Elsevier

Neuropsychologia

Volume 49, Issue 2, January 2011, Pages 238-246
Neuropsychologia

Attention biases the perceived midpoint of horizontal lines

https://doi.org/10.1016/j.neuropsychologia.2010.11.022Get rights and content

Abstract

In patients with right brain damage and left visual neglect, attention tends to be captured by right-sided objects and cannot easily disengage from them. While these phenomena can account for several clinical and experimental patterns of performance such as biased visual search, its role is more controversial for other neglect-related signs, such as the typical rightward shifts in horizontal line bisection. It is thus important to see whether and how attentional orienting can bias line bisection in normal participants using standard clinical bisection stimuli. In 3 experiments, we explored the Attentional Repulsion Effect (ARE, Suzuki & Cavanagh, 1997) on pre-bisected lines. Normal observers saw horizontal lines with a vertical bisection mark near the center, preceded by a cue to the left or right of the line, or by no cue. On each trial, observers indicated whether they saw the bisection mark to the left or at the right of the midpoint. We plotted the proportion of ‘seen-at-right’ responses as a function of the mark's actual position. For uncued lines, the point of subjective equality was slightly at the left of the true center, consistent with the pseudoneglect phenomenon. Right-sided cues shifted the apparent bisection point to the left (and vice versa), as predicted by the ARE. Similar results occurred with different task instructions (compare the length of the left-sided line segment to the right-sided segment) and in the presence or absence of central fixation marks. These results obtained in normal participants support attentional accounts of biased line bisection in neglect patients.

Research highlights

▶ Exogenous cues repelled the perceived position of transectors in horizontal lines. ▶ Without cues, the transector was perceived leftward of its objective position. ▶ Choosing the longer (shorter) line segment gave similar results. ▶ These results are consistent with the pseudoneglect phenomenon. ▶ Results support attentional accounts of line bisection in neglect.

Introduction

Patients with right brain lesions and left neglect demonstrate a directional bias towards the right side of space when perceiving and acting in their environment (Bartolomeo & Chokron, 2002); their attention tends to be captured by right-sided objects and cannot easily disengage from them (Posner et al., 1984, Rastelli et al., 2006). These deficits often provoke clinical signs such as “magnetic attraction” of gaze to right-sided stimuli (Gainotti, D’Erme, & Bartolomeo, 1991) and neglect of left-sided items on visual search (Mark, Kooistra, & Heilman, 1988). When asked to bisect a horizontal line, neglect patients typically err rightwards (Schenkenberg, Bradford, & Ajax, 1980). Even when no manual action is required, as in the landmark task (Harvey, Milner, & Roberts, 1995), patients consider the right segment as being longer than the left one (Milner, Harvey, Roberts, & Forster, 1993), consistent with their bisection behavior.

Neurologically healthy subjects, on the other hand, may make (much smaller) errors in the opposite direction, and bisect lines to the left of the veridical center, a phenomenon dubbed “pseudoneglect” (Bowers & Heilman, 1980; see Jewell & McCourt, 2000 for a review). On the landmark task, when judging lines pre-bisected to the left of their true center, normal participants consider the left segment as being longer than the right one (Milner, Brechmann, & Pagliarini, 1992). This asymmetry likely results from the specialization of networks in the right hemisphere for the deployment of spatial attention (Heilman and Van Den Abell, 1980, McCourt and Jewell, 1999, McCourt and Olafson, 1997, Mesulam, 1999), although reading habits could also contribute by biasing left-to-right readers to explore the line from its left endpoint (Chokron & Imbert, 1993).

This evidence from normal and brain-damaged patients suggests that spatial attention influences the perceptual estimation of horizontal lengths, leading to over-estimation of the portion of the line where attention is focused on. For example, Marshall and Halligan (1990) proposed that during line bisection neglect patients search the line for its midpoint from the right to the left, and subsequently place the bisection point where the two hemi-segments appear to be of equal length. A rightward attentional bias might thus increase the perceptual salience of the right portion relative to the left portion of the line (Anderson, 1996), with consequent overestimation of the right portion of the line (Urbanski & Bartolomeo, 2008).

Alternatively, however, deviations on bisection-like tasks in neglect could result from an acquired anisometry of a mental representation of space that is progressively compressed from the right to the left (Bisiach et al., 1996, Bisiach et al., 1998, Savazzi et al., 2007, but see Doricchi et al., 2008). According to this view, rightward shifts in neglect would be the consequence of a progressive right-to-left relaxation of spatial coordinates, so that the left half of the line would be subjectively perceived as shorter than the right half. This hypothesis originated from the finding that, when neglect patients were asked to place the endpoints of an imaginary line around a given centre, the left-sided imaginary segment was longer than the right-sided one. Importantly, Bisiach, Rusconi, Peretti, and Vallar (1994) claimed that such a pattern of performance was inconsistent with current accounts of neglect, including attentional hypotheses.

In view of this debate, it is important to see whether length estimation can actually be linked to relatively uncontroversial attentional mechanisms. Such a demonstration would provide a proof of principle that biased attentional orienting can at least partially determine neglect patients’ perceptual estimation of horizontal lengths.

In the present study, we investigated these issues by taking advantage of the Attentional Repulsion Effect (ARE), whereby briefly presented visual stimuli appear displaced away from the focus of attention (Suzuki & Cavanagh, 1997). In the original study, a central cross was presented for 1800 ms, attention was then focused at a given location by a transient peripheral cue lasting 30 ms. After an SOA of 180 ms, two vertical lines aligned across a wide gap (a vernier) were flashed for 60 ms and followed by a mask presented for 255 ms. In a two-alternative forced choice procedure, subjects had to decide whether the vernier offset was directed clockwise or counterclockwise. Results showed that the line closer to the attentional focus was seen further away. The effect disappeared rapidly as the vernier was exposed for a longer duration (more than 200 ms), which suggested the involvement of a transient component of spatial attention (Nakayama & Mackeben, 1989). The attentional nature of the ARE has been convincingly demonstrated, and several candidate nonattentional mechanisms (figural after effects, perception of apparent motion from the cue to the nearer vernier line) have been excluded (Suzuki & Cavanagh, 1997). Subsequently, Pratt and Arnott (2008) examined whether the AREs are analogous to effects obtained in temporal attentional tasks. They used onset cues (i.e., cues appearing suddenly and remaining present), offset cues (cues disappearing suddenly and never reappearing) and onset–offset cues (cues suddenly appearing and disappearing), all of which produced attentional repulsion effects that were equivalent in magnitude. However, when simultaneous onset and offset cues always appeared or disappeared on the same side of the display, the magnitude of the repulsion was greater for onset cues than for offset cues. This result is consistent with the response time literature on spatial attention, where onsets have priority for the exogenous allocation of attention in situations when two objects both appear and disappear at the same time. Thus, these experiments confirmed the evidence obtained by Suzuki and Cavanagh (1997), by demonstrating that the repulsion effect can be modulated by attentional manipulations, and proposed that the repulsion effect could be a spatial analogue of temporal effects in response times typical of exogenous spatial cueing.

Although the perceptual consequence of the cue is a repulsion in the test position away from the cue, Suzuki and Cavanagh (1997) attributed the result to the effect of attention on receptive fields activated by the cue. Specifically, the physiological literature has reported that attention leads to a shifting of the effective centers of receptive fields toward the cue and a spatial narrowing of the receptive fields (Connor et al., 1997, Womelsdorf et al., 2006). According to the labeled line view of location perception (e.g. Barlow, 1972) where the activity of each neuron signals a specific feature or location value, this would lead to a magnification of space around the cue, and a resulting shift in the centroid of responses to the subsequent test. These intermediate effects would lead to the perceived offset of a subsequent test away from the initial cue: the Attentional Repulsion Effect. An attentional magnification should also lead to the expansion in the linear extent of the portion of any test that is closer to the cue. For example, with the horizontal lines used in the present experiments, attention would “stretch” the portion of the line that is closer to the cue. In the original Suzuki and Cavanagh (1997) study, there was only blank space between the cue and the test so that study could not directly examine any other effects.

The present study consists of three experiments carried out to assess whether the repulsive effect of an exogenous cue is able to bias the location of perception. The effect of cueing on the perception of a midline has been studied in pseudoneglect (McCourt, Garlinghouse, & Reuter-Lorenz, 2005) and we now extend this work to landmark stimuli similar to those employed with neglect patients and by asking judgments not only of the midpoint but also of horizontal extent. The occurrence of an ARE in these conditions in normal subjects would obviously support accounts of neglect patients’ biased performance on line bisection as resulting, at least in part, from asymmetrical orienting of exogenous attention.

Section snippets

Experiment 1

In Experiment 1, observers judged the position of the transector in pre-bisected lines. Exogenous cues were briefly presented adjacent to the left or right extremity of the line. If an ARE occurs in this setting, then observers should perceive the cued segment as longer, and therefore subjectively displace the transector towards the uncued extremity.

Experiment 2

Five observers (two males and three females, aged 22–31 years) participated in Experiment 2. All were right-handed according to the Edinburgh laterality inventory, (mean score, 83.25; SD, 33.50). The procedure was identical to Experiment 1, with the following exception: at variance with Experiment 1, where the fixation cross was present throughout the experiment, in Experiment 2 the fixation cross was present only at the beginning of the trial for 700 ms, before the sequence started. Then the

Experiment 3

In the first two experiments, observers indicated the side of apparent displacement of the transector. On the landmark test used to test neglect patients, however, other types of responses are more usual. For example, observers are typically asked to state which of the two segments appears to be longer or shorter. In the present context it was thus important to see whether the response type changed the pattern of results. In Experiment 3, seven students (four males and three females, aged 22–37

Summary of results

Three experiments assessed the presence, direction and magnitude of the attentional repulsion effect (ARE, Suzuki & Cavanagh, 1997) with pre-bisected lines. Observers perceived the transector of pre-bisected lines as being shifted contralaterally to a peripheral visual cue, consistent with the ARE phenomenon previously demonstrated with vernier offsets (Pratt and Arnott, 2008, Suzuki and Cavanagh, 1997). Cues-induced exogenous attention repelled the perceived location of the bisection marker

Conclusion

By experimentally manipulating exogenous attention, we biased normal participants’ perception of horizontal lengths, in ways consistent with attentional interpretation of neglect patients’ biased line bisection behavior and of the pseudoneglect phenomenon. Importantly, and at variance with previous similar studies, our stimuli were the pre-bisected lines typically used in the clinical literature. Moreover, we showed that observers’ response pattern did not substantially change with different

Acknowledgments

We would like to thank Florent Caetta and Bastien Oliviero for their help in programming this experiment. We are also grateful to the participants for their patience and suggestions and to an anonymous reviewer for helpful comments.

References (49)

  • M. Harvey et al.

    An investigation of hemispatial neglect using the Landmark Task

    Brain and Cognition

    (1995)
  • G. Jewell et al.

    Pseudoneglect: A review and meta-analysis of performance factors in line bisection tasks

    Neuropsychologia

    (2000)
  • J.C. Marshall et al.

    Within- and between-task dissociations in visuo-spatial neglect: A case study

    Cortex

    (1995)
  • J.B. Mattingley et al.

    To see or not to see: The effects of visible and invisible cues on line bisection judgements in unilateral neglect

    Neuropsychologia

    (1993)
  • M.E. McCourt et al.

    Unilateral visual cueing and asymmetric line geometry share a common attentional origin in the modulation of pseudoneglect

    Cortex

    (2005)
  • M.E. McCourt et al.

    Visuospatial attention in line bisection: Stimulus modulation of pseudoneglect

    Neuropsychologia

    (1999)
  • M.E. McCourt et al.

    Cognitive and perceptual influences on visual line bisection: Psychophysical and chronometric analyses of pseudoneglect

    Neuropsychologia

    (1997)
  • A.D. Milner et al.

    To halve and to halve not: An analysis of line bisection judgements in normal subjects

    Neuropsychologia

    (1992)
  • A.D. Milner et al.

    Line bisection errors in visual neglect: Misguided action or size distortion?

    Neuropsychologia

    (1993)
  • K. Nakayama et al.

    Sustained and transient components of focal visual attention

    Vision Research

    (1989)
  • R.C. Oldfield

    The assessment and analysis of handedness: The Edinburgh Inventory

    Neuropsychologia

    (1971)
  • J. Pratt et al.

    Modulating the attentional repulsion effect

    Acta Psychologica

    (2008)
  • I.H. Robertson

    Do we need the “lateral” in unilateral neglect? Spatially nonselective attention deficits in unilateral neglect and their implications for rehabilitation

    NeuroImage

    (2001)
  • M. Urbanski et al.

    Line bisection in left neglect: The importance of starting right

    Cortex

    (2008)
  • Cited by (0)

    View full text