes

https://doi.org/10.1016/j.biopsycho.2003.08.001Get rights and content

Abstract

To investigate the functional relationship between spatial and non-spatial attentional selectivity in somatosensory processing, event-related potentials (ERPs) were recorded to mechanical tactile stimuli, which were delivered to the right or left hand, and were low or high in frequency (Experiment 1), or soft or strong in intensity (Experiment 2). Participants’ task was to attend to a specific combination of one stimulus location and one non-spatial attribute. Spatial attention was reflected in enhanced N140 components followed by a sustained attentional negativity. ERP effects of non-spatial attention (enhanced negativities to the attended frequency or intensity) were observed in the same latency range, suggesting that the attentional selection of relevant spatial and non-spatial attributes occurs in parallel. Most importantly, ERP correlates of attention directed to stimulus frequency and intensity were unaffected by the current focus of spatial attention. In contrast to vision, where the selective processing of non-spatial attributes is hierarchically dependent on selection by location, but similar to auditory attention, spatial and non-spatial attentional selectivity appear to operate independently in touch.

Introduction

While selective attention has been studied intensively in the visual and auditory modality, the processes underlying attentional selectivity in somatosensation are still poorly understood. The aim of the ERP study reported here was to investigate the attentional selection of spatial and non-spatial attributes of tactile stimuli, in order to obtain insights into the mechanisms involved in the selective attentional processing of different tactile stimulus attributes.

Directing visual attention in space results in faster detection and more accurate discrimination of visual stimuli at currently attended locations (Posner et al., 1980). ERP studies have provided evidence that early stages of visual processing are modulated by visual-spatial attention (Mangun and Hillyard, 1990). Enhanced early sensory-specific ERP components (P1 and N1) are elicited over posterior visual cortical areas in response to visual stimuli at currently attended as compared to unattended locations (e.g. Mangun, 1995, Mangun et al., 1987, Mangun and Hillyard, 1987, Eimer, 1995), and these amplitude modulations start as early as 80 ms after stimulus onset. In contrast, visual attention to non-spatial features, such as colour or shape, result in an enhanced sustained negativity that starts much later (at about 150 ms post-stimulus; e.g. Eimer, 1995, Heslenfeld et al., 1997).

To investigate whether the selective attentional processing of non-spatial visual stimulus attributes depends on prior selection for spatial location, Hillyard and Münte (1984) presented their participants with visual stimuli defined by both location and color. The participants’ task was to detect infrequent target stimuli that matched both the attended color and the attended location. ERP waveforms elicited by non-target stimuli with the attended color and/or location were compared to non-targets with the unattended color and/or location. ERPs elicited by stimuli at attended locations showed enhanced P1, N1 and N2 components compared to stimuli at unattended locations, and these spatial attention effects were independent of stimulus colour. Effects of attention directed to colour were associated with a broad negativity starting around 150 ms after stimulus onset. Most importantly, these ERP effects of selective colour processing were only found in response to stimuli at the attended location, thus suggesting that spatially selective processing not only precedes the selective processing of non-spatial attributes of visual stimuli (i.e. color), but also that non-spatial attention effects are contingent upon the prior selection of visual stimuli on the basis of their location (see also Eimer, 1995, Omoto et al., 2001).

Similar to vision, selective spatial attention also facilitates auditory processing (Spence and Driver, 1994, Spence and Driver, 1998). However, recent behavioral studies have shown that the attentional processing of auditory stimulus locations is influenced by stimulus frequency even when frequency is irrelevant (Mondor et al., 1998), and that auditory targets defined by frequency or by the conjunction of frequency and location are identified faster than targets defined by location only (Woods et al., 2001). These findings suggest that unlike in vision, the attentional selection of spatial does not precede the selection of non-spatial attributes of auditory stimuli. To examine the mechanisms underlying the selective processing of auditory spatial and non-spatial attributes, Woods and Alain (2001) recorded ERPs in response to tones defined by a combination of location, frequency and duration, under conditions where targets were defined by a specific combination of location, frequency and duration attributes. When ERP waveforms elicited by stimuli possessing none of the attended attributes with ERPs elicited by stimuli possessing one, two or all of the attended attributes were compared, both attention to location and attention to frequency were reflected in enhanced early attentional negativities starting at about 60 ms post-stimulus. In line with the behavioural evidence described above, these results suggest that the attentional selection of spatial and non-spatial auditory stimulus attributes takes place in parallel and independently (see also Hansen and Hillyard, 1983, Woods et al., 1994).

Only few ERP studies to date have investigated mechanisms of selective attention in touch. In most studies on tactile spatial attention, participants were instructed to attend to one hand versus the other while electrical or mechanical stimuli were delivered to the left or right hand. Spatial attention was found to modulate the sensory-specific somatosensory N140 component, with enhanced N140 amplitudes for tactile stimuli delivered to the attended hand (Desmedt and Robertson, 1977, Michie et al., 1987, Garcia-Larrea et al., 1995, Eimer and Driver, 2000, Eimer et al., 2001, Eimer et al., 2002, Eimer and Forster, 2003). In addition, effects of tactile spatial attention on the somatosensory P100 component (Josiassen et al., 1982, Michie, 1984, Eimer and Forster, 2003), and a later sustained processing negativity (Desmedt and Robertson, 1977, Michie, 1984) have also been observed. The fact that tactile spatial attention affects early, sensory-specific somatosensory ERP components clearly suggests that, analogous to visual and auditory spatial attention, spatial selectivity in touch can modulate relatively early perceptual stages of somatosensory processing.

ERP correlates of non-spatial attentional selectivity in touch have not yet been investigated systematically. Michie et al. (1987) found that ERP effects of tactile spatial attention were modulated by a non-spatial attribute (intensity) defining infrequent target stimuli. In this experiment, participants had to report infrequent weak stimuli delivered among stronger stimuli, or infrequent strong stimuli delivered among weaker stimuli, when these infrequent targets were presented to the currently attended hand. Spatial attention affected the N80 component when weak stimuli had to be detected among strong non-targets, and an effect of spatial attention on the P105 when infrequent strong targets were presented among weak non-targets. However, ERP correlates of tactile attention directed to stimulus intensity were not measured directly in this study.

The present experiment was designed to investigate for the first time ERP correlates of selective attention directed to non-spatial attributes of tactile stimuli, as well as the relationship between spatial and non-spatial selective attention in somatosensation. Mechanical vibratory tactile stimuli were delivered to the left or right hand, and these stimuli also differed with respect to one non-spatial attribute. In Experiment 1, stimulus frequency was either low or high. In Experiment 2, stimulus intensity was either strong or soft. Participants’ task was to direct attention to one side, and to one non-spatial stimulus attribute, in order to detect and respond to infrequent targets (which were longer than non-targets) when these possessed both relevant attributes. Thus, in separate blocks, selective attention was directed to one of the four possible combinations of location (left versus right) and frequency (high versus low, in Experiment 1) or intensity (strong versus soft, in Experiment 2). ERPs were computed in response to non-targets possessing both, one, or neither of these task-relevant attributes.

In vision, ERP effects of spatial attention precede ERP effects of non-spatial attentional selectivity, and ERP correlates of the selective processing of a non-spatial visual attribute are only observed in response to visual stimuli at currently attended locations, thus demonstrating that the attentional processing of non-spatial visual features is hierarchically dependent on prior selection of stimulus location (Hillyard and Munte, 1984, Eimer, 1995). If the same hierarchical dependency also exists for selective attention in the somatosensory modality, ERP effects of attention directed to stimulus frequency or intensity should emerge later than ERP effects of tactile-spatial attention, and any effects of non-spatial attention should only be present for tactile stimuli presented to the currently attended hand. In audition, the attentional processing of spatial and non-spatial stimulus attributes appears to be based on parallel and independent mechanisms (Woods et al., 1994, Woods et al., 2001, Woods and Alain, 2001). If this was also the case for somatosensation, ERP modulations reflecting attention to spatial and non-spatial tactile stimulus attributes should be elicited concurrently. More importantly, ERP correlates of non-spatial tactile attention should not only be elicited in response to stimuli delivered to the currently attended hand, but also for tactile stimuli presented to the unattended hand.

Section snippets

Participants

Seventeen paid volunteers participated in Experiment 1. Two participants had to be excluded because of too many eye-blinks contaminating more than 70% of trials. Three additional subjects had to be excluded because of a lack of clearly defined early somatosensory components (see below). Thus 12 participants (4 males, 8 females), aged 19–41 years (mean age: 25.6 years) remained in the sample. All participants were right-handed and had normal or corrected vision.

Stimuli and apparatus

Participants sat in a dimly lit

Experiment 2

This experiment was conducted to provide additional evidence for the independence of spatial and non-spatial selectivity in touch. Now, stimulus frequency was kept constant, while stimulus intensity was varied. Thus, participants had to attend to specific combinations of stimulus location (left versus right) and stimulus intensity (strong versus soft) to detect infrequent targets characterized by these two features. In all other respects, design and procedure was identical to Experiment 1.

Acknowledgements

This research was supported by a grant from the Biotechnology and Biological Science Research Council (BBSRC).

References (31)

  • M. Eimer et al.

    Cross-modal links in endogenous spatial attention are mediated by common external locations: evidence from event-related brain potentials

    Experimental Brain Research

    (2001)
  • M. Eimer et al.

    Crossmodal interactions between audition, touch and vision in endogenous spatial attention: ERP evidence on preparatory states and sensory modulations

    Journal of Cognitive Neuroscience

    (2002)
  • L. Garcı́a-Larrea et al.

    Somatosensory responses during selective spatial attention: the N120-to-N140 transition

    Psychophysiology

    (1995)
  • J.C. Hansen et al.

    Selective attention to multidimensional auditory stimuli

    Journal of Experimental Psychology: Human Perception and Performance

    (1983)
  • S.A. Hillyard et al.

    Selective attention to color and location: an analysis with event-related brain potentials

    Perception and Psychophysics

    (1984)
  • Cited by (56)

    • The effect of perceptual load on tactile spatial attention: Evidence from event-related potentials

      2017, Brain Research
      Citation Excerpt :

      For instance, sustained attention tasks in which the same body location was attended to throughout a block of trials resulted in earlier effects of attention as compared to transient studies of attention where the task-relevant body location was cued on a trial-by-trial basis (c.f. Eimer and Forster, 2003; Zopf, et al., 2004). However, even within cuing studies, results appear to be inconsistent, with some reporting the earliest attention effects on the P100 (e.g. Eimer and Forster, 2003) while others on the following N140 component (e.g. Eimer et al., 2004; Forster and Eimer, 2004; Forster and Eimer, 2005a, b; Eardley and van Velzen, 2011), or even at longer latencies (from the descending flank of the N140, e.g. Van Velzen et al., 2002; Forster et al., 2009). Because different types of target vs. non-target discriminations were used in previous ERP studies of tactile spatial attention (gap/non-gap, frequency, intensity, etc.), it is conceivable that they required different amounts of attentional resources, resulting in different onset times of the attention effects.

    View all citing articles on Scopus
    View full text